What do you guys think? Is this a good idea? I decided to make a part 2 for this video where I discuss other ideas that could help prevent steamrolls, should be up in the next week probably. Lastly, look at @chrisbadger8684 ‘s comment to see a good argument for why skill based mm might not be that hard to implement and would not increase queue times much. I think his implementation improves on the one I described in the video. It involves setting up teams as is done currently, but then switching players across teams who are in the same tier/class of tank to equal out the teams based on skill after teams have already been selected by the match-maker.
Steam rolling isn't just about skill, often it happens when one team loses their light tank, especially an EBR on an open map. It's also about tank balance. You have players on sides with even skills one side with Bat Chatillion Bourrasques a the other with AMX Chasseur de chars or M4A1 Revalorisés, you know the outcome before the match has even started. Also the game is full of real BOTs, Yesterday there was a tier 9 bot 2000 games with an average of 1 damage. Wargaming needs to weedle this accounts out and ban them. The most boring thing about WOT for me at the moment is not just the monotony of maps but also mostly knowing the outcome of the game by just looking at the line up of tanks before its even started.
This is a great idea. I’ve thought so for years. Does it fix every steamroll game? Of course not. But it would be a great improvement. I think it would be relatively easy to implement. Agree, WG can come up with an improved WIN8 metric coupled with Winrate to evaluate each player. Let the current matchmaker come up with teams THEN using the new evaluation to shuffle a few players between teams for balance. I’m guessing wait times wouldn’t have to grow much at all adding this last balancing step.
@@thrawn-ys9hf Ok, I read a bit about it and people’s impressions on blitz. As far as I understand their mm system works such that if you have 70% win rate you will be matched with like 2-3 40% teammates to balance things out. I think in blitz this system is going to be much harsher on good players especially because the format is 7v7 where each player has a huge impact on the battle. In wot, solo pro players max out at around 66% win rate, with the very best of the best doing around 70% solo. Realistically 99+% of players will be within 40-60% wr players, and probably 90% within 45-55% wr (prob can check exact numbers on tomato.gg). So there will be way less shock to players who are used to winning or losing more often on avg for WoT pc than in blitz due to the smaller impact each player has in a battle. But I do think looking at that situation I realize that there is a very good chance people will complain about the change even if the battle quality truly is improving, simply because people dislike big changes and hate wargaming. This will be especially true if it has the words “skill based mm” in the patch notes. People will freak out even if it really isn’t true skill based mm but better branded as team skill balancing. Either way that is kind of disappointing to me because it means WG will prob never consider this as an option even if it was tested rigorously and shown to work for improving battle quality.
Thank you for bringing this issue up again DarkNinja, good video. I sent a ticket to WG about this issue long ago, complete with blowout matchup examples. Their response? WG will NOT implement skills based matches . I believe their reasoning is money. They are afraid the "pros" and "unicums" who bring in a lot of revenue would not appreciate facing tougher battles and see their win rate drop. There is definitely no reason why one team gets several green players while the other only has average players. I used to keep track of these stats and it happens the majority of times in blowout games. They fail to realize they are losing players by the thousands and don't gain new ones because it's become a frustrating game for newcomers. Better balance would definitely bring more new players and reduce Q wait times. The other problem I see with MM is that tank roles have become meaningless. Lights used to scout, TDs used to snipe, etc. Nowadays, some heavies are as fast as some meds, meds have better view range than some lights, TDs have better armor than some heavies, lights have same alpha as many meds, and so on. The assumption that as long as each team has the same amount of tank types then it's balanced, is no longer a valid one. A new MM would have to take into account ROF, DPM, Armor, speed, penetration, etc. The game has evolved and MM has not. That is another serious problem.
Before thinking about skill-based matchmaking I would first fix the tank imbalances in the MM. Meaning that MM should consider the type of the vehicle within the class. This would already make it better when for example in Himmelsdorf other team wouldn't have Obj268v4 against Grille 15 or Maus against Type 71.
@@jannekoskinen2050 Yeah I agree that plays a part too, especially Sheridan vs good spotting LTs, tds with armor vs snipers, and hulldown heavies vs 50B/T57 on Westfield for example.
@@jannekoskinen2050 I decided I will be making a part 2 video with my other ideas and ideas I’ve heard from others that I support in the next couple of days. While I think skill based mm is a pretty radical idea I do think it would have the biggest impact if we could change one thing, but there are definitely things that would be easier to change that could still have a significant impact imo.
Right out of the gate, DarkNinja highlights the issue of 15-vs-0 or 15-vs-3 results. How telling is that? Players used to complain about 15-vs-5 results being too frequent. It's a Long-standing issue, getting worse over time. I think that indicates that War Gaming has no interest in averting "steamroll" battles. If War Gaming cared about better game-play, +/-2mm match-making would be switched off. I've tracked my battle results for over 3 years. In about 60% of battles, the winning team got greater than double the number of kills compared to the losing team. For example, a 7-vs-15 result would fall into this category. Very few battles end with a close kill count result.
Hey apologies for writing a short novel. But I think you’ll appreciate what I have to say: I think the problem/Solution you identified at 17:40 is my main gripe with team-pool skilled based MM. Said plainly, it doesn’t account for skill across tiers. An example making Skill pool based MM’s weakness clear: Let’s say you are in a tier 10 match, as a platoon of two 2600-3000 WN8 tier 8’s. You're the only two “good” players & your tier 10’s are 800-1200 WN8 players. Let’s also say the other team has just average 1400-2000 WN8 players across the board T8-10. And the pooled WN* has been calculated as a perfect match. I would wager that the average WN8 team would win in *most* cases simply because the average skill of the enemies Tier 10’s will outplay your “bad” Tier 10’s, eventually leading to a collapse in one flank and a possible loss. This is even in the case that you and your good Tier 8 platoonmate are trying to carry. It’s just hard, because skill is somewhat bottlenecked by the tank you are using. Example: a good player in a T8 light will have major problems dealing with a MAUS. NOW you dealt with this challenge with the solution you identified at 17:46 The idea being: Balancing the top tier’s skill levels (which is different from skill pool based MM). The idea is that because top tiers have the largest impact on the game (because their tanks are better), they should be the priority of the MM. And I think it is a step in the right direction. However the main weakness of this, as you probably already know, is that the MM doesn’t always match vehicle subclasses/types correctly against one another. Meaning that a top tier Grille 15 can be matched up against a Minotauro. So this skill-based solution should be paired with tank type matching to ensure more balance. And as you said this pressure on the MM queue times can be off-loaded by using peak-server/threshold “rules” - Funny enough I mention this exact same idea in my recent “Fixing light tanks’"Video. Good to know great minds think alike. But all this said, great video. Props to you for biting into the huge topic of Match-making. -- Gabriel
@@Tankers4Change I agree with everything you said, well put. After reading another comment (@chrisbadger) I think that there might be an easy compromise that as far as I understand it would not increase queue times much at all, which you can see under his comment.
@@Tankers4Change and while I definitely agree it can be complicated because of multiple tier games, as far as I understand it there is no skill based mm really in place rn so I think even small imperfect corrections could definitely be worth implementing.
What kind of day would it be if I didn't get a 13-0 steamroll by the red team? At least 8-12 times daily. Ahh thats why I don't spend anymore real cash on a game that hates their customers.
17:05 Waiting for another 30-60s during primetime is always worth it if it means you'll have 3 more minutes of playtime inside the actual battle, there's no doubt about it
This wouldn't add even one second to queue times. MM can pick the same 30 tanks and then just even the talent of the two teams as best as possible with a simple routine.
@@chrisbadger8684 actually now that you mention it that makes a lot of sense and is way simpler than what I was thinking they would have to do. They just pick the teams as done currently then they can swap players paired against each other in their tier/class until the teams are more balanced for skill (platoons make it a bit more complicated but it’s a good start). I think it’s a great idea.
I thought of this same idea. It won't work well when there are a bunch of tier 10 platoons including iyouxin and 2 other unicums yoloing in their IS7s, but it would work most of the time including off-peak hours when it's mostly solo players.
One detail problem I can think about is that a player has different levels of talent in different tanks, meaning it will be even more advantageous to play strong tanks. Even if this is countered, it would still be the case for players playing their new tank. Is there a fix to this?
I think wargaming may create a random tank pool mode with higher exp or credit award to counter q time. Like if you pick five tanks in a pool and you allow MM to decide which tank will going to the game for you(and you will play it). And you will earn more exp and credit if you play this mode. It may help reduce q time for more complicated mm(skill based).
I think they should experiment with the standard ammo having a set damage amount and the gold having a lower amount like some of the game modes. Maybe not 50% less but less. This would incentivize using regular ammo and assist the newer or cheaper players. I have no issue with using gold, however in the interests of more even games I think it would help.
One of the counter arguments about skill based matchmaking, is that by forcing a rebalancing of the teams to be equal, by default the average win rate for each side will eventually over time also equalize out. ie, over time, with each team balanced to 50/50, the WR for these games will also start to mirror 50/50. And what this means is that more of the players of these games will also start to move to a ~50% WR.. ie, one of the reasons for the +55% WR, is that these top skilled players have an oversized impact on the outcome of the game against lesser skilled teams. And thus have a higher WR as a result. and it will have the equal impact for the lesser skilled players as they will no longer have as much opportunities to hamper their teams. Ie, the +55% WR players will see a drop over time for their WR, and the ~45% players will start to do better. Arguably this is not in the higher skilled players interest - although you are correct in that the matches will be more enjoyable.... Your arguments at the end of the video are valid though.
@@rpsmason personally that is something I would be fine with, because getting 70% wins is not enjoyable as most are just steamrolls, and even if winrate was dropping to 50% the good players still would have high damage battles which might be good enough for them.
There's some obvious stuff that can be fixed even before getting into skill based stuff. For instance, at load up, 1 team already has a 1000 hp advantage. WTF, that's like half an extra tank of damage. Or 1 team has 4 autoloaders and the other team has none. multiple assault tds vs multiple sniping tds, especially on certain maps where each class clearly has an advantage. In most my blow out games I see, it's pretty obvious to me just switching 2 or 3 of the tanks around from one team to the other would make a huge difference. 2 ELCs on one team vs 2 M41 90, I mean cmon, it's obvious. I do like the idea of an impact score where the goal would be to get to 0 for both sides. For instance, a top player has impact score of 5 and a really bad player has -5. So you add up all the player scores on each side and try to get to an equal score. But this doesn't fix the tank v tank match ups. If you're top player is playing a bottom tier heavy with no pen and now you zero it out by making your top tier heavy suck, what if it's the reverse on the other team, where the top player is the top tier heavy. Bz 176 vs. t150, equally skilled players, who's winning? First they need to rebalance a lot of tanks. This should happen regularly. Again, it's pretty obvious what needs what. Then fix the tank v tank match ups. Then you can do some skilled stuff
Skill wins everytime. An 8k WOR player will happily butcher 4-5k players regardless of what tank and the EU server is normally pretty good a tank match ups.
