The wiki has been partially updated since I recorded this video. Here's the link: hoi4.paradoxwikis.com/Attrition_and_accidents It gives a table of some breakpoints for reliability but not the chart I showed. This video was kind of hell to record because of the changes in the math needed that were discovered half way through. Hopefully its still coherent.
I do understand that maintenance companies are not worth for ONLY RELIABILITY stats, but equipment captured essentially is almost equal to that being destroyed without any support company . So it is definitely worth in divisions which you expect to be involved in lots of combat or ones with 'pricey' equipment.
I really like how you're starting to state all the things you've done in the background. (like the supply depots/maxed railways, etc - I noticed you also had enough fuel, and trucks/trains, for instance) It shows you're getting better at anticipating your audience's nitpicky questions! :) Also, it's sorta a compliment that your viewers are smart enough to think through the options/meaning, and are able/think to _ask_ those questions :) Anyhow, love seeing you grow and develop this channel! ^^
I'm the person who added to the wiki the part about max useful reliability per number of equipment. All I did was transform the formula for minimum losses and plugged in typical numbers of equipment. Thank you for doing actual research and figuring this out.
>"maintenance companies are not that useful" First my recon companies were revealed to be useless, now my maintenence companies are also useless? noooooo
maintenance is useful not for the reliability, but for the equipment capture rate. i use it when i have a weak economy, in a defensive situation, e.g. Finland
cav or motorized recon is useful on portguards for the extra defense. armored recon is good for the breakthrough, armor and movement bonusses on the offense.
Well in their defense we can say that the dev who *actually* program the game and the one who write those wiki pages might not be the same people. Or that they just suck at explaining, who know.
The reliability seems to work as described on the HOI4 wiki, i.e. there's always a minimum absolute rate of attrition for a given equipment, whatever the number of pieces of said equipment in the division, which is kinda dumb. Also, the test does have a couple issues : 1) The departure and arrival provinces aren't the same for all divisions, so this creates different effective attrition durations if all divisions move at the same speed. 2) The armored divisions have slightly different speeds depending on the tank design ; to have an identical speed, it would have been necessary to replace the motorized infantry with regular infantry instead. Though I doubt these issues were significant in the test. I don't know if it's gonna be examined in the future video about maintenance companies, but reliability does raise the number of recovered equipment. IIRC, the percentage of equipment lost in combat that is subsequently recovered is, on average, the reliability of the lost equipment squared, and then divided by 10. So let's say you lost 100 tanks of a specific model in a battle ; with 50% reliability, you'd recover 2.5% of those losses, or 2.5 tanks on average. With 100% reliability, that would jump to 10 tanks. It gets even weirder when you realize you can recover/capture aircraft on CAS mission that have been shot down, whether your own, enemy aircraft that were shot down, or allied aircraft that were in a land battle of yours. Even weirder is I got screenshots were the number of recovered equipment exceeded the losses, such as this screengrab, that also showcases recovered allied aircraft (I don't use Wellingtons as the U.S.) : steamcommunity.com/sharedfiles/filedetails/?id=2724657463 This one shows a recovered CAS aircraft, a Typhoon, which I don't use as I only use tactical bombers for the job, so it's clearly from the allied AI : steamcommunity.com/sharedfiles/filedetails/?id=2724657425
Point 1. Yes it was. I moved them all to Nice and then through the mountains. I had every division selected when I was clicking through the mountains. They took the same path. Point 2. This assumes I upgraded the engine at any point in time which I only did for 50%, and that was a few points to get them to 50%, none of the rest had any engine upgrades. 3. You assume recovery actually works as it is described in the tooltip. That is false. I have recovered more equipment in a battle than what I started with before. In a battle where I should have recovered 9% I actually recovered over 100%.
@@71Cloak 3. Is the 9% expected to be recovered an expected value in the mathematical sense ? Could Paradox have coded stuff oddly so we can sometimes more than we lost, but it somehow averages out to 9% ?
Great video as always and very helpful to understand how the game works. Keep it up with this kind of great content please! Also interesting that light tank support companies can slow down your entire division. It means that light flame tanks should have at least the speed of the slowest tank in the division.
