I really enjoy your videos, I've used this one a few times to help my clanmates and myself better understand the combat system. Spreadsheets and swearing, and good sense of humor, I love it!
Hey Nicky! Great video, thanks for taking the time for the analysis and putting this together. I was wondering if you did any tests in scout troops and whether the mechanics for the same with them?
So what is the point of attack during defense in these equations? If you are a maxed player when would it be useful to use taffies? Wouldn’t you always want HHD?
Hi, The video is very interesting. For me it is the first one pointing to some calculations. One interesting point (correct me if I am wrong) is that the winner in the battle always kills 80% of the other army regardless any health pool calculations. This leads to an interesting conclusion that usually when wining large battles you kill more units than you can. Here is an example. Lets assume you have a single unit army of N units and which effective stats Attack=A, Deffence=D, Health=H. If you can say for your unit stats that 125 *A < D * H then if you face just a tiny bit slightly less army (e.g. with N-1 units) with the same stats then your army will kill 80% of the other army , while the other army, which is almost identical to yours will kill much less of your units because of the health pool formula and the inequality 125 *A < D * H
If I understand, I don't think an N-1 will kill a lot fewer than an N size if that is what you are saying. What seems to be missing in some of this discussion is that both sides get 20% back and that happens when 1 or both sides reach 100% deaths. In the N-1 case, both sides lose pretty much all of their troops and get 20% back, with the N side probably getting the victory.
Here is a tip. Take 1000 T1 vs 1500 T1. The damage done by each side, instead of 52%... etc, use 65% *100*100attack*1000troops. That will give (in round 1), 6 500 000 damage done by the small side and 9 750 000 by the large side. The health pool of the small is 1000*100*100 = 10 000 000 so it almost totally dies. The large side has health pool of 1500*100*100 = 15 000 000. Round 2 starts with large side at 8 500 000 health pool (ie, 850 troops left) and small side at 250 000 health pool (ie, 25 troops left). In round 2 small side gets wiped out but not before doing a damage of .65*100*100*25=162 500 (ie, another 16 troops killed from large army). So small loses 1000 and large loses 650+16=666. Give back 20% on each side and the deaths are 800/1000 and 533/1500. Notice how we are almost at the saturation point. 1538 leads to 520 iirc. This algorithm seems to be exact at least for the troop type against same troop type (but with tiers and stats of any value).
I am saying that the winner is killing the opposite army just because of power and not because of any health pool, which is great imbalance between winning and losing sides
Great Info here. So if I am understanding this. As far as gems are concerned, because it keeps going around and around which is the best to use, If you are attacking, use Taeff Gem, Troop specific attack gem, and then what for the last gem. Health I would assume? And on Defense, almost same thing, all troop health/ all troop defense, and then troop specific defense or health?
CruCial I'd suggest add all three gem types attack defense and health to all your fighting sets, idk if you use a specific troop set or saracen for attack or lynx for defense, but add all three types. Gems for all troop attack are rare so you should save them for your lynx gear.
Have a question: how important is attack during defense ? I should rather use health stones or attack ones during defense ? Basically ruby turmaline and health troop specific ignoring the scouts obvious.
How does this work with multiple troops in consideration? Like, if I have t5 and t6 vs t6? Also, I somehow feel that with multiple troops I can see the attack pattern going round robin.
hi, maybe a stupid question but in your formulas, t5 vs t5 would mean attacker always lose? offense366 , def x health/100 = way more than the offense is?
Hey Nicky, we just tried testing ur theory. But something is wrong, we made sure that the overwhelming was definitely 48% like u say to avoid any headaches. In our tests we added health items on the hero of the attacker that was defeated. The overwhelming was still 48% but we were killing more of the defenders troops. That said, health needs to be in the equation on the attackers end. Get back to us, ur onto something here :).
Hello, I do not speak English, only with the translator, the final conclusion is that it is better to recruit troops lvl 1? Did you detect an error in this?
Thus may have been covered and I missed in all of the equations discussed below but does this saturation ratio of 1.537-1 work for attacking as well as defending or is there a defensive advantage? I watched the video 3 times and my eyes aren't great but it looks like all if the equations were changing the defense numbers and keeping the same attacking force, but I might have missed something so I thought I'd ask.
If you do a battle on a neutral site like a tile. Then the simulator treats both the offense and the defense the same. Meaning if 100 t1s attack 1000 t1s with the same stats. The result will be identical as if the 1000 t1s attack the 100. In a town the wall makes a slight difference but not a big one. So to answer your question that value works both ways. If the attacker overwhelms the defender it reduces the amount of troops lost by the attacker. Anytime you win you overwhelm your opponent to some degree. What the ratio is talking about is the maximum amount of overwhelm that you can achieve.
Thanx for the video..but can u make a video on based of the same theory when the attackers and defenders have done some certain knowledges...how it affects it and gears too.... and how this health pool thing works against a different type of troops.... like seige vs melee, killer vs seige...
I'm working on a video on how siege vs melee etc works. The knowledge and stuff works the same way as the boost % I added. For instance if you have 100% attack from knowledge and 100% attack from gear. You end up with 300 attack on a tier 1. 100 + 200.
XX XX It is good to have lots of t1, focus on one troop type only but you need higher troops to stand behind them too. If you plan on taking hits. I have around 15 mill t1 and mostly attack with them only during kvks just in case stuff goes wrong not much points given to the enemy.
What platform are you playing on? I, along with many others have had issues with Android. That being said, I was wondering if you've heard of a fix for it? I have a work around but it sucks.
So how do reducers work exactly? It's clear that on t1 with 100 stat a 50% reducer halves that stat. But lets say they have a 100% buff, we still have this same old 50% reducer, do they end up with 150 or 100 stat.
I have a question. Is it not allways most important to have balanced stats in attack(defence and health?) with zB 500x500x500 do you have a biger army than with 900x300x300. And at attacks if your army is bigger you allways win and enemy lose 80% no matter how much your damage is isnt it?
Nicky Poo, do you want to collaborate with tests to determine the algorithm for fighting? I just don't have the resources to test very much but do have a few interesting theories on how it works. One reason you see the 48% you mentioned I think is because every fight (the algorithm) has at least 1 round, and during a round each side guarantees damage to the other side of a certain amount (determined by stats of each side). So the winner, if had added 1 billion extra troops, still could not have prevented that minimal round 1 damage. However, if after round 1 both sides have troops left, then you repeat a round 2.. and so on. The 48% might be 50% (are you sure there were no bonuses of any type?) or else the defender gets a small advantage. I was playing for formulas as well, .. anyway, I would love if someone could get data at a faster rate than I can (I have just one account and no willing friends, etc, blah blah sad song blah blah). I'm a software developer by hobby, fwiw.