I like this too, honestly I think an overall win rate aggregate balance , is fast, easy to implement, ez to understand, and could help right away with minimal dev costs. It could be gamed, but you could just throw out th
This is a great conversation (I would like to see WarGaming included in these if possible) because it is a problem in all of there games. So this game is so old, and yet they have done nothing to adjust this even though they tell us how the MM is decided to the best of their ability. They are not even looking at the tanks they put into battle, I have sent them screen shots of the MM and have not heard back from them, in one instance we had no light they had one. I have had battles where they put 3 Batchats, 1 T57, 2 TVP's, and two lights, we had no Batchats, 1 TVP, and one light. I mean if they can't match up even the tanks to average out the fire power they will never (and I mean Never) try to correct a skill based MM. This was and still is a big thing in WOTB and that is why so many people have left that game. They are only concentrating on making money by there modes, and using outdated (cheaply done) rewards for these modes. The bell curve mechanics is a simple thing to do in this game just because (NA) the player base is small. Que times are long now in many cases, especially if you're in platoon. I like where your going but I don't believe WG is interested.
I agree that having roughly equally skilled teams should be the goal. Blowouts aren't fun for anybody. Speculating here, but I think WG is moving toward implementing skill consideration into the MM when they made the shift to featuring/highlighting WOT rating instead of PR and winrate. WOT rating isn't perfect just like WN8 isn't perfect, but I think it does a bit better job than WN8 does capturing overall effectiveness because WN8 is based heavily in just damage dealing and kills. An additional thing that WG could do is match tanks more on their specialization than just class. For instance, it doesn't make much sense to me to put a 50B and T57 on one team versus a Maus and 279 on the other team.
In games like xDefiant, they implemented not skill based matchmaking, but match based matchmaking. So they would select 30 random players, then try to average out the skill across the match. So its still random MM, but during loading, the teams would be balanced before the match starts.
I thought about this some more. I'd love to hear from you guys and DarkNinja too, about introducing a new feature that might help with blowouts. The one characteristic of blowouts is that one team gets overun quickly, especially when a flank falls rapidly. What if WG introduced a rule whereby the team that falls behind, by say 5 tanks, in the first 5 minutes gets a few AI-driven reinforcements (bots) that automatically spawn on the losing flanks or at base??? That should be really easy to implement as many events used AI tanks already. I figure this is a bonus for BOTH sides. The winning side would welcome the opportunity to rack up more damage/kills and the losing side now has a chance to make a comeback. I realize the bots won't win against a good team but it mainly acts as a distraction and allows the defenders to regroup, return to base, and even flank the opponent to stage a possible comeback, or at least make it a closer, longer battle. The beauty of this new development is that it does not call for a major re-write of MM and could be implemented almost immediately. What do you guys think? Thanks.
I'm about a 54% WR player. Not the worst and not the best. I use XVM win chance. It most definitely corelates is my observation. I do not give up or drown when I see a 30% chance. I do play more conservative because you can't count on support when being aggressive. Some of my favorite games are winning when it says 30%. When I see 80% I know I have to be aggressive to get any damage. I don't want every game to be 50/50 but it would be nice if 80% of the games were in the 35 to 65 range. I think that can be done without out too much delay in que.
why u use xvm win chance whan thing dont rly work because if u have 3 with anonymizer on one team for example it will show wrong win% becasue those 3 xvm will not take into account
Yes I agree with your proposal. Also I dont think its too complecated. It would be enough to program an additional routine at the end of the MM. Lets say the MM is done as its now, then the routine looks at the skills and swaps some of tanks to balance the skill level. Maybe takes +5 seconds in the queue but not more.
For Random Battles I think the below metrics would be important in creating a balanced game. Tank winrate relative to map side and battle type Tank performance relative to map side and battle type (Damage dealt, kills, Damage taken, Damage prevented, penrate, assisted spotting, assisted tracking) Crew skill level Equipment modifiers Player tank performance relative to map side and battle type Player experience to map side and battle type (with other similar category of tanks like Assualt tank perhaps) Players perform in other battle types and may be stronger in those than randoms or vice versa so this would needed to be taken into consideration as well. The risk though is that games stalemate more often and the amount of draws increases. I personally would like to see a Commander voting system, where one player is voted by majority of players as a commander of the whole battle during loading, whom gets extra abilities like map ping rights and inspire, where players would follow advice from. I sometimes thing the 15-0 wins is because another side is better at influencing stronger positions and plays and inspiring other team members.
Skill based MM was in the game though. Old ranked was as close to performance based MM we could get and people complained about it enough that WG removed it. There are also things that screw up MM like tank balance/map balance. If your team has 3 tier 9 Patton meatballs and a wz-114 vrs three kpzs and an e75, your teams chance of winning just went down considerably. And these match ups do happen, I had that exact scenario on Sunday. There is also the problem of high win rate players at tier 6-7-8 going into tier 10 and under performing. So for real skill based MM you would need to balance that skill across tiers and at least on NA there is not enough players to make that happen. I have been playing since 2016 and there has always been blowouts. I think what people are noticing is that those blowouts are ~5 minutes now because WG has upped the tempo on everything, DPM, Mobility and more accurate high alpha cannons. Because everything is so fast, the individual player has less time to make carry plays.
So my worry is other games have tried this and failed. CoD is the prime example they tried to implement it into a normal casual lobby and it resualt in an increase in blowout lobbies. But an even bigger problem was the number or people rage quitting and leaving empty lobbies/one sided games. Honestly, NickShox makes an extremely calm and factual arguement againts skill based MM. Fix the tank matching as you said in a comment above first but other than that not much can be done.
I mean the QUEUE-time in EU servers is typically couple of seconds to 15sec. THEN the wait time before battle is always full 30sec. I would not mind AT ALL even if the queue time was +-30sec on every battle. If the battles were more even, heck give me one minute I don´t mind
They should use WTR. Just add up each sides WTR and make them relatively equal. The real issue is that it feels like the goal is to make games fast. Steamrolls are a by product of that.
There are SO many players constantly complaining about the match making. Which I agree with both points, of skill matching AND class-role matching. I would happily wait a couple minutes if every match would be very fair. I'm pretty sure WG has said players would prefer this instant matching. But You CANNOT complain about the matchmaker if you then want instant matching. This is either not true, or most players are dumb. Im pretty sure most players are dumb (except me obviously) so whatcha gonna do I guess
First: Winrate or WN8 are not a real measure of performance. Can just mean you play lower tiers and you win a lot in those. Second: Can't remember how many times WG has already refused to implement skill based MM . It's just not going to happen. And it's also not MMs fault, you have to remember MM works by "best effort" meaning it has a set of rules but if teams won't fit in a certain time it'll just launch with what it has. Meaning sometimes matches will be completely imbalanced. Nothing to do...
Thats nonsense. Even an imperfect solution is better then nothing. To say WN8 or WR is no real measure only bec there are some exceptions doesnt make any sense. Yes these things can be manipulated but in average it works. If you want a 100 % perfect solution you will never get one. Not only it WOT btw.
I agree to an extent but I believe they are working on it through map changes. I am pretty sure they said only skill based mm in special battle modes the more they looked at it. But the certain randomness magic about random battles is something people cherish too so while there are an amount of 15-0 matches I would have to see the numbers to not be biased as I would think it is much lower, but I also have a sample size, and am also yellow ish so I might be more of a problem on some maps but am very strong on others, which is also a factor. I don't think you could use overall very effectively but they would most likely go by your rating. But it isn't as straightforward as you would hope because a lot of people are casuals. Then I would like your help in understanding if rigging occurs later at night. Que times are a bit long and that is when I see more flop battles, and weird behavior out of tanks that it would seem like people buy accounts to run on second computers and side by side monitors. Last night we were on the map with the big hill on the corner and our T10 drove down to the field and just sat there, meanwhile i took my BZ up on the hill and they knew to push me as I was by the windmill and didnt notice no one was following me, so I observed the what I thought to be a bot but all the sudden it drove down to the right side of field and just sat there. Now you can make your own chat rooms. You ever check those out? just hit search without filling anything in, and there will be names like ebay no wins since 2020, or filthy player or the top three are always weird names with different punctuation to make sure they are at the top. I might guess that you could easily actively throw because often you have one strong heavy at least on one of the teams at tier 9 and then whatever qued against, and if you make that two coordinated heavies knowing they are overmatching a flank, not sure what all can be done. That is IF the lower tiers aren't competent. I have seen a lot of games that start to snowball and then overconfidence costs the team. I think those are some of my favorite matches. I think that is a great challenge to have such disadvantage but still control vision, use your guns and just punish them. Again though this is later game map construction so that they can do that. You do that a lot with your lights I have noticed, keep your team in it. For that though you need a garbage team. But just garbage enough to keep them busy long enough while you pick them apart because your dpm isnt there. But that is why I prefer medium play because you can tend to issues like that. Whatever the case, this is a well presented idea and great execution to hopefully instill some change! Kudos!
This is the fix I have proposed to WG often. WG should give you Skill Based/Balanced MM insisters a separate MM. Choose your own players from those you deem worthy to play with/against. You Darkninja and I agree to be side captains, agree on a map and battle type. We each then choose players with our desired qualities from a list of those who joined this MM as players only. Once that is done, we click Battle and let the chips fall as they may.
I think skill based MM could work. An easier start would be to slow the games down. reduce the top speed of heavies so they are not as fast as meds for a start. Reduce the DPM across the board as well. this should mean, meds a back to there roll of repositioning and outflanking and the reduced DPM will slow down how fast the bad players are punished for making mistakes especially at T10. The other thing is to require players to reach a certain skill level before progressing past tier 8
You need to split the ultrabad and great players more even. The very bad players do nothing, or die instant so just as important as the dominant ones in my opinion. I use XVM and its way too many games of one team 10 red/orange, vs 10 yellow/green and above. Thats easy to fix if WG wants. Doesnt have to be equal, just better than this randomness.
Ok, I feel you’ve put forth some sensible ideas. However… Wargaming already does this. You should read the account of their lawsuits regarding their matchmaking engine and how they roll out games that will for some give easy games, and eventually harder and harder until the player is at several losses, and then flips this idea by giving the player statistically likely wins. Ostensibly it’s to help the learning metrics, where the player will be challenged, will either learn (or in some cases learn more slowly or not at all as I see in some players), or become over-challenged. Acc to WG this creates a player base that becomes more entrenched and engaged; no one loses too much to become frustrated, nor win too much also, while being challenged to develop skill at the game. While I feel the pain of being in a JagE100 in a rolling match, I’ve equally been the t100 LT stuck in yet another city map, catered to slow, armoured, heavy tanks. So, my bes advice is, learn the game mechanics and legal patents (how matchmaker works); it isn’t conspiracy. As for rollers, don’t ‘get good’, get BETTER. It’s challenge and change.
All good points here, but you are wishing WG cared. They have had this match making issue for many years and done nothing. WG has stated they like fast matches, so I do not see them changing the match making. WG has limited the number of lites and Arty in a game, which was good, but they have not gone further to improve the game.
@@eax2010EA I agree for BZ, and elc 90 for obvious reasons, and for borrasque as well since it is basically an extra LT on your team. For other tanks I just don’t think it’s feasible because of queue restrictions and rare tanks
I get this logic, but I don't think it works that well. Darkninja on one team with a BZ or EBR105 is not equivalent to me on the other side with the same tank. Skill-based MM is actually more important than mirrored tanks.