Wow. I've actually always just assumed 80% for all tanks. If a tank has less than 80% reliability, I don't save the design, because I don't want to have to replace them frequently. Looks like I was very wrong. Lol
Well, it IS all relative. As he mentioned, if you aren't expecting to suffer from attrition (not training or fighting in bad terrain), then it doesn't matter for out-of-combat losses. It would still matter for post-combat recovery (although THAT only matters for battles you win, so it's not as simple as a straight % reduction in losses), and for org recovery rate. Similar to air reliability, you could have scenarios where your combat losses exceed your reliability losses by an order of magnitude. In that case, you would have to try to take into account the value of the higher post-combat recovery versus increasing breakthrough/attack/defence to lower your base level of combat losses. It'll probably take some large-scale testing to determine how impactful it is overall.
Hey, cloak I had an idea that I tested out making training tanks. Assuming you are using mediums you can use the chassis after inter-war that has 100% reliability with a support gun to train. They only cost 4.0 IC for each one. If you made 5000 of them (20000IC) and trained 10 tank divisions at a time it could save IC in the long run. The test I did used 18.8 IC medium tanks and if you trained about 15 tank divisions you would save IC by training them to full then swapping to the non-training tanks (you don't lose any experience). You could also be producing these at the start of the game too.
I thought that if I add 100% reliability flame tank to my infantry division, it will increase weapons and equipment reliability, like the tooltip said. How wrong was I.
This might be a stupid question that you answered in a different video, but does reliability affect combat losses/stats or reinforcement rate/org at all? Loving these videos, thanks
He has another video which addresses this which you can search for. Reliability has zero impact on any combat statistics, 0% and 100% reliability tanks will perform exactly the same in a battle. However, after the battle you will recover a portion of your lost equipment. I believe at base it is 30% of losses are recovered + bonus recovery for any reliability over 0%. You will also have faster org recovery based on your average equipment reliability for the entire division. For a theoretical 100% reliability division it has no effect, down to something like -20% recovery for a division with 0% average reliability.
Are any of the support companies other than engineer battalions worth it? I mean logistics are ok if you need the little boost, but it seems all of the other ones are so situational or not working as intended so it’s not worth the tech investment
Art for more soft attack. AA if your operating under enemy air superiority and with the added bonus of some piercing capabilities. Flame tanks for the attack bonuses.
Engineers are SO worth it for the entrenchment bonus, especially for infantry that's just meant to hold the line. Also, Arty for soft attack, and AA for piercing (against the AI, at least - although I don't even bother) I also really like armored recon for a bit extra breakthrough and armor. And motorized recon gives a very nice edge for defense! However, I've never rigorously tested the above - I've just looked at the stats in the division designer. (I also haven't bought NSB yet, because I refuse to pay 20 euros for a game that's started to randomly crash _now,_ whereas I _never_ had that issue for the first year or two)
@@mainman879 MP as a support company gives a suprising amount of defence (34 at lvl 4) whilst engineer gives 37 (lvl 4) But MP doesnt give the terrain buffs engineer offers.
According to the wiki "Attrition is the loss of equipment [...] when fighting on the offensive." But you lost equipment to attrition when simply moving your troops around without attacking. So is the wiki wrong?
Hello, I made a tank division that had 100% reliablity which did not lose any tanks for a long time combat after combat unless I fuck up like getting into a bad battle or having low supply because that division was glass cannon as it had high harness and good breakthrough but low org and hp. I've lost a couple hundereds to combat until losing a couple to attrition. So my question is rather than going for low reliablity support companies as they suffer less losses due to their low numbers, how feasible is it to go 100% reliablity for them just to boost the average reliablity of the division for maybe other equipment like guns and support companies which has less than 100% reliablity suffer less loesses.
Reliability of one equipment type does not impact the reliability of other equipment types. The only time you are going to lose equipment to attrition is when you are taking attrition. If you completely avoided attrition at all times then you would lose the same amount of tanks with 0% reliability as you would with 100% reliability. With all equipment types regardless of amount there is a base amount of attrition that is unavoidable regardless of reliability. The smaller the amount of equipment used the lower the reliability needs to be to minimize equipment losses. At 24 tanks for a tank recon support company, you only need 16.67% reliability to minimize the losses to attrition.
@@71Cloak I tested a 10 width tank division with a recon tank. Everything had 100% reliablity. I don't think it was possible to not suffer from attrition for some time as I had more than 400 tanks lost to combat but only 1 to attrition. But I guess your statement is also correct because I made a division with 91% relaiablity and added a maintanence 2 to have the aberage realiablity to up to 100% and still lost condireable equipment.