Observation from the data you provided, eg at 9:10: The second round of fighting, if there is one, appears to use 1/3 ratio of swing hit vs swing miss. First round uses 1/2 ratio (1:1). This, along with one more item, is why during a saturation, the loser kills about 50% (52 or whatever). My rationalization of this is that half the time you miss and half the time you hit in a fight. It's arbitrary, but it seems to be what plarium chose. [there is one more important swing/miss/hit point below.. where unpaired troops hit 100% of the time.. see example 2] The "one more item" is that each side inflicts its full damage for its hits during that round. An alternative is that the weaker side gets killed fast enough that it doesn't inflict its full damage. Let me explain. Say one side can inflict damage of 1000 healthpool and the other can do 1 000 000. The 1M is way superior and will flatten small army in round 1; however, the deaths would come at the very end of the round after the small army has already inflicted its full damage on strong opponent for the round. This leads to the saturation effect. No matter how much bigger, the smaller always does its guaranteed first round damage. Rationalization can be that the excess troops can't do their damage until they make their way past the battle line, by which time the small army have already done their attack, even if the smaller army then gets demolished right afterward by the excess unpaired fighters or the larger army (eg, while they are reloading crossbows, loading catapult, picking up swords or pulling them out of bodies, etc). Round 2 exists only in cases where army "sizes" are similar enough in size or where both are weak enough (eg, via enemy weakening spells) that more rounds are required to finish off a side. Anyway, in round 2, the data shows that for every 100 extra in the larger side (Eg, see adjacent lines 1300 and 1400 in the table), the loss is about 33 troops. This may only apply to the losing side or to both. I am working on algorithm and assumed both but will only be able to test it if have more data. In any case, a rationalization is that in further rounds, a fighter is more fatigued than in the first round so you miss 2x as much as you hit. [see example 2 below] In the algorithm I am writing (ie, trying to reverse engineer plarium) at the end of the round, you have to take into account the massive power of the larger army to kill off the weaker as necessary. [see next comment that mentions healthpool] Example 1, both armies are the same size and strength. Round 1 they each kill about 50% of the other (swing miss the other half). Round 2 starts with each side half the size and that leads to a further killing off the enemy of 500/3=166 troops. So they each get reduced to 1/2*2/3 = 1/3 of original size (1000-500-166=1000/3). Round 3 would be similar although I don't know what ratio to use. In any case, few battles would make it past round 2 since the numbers must be close to each other to do so (on your table, it's the 1000 to 1100 region). Regardless of the fatigue ratio in further rounds, eventually both sides go to 0 after enough rounds (more or less, obviously something has to be done to allow one side to have slight advantage). Then plarium gives back 20% to end up with 80% loss on each side.. for the balanced army scenario. Recap: each side kills 50% of other side in first round but after enough rounds have killed 100% of each other. Example 2, as discussed with saturation, if one army is much larger, the weaker still does its damage, meaning in the case of same strength troops that half of the size of the smaller army (eg, 500/1000) is what the stronger loses. Then at the end of the round, if the large is large enough (larger than 150% of smaller), those extra troops hit **without miss**. The without miss part explains why 1538, ie, 1500 more or less, ie, 150% of 1000, is enough to kill off the weaker in round 1. Now, let's look more closely at this 1500 vs 1000 case: 1000 on each side swing miss/hit in 1:1 ratio killing 500 of the enemy (the deaths occur only at end of round). That leaves 500 unpaired troops from larger army. These can hit without miss for 500 more kills, killing the total 1000 of the smaller army (500 deaths from 1000 matched troops at 50% efficiency and 500 more deaths from remaining 500 unmatched troops at 100% efficiency). In the range of 1000 to 1500: the leftovers above 1000 kill that many of the weaker army but will leave some alive for round 2 (ie, 500+N, where Large army - Small army = N < 500, totals less than 1000, the army size of smaller army, so some troops of smaller army remain for round 2). With this algorithm you approximate everything on your table at least in the 1100 to 1500 range, in two rounds. The closer army sizes require 3 rounds or more. Quick rough 3rd example. Say 1200 vs 1000. The damage in round 1 from the 1000 paired off troops leads to 500 deaths from each side. Further, the remaining unmatched 200 troops from larger army don't miss at all so kill 200 more. Round 2 starts with larger army having lost 500, now having 700 alive and smaller army having lost 700, now having 300 alive. In round 2, 100 more are killed by the 300 matched troops (1:2 ratio of hit:miss). The larger army has 400 unpaired troops, inflicting 400 more kills on weaker army, demolishing what was left since 400+100>300. Meanwhile, the larger army went from 700 down to 600 in round 2. Round 3 won't be reached. To reach round 3, the sizes have to be closer so that the unmatched troops after round 2 aren't enough to finish off what remains of the small army. I don't know the fatigue ratio for rounds greater than 2. I didn't try to pluck it from your data since there is noise in the data (thanks, plarium and small sizes used in the data). I think 1/3 also should work though but not a big deal to know for most scenarios since most damage is done in first 2 rounds. Also, by round 3, perhaps it's important to understand the table/detail of 52% vs simplified 50%. In next comment below this one (this comment is large enough already), I'll mention the total damage done per round. It's related closely to the discussion you have on healthpool. Again, I don't know if details are correct, but you'll probably/maybe agree it makes sense.
Upon further review, the play is called back.. somewhat. OK, use 65% as the hit ratio for everyone for round 1 and round 2 and every other round. Also ignore the idea that unmatched troops hit at 100%. Let's look at two examples. Example 1, 1538 vs 1000 (saturation boundary): 1538 * 65% = 1000. Since this kills off everyone on other side, we are done, and Then give back 20% to give 800 deaths. On the other side, 1000*65% = 650 Since we know the battle ends after this round, Then give back 20% to give 520 deaths. This matches the table, 800 deaths to 520 deaths. Example 2, 1200 vs 1000. Round 1: 1200 * 65% = 780. We need round 2, and there are 1000-780=220 troops left on other side. On other side, 1000 * 65% = 650 We need round 2, and there are 1200 - 650 = 550 troops left on stronger side. Round 2, 550 vs 220: 550 * 65% = 358 No more rounds needed since other side had only 220 troops, so all are dead, all 1000. Give back 20% of 1000 deaths = 800 deaths. On other side, 220 * 65% = 143 No more rounds needed we already know, and 550-143=407 troops left on stronger side, implying 793 deaths out of 1200. Give back 20% of 793 deaths = 634 deaths. These numbers of deaths, 800 and 634 match what your table shows on the 1200 vs 1000 case (well, 635 but I rounded along the way instead of at the end so got slightly different 634).