WG has said many times over the years that they will NEVER implement any SBMM. So don't hold your breath. RN the mm cannot even divide by two. Nor does it attempt to balance by category within a class.. such as 'assault' td vs 'sniper' td. When badgered about such things, over these many years, Eekaboo (Official WoT channel) always says the same thing -> "We don't want to create a 'que time simulator' game." When told that we would gladly wait in que longer for decent games. he says the statistics do not support that, but say the opposite." To be clear: I would ABSOLUTELY wait longer for decent games. What we have had for years is trash, and practically the entire population has left the game.
Its funny how everyone is talking about AI. Well its this a perfect use case for it, got all these games all that data, replays , feed AI and come up with some models aka solutions to fix this. It wont be a fixed by just adjusting it once, but after 4 or 5 waves you could come to a something that's close to an idea scenario.
The problem is they dont care, i dont mind to play with tomatoes. My biggest problem is that 99% of games i play as low or bottom tier which makes for me hard to influence the game as much. But yes about 75% games are 15 to 3 blowouts which makes no fun for either side.
*I've solved this problem for YEARS....* And yes, multiple things need to be done, there is no "one thing fixes all", SBMM is just the start. Here's my basic list. 1. Skill "Balanced" MM.... Meaning whatever's in the Que, each side get's an Equal # of similar skilled players. i.e. there are 6 Unicorns, each side get's 3, 8 Green, each side get's 4, etc. And just split using the games Skill Rating system. 2. Remove Gold.... Guns have proper pen that can pen designated weakspots. For example, an armored tank that you can't pen almost anywhere else should allow you to pen the lower plate. 3. At least DOUBLE the HP of all tanks. 4. No tank does over 700 Damage, and it's ONLY Derps doing that high. 5. One Arty per side, and it does Max 300 Damage, but no shots lower than 100 if you actually HIT the tank. STUN is removed from the game, or given the same "time" as the Low-Tier Stun has which is actually reasonable. 6 Auto-Aim works like the Auto-Aim Mod, given to ALL tanks, EBR's no longer "snap". 7. Remove Round Placement Nerf put in years ago. Unacceptable that rounds shoot so inaccurately. 8. Power-Spread between tiers of the same Class is Max 7% in Armor, Pen, Damage, etc. In other words, max powerful tank you will face of your class with +2-2MM will be 14% stronger. Do all the above to START with as an essential thing, and the game will finally be more enjoyable. The "Quick Death" is removed, we can actually play some before dying. Mistakes are allowed, and we can try something different. The game becomes more dynamic, people can take chances instead of "afraid to move", aka Camp as much, etc. There is MORE I would do to improve the game, for example I would make Arty FUN again, make them FASTER, for example make the FV304 FUN again, make them more accurate. I would give them a "TD Mode" in which they have 3 Rounds that can be used to shoot as a TD to again create a more dynamic game. Many tanks I would make them faster on the "Low-End" so they can try to circle attack tanks and avoid Arty shots properly. Oh, my ultimate plan is to BUFF all the massive CRAP of the game to play as well as other tanks that are considered "good", but also some of those good tanks have crap aspects that tick us off, I would get rid of those also. When ANY tank of the game is fully loaded, modded, etc. it should perform almost PERFECT per it's class, style, etc. When you've spent ALL the MONEY you can on it to make it good, it SHOULD be good. It should FEEL like LOVE when playing. I'm not talking "op", I'm talking just love. Further, other tanks will be just like you in their capacity, so it's LOVE.... nobody will be "op" anymore. BTW, with certain changes to the game, certain "performance" characteristics would need to be buffed, Bloom/Aim-Times reduced, speeds improved, etc. For example the ShipBarn.... multiple characteristics on it would need to be buffed since the max damage it now would do is 700.
I think additionally, while not as impactful as your proposed idea, the three tier spread of tanks needs to be changed to 2 tiers, ie 9 & 10's, 8 & 9's, 7 & 8's and so on. Though I did just have a blowout 4:45 game in all tens. Shrug
Some of these blowouts are due to personality types. You get a team of docile campers against an aggressive Team that sacrifices a tank for spotting and they get spotted and rolled. Happens all the time. Bases in the game are not fortified or set-up to be easily defended once spotted.
I'd kill to have skill-based MM in this game. It likely wouldn't fix the blow outs completely, but it would make a much noticed dent in them appearing. I'd be VERY willing to wait a little longer in the queue for the MM to find me a match if it meant I wasn't facing off against 2 platoons of super unicums that just steamroller my team, as well as vice-versa. Three main fixes that need to be implemented to fix the steamrolling: 1. Skill-based MM (just like in Onslaught, they do a type of skill-based there all the freaking time for that game mode), and further, get rid of the +2/-2 MM rule. 2. Making gold rounds a choice between higher damage (normal ammo) and higher pen (gold ammo), and making all shells cost the same. They're already doing this with the new Czech light tanks. 3. Nerfing overpowered premium tanks like the BZ-176 (seriously, just give it the tier 8 tech tree's gun, nerf the lower plate a little, and call it a day), the premium EBR (i'd like wheelies removed completely, but I digress), and overperforming premiums like the Borrasque/Miel (increase the intra-magazine time between shells). Another thing to fix MM at high tiers: FORCE 279e AND CHIEFTAINs TO HAVE MIRROR MATCHMAKING. Seriously, even after the "nerfs" they're still fucking horrible to fight against.
Yes. We need this and also we need less -2 tier games. For two days I have been lowest tier in well above 80% of the games. Playing tier 6 and 7 is horrible, because you meet pros in tier 8 premimums all day long. Also why does mm assume you are the best players in the world when you team up with a friend?
Back in the day several years ago when I used XVM, I noticed the "all-round average win%" it gave me was at around 39-41%. Im average at best player at 52%, so I threw the XVM away, it was useless for me
I think their algorithm is pretty basic, It's based on your win rate, and the higher the win rate you have, the less 15 to 1 blowouts you will have. 25k+ battles, Just watch the patterns. Dont stress out on the night you have the designated L
I don't mind waiting 2-3 mins just to get more balance, I'm tired of being on the winning, or losing side knowing that most probably this was already predetermined by our lords at WG high command
the main question here is, if WG is even willing to adress the problem of 15:0 games. those short games will in the short term lead to more consumables used and therefore to more money earned overall. so if players still play this game and buy stuff, why on earth should WG change anything?
FINALLY SOMEONE SAID IT. Theres an elephant in the room,and its called MM ALGORITHM,which WG controls. Tank imblance is nothing compared to player skill imbalance. Concider a 40k games 400 WN8 46%wr T-100 player vs a 40k 2500 WN8 57%wr Sheridan player on open map.I bet you 7/10 times the good Sheridan player will win the match for his team...... Why put the five 2500+ WN8 CW veteran try-hards in the same team as top tiers,and the five casual-drunk-''i dont care'' 500 WN8 bots all in the other teams top tiers?
@@paulrichards1921 yeah xvm is prob very inaccurate now but anon didn’t exist in 2015. WG wouldn’t have that problem I think because even if you are displayed as anon they can see your stats
The matchmaker should only take skill that has been 7-30 days from the battle, beacuse of returning players. I have low win rate but I started playing again after like 3-4 years and got a lot better. Sorry for any spelling errors I am not from english speaking country.
Removing +2 matches would already be a huge step forward to better matchmaking! Frankly I really love this game even when I get a loosing streak but +2 matches when in the lowest tier is just stupid imho
recorded at 6am, dedication... but try to reverse it, its gonna fck u up in long term. you can try to stay up till 10pm and reset it that way :D it worked for me
short term this may be a solution but in the long run, this wouldn't work as all players would eventually end up with close to 50/50 win rates. in addition, platooning completely wrecks this system as a platoon of good players will run at 90%+ win rate regardless who is on the other side. to quote a game developer: you think you do but you dont........
I think also they should stop putting 2 platoons of Cheifs and IS-7s against a single platoon of stock E100’s who are average players were those two platoons have 4K WN8.
No Dark this wouldn't work. From the company perspective AND the player perspective but I do agree that the very best players would like it, not because it benefits them, because they don't need stats to feel good about their gameplay and here's why: 1. What I mean is that WG will not give up good queue times and average time per match for anything in the world, It benefits them by having a player login 20 minutes before going to work. That's the main argument but not the only one, a lot of players wouldn't still be playing just because of how much time they need to invest in the game. 2. My next argument is personal for me but I think most people agree with me even if they don't realize it, this game already lacks casual gameplay (which they somewhat fixed with all those new game modes but they aren't permanent) and what I mean by that is that I would never use random battles as an argument to a casual game mode because it isn't. You need random battle stats to be seen and the worst thing about that is that people don't care about win rate (with reasons) and what is ironic about all that is that very bad players are the ones caring about wins somehow. Win rate has no value because of a lot of factors; tiers, platoons, ridiculous time to cap (that literally hinders your learning curve by wasting time on cap but i understand why capping does take long), etc.. 3. The old WG 100% thought of that and I'd say the new WG did too and the ''solution is no solution'' is most likely what they learned from that. There is no way to give us a skilled base matchmaking in any real way without hurting some other part of the game too heavily (that is a HUGE risk for WG). Your solution is no exception to that. Having a different matchmaking will make stats irrelevant (personally I don't care) but I know A HUGE PART of the players will not like this change. Some players will avoid those hours, some will only play those hours and that brings problems left and right and yes there will be conspiration in a huge amount too. What are the people with small amount of time to play decides that 1 type of mm isn't worth it and is disgusting and it just happens to be when he's available to play? This is gonna be a mess and for nothing in the world they would let down whales from NA because they would be in trouble. We should remember that this game is wargaming's survival. They have too many employees to take such risks when they know they can keep getting money from greedy ass patches by doing 1 step forward and 2 backwards over and over. That is just a fact by the way. Remember 1rst iteration of crew 2.0? It really felt like they not only didn't care about what we should get but they also don't care about us at all if it wasn't for money. They are not here to make this game good they are here to make us feel like we need to come back to the game when we decide to leave. It's intentionally made that way to be an average game with TERRIBLE monetization. And if you disagree with what i just said I'll just say look at Path of Exile devs making a joke out of the already best devs there are in the world. WG is literally morally bankrupt. EDIT: I don't think all the WG devs are bad. I think the executives are.
The MM got even worse with the new Tour of duty concept since there is now platoon missions to be done. What does a very goodplayer with a 60% or higher win rate do when he has to play in a platoon? Does he invite a 45% WR player? No way, he will only platoon with like 55%+ WR players and the team with 3 high skilled players that works together has a huge advantage vs the 45-50% WR platoon in the other team. I have started to use a mod lately (not XVM) that shows the WR of all the players in a battle and I'm surprised to see that it's usually 5-6 players in total for both teams with a WR above 50%. This means a unicum platoon will have a party. This is far from being the main issue with the MM but it does affect the steam rolls for sure.
This is so obvious if you look at the extremes, watched a vid with 2x players in a platoon with a total WOR of 22k. They just butchered team after team until they ran into another platoon of 22k skill level. End of the game and only those 4 tanks left to fight it out.
WG looking to max revenue. Short games, high premium consumables use, high premium ammo use pushes players to buy premium time and gold. WOT is a dying game and WG looking to extract the last bit of cash from it No map improvements or development, op premium tanks, etc. Can’t see WG upping queue times to balance skill
blowout matches rarely if ever happen at high mmr. i got the game last week i ranked in as a super unicum and i just finished a match that was 15-10. play better and you'll get closer matches. the players WG uses to rig the 50% win rate matchmaking methods are the people that are terrible at video games and have no right to complain. put 10 of them on a team and that team loses.