You've messed up your test. The attrition is based on the Av. Reliability, not on the reliability of a singular piece of item, or equipment. For instance at 3:45 you can see that your supposed "75%" reliability division actually has a reliability of 82,1%. At 0:47 you can see that a supposed 100% reliability division has an actual reliability of 91,0%. At 0:01 even the supposed 0% reliability division has an actual reliability of 81,0%. Your divisions also have a widely different supply use, which also affects attrition.
Literally nothing you just said is accurate. Attrition is based on actual reliability of the equipment not average reliability of the division. Feedback is wrong on this and he repeats it constantly.
The wiki has been partially updated since I recorded this video. Here's the link: hoi4.paradoxwikis.com/Attrition_and_accidents
It gives a table of some breakpoints for reliability but not the chart I showed.
This video was kind of hell to record because of the changes in the math needed that were discovered half way through. Hopefully its still coherent.
I do understand that maintenance companies are not worth for ONLY RELIABILITY stats, but equipment captured essentially is almost equal to that being destroyed without any support company . So it is definitely worth in divisions which you expect to be involved in lots of combat or ones with 'pricey' equipment.
What about infantry guns? does reliability matter?
Once again we're finding out that HOI4 reliability is perfectly balanced as all things should be.
XD good old spiff co
My tank waiting to break down the last 5 minutes of a battle
Community: "How did we learn the meta for HoI4?"
71Cloak: "Well, ya got me. By all accounts, it doesn't make sense"
I really like how you're starting to state all the things you've done in the background.
(like the supply depots/maxed railways, etc - I noticed you also had enough fuel, and trucks/trains, for instance)
It shows you're getting better at anticipating your audience's nitpicky questions! :)
Also, it's sorta a compliment that your viewers are smart enough to think through the options/meaning, and are able/think to _ask_ those questions :)
Anyhow, love seeing you grow and develop this channel! ^^
It's a very precise theme, the perfect reliability for light tanks of Hoi4, nitpickyness is of essence if you want to get it right
I'm the person who added to the wiki the part about max useful reliability per number of equipment. All I did was transform the formula for minimum losses and plugged in typical numbers of equipment.
Thank you for doing actual research and figuring this out.
>"maintenance companies are not that useful"
First my recon companies were revealed to be useless, now my maintenence companies are also useless? noooooo
sucks feel the same first signal then recon now maintenance
maintenance is useful not for the reliability, but for the equipment capture rate. i use it when i have a weak economy, in a defensive situation, e.g. Finland
@@dermajor4472 signal gives you coordination, it's something. But the question is if you really want it in the first place.
cav or motorized recon is useful on portguards for the extra defense. armored recon is good for the breakthrough, armor and movement bonusses on the offense.
@@melonenlord6901 recon is never worth it, it's bad designed and is only for countering tactics. Armored recon is good against AI
It's both funny and disturbing that the devs know nothing about their game
that's paradox for you
i feel the same way, but more on the disturbing side, about ck3
@@Bleilock1 Vicky 2 has artisans that turn fruit into tanks. The devs of it have no idea why.
@@TheTripleAce3 Maybe they use tractors
Well in their defense we can say that the dev who *actually* program the game and the one who write those wiki pages might not be the same people. Or that they just suck at explaining, who know.
Babe wake up 71Cloak made a video
Good to know that my meta of adding maintainence to expensive units is completely useless.
I love how great this game is but under the surface it barely works intended
Your channel is the best thing to happen to a paradox game since remans paradox
Very, very helpful! Thank you!
The reliability seems to work as described on the HOI4 wiki, i.e. there's always a minimum absolute rate of attrition for a given equipment, whatever the number of pieces of said equipment in the division, which is kinda dumb.
Also, the test does have a couple issues :
1) The departure and arrival provinces aren't the same for all divisions, so this creates different effective attrition durations if all divisions move at the same speed.
2) The armored divisions have slightly different speeds depending on the tank design ; to have an identical speed, it would have been necessary to replace the motorized infantry with regular infantry instead.
Though I doubt these issues were significant in the test.
I don't know if it's gonna be examined in the future video about maintenance companies, but reliability does raise the number of recovered equipment. IIRC, the percentage of equipment lost in combat that is subsequently recovered is, on average, the reliability of the lost equipment squared, and then divided by 10. So let's say you lost 100 tanks of a specific model in a battle ; with 50% reliability, you'd recover 2.5% of those losses, or 2.5 tanks on average. With 100% reliability, that would jump to 10 tanks.