I adjusted the algorithm to use 65%. It can generate values, but I don't have resources to test them. Do you have more data? Specifically, atm, vanilla troops (one type against one type only) in any quantities. The alg can give the final troop losses for each side. The alg can handle nonvanilla (ie, tier 4 is handled as if just tier 1 with 300 attack,defense,health), so if you have a detail of the net attack,defense,health (including spells, etc), I can test that as well I guess.
At 15:40 I get 496 defender deaths not 492 for the first entry. To be clear, I am reading your table as follows: attack 102, defense 103.75 , health 103.75 for **both** attacker and defender. Also 1000 vs 5000 and no other adjustments. Is that correct? [BTW, if I round upward the 103.75 to 104 in all cases, I get 495 which is a little closer but still not exact.]
The algorithm generated the exact same death count as you showed in the saturation table (9:10) for the several lines I tested (1001,1050,1100,1200,1300,1400,1500,1538). This was a T1 vs T1 vanilla case. While that table gave exact matches (once I switched over to the 65% efficiency modification), I have not gotten this match in the table at 15:40 yet. I am not sure I understand the set up in this last table for each attacker and defender or else the algorithm or something is a bit off. A brief description of the algorithm follows. It only handles one troop type vs one other troop type. I don't use healthpool formula you mentioned but use something similar. I calculate total health for each team (health of troop times number of troops). The defense is used elsewhere though. I have a net attack value defined as the attack divided by the defense of the enemy times 100. The times 100 is to get the units correct (since 100 health is used as the base value instead of 1 health). The other part is a ratio of attack over defense. That's how defense helps, it reduces the attack on you by that factor (with the 100 multiplier added in). This boils down to the healthpool you mentioned, but it's in a more broken down form used in cycles of the algorithm to calculate "exact" death counts for each side, presumably no matter the stats of each side (that's the hope). OK, basic algorithm (which is done in javascript but I won't get that detailed here): There is an outer and inner loop. The outer is over as many rounds as it takes. You break out of that whenever the healthpool of either side reaches or goes below 0 in a test done at the end of each round. That event would signal the last round and the end of combat (with one or both side's troops all dead). The inner loop (comprising one round) iterates over troop count (infinite loop you break out of when both sides reach the number of troops they started at in that round). During each cycle you calculate the attack by each team on the other. If the team is still able to attack, you do an attack (65%*(yourattack/enemydefense)*100) and hold that accumulated sum until the round ends and then subtract that from enemy healthpool. If exactly one team has run out of troops while the round is still running, it effectively adds 0 to its accumulated attack damage. Each cycle increments the troop counter. To repeat, at the end of the inner cycle, *after* every troop from both sides has attacked using the formula above, then for each team you take the accumulated damage to enemy healthpools and subtract from the actual enemy health pool to start the next round. This healthpool value defines the number of troops for that new round (just divide by health of one troop to get the count). With this approach of subtracting the damage at the *very end* of the round, even if the small army gets blasted at the end of the round, it still got all of its troops to attack during the round (this is seen easily in the saturation effect, where adding more to the large army doesn't impact the damage they take). At the end of combat take 80% of the deaths of each side and that is the final answer. Summarizing, this algorithm, a loop over rounds, each round a loop over troop attacks on healthpool, uses a formula similar to the healthpool formula you gave and is configured for attack, defense, and health of each side. Each side can only have one troop type. And the answer is the death counts if you play those size/attribute armies against each other. I tested and verified the exact values you gave in one of your tables (for T1 vs T1 at minute 9). I also verified some random T1 vs T4 examples you gave. What it has not done precisely is give the table at minute 15 of the video. I don't know if I understand the set up for that one. Still working on it... [Thanks for the video since I had been wanting to make progress with this algorithm for a while but was lacking larger quantity of data to complement my tests and give extra insight details and confirmation, which this video provided. Hope in return, the algorithm described is found interesting. It's all pretty simple. The key details are that there are rounds, that the attack calculated for a round (for each side) is accumulated and doesn't go into affect until the end of the round, and that net attack is precisely 65% of yourattack/enemydefense*100, done onto the total enemy's health (similar to healthpool), and this all seems to work no matter the troop type and attributes.] [Clarification: this handles any troop type but against that same troop type (but they can be different tiers against each other). Using different troop types requires adding in a bonus/penalty. Mixing troop types and having more than one type for an attack is next on the list once I can be more sure that the algorithm works for the easier cases.]
godmode In the table at minute 15 and in general in other parts of the game, different applications of percent (boosts, gain, etc) tend to group into more than one group. This is likely why 50% enemy reduction from skills and 30% reduction from a different source would not equal 80% net reduction. For example, let's say that the base is 100. Applying 50% takes it to 50. Applying 30% reduction on 50 gives 35. Having a single group (80%) vs two groups (50% and 30%) is the difference between (100)*(1.00-0.50)*(1.00-0.30) = 35 vs (100)*(1.00-0.50-0.30)=20. In one case they add up and then multiply. In the other they multiply separately. This happens in many other areas. For example, all weapons/armor combine additively with skills and knowledge when it comes to things like time boosts. The total is then applied multiplicatively. Meanwhile, VIP combines separately with other things when that applies (eg, in production speedups). And iirc the daily 10% boosts and hourly "spells" and townskin all combine into a group. [or something like that]. This is why adding more skill points or weapons percent when you already have a lot of those doesn't contribute much as adding a simple 10% time boost. If you want a big effect, add more % gains to a group that has a small value. However, also contrast with things that increase speed vs things that reduce time. Increase speed is based upwards from 100%. Eg, 500% would allow you to do 6 total times the original amount of work in the same amount of time. That is you are increasing the rate (increasing the work done per unit time). Alternatively an -80% time reduction boost would achieve the same effect as that 500% rate boost. The 80 is a much smaller number but applied differently mathematically. In fact the "god mode" might have been because of something like the above, eg, 53% enemy reduction (skill) and 45% enemy reduction (spell) if combined additive before applying as a reduction, would give a 98% reduction, effectively reducing the enemy's level by a factor of 50 (ie, to 2% from 100%)! If they combine separately but also as a reduction, then you reduce effectively by multiplying by a factor of .5*.55=.275=27.5% which is approximately a reduction by a factor of 4. Compare 4 vs 50! Finally, if there were to apply inversely (eg, rate and time examples above, where one increases above 100% and then you divide by that as opposed to just multiply from the onset and move towards 0), then it would be even less effective. Eg, even as a whole (98%), it that might lead to dividing by 1.98 which is just a factor reduction of about 2.