I was trying to 2 mark type 71 and got to 83.7% …. For the next 2-3 games I played a STAKD platoon on enemy team with no ‘good clan players’ on my team .. I don’t use xvm … they were blowout games and I am now back to sub 80% on type …. Def not fun and skill based mm would be nice
@@nukky2796 ? How are you supposed to play good when the score is 0-5 and your down 5-7k in the first 2 minutes lol. Edit: please explain lol?… even darkninja said in this video, that in his blowout game he didn’t feel like he contributed to the win… so does darkninja have a skill issue?
@@nukky2796 and skill based mm would help my ‘skill issue’ by matching me up against people who also have a ‘skill issue’ so this statement makes no sense lol.
The reason I quit the game was due to these blow out unbalanced games. For some weird reason WG totally ignore this. I think they have lost alot of the player base for the same reason.
i think that wg could do this but im not a fan of this, in my opinion the random queue would get to sweaty and a lot of people would not like this (or at least me) I do think they should try to match platoons a bit (not an 60% toon vs a 45% platoon in top tier tanks) and try to balance the tanks on both teams (not 50B against mouse or leo vs 430U)
The WOT player rating is a perfectly good metric. In a blow out have a quick look and compare the drivers of the top tanks wot ratings. When the top tier Medium has a rating of 8k compared to his opposite number with 4.5k that game is over. Even worse if the LT in one team is much better than his competitor, game over. The only problem is that high skill players would have a long wait to get a match in lower tier games, but is that really a problem (seal clubbers)
You'll get no argument here, I think like you said it isn't the cure all but I don't see how it could be anything but helpful. They need to DO something. One thing that would help remove the complication to the problem is eliminate the +/-2mm, problem with that is that is a large portion of the PLAYER base wants an unfair advantage when they play...
Digging the DN content, not overly polished and honest. But the NA server's pop is so low, post midnight queues already so long; skill based mm would often leave us with 4+ min queues would be my fear.
as a EU1 veteran player, with such bad red-orange playerbase that makes up like up to 90% of currently active players on servers, probably going to 98% on EU3 u already have skill based MM with bad players no need to fix anything really - it is what it is, the game is crap.
@@scoutboy776 for me the problem is not when 90-95% of team is red or orange players u can still do damage and be on top even if you lose if you are a good player, simply because of situational awareness and good map reading u can even help your red team win but the problem on EU servers is when u are put on almost all red team with 1-2 good players and the enemy team has 7-8 good players of whom a few are purple players or blue players and then u get such games in streaks and u know u are meant to lose those games.Lately, people just desert such games or disconnect and change server or just go yolo and die at the beginning with curses in chat....There is no way it is not intentional, such MM, there is no way that they at WG dont know what they are doing with such mm. Also try playing tier 9 and 10 games during past few weeks on EU servers u will end up with 2 arty games game after game after game and then 3 arty game will happen as well...clickers are back and if you are an active heavy tank player they will focus u and make u go to garage quickly. Current MM state on EU servers is NOT FUN to be as polite as it is possible ;)
This software we use, WOT, is just that. The MM is set up to make WG the most profit and to keep the games rolling. There is no way there are enough people logged in at any one time to have a MM based on skill, it is just not going to happen. The amount of players in the queue would have to be in the thousands to get even close to a fair, skilled-based MM. It won't happen, mainly, because WG doesn't want that to happen. We are only guessing at how it works by what they are telling us, but I think there are factors involved that we are not privy too.
Not so obvious, but after thinking about it, you could simply add the aggregate wr on each side and balance on that. If you do a new type mm balance then wn8 would change. You don't need wn8 you just need win rate period, nobody give af about wn8.
@@msromike123 yeah I agree winrate would prob be good enough. Only reason I mention other metrics is they could test several or weighted avgs of several if it would be more accurate similar to what wn8 does. I think a good metric would be avg winrate over past 500 battles in that tier specifically if you have that many, which would prevent low tier 55% wr players who suck in tier 10s from getting matched against 55% wr tier 10 players who will perform much better, or something along those lines.
well it's all wrong the MM was the same before and steamroll was not an issue that means its got everything to do with speeding the game up giving heavies more speed and view range and less frontal waypoints and so many autoloaders destroyed the game
Dude there's alot of "only" arty players with wn8's of 50% and above. The players with the "conspiracy theories" as you call them do have valid arguments as my above example. I mean how are high skill rates achieved in playing mostly SPG's as even you have seen over the years of playing WOT. So with that said' i don't see MM with the same skill levels contributing to eliminating these blowouts. Do yourself a favour and look at the map changes especily Airfield (a bad map) where alot of the bushes are getting removed to increase these stupid blowouts even more. WG knows exactly why they are doing this. NB tell me what skill levels are required in battles like this one which are rapidly increasing and don't forget, have contributed to our WN8s
The sbmm ruins random battles in my opinion. Putting the same classes with the same roles tier per tier against each other would be enough here. If you want to play sbmm, play onslaught or advances. This is exactly why the battles are random, because they are called "Random battles". Also, why are you trying to push 2014-2015 data on the 2024 random queue? It makes no sense to me
That article is that of 1 persons side of 1000 battles, and even like u said at predicting 50% chance to win it's pretty accurate at extremes it was not. Not sure about you I was able to bullshit myself through True False questions alot too. I mean famn u got a 50% chance to get it right. And it does that well.
Bad players will still get themselves deleted early game or camp at the back until the game is over and they are the reason for blow out games. Eu servers since the merge of ru accounts is a prime example. You regularly have 75% of the teams being 30k+ battle players with sub 2k WTR and they die in the first 2 minutes of the game and the game is now unplayable 90% of the time.
I basically quit playing the game over it since the merge on EU. (+100 games a week -> +-30 a week) It wasnt great before, but now its just ridiculous, and im reminded of this every time i log on and so i log back off.
@@Tritriumchannel ... Same as that for me, gone from it being a main hobby that I played almost every evening to something I play a couple of games a week just to realise it's the same rubbish experience.
@@ziginaigra... skill based MM has been tested again and again in the "Ranked" game mode and has proven again and again it fixes nothing, you still get 15-0 blowouts in almost every game.
YOU CANT FIX A SKILL PROBLEM BY BOOTING OUT PLAYERS THAT KNOW HOW TO SURVIVE.....THAT IS EXACTLY WHAT HAPPENS TO THE BEST PLAY FOR FREE PLAYERS...THEY ARE SO GOOD AT STAYING ALIVE THE PREMIUMS COMPLAIN THEM OUT OF THE GAME BECAUSE THEY WONT DIE FOR THEM...
Dark i had a question. My back hurts when i see you rockin on this piece of shit chair you are sit on. Would you order a gaming chair for you when i send you a superthanks for this? Sure, a cheap one but i think there are many affordable offers in the US for an gaming chair. Unconditional offer from my side. Yes or No?
@@johannesnowak84 I really appreciate the offer and your support. Your tips have all gone towards me meeting my basic needs. That being said, I actually think this chair is pretty nice 🤣. I don’t know if I would like the headrest of a gaming chair or not to be honest, haven’t tried one before. But I am more than happy with the chair I have now, so thanks, but I’m happy with the one I have now.
@@darkninja___ Don' t fix what's not broken :-) I'm fine with that. Anyways i send a bit. Maybe it helps otherwise. :-) Thanks for your priceless videos and tips. Greetings from Austria!
We already have skill based MM. If you lose multiple battles in a row you get put in top tier. That's why we get so many stock top tier tanks. And yes that's crap WG thinking.
WoT does not have MM in terms of taking into account skill or something along those lines. WG once explicitly said they didnt want any kind of MM of that kind. That "random battles" should be truly "random". Hence, an already snowbally game to begin with, has tons of unfun blowout games as a result.
@@darkninja___i said i think ,i didnt say mm works like that 100%..but its little strange to me that almost every game one team have a lot better players,that doesnt look random to me+ usually u get streaks of good and bad teams
how about wg does a toutrial of how mm is made, if they show this it will show the scam or no scam, i think wg is refusing to do this because players will see the lie of how wg say its random, this mm in this game is not random, the way i think wg does mm is make one team first and which does this stack one side, and what ever that is left over in the que is on the other side, if wg would just show how mm is made in real time, not just one but 2 or 3 different games teams and how they are made
And also the issue with bz176 is not to nerf it (not gonna happen anyway) or to not sell it again. I would argue that it should be made available again for others to purchase at an even better deal. That will level out the playing field regarding bz176 in my estimation.
What do you guys think? Is this a good idea? I decided to make a part 2 for this video where I discuss other ideas that could help prevent steamrolls, should be up in the next week probably. Lastly, look at @chrisbadger8684 ‘s comment to see a good argument for why skill based mm might not be that hard to implement and would not increase queue times much. I think his implementation improves on the one I described in the video. It involves setting up teams as is done currently, but then switching players across teams who are in the same tier/class of tank to equal out the teams based on skill after teams have already been selected by the match-maker.
Steam rolling isn't just about skill, often it happens when one team loses their light tank, especially an EBR on an open map. It's also about tank balance. You have players on sides with even skills one side with Bat Chatillion Bourrasques a the other with AMX Chasseur de chars or M4A1 Revalorisés, you know the outcome before the match has even started.
Also the game is full of real BOTs, Yesterday there was a tier 9 bot 2000 games with an average of 1 damage. Wargaming needs to weedle this accounts out and ban them.
The most boring thing about WOT for me at the moment is not just the monotony of maps but also mostly knowing the outcome of the game by just looking at the line up of tanks before its even started.
This is a great idea. I’ve thought so for years. Does it fix every steamroll game? Of course not. But it would be a great improvement.
I think it would be relatively easy to implement. Agree, WG can come up with an improved WIN8 metric coupled with Winrate to evaluate each player. Let the current matchmaker come up with teams THEN using the new evaluation to shuffle a few players between teams for balance.
I’m guessing wait times wouldn’t have to grow much at all adding this last balancing step.
they tried this in World of tanks: Blitz
. . .it uhhhh, it did not go well, to put it mildly.
@@thrawn-ys9hf what was the bad outcome? Did they implement it in the same way as described above or was it a bit different?
@@thrawn-ys9hf Ok, I read a bit about it and people’s impressions on blitz. As far as I understand their mm system works such that if you have 70% win rate you will be matched with like 2-3 40% teammates to balance things out. I think in blitz this system is going to be much harsher on good players especially because the format is 7v7 where each player has a huge impact on the battle. In wot, solo pro players max out at around 66% win rate, with the very best of the best doing around 70% solo. Realistically 99+% of players will be within 40-60% wr players, and probably 90% within 45-55% wr (prob can check exact numbers on tomato.gg). So there will be way less shock to players who are used to winning or losing more often on avg for WoT pc than in blitz due to the smaller impact each player has in a battle. But I do think looking at that situation I realize that there is a very good chance people will complain about the change even if the battle quality truly is improving, simply because people dislike big changes and hate wargaming. This will be especially true if it has the words “skill based mm” in the patch notes. People will freak out even if it really isn’t true skill based mm but better branded as team skill balancing. Either way that is kind of disappointing to me because it means WG will prob never consider this as an option even if it was tested rigorously and shown to work for improving battle quality.
