EchelonIII

Trendlines of Various Tanks: Player win rate vs Win rate of tank

245 posts in this topic

That's sounds about right.

 

I think I am hurting myself by grinding all the German tree at once and not platooning with better players at all, but that's another topic.

Share this post


Link to post
Share on other sites

I wonder if using information such as these charts, and more data of corse, a better way would be to change MM weights to balance tanks. A tank would still see the same tier spread, but just get a different mix of matches, more of higher tiers in its spread if it over performs, less if it's a proven underperformer.

The tank itself doesn't change, thus heading off player complaints. Players would still complain about their Hellcat vehicle seeing too many T8 matches, but then their M60 would see less higher tier matches, so the effect on their stats would be minimal. This is already done for vehicles such as the Tiger, which sees much higher tier vehicles TIRL to balance it. Since it was built in early for that vehicle, it was accepted and the complaining about it produced a minor tweak that seemed to work for everyone, so the complaints died off.

This might also improve the MM, making the matches more balanced. You might have to wait a little longer, but I have rarely waited more than 30 seconds for a match.

It would be a shame for such useful information to not be used. Serb stated that 2014 the focus will be on game engine, maybe this is a good way to proceed.

Share this post


Link to post
Share on other sites

their M60 would see less higher tier matches

...Wat.

Share this post


Link to post
Share on other sites

My argument is that the Su-122 is OP, but that's simply WG's shit ability to balance the game, not a case of pay to win.

 

Looking at those charts I'd guess that armour helps players < 55%... mobility helps players > 55% (and of course the whole tank WR line shifs up and down based on other criteria, DPM?)

 

I don't know how WG can keep up the pretense of historical accuracy and keeping tank diversity w/o making everything balanced clones (ie E75 ,etc)

Share this post


Link to post
Share on other sites

...Wat.

 

 

I would guess that he means that the M60 would see more matches where it would be the only or only 1 of 2 tier X's on the team? But I am not sure if WGNA would be able to code what he is suggesting....

Share this post


Link to post
Share on other sites

Correct.  But the M60 was a poor example since it is a premium T10 tank.

 

Lets pick the Hellcat, it seems to do well.  Its current spread in T6-8, and lets assume the MM will put it equally in those tiers.  So instead of %33/%34%/33 split in those tiers maybe change that to maybe %20 T6, %50 T7, and then %30 T8.  You see less T6 where the vehicle is very good for that Tier, and a little less T8, where it is a good support vehicle.

 

The vehicles themselves are fairly historically accurate, the Hellcat did have that armor and was that fast and did have those guns and could hide in an open field (wait...)  so WOT attempts to balance tanks by tiering them.  If the Hellcat was moved to T7-8 only, I am not sure it would do so well and it would also hurt players who like playing it. This way though they would get some matches that they can feel strong in, but in most matches it will test how good they are as opposed to the vehicle.

 

Weaker vehicles go the other way of course.

 

Maybe for platoons, say 2 Hellcats, that would increase the chance of being in a T7-8 match.  Platoons maybe should get into a little tougher matches, I saw a 3 man Excel platoon shooting all gold that destroyed a T6 match they were in.

 

Vehicles have MM weights by class now, you would just have to adjust those for each vehicle a little.  I think there is a large enough player base that you wouldn't have to wait on the MM too long.

 

Just brainstorming

Share this post


Link to post
Share on other sites

I would guess that he means that the M60 would see more matches where it would be the only or only 1 of 2 tier X's on the team? But I am not sure if WGNA would be able to code what he is suggesting....

 

As far as I know it has the same MM spread as every other tier 10 (BT10-12).

Share this post


Link to post
Share on other sites

Correct. But the M60 was a poor example since it is a premium T10 tank.

Lets pick the Hellcat, it seems to do well. Its current spread in T6-8, and lets assume the MM will put it equally in those tiers. So instead of %33/%34%/33 split in those tiers maybe change that to maybe %20 T6, %50 T7, and then %30 T8. You see less T6 where the vehicle is very good for that Tier, and a little less T8, where it is a good support vehicle.

The vehicles themselves are fairly historically accurate, the Hellcat did have that armor and was that fast and did have those guns and could hide in an open field (wait...) so WOT attempts to balance tanks by tiering them. If the Hellcat was moved to T7-8 only, I am not sure it would do so well and it would also hurt players who like playing it. This way though they would get some matches that they can feel strong in, but in most matches it will test how good they are as opposed to the vehicle.

Weaker vehicles go the other way of course.

Maybe for platoons, say 2 Hellcats, that would increase the chance of being in a T7-8 match. Platoons maybe should get into a little tougher matches, I saw a 3 man Excel platoon shooting all gold that destroyed a T6 match they were in.

Vehicles have MM weights by class now, you would just have to adjust those for each vehicle a little. I think there is a large enough player base that you wouldn't have to wait on the MM too long.

Just brainstorming

I am so happy that I read this thread to it's conclusion, despite my eyes getting blurry at the half way mark. Three pages ago it occurred to me that some of the offset seen in tanks that were universally "OP" or "UP" could be a function of the closely guarded MM weights that WG assigns to it's vehicles. While I have no basis to gauge how much influence it would have on a tanks performance, having that data in addition to the graphs would be revealing.

I would have looked like such a band wagon jumper if I hadn't read smbakers last post and just threw that out there.