godmode: math subtleties. a) There is the effect of combining into one vs different groups before applying. b) There is the effect of multiplying directly vs directly but after subtraction from 100%. c) There is the effect of multiplying inversely or directly. (eg, when talking about rates, work done (distance covered, food make, etc) are inverses with time) Let's take 60% and 40% and consider the possible ways to affect 100. a) i) Here we could be talking about the difference between applying as 60% and then as 35% (or vice-versa if we are just multiplying since multiplication is commutative.. and also associative). 100 * 0.6 * 0.35 = 0.21 ii) In contrast we could be talking about applying as one unit (60%+35%=95%). 100 * 0.95 = 0.95 b) i) An example of the case where we first subtract from 100% is if we are talking about a reduction. 60% reduction is the same as multipying by 40%. We take 100% and reduce it by 60% to get 40%. 100 reduced by 60% = 100 * 0.4 = 40. ii) In contrast, other uses of percents are applied directly, simply times 60%. 100 * 0.6 = 60. c) i) Here consider boost that reduces the amount of time something takes by 60% (to 40% of the whole). In this case, we multiply the total time by 0.40. If something took 10 hrs, it would now take 4 hrs. ii) In contrast, a 60% improvement in the speed your town produces food would lead to the new time being calculated differently. We increase above 100% by 60% to get 160%, aka, 1.6. That is what it means to improve the rate of something by 60%. Then that value affects the total time inversely, eg, if we could produce 100 units of food in 1 hour, now we can produce 160 units in that same hour. Notice we are talking about increasing food production (aka, that rate) by a certain amount instead of talking about reducing the time by that amount. Taking 100 units/hr as the base rate, if we have only 1000 units to produce, the time is cut from 1000/100 = 10 hrs to 1000/160 = 6.25 hrs. This is different from the 4 hrs if we had applied the 60% as a reduction to time instead of as an increase to the rate. As a special case, let's consider 60% and 40%. In a godmode, we could combine these to get 100% and then apply that as a reduction to time (or to strength or anything else), reducing the time (or strength, etc) to 0. Obviously, if you find yourself doing something like that, reducing something close to 0, you know you are probably misapplying the percent.. or plarium goofed. Or exploit exploit exploit for glory in the game!
little goof. I said near the top to consider 60% and 40%, but most of the examples used only 60% (40% being incidental since 100-60=40) and 35% in the case where we were considering 2 groups vs 1. At the bottom, I do 60 and 40 however. 60 and 35 would have given 95% not 100%.
There is offense, offense during defense and offense during attack. See, defense is the opposite of attack. If someone attacks you, you have to defend and your offense during defense will be used to compare with the defense during attack of your enemy. If you have a choice between 1% increase in offense, 1% in increase in offensive during defense and 1% increase in offensive during attack, you always prefer 1% in offense cause that's equal to 1% increase in offense during defense together with 1% increase in offense during attack.
Question 35000 vs 4000 .. the 35 k were fenrir rank 5 vs 4000 rank 1 poly. And 4000 was minimal and 35000 limp away with maybe 2k unwounded or killed .... How is that even possible
Якщо коротко: марення сивої кобили. Ідея цікава. А ось цифри підганяються, щоб отимати необхідне. Імперічне усе - логічно не повязане. Намішав кашу з вінегретом, атоді підгонка до неповязаного. І, вуаля - цифри начебто сходяться....
Before the T6 update. T1's were defensively dominating regardless the size or stats of incoming attacks. 10m T1's with an average stat will hold against a massive OS. After the update, that shit no longer exist.
Yes and no. Even before the t6 update the t1s were somewhat on their way out. The reason for this is because of the new knowledge and march levels. They are allowing bigger guys to cut through way more t1s than before. On top of that a big onslaught is much easier to make with bigger Mead sizes and 150% enlargements. But t1 masses still work. You just need a lot more than before. 10mil or whatever wont work. You need like 30-50.
I agree, but before the update I've been dominating against massive OS multiple times with just 10m T1's and 500k T5's. Also, had an additional 6M+ rein by clansmen. Maybe the opposing skills were shit, though it's ashame we can't see all participants stats other than the one who is hosting. After the update, I was still convinced my wall will hold strong. Fuckers blast right thru it with ease lol. Now, I definitely need 30-50.
They basically just screwed us with the massive new marches they can send out. Train now t1 at least 20 mill but behind them have t4 and t5 at least 5 mill each, use t1 just as disposal if you ever face a monster sending a full t6 at you. I use siege only so I'm cool cos they all focus on mele it's easier to do finish it and get the t6s ready for action.
One more piece of advice if you have a huge army, don't fuck around with double hero, have it always on when going on multiple tile hits at start of kvk or cvc be in your lynx gear, farmers will be murdered anyhow. I can hit in 5 seconds but there are ppl out there who hit in two you won't even be able to tap items for shield even if you shield after each attack momentarily.
Mrclouder+ I agree, that's my strategy when hitting tiles/towns. I leave 1 hero home and send second for battle. Speaking of T6 those boys are fucken deadly especially the high skills. I've seen some reports of massive OS 9M against 3M T6 and lost. They are the NEW god mode
What is the best troops to use as main troops ? and the best set skill points ?
I really enjoy your videos, I've used this one a few times to help my clanmates and myself better understand the combat system. Spreadsheets and swearing, and good sense of humor, I love it!
Hey Nicky! Great video, thanks for taking the time for the analysis and putting this together. I was wondering if you did any tests in scout troops and whether the mechanics for the same with them?
Nick, have you had an opportunity to make any more progress on these calculations? Or worked with Hozelda with his variations?
What is Hozelda's variations? Do you have a link?
Nicky what is your take on building resource towns to support your main town
you from the burgh?
you say toward the end to add Taffies to your attack set - what are 'taffies' ? tia
Taaffeite gems for Total Offense.