It gets even weirder when you realize you can recover/capture aircraft on CAS mission that have been shot down, whether your own, enemy aircraft that were shot down, or allied aircraft that were in a land battle of yours. Even weirder is I got screenshots were the number of recovered equipment exceeded the losses, such as this screengrab, that also showcases recovered allied aircraft (I don't use Wellingtons as the U.S.) : steamcommunity.com/sharedfiles/filedetails/?id=2724657463
This one shows a recovered CAS aircraft, a Typhoon, which I don't use as I only use tactical bombers for the job, so it's clearly from the allied AI : steamcommunity.com/sharedfiles/filedetails/?id=2724657425
Point 1. Yes it was. I moved them all to Nice and then through the mountains. I had every division selected when I was clicking through the mountains. They took the same path.
Point 2. This assumes I upgraded the engine at any point in time which I only did for 50%, and that was a few points to get them to 50%, none of the rest had any engine upgrades.
3. You assume recovery actually works as it is described in the tooltip. That is false. I have recovered more equipment in a battle than what I started with before. In a battle where I should have recovered 9% I actually recovered over 100%.
Before NSB, you could already capture ennemy cas using maintenance company.
@@71Cloak 3. Is the 9% expected to be recovered an expected value in the mathematical sense ? Could Paradox have coded stuff oddly so we can sometimes more than we lost, but it somehow averages out to 9% ?
Great video as always and very helpful to understand how the game works. Keep it up with this kind of great content please!
Also interesting that light tank support companies can slow down your entire division. It means that light flame tanks should have at least the speed of the slowest tank in the division.
Wow. I've actually always just assumed 80% for all tanks. If a tank has less than 80% reliability, I don't save the design, because I don't want to have to replace them frequently.
Looks like I was very wrong. Lol
Well, it IS all relative. As he mentioned, if you aren't expecting to suffer from attrition (not training or fighting in bad terrain), then it doesn't matter for out-of-combat losses. It would still matter for post-combat recovery (although THAT only matters for battles you win, so it's not as simple as a straight % reduction in losses), and for org recovery rate. Similar to air reliability, you could have scenarios where your combat losses exceed your reliability losses by an order of magnitude. In that case, you would have to try to take into account the value of the higher post-combat recovery versus increasing breakthrough/attack/defence to lower your base level of combat losses. It'll probably take some large-scale testing to determine how impactful it is overall.
Hey, cloak I had an idea that I tested out making training tanks. Assuming you are using mediums you can use the chassis after inter-war that has 100% reliability with a support gun to train. They only cost 4.0 IC for each one. If you made 5000 of them (20000IC) and trained 10 tank divisions at a time it could save IC in the long run. The test I did used 18.8 IC medium tanks and if you trained about 15 tank divisions you would save IC by training them to full then swapping to the non-training tanks (you don't lose any experience). You could also be producing these at the start of the game too.
This channel needs a million subscribers and any of these videos need +10MM views.
Oh fuck, I didn't know this. Maybe I need to rethink the 80% heavys I made
I thought that if I add 100% reliability flame tank to my infantry division, it will increase weapons and equipment reliability, like the tooltip said.
How wrong was I.
A vídeo about companys in general would be nice (if they actually work as planned etc)
waiting the maintenance company video. thank you a lot.
some of them take a longer path through the mountain
Too bad I am an OCD person regarding reliability I HAVE TO GET 100 %
Holy shit, I always limit my reliability around 50-60%, I now see how wrong I was.
Cloak back again with tha facts
> i had to use the petrol electric engine and nobody uses that
You underestimate my larp
This might be a stupid question that you answered in a different video, but does reliability affect combat losses/stats or reinforcement rate/org at all? Loving these videos, thanks
He has another video which addresses this which you can search for. Reliability has zero impact on any combat statistics, 0% and 100% reliability tanks will perform exactly the same in a battle. However, after the battle you will recover a portion of your lost equipment. I believe at base it is 30% of losses are recovered + bonus recovery for any reliability over 0%. You will also have faster org recovery based on your average equipment reliability for the entire division. For a theoretical 100% reliability division it has no effect, down to something like -20% recovery for a division with 0% average reliability.
@@Strait_Raider cheers, that really helps thanks :)
Awesome again!!!
Are any of the support companies other than engineer battalions worth it? I mean logistics are ok if you need the little boost, but it seems all of the other ones are so situational or not working as intended so it’s not worth the tech investment
Art for more soft attack. AA if your operating under enemy air superiority and with the added bonus of some piercing capabilities. Flame tanks for the attack bonuses.
Engineers are SO worth it for the entrenchment bonus, especially for infantry that's just meant to hold the line.