Thank you for bringing this issue up again DarkNinja, good video. I sent a ticket to WG about this issue long ago, complete with blowout matchup examples. Their response? WG will NOT implement skills based matches . I believe their reasoning is money. They are afraid the "pros" and "unicums" who bring in a lot of revenue would not appreciate facing tougher battles and see their win rate drop. There is definitely no reason why one team gets several green players while the other only has average players. I used to keep track of these stats and it happens the majority of times in blowout games. They fail to realize they are losing players by the thousands and don't gain new ones because it's become a frustrating game for newcomers. Better balance would definitely bring more new players and reduce Q wait times.
The other problem I see with MM is that tank roles have become meaningless. Lights used to scout, TDs used to snipe, etc. Nowadays, some heavies are as fast as some meds, meds have better view range than some lights, TDs have better armor than some heavies, lights have same alpha as many meds, and so on. The assumption that as long as each team has the same amount of tank types then it's balanced, is no longer a valid one. A new MM would have to take into account ROF, DPM, Armor, speed, penetration, etc. The game has evolved and MM has not. That is another serious problem.
Before thinking about skill-based matchmaking I would first fix the tank imbalances in the MM. Meaning that MM should consider the type of the vehicle within the class. This would already make it better when for example in Himmelsdorf other team wouldn't have Obj268v4 against Grille 15 or Maus against Type 71.
@@jannekoskinen2050 Yeah I agree that plays a part too, especially Sheridan vs good spotting LTs, tds with armor vs snipers, and hulldown heavies vs 50B/T57 on Westfield for example.
@@jannekoskinen2050 I decided I will be making a part 2 video with my other ideas and ideas I’ve heard from others that I support in the next couple of days. While I think skill based mm is a pretty radical idea I do think it would have the biggest impact if we could change one thing, but there are definitely things that would be easier to change that could still have a significant impact imo.
Sure thats a problem too. I had often even 2 Grille against 2 Mino or so. Would be enough to swap one of them to make it more balanced.
Right out of the gate, DarkNinja highlights the issue of 15-vs-0 or 15-vs-3 results. How telling is that? Players used to complain about 15-vs-5 results being too frequent. It's a Long-standing issue, getting worse over time. I think that indicates that War Gaming has no interest in averting "steamroll" battles.
If War Gaming cared about better game-play, +/-2mm match-making would be switched off.
I've tracked my battle results for over 3 years. In about 60% of battles, the winning team got greater than double the number of kills compared to the losing team. For example, a 7-vs-15 result would fall into this category. Very few battles end with a close kill count result.
Hey apologies for writing a short novel. But I think you’ll appreciate what I have to say:
I think the problem/Solution you identified at 17:40 is my main gripe with team-pool skilled based MM. Said plainly, it doesn’t account for skill across tiers.
An example making Skill pool based MM’s weakness clear: Let’s say you are in a tier 10 match, as a platoon of two 2600-3000 WN8 tier 8’s. You're the only two “good” players & your tier 10’s are 800-1200 WN8 players. Let’s also say the other team has just average 1400-2000 WN8 players across the board T8-10. And the pooled WN* has been calculated as a perfect match.
I would wager that the average WN8 team would win in *most* cases simply because the average skill of the enemies Tier 10’s will outplay your “bad” Tier 10’s, eventually leading to a collapse in one flank and a possible loss. This is even in the case that you and your good Tier 8 platoonmate are trying to carry. It’s just hard, because skill is somewhat bottlenecked by the tank you are using. Example: a good player in a T8 light will have major problems dealing with a MAUS.
NOW you dealt with this challenge with the solution you identified at 17:46 The idea being: Balancing the top tier’s skill levels (which is different from skill pool based MM). The idea is that because top tiers have the largest impact on the game (because their tanks are better), they should be the priority of the MM. And I think it is a step in the right direction.
However the main weakness of this, as you probably already know, is that the MM doesn’t always match vehicle subclasses/types correctly against one another. Meaning that a top tier Grille 15 can be matched up against a Minotauro. So this skill-based solution should be paired with tank type matching to ensure more balance.
And as you said this pressure on the MM queue times can be off-loaded by using peak-server/threshold “rules” - Funny enough I mention this exact same idea in my recent “Fixing light tanks’"Video. Good to know great minds think alike. But all this said, great video. Props to you for biting into the huge topic of Match-making. -- Gabriel
@@Tankers4Change I agree with everything you said, well put. After reading another comment (@chrisbadger) I think that there might be an easy compromise that as far as I understand it would not increase queue times much at all, which you can see under his comment.
@@Tankers4Change and while I definitely agree it can be complicated because of multiple tier games, as far as I understand it there is no skill based mm really in place rn so I think even small imperfect corrections could definitely be worth implementing.
@@darkninja___ I agree. Even imperfect fixes will do something to help 👍
What kind of day would it be if I didn't get a 13-0 steamroll by the red team? At least 8-12 times daily. Ahh thats why I don't spend anymore real cash on a game that hates their customers.
17:05
Waiting for another 30-60s during primetime is always worth it if it means you'll have 3 more minutes of playtime inside the actual battle, there's no doubt about it
Dont think it takes more than 5 sec additional waiting time.
This wouldn't add even one second to queue times. MM can pick the same 30 tanks and then just even the talent of the two teams as best as possible with a simple routine.
@@chrisbadger8684 actually now that you mention it that makes a lot of sense and is way simpler than what I was thinking they would have to do. They just pick the teams as done currently then they can swap players paired against each other in their tier/class until the teams are more balanced for skill (platoons make it a bit more complicated but it’s a good start). I think it’s a great idea.
I thought of this same idea. It won't work well when there are a bunch of tier 10 platoons including iyouxin and 2 other unicums yoloing in their IS7s, but it would work most of the time including off-peak hours when it's mostly solo players.
One detail problem I can think about is that a player has different levels of talent in different tanks, meaning it will be even more advantageous to play strong tanks. Even if this is countered, it would still be the case for players playing their new tank. Is there a fix to this?
I think wargaming may create a random tank pool mode with higher exp or credit award to counter q time. Like if you pick five tanks in a pool and you allow MM to decide which tank will going to the game for you(and you will play it). And you will earn more exp and credit if you play this mode. It may help reduce q time for more complicated mm(skill based).
@@frankg189 interesting idea
I think they should experiment with the standard ammo having a set damage amount and the gold having a lower amount like some of the game modes. Maybe not 50% less but less. This would incentivize using regular ammo and assist the newer or cheaper players. I have no issue with using gold, however in the interests of more even games I think it would help.
One of the counter arguments about skill based matchmaking, is that by forcing a rebalancing of the teams to be equal, by default the average win rate for each side will eventually over time also equalize out. ie, over time, with each team balanced to 50/50, the WR for these games will also start to mirror 50/50. And what this means is that more of the players of these games will also start to move to a ~50% WR.. ie, one of the reasons for the +55% WR, is that these top skilled players have an oversized impact on the outcome of the game against lesser skilled teams. And thus have a higher WR as a result. and it will have the equal impact for the lesser skilled players as they will no longer have as much opportunities to hamper their teams. Ie, the +55% WR players will see a drop over time for their WR, and the ~45% players will start to do better. Arguably this is not in the higher skilled players interest - although you are correct in that the matches will be more enjoyable.... Your arguments at the end of the video are valid though.
@@rpsmason personally that is something I would be fine with, because getting 70% wins is not enjoyable as most are just steamrolls, and even if winrate was dropping to 50% the good players still would have high damage battles which might be good enough for them.
@@darkninja___Yes thats true. Also I think as the skill balanced mm cant be perfect the WR would still differ between players.
Danke!
@@johannesnowak84 really appreciate it
There's some obvious stuff that can be fixed even before getting into skill based stuff. For instance, at load up, 1 team already has a 1000 hp advantage. WTF, that's like half an extra tank of damage. Or 1 team has 4 autoloaders and the other team has none. multiple assault tds vs multiple sniping tds, especially on certain maps where each class clearly has an advantage. In most my blow out games I see, it's pretty obvious to me just switching 2 or 3 of the tanks around from one team to the other would make a huge difference. 2 ELCs on one team vs 2 M41 90, I mean cmon, it's obvious.
I do like the idea of an impact score where the goal would be to get to 0 for both sides. For instance, a top player has impact score of 5 and a really bad player has -5. So you add up all the player scores on each side and try to get to an equal score. But this doesn't fix the tank v tank match ups. If you're top player is playing a bottom tier heavy with no pen and now you zero it out by making your top tier heavy suck, what if it's the reverse on the other team, where the top player is the top tier heavy. Bz 176 vs. t150, equally skilled players, who's winning?
First they need to rebalance a lot of tanks. This should happen regularly. Again, it's pretty obvious what needs what.
Then fix the tank v tank match ups.
Then you can do some skilled stuff
Skill wins everytime. An 8k WOR player will happily butcher 4-5k players regardless of what tank and the EU server is normally pretty good a tank match ups.
I like this too, honestly I think an overall win rate aggregate balance , is fast, easy to implement, ez to understand, and could help right away with minimal dev costs. It could be gamed, but you could just throw out th
This is a great conversation (I would like to see WarGaming included in these if possible) because it is a problem in all of there games. So this game is so old, and yet they have done nothing to adjust this even though they tell us how the MM is decided to the best of their ability. They are not even looking at the tanks they put into battle, I have sent them screen shots of the MM and have not heard back from them, in one instance we had no light they had one. I have had battles where they put 3 Batchats, 1 T57, 2 TVP's, and two lights, we had no Batchats, 1 TVP, and one light. I mean if they can't match up even the tanks to average out the fire power they will never (and I mean Never) try to correct a skill based MM. This was and still is a big thing in WOTB and that is why so many people have left that game. They are only concentrating on making money by there modes, and using outdated (cheaply done) rewards for these modes. The bell curve mechanics is a simple thing to do in this game just because (NA) the player base is small. Que times are long now in many cases, especially if you're in platoon. I like where your going but I don't believe WG is interested.
I've been promoting a similar idea for years. Thanks for making a comprehensive video with great data to back up your claims! o7
I agree that having roughly equally skilled teams should be the goal. Blowouts aren't fun for anybody.
Speculating here, but I think WG is moving toward implementing skill consideration into the MM when they made the shift to featuring/highlighting WOT rating instead of PR and winrate. WOT rating isn't perfect just like WN8 isn't perfect, but I think it does a bit better job than WN8 does capturing overall effectiveness because WN8 is based heavily in just damage dealing and kills.
An additional thing that WG could do is match tanks more on their specialization than just class. For instance, it doesn't make much sense to me to put a 50B and T57 on one team versus a Maus and 279 on the other team.
In games like xDefiant, they implemented not skill based matchmaking, but match based matchmaking. So they would select 30 random players, then try to average out the skill across the match. So its still random MM, but during loading, the teams would be balanced before the match starts.
nice vid dark Goodluck in WCI
I thought about this some more. I'd love to hear from you guys and DarkNinja too, about introducing a new feature that might help with blowouts. The one characteristic of blowouts is that one team gets overun quickly, especially when a flank falls rapidly. What if WG introduced a rule whereby the team that falls behind, by say 5 tanks, in the first 5 minutes gets a few AI-driven reinforcements (bots) that automatically spawn on the losing flanks or at base??? That should be really easy to implement as many events used AI tanks already.