So what is the point of attack during defense in these equations? If you are a maxed player when would it be useful to use taffies? Wouldn’t you always want HHD?
Hi, The video is very interesting. For me it is the first one pointing to some calculations. One interesting point (correct me if I am wrong) is that the winner in the battle always kills 80% of the other army regardless any health pool calculations. This leads to an interesting conclusion that usually when wining large battles you kill more units than you can. Here is an example. Lets assume you have a single unit army of N units and which effective stats Attack=A, Deffence=D, Health=H. If you can say for your unit stats that 125 *A < D * H then if you face just a tiny bit slightly less army (e.g. with N-1 units) with the same stats then your army will kill 80% of the other army , while the other army, which is almost identical to yours will kill much less of your units because of the health pool formula and the inequality 125 *A < D * H
If I understand, I don't think an N-1 will kill a lot fewer than an N size if that is what you are saying. What seems to be missing in some of this discussion is that both sides get 20% back and that happens when 1 or both sides reach 100% deaths. In the N-1 case, both sides lose pretty much all of their troops and get 20% back, with the N side probably getting the victory.
Here is a tip. Take 1000 T1 vs 1500 T1. The damage done by each side, instead of 52%... etc, use 65% *100*100attack*1000troops. That will give (in round 1), 6 500 000 damage done by the small side and 9 750 000 by the large side. The health pool of the small is 1000*100*100 = 10 000 000 so it almost totally dies. The large side has health pool of 1500*100*100 = 15 000 000. Round 2 starts with large side at 8 500 000 health pool (ie, 850 troops left) and small side at 250 000 health pool (ie, 25 troops left). In round 2 small side gets wiped out but not before doing a damage of .65*100*100*25=162 500 (ie, another 16 troops killed from large army). So small loses 1000 and large loses 650+16=666. Give back 20% on each side and the deaths are 800/1000 and 533/1500. Notice how we are almost at the saturation point. 1538 leads to 520 iirc. This algorithm seems to be exact at least for the troop type against same troop type (but with tiers and stats of any value).
I am saying that the winner is killing the opposite army just because of power and not because of any health pool, which is great imbalance between winning and losing sides
Great Info here. So if I am understanding this. As far as gems are concerned, because it keeps going around and around which is the best to use, If you are attacking, use Taeff Gem, Troop specific attack gem, and then what for the last gem. Health I would assume? And on Defense, almost same thing, all troop health/ all troop defense, and then troop specific defense or health?
CruCial I'd suggest add all three gem types attack defense and health to all your fighting sets, idk if you use a specific troop set or saracen for attack or lynx for defense, but add all three types. Gems for all troop attack are rare so you should save them for your lynx gear.
One question . if you have several level of toops how it is calculated ?
Have a question: how important is attack during defense ? I should rather use health stones or attack ones during defense ? Basically ruby turmaline and health troop specific ignoring the scouts obvious.
How does this work with multiple troops in consideration? Like, if I have t5 and t6 vs t6? Also, I somehow feel that with multiple troops I can see the attack pattern going round robin.
hi, maybe a stupid question but in your formulas, t5 vs t5 would mean attacker always lose? offense366 , def x health/100 = way more than the offense is?
Hey Nicky, we just tried testing ur theory. But something is wrong, we made sure that the overwhelming was definitely 48% like u say to avoid any headaches. In our tests we added health items on the hero of the attacker that was defeated. The overwhelming was still 48% but we were killing more of the defenders troops. That said, health needs to be in the equation on the attackers end. Get back to us, ur onto something here :).
Hello, I do not speak English, only with the translator, the final conclusion is that it is better to recruit troops lvl 1? Did you detect an error in this?
Thus may have been covered and I missed in all of the equations discussed below but does this saturation ratio of 1.537-1 work for attacking as well as defending or is there a defensive advantage? I watched the video 3 times and my eyes aren't great but it looks like all if the equations were changing the defense numbers and keeping the same attacking force, but I might have missed something so I thought I'd ask.
If you do a battle on a neutral site like a tile. Then the simulator treats both the offense and the defense the same. Meaning if 100 t1s attack 1000 t1s with the same stats. The result will be identical as if the 1000 t1s attack the 100. In a town the wall makes a slight difference but not a big one.
So to answer your question that value works both ways. If the attacker overwhelms the defender it reduces the amount of troops lost by the attacker. Anytime you win you overwhelm your opponent to some degree. What the ratio is talking about is the maximum amount of overwhelm that you can achieve.
Nicky Poo thank you that's helps a lot. Keep up the good work👍
Thanx for the video..but can u make a video on based of the same theory when the attackers and defenders have done some certain knowledges...how it affects it and gears too....
and how this health pool thing works against a different type of troops.... like seige vs melee, killer vs seige...
I'm working on a video on how siege vs melee etc works.
The knowledge and stuff works the same way as the boost % I added.
For instance if you have 100% attack from knowledge and 100% attack from gear.
You end up with 300 attack on a tier 1. 100 + 200.
Hello, I do not speak English, only with the translator, this video shows that it is better to recruit troops lvl1?
XX XX It is good to have lots of t1, focus on one troop type only but you need higher troops to stand behind them too. If you plan on taking hits. I have around 15 mill t1 and mostly attack with them only during kvks just in case stuff goes wrong not much points given to the enemy.
What platform are you playing on? I, along with many others have had issues with Android. That being said, I was wondering if you've heard of a fix for it? I have a work around but it sucks.
It would be wise to learn a little more about troops because of all the money spent in this game!
So how do reducers work exactly? It's clear that on t1 with 100 stat a 50% reducer halves that stat. But lets say they have a 100% buff, we still have this same old 50% reducer, do they end up with 150 or 100 stat.
I have a question. Is it not allways most important to have balanced stats in attack(defence and health?) with zB 500x500x500 do you have a biger army than with 900x300x300. And at attacks if your army is bigger you allways win and enemy lose 80% no matter how much your damage is isnt it?