Also, Arty for soft attack, and AA for piercing (against the AI, at least - although I don't even bother)
I also really like armored recon for a bit extra breakthrough and armor.
And motorized recon gives a very nice edge for defense!
However, I've never rigorously tested the above - I've just looked at the stats in the division designer.
(I also haven't bought NSB yet, because I refuse to pay 20 euros for a game that's started to randomly crash _now,_ whereas I _never_ had that issue for the first year or two)
Armored recon on infantry turns them into space Marines
Artillery, anti air, logistics, flame tank supports, armored recon
Engineers dont even have a slot
Flame support is best for offensive division
Very interesting, good video
Can you test this with maintenance companies, I have no idea if they do something
i like percentages so i always use maintanence companies please do a video on that.
This is what that Soviet general meant when he said the German tanks were too reliable lol
Really interesting. But what is about the aircraft?
I dont need reliability on my interwar light flamethrower tank when i can build like a million per day and only care about the attack buffs
We should also assume that this is also true for medium and heavy tanks?
Tanks are tanks. The only difference is light tanks use slightly more per battalion.
here's a potentially fruitless challenge: build a (defensive) division that maximises defence with MPs and see how well it fares
MP only gives suppression bonuses not defense bonuses. Why would you ever want to put it into a fighting division?
@@mainman879 MP as a support company gives a suprising amount of defence (34 at lvl 4) whilst engineer gives 37 (lvl 4) But MP doesnt give the terrain buffs engineer offers.
According to the wiki "Attrition is the loss of equipment [...] when fighting on the offensive." But you lost equipment to attrition when simply moving your troops around without attacking. So is the wiki wrong?
Hello, I made a tank division that had 100% reliablity which did not lose any tanks for a long time combat after combat unless I fuck up like getting into a bad battle or having low supply because that division was glass cannon as it had high harness and good breakthrough but low org and hp. I've lost a couple hundereds to combat until losing a couple to attrition.
So my question is rather than going for low reliablity support companies as they suffer less losses due to their low numbers, how feasible is it to go 100% reliablity for them just to boost the average reliablity of the division for maybe other equipment like guns and support companies which has less than 100% reliablity suffer less loesses.
Reliability of one equipment type does not impact the reliability of other equipment types. The only time you are going to lose equipment to attrition is when you are taking attrition. If you completely avoided attrition at all times then you would lose the same amount of tanks with 0% reliability as you would with 100% reliability.
With all equipment types regardless of amount there is a base amount of attrition that is unavoidable regardless of reliability. The smaller the amount of equipment used the lower the reliability needs to be to minimize equipment losses. At 24 tanks for a tank recon support company, you only need 16.67% reliability to minimize the losses to attrition.
@@71Cloak I tested a 10 width tank division with a recon tank. Everything had 100% reliablity. I don't think it was possible to not suffer from attrition for some time as I had more than 400 tanks lost to combat but only 1 to attrition.
But I guess your statement is also correct because I made a division with 91% relaiablity and added a maintanence 2 to have the aberage realiablity to up to 100% and still lost condireable equipment.
You've messed up your test.
The attrition is based on the Av. Reliability, not on the reliability of a singular piece of item, or equipment.
For instance at 3:45 you can see that your supposed "75%" reliability division actually has a reliability of 82,1%.
At 0:47 you can see that a supposed 100% reliability division has an actual reliability of 91,0%.
At 0:01 even the supposed 0% reliability division has an actual reliability of 81,0%.
Your divisions also have a widely different supply use, which also affects attrition.
Devs themselves said the average reliability is not used for attrition calculation. Instead each equipment type is evaluated separately.
Literally nothing you just said is accurate. Attrition is based on actual reliability of the equipment not average reliability of the division. Feedback is wrong on this and he repeats it constantly.
@@PadaV4 But isn't that inconsistant with your test results?
@@Caldera01 i think you wanted to reply to 71Cloak not to me
@@PadaV4 Yes
How about making precisely the same design tank but saving it as a different one, this will require you to have different production lines tough.
So you're telling me that making my tanks have a minimum of 80% reliability and throwing maintenance for the 20% boost was a complete waste? REEEEEEE
New mic?
Just messing around with the settings to try and prevent peaking while not being too quiet. Didn't work too well.
@@71Cloak I liked the new settings, sounded better imo.
@@sakiqaci3746 👍
Try if reliability matters in combat
i love this videos but make me more mad about HOI4
so is realiabilitu broken? it doesnt matter at all?