I figure this is a bonus for BOTH sides. The winning side would welcome the opportunity to rack up more damage/kills and the losing side now has a chance to make a comeback. I realize the bots won't win against a good team but it mainly acts as a distraction and allows the defenders to regroup, return to base, and even flank the opponent to stage a possible comeback, or at least make it a closer, longer battle.
The beauty of this new development is that it does not call for a major re-write of MM and could be implemented almost immediately. What do you guys think? Thanks.
I'm about a 54% WR player. Not the worst and not the best. I use XVM win chance. It most definitely corelates is my observation. I do not give up or drown when I see a 30% chance. I do play more conservative because you can't count on support when being aggressive. Some of my favorite games are winning when it says 30%. When I see 80% I know I have to be aggressive to get any damage. I don't want every game to be 50/50 but it would be nice if 80% of the games were in the 35 to 65 range. I think that can be done without out too much delay in que.
why u use xvm win chance whan thing dont rly work because if u have 3 with anonymizer on one team for example it will show wrong win% becasue those 3 xvm will not take into account
btw i use wvm too and i never give up or drown(on purpose lol) because of bad team
Yes I agree with your proposal. Also I dont think its too complecated. It would be enough to program an additional routine at the end of the MM. Lets say the MM is done as its now, then the routine looks at the skills and swaps some of tanks to balance the skill level. Maybe takes +5 seconds in the queue but not more.
For Random Battles I think the below metrics would be important in creating a balanced game.
Tank winrate relative to map side and battle type
Tank performance relative to map side and battle type (Damage dealt, kills, Damage taken, Damage prevented, penrate, assisted spotting, assisted tracking)
Crew skill level
Equipment modifiers
Player tank performance relative to map side and battle type
Player experience to map side and battle type (with other similar category of tanks like Assualt tank perhaps)
Players perform in other battle types and may be stronger in those than randoms or vice versa so this would needed to be taken into consideration as well. The risk though is that games stalemate more often and the amount of draws increases.
I personally would like to see a Commander voting system, where one player is voted by majority of players as a commander of the whole battle during loading, whom gets extra abilities like map ping rights and inspire, where players would follow advice from. I sometimes thing the 15-0 wins is because another side is better at influencing stronger positions and plays and inspiring other team members.
Skill based MM was in the game though. Old ranked was as close to performance based MM we could get and people complained about it enough that WG removed it. There are also things that screw up MM like tank balance/map balance. If your team has 3 tier 9 Patton meatballs and a wz-114 vrs three kpzs and an e75, your teams chance of winning just went down considerably. And these match ups do happen, I had that exact scenario on Sunday. There is also the problem of high win rate players at tier 6-7-8 going into tier 10 and under performing. So for real skill based MM you would need to balance that skill across tiers and at least on NA there is not enough players to make that happen.
I have been playing since 2016 and there has always been blowouts. I think what people are noticing is that those blowouts are ~5 minutes now because WG has upped the tempo on everything, DPM, Mobility and more accurate high alpha cannons. Because everything is so fast, the individual player has less time to make carry plays.
WG was talking about skill based MM and then gave rental tanks to players with zero tier 10 battles. How does that work out?
So my worry is other games have tried this and failed. CoD is the prime example they tried to implement it into a normal casual lobby and it resualt in an increase in blowout lobbies. But an even bigger problem was the number or people rage quitting and leaving empty lobbies/one sided games. Honestly, NickShox makes an extremely calm and factual arguement againts skill based MM.
Fix the tank matching as you said in a comment above first but other than that not much can be done.
I ate food while watching this video, therefore it's peak content creation
I mean the QUEUE-time in EU servers is typically couple of seconds to 15sec. THEN the wait time before battle is always full 30sec. I would not mind AT ALL even if the queue time was +-30sec on every battle. If the battles were more even, heck give me one minute I don´t mind
They should use WTR. Just add up each sides WTR and make them relatively equal. The real issue is that it feels like the goal is to make games fast. Steamrolls are a by product of that.
There are SO many players constantly complaining about the match making. Which I agree with both points, of skill matching AND class-role matching. I would happily wait a couple minutes if every match would be very fair. I'm pretty sure WG has said players would prefer this instant matching. But You CANNOT complain about the matchmaker if you then want instant matching. This is either not true, or most players are dumb. Im pretty sure most players are dumb (except me obviously) so whatcha gonna do I guess
First: Winrate or WN8 are not a real measure of performance. Can just mean you play lower tiers and you win a lot in those. Second: Can't remember how many times WG has already refused to implement skill based MM . It's just not going to happen. And it's also not MMs fault, you have to remember MM works by "best effort" meaning it has a set of rules but if teams won't fit in a certain time it'll just launch with what it has. Meaning sometimes matches will be completely imbalanced. Nothing to do...
Thats nonsense. Even an imperfect solution is better then nothing. To say WN8 or WR is no real measure only bec there are some exceptions doesnt make any sense. Yes these things can be manipulated but in average it works. If you want a 100 % perfect solution you will never get one. Not only it WOT btw.
First: dumbest thing I’ve read today
I agree to an extent but I believe they are working on it through map changes. I am pretty sure they said only skill based mm in special battle modes the more they looked at it. But the certain randomness magic about random battles is something people cherish too so while there are an amount of 15-0 matches I would have to see the numbers to not be biased as I would think it is much lower, but I also have a sample size, and am also yellow ish so I might be more of a problem on some maps but am very strong on others, which is also a factor. I don't think you could use overall very effectively but they would most likely go by your rating. But it isn't as straightforward as you would hope because a lot of people are casuals.
Then I would like your help in understanding if rigging occurs later at night. Que times are a bit long and that is when I see more flop battles, and weird behavior out of tanks that it would seem like people buy accounts to run on second computers and side by side monitors. Last night we were on the map with the big hill on the corner and our T10 drove down to the field and just sat there, meanwhile i took my BZ up on the hill and they knew to push me as I was by the windmill and didnt notice no one was following me, so I observed the what I thought to be a bot but all the sudden it drove down to the right side of field and just sat there.
Now you can make your own chat rooms. You ever check those out? just hit search without filling anything in, and there will be names like ebay no wins since 2020, or filthy player or the top three are always weird names with different punctuation to make sure they are at the top. I might guess that you could easily actively throw because often you have one strong heavy at least on one of the teams at tier 9 and then whatever qued against, and if you make that two coordinated heavies knowing they are overmatching a flank, not sure what all can be done.
That is IF the lower tiers aren't competent. I have seen a lot of games that start to snowball and then overconfidence costs the team. I think those are some of my favorite matches. I think that is a great challenge to have such disadvantage but still control vision, use your guns and just punish them. Again though this is later game map construction so that they can do that. You do that a lot with your lights I have noticed, keep your team in it.
For that though you need a garbage team. But just garbage enough to keep them busy long enough while you pick them apart because your dpm isnt there. But that is why I prefer medium play because you can tend to issues like that.
Whatever the case, this is a well presented idea and great execution to hopefully instill some change! Kudos!
This is the fix I have proposed to WG often. WG should give you Skill Based/Balanced MM insisters a separate MM. Choose your own players from those you deem worthy to play with/against. You Darkninja and I agree to be side captains, agree on a map and battle type. We each then choose players with our desired qualities from a list of those who joined this MM as players only. Once that is done, we click Battle and let the chips fall as they may.
This seems like a good idea. I'd like to be select players who are team focused and play with them more without the commitment of going a clan
The tanks that get matched up is a bit messed up too.
E.g. 4 top tier auto loader heavies vs. 4 Regular heavies.
Grille vs Jagdpanzer on a city map in tier 8s is crazy
I think skill based MM could work. An easier start would be to slow the games down. reduce the top speed of heavies so they are not as fast as meds for a start. Reduce the DPM across the board as well. this should mean, meds a back to there roll of repositioning and outflanking and the reduced DPM will slow down how fast the bad players are punished for making mistakes especially at T10. The other thing is to require players to reach a certain skill level before progressing past tier 8
You need to split the ultrabad and great players more even. The very bad players do nothing, or die instant so just as important as the dominant ones in my opinion. I use XVM and its way too many games of one team 10 red/orange, vs 10 yellow/green and above. Thats easy to fix if WG wants. Doesnt have to be equal, just better than this randomness.
u think that is random that every game one team is better lol
@@ziginaigra no, I think the teams are uneven because the working mm makes them *randomly* uneven. it works as intended at the momoent
@@imwillingandable so u think that somehow randomly every game one team is a lot better ok lol
Ok, I feel you’ve put forth some sensible ideas. However…
Wargaming already does this.
You should read the account of their lawsuits regarding their matchmaking engine and how they roll out games that will for some give easy games, and eventually harder and harder until the player is at several losses, and then flips this idea by giving the player statistically likely wins. Ostensibly it’s to help the learning metrics, where the player will be challenged, will either learn (or in some cases learn more slowly or not at all as I see in some players), or become over-challenged. Acc to WG this creates a player base that becomes more entrenched and engaged; no one loses too much to become frustrated, nor win too much also, while being challenged to develop skill at the game.
While I feel the pain of being in a JagE100 in a rolling match, I’ve equally been the t100 LT stuck in yet another city map, catered to slow, armoured, heavy tanks.
So, my bes advice is, learn the game mechanics and legal patents (how matchmaker works); it isn’t conspiracy.
As for rollers, don’t ‘get good’, get BETTER. It’s challenge and change.
All good points here, but you are wishing WG cared. They have had this match making issue for many years and done nothing. WG has stated they like fast matches, so I do not see them changing the match making.
WG has limited the number of lites and Arty in a game, which was good, but they have not gone further to improve the game.
I just want mirrored tanks in each team. If one team gets BZ, the other should get a BZ as well.
elc vs elc and bourrasque vs bourrasque too :)
@@eax2010EA I agree for BZ, and elc 90 for obvious reasons, and for borrasque as well since it is basically an extra LT on your team. For other tanks I just don’t think it’s feasible because of queue restrictions and rare tanks
I get this logic, but I don't think it works that well. Darkninja on one team with a BZ or EBR105 is not equivalent to me on the other side with the same tank. Skill-based MM is actually more important than mirrored tanks.
@@gdalmuti Both factors could be applied
WG has said many times over the years that they will NEVER implement any SBMM. So don't hold your breath.
RN the mm cannot even divide by two.
Nor does it attempt to balance by category within a class.. such as 'assault' td vs 'sniper' td.
When badgered about such things, over these many years, Eekaboo (Official WoT channel) always says the same thing -> "We don't want to create a 'que time simulator' game." When told that we would gladly wait in que longer for decent games. he says the statistics do not support that, but say the opposite."
To be clear: I would ABSOLUTELY wait longer for decent games. What we have had for years is trash, and practically the entire population has left the game.
Its funny how everyone is talking about AI. Well its this a perfect use case for it, got all these games all that data, replays , feed AI and come up with some models aka solutions to fix this.
It wont be a fixed by just adjusting it once, but after 4 or 5 waves you could come to a something that's close to an idea scenario.