Nicky Poo, do you want to collaborate with tests to determine the algorithm for fighting? I just don't have the resources to test very much but do have a few interesting theories on how it works. One reason you see the 48% you mentioned I think is because every fight (the algorithm) has at least 1 round, and during a round each side guarantees damage to the other side of a certain amount (determined by stats of each side). So the winner, if had added 1 billion extra troops, still could not have prevented that minimal round 1 damage. However, if after round 1 both sides have troops left, then you repeat a round 2.. and so on. The 48% might be 50% (are you sure there were no bonuses of any type?) or else the defender gets a small advantage. I was playing for formulas as well, .. anyway, I would love if someone could get data at a faster rate than I can (I have just one account and no willing friends, etc, blah blah sad song blah blah). I'm a software developer by hobby, fwiw.
Observation from the data you provided, eg at 9:10:
The second round of fighting, if there is one, appears to use 1/3 ratio of swing hit vs swing miss. First round uses 1/2 ratio (1:1). This, along with one more item, is why during a saturation, the loser kills about 50% (52 or whatever). My rationalization of this is that half the time you miss and half the time you hit in a fight. It's arbitrary, but it seems to be what plarium chose. [there is one more important swing/miss/hit point below.. where unpaired troops hit 100% of the time.. see example 2]
The "one more item" is that each side inflicts its full damage for its hits during that round. An alternative is that the weaker side gets killed fast enough that it doesn't inflict its full damage. Let me explain. Say one side can inflict damage of 1000 healthpool and the other can do 1 000 000. The 1M is way superior and will flatten small army in round 1; however, the deaths would come at the very end of the round after the small army has already inflicted its full damage on strong opponent for the round. This leads to the saturation effect. No matter how much bigger, the smaller always does its guaranteed first round damage. Rationalization can be that the excess troops can't do their damage until they make their way past the battle line, by which time the small army have already done their attack, even if the smaller army then gets demolished right afterward by the excess unpaired fighters or the larger army (eg, while they are reloading crossbows, loading catapult, picking up swords or pulling them out of bodies, etc).
Round 2 exists only in cases where army "sizes" are similar enough in size or where both are weak enough (eg, via enemy weakening spells) that more rounds are required to finish off a side. Anyway, in round 2, the data shows that for every 100 extra in the larger side (Eg, see adjacent lines 1300 and 1400 in the table), the loss is about 33 troops. This may only apply to the losing side or to both. I am working on algorithm and assumed both but will only be able to test it if have more data. In any case, a rationalization is that in further rounds, a fighter is more fatigued than in the first round so you miss 2x as much as you hit. [see example 2 below]
In the algorithm I am writing (ie, trying to reverse engineer plarium) at the end of the round, you have to take into account the massive power of the larger army to kill off the weaker as necessary. [see next comment that mentions healthpool]
Example 1, both armies are the same size and strength. Round 1 they each kill about 50% of the other (swing miss the other half). Round 2 starts with each side half the size and that leads to a further killing off the enemy of 500/3=166 troops. So they each get reduced to 1/2*2/3 = 1/3 of original size (1000-500-166=1000/3). Round 3 would be similar although I don't know what ratio to use. In any case, few battles would make it past round 2 since the numbers must be close to each other to do so (on your table, it's the 1000 to 1100 region). Regardless of the fatigue ratio in further rounds, eventually both sides go to 0 after enough rounds (more or less, obviously something has to be done to allow one side to have slight advantage). Then plarium gives back 20% to end up with 80% loss on each side.. for the balanced army scenario. Recap: each side kills 50% of other side in first round but after enough rounds have killed 100% of each other.
Example 2, as discussed with saturation, if one army is much larger, the weaker still does its damage, meaning in the case of same strength troops that half of the size of the smaller army (eg, 500/1000) is what the stronger loses. Then at the end of the round, if the large is large enough (larger than 150% of smaller), those extra troops hit **without miss**. The without miss part explains why 1538, ie, 1500 more or less, ie, 150% of 1000, is enough to kill off the weaker in round 1. Now, let's look more closely at this 1500 vs 1000 case: 1000 on each side swing miss/hit in 1:1 ratio killing 500 of the enemy (the deaths occur only at end of round). That leaves 500 unpaired troops from larger army. These can hit without miss for 500 more kills, killing the total 1000 of the smaller army (500 deaths from 1000 matched troops at 50% efficiency and 500 more deaths from remaining 500 unmatched troops at 100% efficiency). In the range of 1000 to 1500: the leftovers above 1000 kill that many of the weaker army but will leave some alive for round 2 (ie, 500+N, where Large army - Small army = N < 500, totals less than 1000, the army size of smaller army, so some troops of smaller army remain for round 2).
With this algorithm you approximate everything on your table at least in the 1100 to 1500 range, in two rounds. The closer army sizes require 3 rounds or more. Quick rough 3rd example. Say 1200 vs 1000. The damage in round 1 from the 1000 paired off troops leads to 500 deaths from each side. Further, the remaining unmatched 200 troops from larger army don't miss at all so kill 200 more. Round 2 starts with larger army having lost 500, now having 700 alive and smaller army having lost 700, now having 300 alive. In round 2, 100 more are killed by the 300 matched troops (1:2 ratio of hit:miss). The larger army has 400 unpaired troops, inflicting 400 more kills on weaker army, demolishing what was left since 400+100>300. Meanwhile, the larger army went from 700 down to 600 in round 2. Round 3 won't be reached. To reach round 3, the sizes have to be closer so that the unmatched troops after round 2 aren't enough to finish off what remains of the small army. I don't know the fatigue ratio for rounds greater than 2. I didn't try to pluck it from your data since there is noise in the data (thanks, plarium and small sizes used in the data). I think 1/3 also should work though but not a big deal to know for most scenarios since most damage is done in first 2 rounds. Also, by round 3, perhaps it's important to understand the table/detail of 52% vs simplified 50%.
In next comment below this one (this comment is large enough already), I'll mention the total damage done per round. It's related closely to the discussion you have on healthpool. Again, I don't know if details are correct, but you'll probably/maybe agree it makes sense.
Upon further review, the play is called back.. somewhat.
OK, use 65% as the hit ratio for everyone for round 1 and round 2 and every other round. Also ignore the idea that unmatched troops hit at 100%.
Let's look at two examples.
Example 1, 1538 vs 1000 (saturation boundary):
1538 * 65% = 1000.
Since this kills off everyone on other side, we are done, and
Then give back 20% to give 800 deaths.
On the other side, 1000*65% = 650
Since we know the battle ends after this round,
Then give back 20% to give 520 deaths.
This matches the table, 800 deaths to 520 deaths.
Example 2, 1200 vs 1000.
Round 1:
1200 * 65% = 780.
We need round 2, and there are 1000-780=220 troops left on other side.