As a last thought, don't try and employ game theory in a bar fight, guys rarely wait for you to finish your multiquote before acting.

Share this post


Link to post
Share on other sites

 

 

Maybe for platoons, say 2 Hellcats, that would increase the chance of being in a T7-8 match.  Platoons maybe should get into a little tougher matches, I saw a 3 man Excel platoon shooting all gold that destroyed a T6 match they were in.

 

Vehicles have MM weights by class now, you would just have to adjust those for each vehicle a little.  I think there is a large enough player base that you wouldn't have to wait on the MM too long.

 

Just brainstorming

 

No need to adjust the MM tiers. FYI if you didn't know, vehicles already have different MM weight and WG sometimes "stealth nerfs" tanks this way by increasing their MM weight. This puts more powerful tanks on the other team to a certain extent and works out similarly to the way you suggest without messing with the MM tiers. 

 

Please no platoons should get into tougher matches. Anyone can form a platoon and if they don't want to it's their loss. In any case, most of the time MM seems to put an equivalent amount of platoons on both sides, you'll rarely see 3 on one side and none on the other, so maybe something like that is already in the MM system. 

Share this post


Link to post
Share on other sites

 

On to WOT:

 

 

I wonder what your winrate would be if you are a player like me (ever so slightly above average) but using these charts Echelon has provided limited your self to ONLY the tanks that are OP? No Douchebagwagons, no M3's, just the stronk tanks.

 

What does everyone think? %53-%54? 

 

What if you did that and only platooned?  I have only solo play so I don't know but would like to find out what people think....

 

The platooning outcome is going to depend heavily on who you're platooning with.

 

Purple tankers are going to take literally anyone that isn't actively trying to make them lose, put them in any tank you want that is appropriate to your MM tier, and carry them to 60+% win rate whether they want it or not.   

 

If they're motivated and a team player then it's a piece of cake - compared to the random sea of mostly reds, having a reliably cooperative 52%'er is a vast improvement.

 

The only real caveat to this is that doing it when they're in arty will - for now - seriously bork the number of games in which you see arty, for obvious reasons.

 

Now, if everyone in your platoon is like our hypothetical tanker, your results may be all over the map.  As you start getting up into the higher win rates, you're making more of those marginal wins by consistently not potato-ing.   Your random "oops" games are going to combine with other normal factors like server pop, time of day, tier distribution in the MM, etc, during any particular tanking session.

Share this post


Link to post
Share on other sites

The platooning outcome is going to depend heavily on who you're platooning with.

 

Purple tankers are going to take literally anyone that isn't actively trying to make them lose, put them in any tank you want that is appropriate to your MM tier, and carry them to 60+% win rate whether they want it or not.   

 

 

 

I don't necessarily think this requires purples. I've platooned probably 99% of my matches since 2.5k battles, just because it's more fun. Most of these matches have been with greens, yellows, and blues. I've only occasionally platooned with a purple player. The end result: a heavily platoon padded win rate. I can't even imagine what my overall win rate would be if I only platooned with purples.

Share this post


Link to post
Share on other sites

I'd agree, not required for every tank... but some tanks are really, really bad.

Share this post


Link to post
Share on other sites

In any case, most of the time MM seems to put an equivalent amount of platoons on both sides, you'll rarely see 3 on one side and none on the other, so maybe something like that is already in the MM system.

The devs have already acknowledged that the MM will try to balance platoons on each team.

Share this post


Link to post
Share on other sites

So after all this talk, is the data still viable?  I mean sure, LT's in particular would have suiscouting and unlucky oneshots go against their stats.  And Platooning or playstyle would influence tank stats.  But WG has agreed the Hellcat is OP, and the data supplied by the players confirmed it.

 

BTW, will there be a new data list made? Either a new one for RU or NA servers?  Is that even possible?

Share this post


Link to post
Share on other sites

Feels about right for the 113. I feel like if I were just a bit better, I'd really start to be able to use it right.

 

113.png

 

And poor IS-7-kun T~T

Share this post


Link to post
Share on other sites

The data and conclusions are still valid, they haven't changed the tanks all that much.  Things like the HEAT nerf to the M4 Sherman was a big move, but the main over-performers and under performers have not changed much.

 

In my little time here it is clear that WG is just moving around the edges.  I do not think any premium vehicles that over-perform will get cut back a little, although now that some vehicles have been given away for free can be nerfed as not everyone paid real money for the vehicle.  I do think lagging sales on some premiums will make some of them get buffed a little, like the Lowe.  That's just good business.

 

This is of course excepting all Russian vehicles.  They have their own rules.

Share this post


Link to post
Share on other sites

Sorry for the necro. I was arguing with a friend regarding some methodology used to define are certain tanks OP or UP, and I remembered this analysis. Do any of you know, is there something similar done lately? Is it possible to create something similar today with the API that WG provides? (meaning is the data available to us - players). 

Share this post


Link to post
Share on other sites
3 hours ago, 3MAJ86 said:

Sorry for the necro. I was arguing with a friend regarding some methodology used to define are certain tanks OP or UP, and I remembered this analysis. Do any of you know, is there something similar done lately? Is it possible to create something similar today with the API that WG provides? (meaning is the data available to us - players). 

http://wot-news.com/game/tankinfo/en/eu just pick the tank and click on the tank curves tab.

Maus gud balance komrad:

41qpF0X.png

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.