The problem is they dont care, i dont mind to play with tomatoes. My biggest problem is that 99% of games i play as low or bottom tier which makes for me hard to influence the game as much. But yes about 75% games are 15 to 3 blowouts which makes no fun for either side.
*I've solved this problem for YEARS....* And yes, multiple things need to be done, there is no "one thing fixes all", SBMM is just the start. Here's my basic list.
1. Skill "Balanced" MM.... Meaning whatever's in the Que, each side get's an Equal # of similar skilled players. i.e. there are 6 Unicorns, each side get's 3, 8 Green, each side get's 4, etc. And just split using the games Skill Rating system.
2. Remove Gold.... Guns have proper pen that can pen designated weakspots. For example, an armored tank that you can't pen almost anywhere else should allow you to pen the lower plate.
3. At least DOUBLE the HP of all tanks.
4. No tank does over 700 Damage, and it's ONLY Derps doing that high.
5. One Arty per side, and it does Max 300 Damage, but no shots lower than 100 if you actually HIT the tank. STUN is removed from the game, or given the same "time" as the Low-Tier Stun has which is actually reasonable.
6 Auto-Aim works like the Auto-Aim Mod, given to ALL tanks, EBR's no longer "snap".
7. Remove Round Placement Nerf put in years ago. Unacceptable that rounds shoot so inaccurately.
8. Power-Spread between tiers of the same Class is Max 7% in Armor, Pen, Damage, etc. In other words, max powerful tank you will face of your class with +2-2MM will be 14% stronger.
Do all the above to START with as an essential thing, and the game will finally be more enjoyable.
The "Quick Death" is removed, we can actually play some before dying. Mistakes are allowed, and we can try something different.
The game becomes more dynamic, people can take chances instead of "afraid to move", aka Camp as much, etc.
There is MORE I would do to improve the game, for example I would make Arty FUN again, make them FASTER, for example make the FV304 FUN again, make them more accurate. I would give them a "TD Mode" in which they have 3 Rounds that can be used to shoot as a TD to again create a more dynamic game.
Many tanks I would make them faster on the "Low-End" so they can try to circle attack tanks and avoid Arty shots properly.
Oh, my ultimate plan is to BUFF all the massive CRAP of the game to play as well as other tanks that are considered "good", but also some of those good tanks have crap aspects that tick us off, I would get rid of those also.
When ANY tank of the game is fully loaded, modded, etc. it should perform almost PERFECT per it's class, style, etc. When you've spent ALL the MONEY you can on it to make it good, it SHOULD be good. It should FEEL like LOVE when playing. I'm not talking "op", I'm talking just love. Further, other tanks will be just like you in their capacity, so it's LOVE.... nobody will be "op" anymore.
BTW, with certain changes to the game, certain "performance" characteristics would need to be buffed, Bloom/Aim-Times reduced, speeds improved, etc.
For example the ShipBarn.... multiple characteristics on it would need to be buffed since the max damage it now would do is 700.
I think additionally, while not as impactful as your proposed idea, the three tier spread of tanks needs to be changed to 2 tiers, ie 9 & 10's, 8 & 9's, 7 & 8's and so on. Though I did just have a blowout 4:45 game in all tens. Shrug
Some of these blowouts are due to personality types. You get a team of docile campers against an aggressive Team that sacrifices a tank for spotting and they get spotted and rolled. Happens all the time. Bases in the game are not fortified or set-up to be easily defended once spotted.
Thats often related to skill level.
I'd kill to have skill-based MM in this game. It likely wouldn't fix the blow outs completely, but it would make a much noticed dent in them appearing. I'd be VERY willing to wait a little longer in the queue for the MM to find me a match if it meant I wasn't facing off against 2 platoons of super unicums that just steamroller my team, as well as vice-versa.
Three main fixes that need to be implemented to fix the steamrolling:
1. Skill-based MM (just like in Onslaught, they do a type of skill-based there all the freaking time for that game mode), and further, get rid of the +2/-2 MM rule.
2. Making gold rounds a choice between higher damage (normal ammo) and higher pen (gold ammo), and making all shells cost the same. They're already doing this with the new Czech light tanks.
3. Nerfing overpowered premium tanks like the BZ-176 (seriously, just give it the tier 8 tech tree's gun, nerf the lower plate a little, and call it a day), the premium EBR (i'd like wheelies removed completely, but I digress), and overperforming premiums like the Borrasque/Miel (increase the intra-magazine time between shells).
Another thing to fix MM at high tiers: FORCE 279e AND CHIEFTAINs TO HAVE MIRROR MATCHMAKING. Seriously, even after the "nerfs" they're still fucking horrible to fight against.
Don't forget to link the article in the description
@@redgui-of-tar8776 fixed
Yes. We need this and also we need less -2 tier games. For two days I have been lowest tier in well above 80% of the games. Playing tier 6 and 7 is horrible, because you meet pros in tier 8 premimums all day long. Also why does mm assume you are the best players in the world when you team up with a friend?
Back in the day several years ago when I used XVM, I noticed the "all-round average win%" it gave me was at around 39-41%. Im average at best player at 52%, so I threw the XVM away, it was useless for me
45% matches be like: *Tumble weeds*
I think their algorithm is pretty basic, It's based on your win rate, and the higher the win rate you have, the less 15 to 1 blowouts you will have.
25k+ battles, Just watch the patterns. Dont stress out on the night you have the designated L
I don't mind waiting 2-3 mins just to get more balance, I'm tired of being on the winning, or losing side knowing that most probably this was already predetermined by our lords at WG high command
the main question here is, if WG is even willing to adress the problem of 15:0 games. those short games will in the short term lead to more consumables used and therefore to more money earned overall. so if players still play this game and buy stuff, why on earth should WG change anything?
FINALLY SOMEONE SAID IT.
Theres an elephant in the room,and its called MM ALGORITHM,which WG controls.
Tank imblance is nothing compared to player skill imbalance.
Concider a 40k games 400 WN8 46%wr T-100 player vs a 40k 2500 WN8 57%wr Sheridan player on open map.I bet you 7/10 times the good Sheridan player will win the match for his team......
Why put the five 2500+ WN8 CW veteran try-hards in the same team as top tiers,and the five casual-drunk-''i dont care'' 500 WN8 bots all in the other teams top tiers?
I'd make a separate mode for platooning and separate it from normal MM.
@@eax2010EA might make part 2 with my other ideas in the future.
@@darkninja___ do it
@@ziginaigra confirmed I’ll make a vid in the next few days.
The XVM win rate prediction counts anon tanks as a zero, or unknown, for the calculation. The more anon tanks, the more inaccurate the number is.
@@paulrichards1921 yeah xvm is prob very inaccurate now but anon didn’t exist in 2015. WG wouldn’t have that problem I think because even if you are displayed as anon they can see your stats
Game seems to be on the way out. FIRESALE ?
The matchmaker should only take skill that has been 7-30 days from the battle, beacuse of returning players. I have low win rate but I started playing again after like 3-4 years and got a lot better.
Sorry for any spelling errors I am not from english speaking country.
Removing +2 matches would already be a huge step forward to better matchmaking! Frankly I really love this game even when I get a loosing streak but +2 matches when in the lowest tier is just stupid imho
recorded at 6am, dedication...
but try to reverse it, its gonna fck u up in long term.
you can try to stay up till 10pm and reset it that way :D it worked for me
@@dannys7773 my schedule is nonexistent right now. Woke up at 1am today
@@darkninja___ 🤣
short term this may be a solution but in the long run, this wouldn't work as all players would eventually end up with close to 50/50 win rates. in addition, platooning completely wrecks this system as a platoon of good players will run at 90%+ win rate regardless who is on the other side. to quote a game developer: you think you do but you dont........
So if XVM said you had a 65% chance to win and it was actually 85% correct…..so what are you trying to say….at 65% win chance it is 85% correct…
I think also they should stop putting 2 platoons of Cheifs and IS-7s against a single platoon of stock E100’s who are average players were those two platoons have 4K WN8.
The NA and ASIA player base is way too tiny to support SBMM, it would kill queue time
both teams should have roughly the same WTR, within a few thousand
Needs to be a bit better than that because a really good player with a bunch of newbees will still run through a team of average Joes.
@@derekaarts4997 yeah i agree but it would be much more rare then it is now
No Dark this wouldn't work. From the company perspective AND the player perspective but I do agree that the very best players would like it, not because it benefits them, because they don't need stats to feel good about their gameplay and here's why:
1. What I mean is that WG will not give up good queue times and average time per match for anything in the world, It benefits them by having a player login 20 minutes before going to work. That's the main argument but not the only one, a lot of players wouldn't still be playing just because of how much time they need to invest in the game.
2. My next argument is personal for me but I think most people agree with me even if they don't realize it, this game already lacks casual gameplay (which they somewhat fixed with all those new game modes but they aren't permanent) and what I mean by that is that I would never use random battles as an argument to a casual game mode because it isn't. You need random battle stats to be seen and the worst thing about that is that people don't care about win rate (with reasons) and what is ironic about all that is that very bad players are the ones caring about wins somehow. Win rate has no value because of a lot of factors; tiers, platoons, ridiculous time to cap (that literally hinders your learning curve by wasting time on cap but i understand why capping does take long), etc..
3. The old WG 100% thought of that and I'd say the new WG did too and the ''solution is no solution'' is most likely what they learned from that. There is no way to give us a skilled base matchmaking in any real way without hurting some other part of the game too heavily (that is a HUGE risk for WG). Your solution is no exception to that. Having a different matchmaking will make stats irrelevant (personally I don't care) but I know A HUGE PART of the players will not like this change. Some players will avoid those hours, some will only play those hours and that brings problems left and right and yes there will be conspiration in a huge amount too. What are the people with small amount of time to play decides that 1 type of mm isn't worth it and is disgusting and it just happens to be when he's available to play? This is gonna be a mess and for nothing in the world they would let down whales from NA because they would be in trouble. We should remember that this game is wargaming's survival. They have too many employees to take such risks when they know they can keep getting money from greedy ass patches by doing 1 step forward and 2 backwards over and over. That is just a fact by the way. Remember 1rst iteration of crew 2.0? It really felt like they not only didn't care about what we should get but they also don't care about us at all if it wasn't for money. They are not here to make this game good they are here to make us feel like we need to come back to the game when we decide to leave. It's intentionally made that way to be an average game with TERRIBLE monetization.
And if you disagree with what i just said I'll just say look at Path of Exile devs making a joke out of the already best devs there are in the world. WG is literally morally bankrupt.
EDIT: I don't think all the WG devs are bad. I think the executives are.
I may not think that your solution is good Dark but I do like the fact that some people want a better game and think about it thoroughly.
The MM got even worse with the new Tour of duty concept since there is now platoon missions to be done. What does a very goodplayer with a 60% or higher win rate do when he has to play in a platoon? Does he invite a 45% WR player? No way, he will only platoon with like 55%+ WR players and the team with 3 high skilled players that works together has a huge advantage vs the 45-50% WR platoon in the other team. I have started to use a mod lately (not XVM) that shows the WR of all the players in a battle and I'm surprised to see that it's usually 5-6 players in total for both teams with a WR above 50%. This means a unicum platoon will have a party. This is far from being the main issue with the MM but it does affect the steam rolls for sure.