On other side, 1000 * 65% = 650
We need round 2, and there are 1200 - 650 = 550 troops left on stronger side.
Round 2, 550 vs 220:
550 * 65% = 358
No more rounds needed since other side had only 220 troops, so all are dead, all 1000.
Give back 20% of 1000 deaths = 800 deaths.
On other side, 220 * 65% = 143
No more rounds needed we already know, and 550-143=407 troops left on stronger side, implying 793 deaths out of 1200.
Give back 20% of 793 deaths = 634 deaths.
These numbers of deaths, 800 and 634 match what your table shows on the 1200 vs 1000 case (well, 635 but I rounded along the way instead of at the end so got slightly different 634).
I adjusted the algorithm to use 65%. It can generate values, but I don't have resources to test them. Do you have more data? Specifically, atm, vanilla troops (one type against one type only) in any quantities. The alg can give the final troop losses for each side. The alg can handle nonvanilla (ie, tier 4 is handled as if just tier 1 with 300 attack,defense,health), so if you have a detail of the net attack,defense,health (including spells, etc), I can test that as well I guess.
At 15:40 I get 496 defender deaths not 492 for the first entry. To be clear, I am reading your table as follows: attack 102, defense 103.75 , health 103.75 for **both** attacker and defender. Also 1000 vs 5000 and no other adjustments. Is that correct? [BTW, if I round upward the 103.75 to 104 in all cases, I get 495 which is a little closer but still not exact.]
The algorithm generated the exact same death count as you showed in the saturation table (9:10) for the several lines I tested (1001,1050,1100,1200,1300,1400,1500,1538). This was a T1 vs T1 vanilla case. While that table gave exact matches (once I switched over to the 65% efficiency modification), I have not gotten this match in the table at 15:40 yet. I am not sure I understand the set up in this last table for each attacker and defender or else the algorithm or something is a bit off.
A brief description of the algorithm follows. It only handles one troop type vs one other troop type. I don't use healthpool formula you mentioned but use something similar. I calculate total health for each team (health of troop times number of troops). The defense is used elsewhere though. I have a net attack value defined as the attack divided by the defense of the enemy times 100. The times 100 is to get the units correct (since 100 health is used as the base value instead of 1 health). The other part is a ratio of attack over defense. That's how defense helps, it reduces the attack on you by that factor (with the 100 multiplier added in). This boils down to the healthpool you mentioned, but it's in a more broken down form used in cycles of the algorithm to calculate "exact" death counts for each side, presumably no matter the stats of each side (that's the hope).
OK, basic algorithm (which is done in javascript but I won't get that detailed here):
There is an outer and inner loop. The outer is over as many rounds as it takes. You break out of that whenever the healthpool of either side reaches or goes below 0 in a test done at the end of each round. That event would signal the last round and the end of combat (with one or both side's troops all dead). The inner loop (comprising one round) iterates over troop count (infinite loop you break out of when both sides reach the number of troops they started at in that round). During each cycle you calculate the attack by each team on the other. If the team is still able to attack, you do an attack (65%*(yourattack/enemydefense)*100) and hold that accumulated sum until the round ends and then subtract that from enemy healthpool. If exactly one team has run out of troops while the round is still running, it effectively adds 0 to its accumulated attack damage. Each cycle increments the troop counter. To repeat, at the end of the inner cycle, *after* every troop from both sides has attacked using the formula above, then for each team you take the accumulated damage to enemy healthpools and subtract from the actual enemy health pool to start the next round. This healthpool value defines the number of troops for that new round (just divide by health of one troop to get the count). With this approach of subtracting the damage at the *very end* of the round, even if the small army gets blasted at the end of the round, it still got all of its troops to attack during the round (this is seen easily in the saturation effect, where adding more to the large army doesn't impact the damage they take). At the end of combat take 80% of the deaths of each side and that is the final answer.
Summarizing, this algorithm, a loop over rounds, each round a loop over troop attacks on healthpool, uses a formula similar to the healthpool formula you gave and is configured for attack, defense, and health of each side. Each side can only have one troop type. And the answer is the death counts if you play those size/attribute armies against each other. I tested and verified the exact values you gave in one of your tables (for T1 vs T1 at minute 9). I also verified some random T1 vs T4 examples you gave. What it has not done precisely is give the table at minute 15 of the video. I don't know if I understand the set up for that one. Still working on it...
[Thanks for the video since I had been wanting to make progress with this algorithm for a while but was lacking larger quantity of data to complement my tests and give extra insight details and confirmation, which this video provided. Hope in return, the algorithm described is found interesting. It's all pretty simple. The key details are that there are rounds, that the attack calculated for a round (for each side) is accumulated and doesn't go into affect until the end of the round, and that net attack is precisely 65% of yourattack/enemydefense*100, done onto the total enemy's health (similar to healthpool), and this all seems to work no matter the troop type and attributes.]
[Clarification: this handles any troop type but against that same troop type (but they can be different tiers against each other). Using different troop types requires adding in a bonus/penalty. Mixing troop types and having more than one type for an attack is next on the list once I can be more sure that the algorithm works for the easier cases.]
godmode
In the table at minute 15 and in general in other parts of the game, different applications of percent (boosts, gain, etc) tend to group into more than one group. This is likely why 50% enemy reduction from skills and 30% reduction from a different source would not equal 80% net reduction. For example, let's say that the base is 100. Applying 50% takes it to 50. Applying 30% reduction on 50 gives 35. Having a single group (80%) vs two groups (50% and 30%) is the difference between (100)*(1.00-0.50)*(1.00-0.30) = 35 vs (100)*(1.00-0.50-0.30)=20. In one case they add up and then multiply. In the other they multiply separately. This happens in many other areas. For example, all weapons/armor combine additively with skills and knowledge when it comes to things like time boosts. The total is then applied multiplicatively. Meanwhile, VIP combines separately with other things when that applies (eg, in production speedups). And iirc the daily 10% boosts and hourly "spells" and townskin all combine into a group. [or something like that]. This is why adding more skill points or weapons percent when you already have a lot of those doesn't contribute much as adding a simple 10% time boost. If you want a big effect, add more % gains to a group that has a small value. However, also contrast with things that increase speed vs things that reduce time. Increase speed is based upwards from 100%. Eg, 500% would allow you to do 6 total times the original amount of work in the same amount of time. That is you are increasing the rate (increasing the work done per unit time). Alternatively an -80% time reduction boost would achieve the same effect as that 500% rate boost. The 80 is a much smaller number but applied differently mathematically.