This is so obvious if you look at the extremes, watched a vid with 2x players in a platoon with a total WOR of 22k. They just butchered team after team until they ran into another platoon of 22k skill level. End of the game and only those 4 tanks left to fight it out.
WG looking to max revenue. Short games, high premium consumables use, high premium ammo use pushes players to buy premium time and gold. WOT is a dying game and WG looking to extract the last bit of cash from it No map improvements or development, op premium tanks, etc. Can’t see WG upping queue times to balance skill
blowout matches rarely if ever happen at high mmr. i got the game last week i ranked in as a super unicum and i just finished a match that was 15-10. play better and you'll get closer matches. the players WG uses to rig the 50% win rate matchmaking methods are the people that are terrible at video games and have no right to complain. put 10 of them on a team and that team loses.
I was trying to 2 mark type 71 and got to 83.7% …. For the next 2-3 games I played a STAKD platoon on enemy team with no ‘good clan players’ on my team .. I don’t use xvm … they were blowout games and I am now back to sub 80% on type …. Def not fun and skill based mm would be nice
skill based mm wont help with your own skill issue
@@nukky2796 ? How are you supposed to play good when the score is 0-5 and your down 5-7k in the first 2 minutes lol.
Edit: please explain lol?…
even darkninja said in this video, that in his blowout game he didn’t feel like he contributed to the win… so does darkninja have a skill issue?
@@nukky2796 and skill based mm would help my ‘skill issue’ by matching me up against people who also have a ‘skill issue’ so this statement makes no sense lol.
What if WG removes vents, bia, gun rammer and so on...
The reason I quit the game was due to these blow out unbalanced games. For some weird reason WG totally ignore this. I think they have lost alot of the player base for the same reason.
i think that wg could do this but im not a fan of this, in my opinion the random queue would get to sweaty and a lot of people would not like this (or at least me) I do think they should try to match platoons a bit (not an 60% toon vs a 45% platoon in top tier tanks) and try to balance the tanks on both teams (not 50B against mouse or leo vs 430U)
The WOT player rating is a perfectly good metric. In a blow out have a quick look and compare the drivers of the top tanks wot ratings. When the top tier Medium has a rating of 8k compared to his opposite number with 4.5k that game is over. Even worse if the LT in one team is much better than his competitor, game over.
The only problem is that high skill players would have a long wait to get a match in lower tier games, but is that really a problem (seal clubbers)
wtr isn't a good player metric some people in my clan have 5k and if you look at their wn8 it shows purple which is at minimum 8k wtr
@@scoutboy776 if they have purple stats with 5k wtr they probably farming from red line nonstop
You'll get no argument here, I think like you said it isn't the cure all but I don't see how it could be anything but helpful. They need to DO something. One thing that would help remove the complication to the problem is eliminate the +/-2mm, problem with that is that is a large portion of the PLAYER base wants an unfair advantage when they play...
SBMM ( Skill Based Match Making) bring balance in every game . On the other hand reduces the income of the content creators so ...
Digging the DN content, not overly polished and honest. But the NA server's pop is so low, post midnight queues already so long; skill based mm would often leave us with 4+ min queues would be my fear.
@@syrupusurper3774 my concern as well, but checked pinned comment
as a EU1 veteran player, with such bad red-orange playerbase that makes up like up to 90% of currently active players on servers, probably going to 98% on EU3 u already have skill based MM with bad players no need to fix anything really - it is what it is, the game is crap.
and super sweats fighting each other isn't a fun experience either
@@scoutboy776 for me the problem is not when 90-95% of team is red or orange players u can still do damage and be on top even if you lose if you are a good player, simply because of situational awareness and good map reading u can even help your red team win but the problem on EU servers is when u are put on almost all red team with 1-2 good players and the enemy team has 7-8 good players of whom a few are purple players or blue players and then u get such games in streaks and u know u are meant to lose those games.Lately, people just desert such games or disconnect and change server or just go yolo and die at the beginning with curses in chat....There is no way it is not intentional, such MM, there is no way that they at WG dont know what they are doing with such mm. Also try playing tier 9 and 10 games during past few weeks on EU servers u will end up with 2 arty games game after game after game and then 3 arty game will happen as well...clickers are back and if you are an active heavy tank player they will focus u and make u go to garage quickly. Current MM state on EU servers is NOT FUN to be as polite as it is possible ;)
This software we use, WOT, is just that. The MM is set up to make WG the most profit and to keep the games rolling. There is no way there are enough people logged in at any one time to have a MM based on skill, it is just not going to happen. The amount of players in the queue would have to be in the thousands to get even close to a fair, skilled-based MM. It won't happen, mainly, because WG doesn't want that to happen. We are only guessing at how it works by what they are telling us, but I think there are factors involved that we are not privy too.
this would add smurfing in random que. but that probably better then the current state
Not so obvious, but after thinking about it, you could simply add the aggregate wr on each side and balance on that. If you do a new type mm balance then wn8 would change. You don't need wn8 you just need win rate period, nobody give af about wn8.
@@msromike123 yeah I agree winrate would prob be good enough. Only reason I mention other metrics is they could test several or weighted avgs of several if it would be more accurate similar to what wn8 does. I think a good metric would be avg winrate over past 500 battles in that tier specifically if you have that many, which would prevent low tier 55% wr players who suck in tier 10s from getting matched against 55% wr tier 10 players who will perform much better, or something along those lines.
I don't play much anymore because the game play sucks...
well it's all wrong the MM was the same before and steamroll was not an issue that means its got everything to do with speeding the game up giving heavies more speed and view range and less frontal waypoints and so many autoloaders destroyed the game
slow down the speed of all tanks, and remove cvs / add equipment that counters cvs / make cvs work only stationary
Dude there's alot of "only" arty players with wn8's of 50% and above. The players with the "conspiracy theories" as you call them do have valid arguments as my above example. I mean how are high skill rates achieved in playing mostly SPG's as even you have seen over the years of playing WOT. So with that said' i don't see MM with the same skill levels contributing to eliminating these blowouts. Do yourself a favour and look at the map changes especily Airfield (a bad map) where alot of the bushes are getting removed to increase these stupid blowouts even more. WG knows exactly why they are doing this. NB tell me what skill levels are required in battles like this one which are rapidly increasing and don't forget, have contributed to our WN8s
they tried this in World of tanks: Blitz
. . .it uhhhh, it did not go well, to put it mildly
The sbmm ruins random battles in my opinion. Putting the same classes with the same roles tier per tier against each other would be enough here. If you want to play sbmm, play onslaught or advances. This is exactly why the battles are random, because they are called "Random battles". Also, why are you trying to push 2014-2015 data on the 2024 random queue? It makes no sense to me
That article is that of 1 persons side of 1000 battles, and even like u said at predicting 50% chance to win it's pretty accurate at extremes it was not.
Not sure about you I was able to bullshit myself through True False questions alot too. I mean famn u got a 50% chance to get it right. And it does that well.
skill base mm ..is one way but they dont want do that
Win rate means skill?
How does it mean anything with this god awful mm.
I bet there was probably at least one unicom on the enemy team, but he lost.
Bad players will still get themselves deleted early game or camp at the back until the game is over and they are the reason for blow out games. Eu servers since the merge of ru accounts is a prime example. You regularly have 75% of the teams being 30k+ battle players with sub 2k WTR and they die in the first 2 minutes of the game and the game is now unplayable 90% of the time.
I basically quit playing the game over it since the merge on EU. (+100 games a week -> +-30 a week)
It wasnt great before, but now its just ridiculous, and im reminded of this every time i log on and so i log back off.
@@Tritriumchannel ... Same as that for me, gone from it being a main hobby that I played almost every evening to something I play a couple of games a week just to realise it's the same rubbish experience.
no that is not the reason for blowouts lol whan one team have much better players =blowout almost every time
@@ziginaigra... skill based MM has been tested again and again in the "Ranked" game mode and has proven again and again it fixes nothing, you still get 15-0 blowouts in almost every game.
@@chucklesx ranked dont have sbmm
Every other online game has skill based MM. It's not that hard to do. If it was, WG wouldn't be the outlier.
There is no fixing. The games are short as intended, so you spend more money in a smaller amount of time
YOU CANT FIX A SKILL PROBLEM BY BOOTING OUT PLAYERS THAT KNOW HOW TO SURVIVE.....THAT IS EXACTLY WHAT HAPPENS TO THE BEST PLAY FOR FREE PLAYERS...THEY ARE SO GOOD AT STAYING ALIVE THE PREMIUMS COMPLAIN THEM OUT OF THE GAME BECAUSE THEY WONT DIE FOR THEM...
nice idea.. unfortunately players and youtubers should think about it instead of WG
Dark i had a question. My back hurts when i see you rockin on this piece of shit chair you are sit on. Would you order a gaming chair for you when i send you a superthanks for this? Sure, a cheap one but i think there are many affordable offers in the US for an gaming chair. Unconditional offer from my side. Yes or No?
@@johannesnowak84 I really appreciate the offer and your support. Your tips have all gone towards me meeting my basic needs. That being said, I actually think this chair is pretty nice 🤣. I don’t know if I would like the headrest of a gaming chair or not to be honest, haven’t tried one before. But I am more than happy with the chair I have now, so thanks, but I’m happy with the one I have now.
@@darkninja___ Don' t fix what's not broken :-) I'm fine with that. Anyways i send a bit. Maybe it helps otherwise. :-) Thanks for your priceless videos and tips. Greetings from Austria!
i need gaming bed do u have that too?😂with min 4 usb
@@ziginaigra at least i like the idea 😂
We already have skill based MM. If you lose multiple battles in a row you get put in top tier. That's why we get so many stock top tier tanks. And yes that's crap WG thinking.
Who said this I've lost 10 in a row b4. I just turned it off after that.
Nope
LOL had 8 blowouts yesterday before a single win. #nicetry
Doesn't exist, stop coping
@@thedude1553 I won 8 in a row yesterday. What time were you on LOL?
i think mm already take into account players skill :)
@@ziginaigra I never heard that before, I think it’s basically random
WoT does not have MM in terms of taking into account skill or something along those lines.
WG once explicitly said they didnt want any kind of MM of that kind.
That "random battles" should be truly "random".
Hence, an already snowbally game to begin with, has tons of unfun blowout games as a result.
@@Tritriumchannel thing is nobody rly knows 100% how mm algorithm works,apart from wg devs,u can believe them or not:)
@@darkninja___i said i think ,i didnt say mm works like that 100%..but its little strange to me that almost every game one team have a lot better players,that doesnt look random to me+ usually u get streaks of good and bad teams
how about wg does a toutrial of how mm is made, if they show this it will show the scam or no scam, i think wg is refusing to do this because players will see the lie of how wg say its random, this mm in this game is not random, the way i think wg does mm is make one team first and which does this stack one side, and what ever that is left over in the que is on the other side, if wg would just show how mm is made in real time, not just one but 2 or 3 different games teams and how they are made
And also the issue with bz176 is not to nerf it (not gonna happen anyway) or to not sell it again. I would argue that it should be made available again for others to purchase at an even better deal. That will level out the playing field regarding bz176 in my estimation.
@@ismailmiley4410 I get where you’re coming from but I still really do hope they nerf it, and I own one lol
play war thunder🗿
Doing that now screw WOT