In fact the "god mode" might have been because of something like the above, eg, 53% enemy reduction (skill) and 45% enemy reduction (spell) if combined additive before applying as a reduction, would give a 98% reduction, effectively reducing the enemy's level by a factor of 50 (ie, to 2% from 100%)! If they combine separately but also as a reduction, then you reduce effectively by multiplying by a factor of .5*.55=.275=27.5% which is approximately a reduction by a factor of 4. Compare 4 vs 50! Finally, if there were to apply inversely (eg, rate and time examples above, where one increases above 100% and then you divide by that as opposed to just multiply from the onset and move towards 0), then it would be even less effective. Eg, even as a whole (98%), it that might lead to dividing by 1.98 which is just a factor reduction of about 2.
godmode: math subtleties.
a) There is the effect of combining into one vs different groups before applying.
b) There is the effect of multiplying directly vs directly but after subtraction from 100%.
c) There is the effect of multiplying inversely or directly. (eg, when talking about rates, work done (distance covered, food make, etc) are inverses with time)
Let's take 60% and 40% and consider the possible ways to affect 100.
a)
i) Here we could be talking about the difference between applying as 60% and then as 35% (or vice-versa if we are just multiplying since multiplication is commutative.. and also associative). 100 * 0.6 * 0.35 = 0.21
ii) In contrast we could be talking about applying as one unit (60%+35%=95%). 100 * 0.95 = 0.95
b)
i) An example of the case where we first subtract from 100% is if we are talking about a reduction. 60% reduction is the same as multipying by 40%. We take 100% and reduce it by 60% to get 40%. 100 reduced by 60% = 100 * 0.4 = 40.
ii) In contrast, other uses of percents are applied directly, simply times 60%. 100 * 0.6 = 60.
c)
i) Here consider boost that reduces the amount of time something takes by 60% (to 40% of the whole). In this case, we multiply the total time by 0.40. If something took 10 hrs, it would now take 4 hrs.
ii) In contrast, a 60% improvement in the speed your town produces food would lead to the new time being calculated differently. We increase above 100% by 60% to get 160%, aka, 1.6. That is what it means to improve the rate of something by 60%. Then that value affects the total time inversely, eg, if we could produce 100 units of food in 1 hour, now we can produce 160 units in that same hour. Notice we are talking about increasing food production (aka, that rate) by a certain amount instead of talking about reducing the time by that amount. Taking 100 units/hr as the base rate, if we have only 1000 units to produce, the time is cut from 1000/100 = 10 hrs to 1000/160 = 6.25 hrs. This is different from the 4 hrs if we had applied the 60% as a reduction to time instead of as an increase to the rate.
As a special case, let's consider 60% and 40%. In a godmode, we could combine these to get 100% and then apply that as a reduction to time (or to strength or anything else), reducing the time (or strength, etc) to 0. Obviously, if you find yourself doing something like that, reducing something close to 0, you know you are probably misapplying the percent.. or plarium goofed. Or exploit exploit exploit for glory in the game!
little goof. I said near the top to consider 60% and 40%, but most of the examples used only 60% (40% being incidental since 100-60=40) and 35% in the case where we were considering 2 groups vs 1. At the bottom, I do 60 and 40 however. 60 and 35 would have given 95% not 100%.
Subbed
Then what is the point of «attack during defense» knowledge ?
Interesting video nonetheless.
There is offense, offense during defense and offense during attack.
See, defense is the opposite of attack. If someone attacks you, you have to defend and your offense during defense will be used to compare with the defense during attack of your enemy.
If you have a choice between 1% increase in offense, 1% in increase in offensive during defense and 1% increase in offensive during attack, you always prefer 1% in offense cause that's equal to 1% increase in offense during defense together with 1% increase in offense during attack.
Question 35000 vs 4000 .. the 35 k were fenrir rank 5 vs 4000 rank 1 poly. And 4000 was minimal and 35000 limp away with maybe 2k unwounded or killed .... How is that even possible
if you are still playing the game please contact me. I have the true theory that i based on your theory, but you are missing some very key things.
Hey, let's talk
Me also I have things I would like to apply also do you have contact information or line app mitchell00
Me too. Do what email can i contact you at?
Якщо коротко: марення сивої кобили. Ідея цікава. А ось цифри підганяються, щоб отимати необхідне. Імперічне усе - логічно не повязане. Намішав кашу з вінегретом, атоді підгонка до неповязаного. І, вуаля - цифри начебто сходяться....
Before the T6 update. T1's were defensively dominating regardless the size or stats of incoming attacks. 10m T1's with an average stat will hold against a massive OS. After the update, that shit no longer exist.
Yes and no. Even before the t6 update the t1s were somewhat on their way out. The reason for this is because of the new knowledge and march levels. They are allowing bigger guys to cut through way more t1s than before. On top of that a big onslaught is much easier to make with bigger Mead sizes and 150% enlargements.
But t1 masses still work. You just need a lot more than before. 10mil or whatever wont work. You need like 30-50.
I agree, but before the update I've been dominating against massive OS multiple times with just 10m T1's and 500k T5's. Also, had an additional 6M+ rein by clansmen. Maybe the opposing skills were shit, though it's ashame we can't see all participants stats other than the one who is hosting.
After the update, I was still convinced my wall will hold strong. Fuckers blast right thru it with ease lol. Now, I definitely need 30-50.
They basically just screwed us with the massive new marches they can send out. Train now t1 at least 20 mill but behind them have t4 and t5 at least 5 mill each, use t1 just as disposal if you ever face a monster sending a full t6 at you. I use siege only so I'm cool cos they all focus on mele it's easier to do finish it and get the t6s ready for action.
One more piece of advice if you have a huge army, don't fuck around with double hero, have it always on when going on multiple tile hits at start of kvk or cvc be in your lynx gear, farmers will be murdered anyhow. I can hit in 5 seconds but there are ppl out there who hit in two you won't even be able to tap items for shield even if you shield after each attack momentarily.
Mrclouder+
I agree, that's my strategy when hitting tiles/towns. I leave 1 hero home and send second for battle. Speaking of T6 those boys are fucken deadly especially the high skills. I've seen some reports of massive OS 9M against 3M T6 and lost. They are the NEW god mode
do not spoil the people
When I see a spread sheet it’s an automatic pause, comment and close out. No bueno!