Jump to content

Top vs Bottom Tier tracking for a month. What I found out surprised me.

Top Tier Bottom Tier Win Rate

  • Please log in to reply
25 replies to this topic

dindincf #21 Posted 07 September 2017 - 02:34 PM

    Potato Rights Activist

  • Players
  • 25889 battles
  • 2,407
  • [CLEAR]
  • Member since:

View Postecpgieicg, on 06 September 2017 - 08:04 PM, said:

Good post.


A minor objection: 



It totally is. I am top tier a lot more. And apparently, you are too.



Tier 8 is one of the most popular tiers due to credit grinding.  Tier 9s get tossed in with tier X who are always top tier.  It makes sense to me that tier 8 is consistently top tier.  Not rigged



Tier X's: Leopard 1 / T110E4 /  T110E5 /  M48A1 Patton / Object 263 / IS-7 / Object 140 / T-62A / JPE-100 / AMX 50B

“At the moment, you should forget about the M60.”        -ZIoydd, on 01 August 2017 - 06:01 AM


Wombeer #22 Posted 07 September 2017 - 03:01 PM

    First Sergeant

  • Players
  • 38390 battles
  • 1,559
  • [HRNET]
  • Member since:
Very interesting, thanks for all the work and analysis! I suspect the higher winrate at bottom tier may be due to the fact that if a good player (you) is at bottom tier, they are in effect displacing a worse player who would have been there. Would you rather play at tier 10 with a weak player using a FV215B 183 supporting you in a T54, or the other way around? I guess it's a subjective thing, but the stats in OP's post support the idea that it's best if the weaker player is in the stronger tank.

BorisBaddenov #23 Posted 07 September 2017 - 03:10 PM

    3.8 Made me quit

  • Players
  • 28451 battles
  • 9,556
  • [SPUD]
  • Member since:

View Postecpgieicg, on 06 September 2017 - 08:04 PM, said:

Good post.


A minor objection: 



It totally is. I am top tier a lot more. And apparently, you are too.



Have you done the same study as the OP in order to be able to make this claim; "I am top tier a lot more."?


If you have not - then you cannot make a claim that the MM is rigged.


OP - Fantastic post.

I’ve noticed a strong lack of Spare Parts. Is that intended?

No, it is not. This problem will solve itself with time. Many players have several dozens of vehicles in their Garages, so they don’t have enough Spare Parts for the entire vehicle fleet. But you will gradually collect enough Spare Parts to unlock new slots for purchased tanks in time.

TonyRomano #24 Posted 23 March 2018 - 01:40 PM


  • Players
  • 24138 battles
  • 97
  • [ODKS]
  • Member since:

View Post_Weeping_Angel_, on 06 September 2017 - 12:31 AM, said:

I took a month and compiled some simple stats about the games I played. Every once in awhile we see people on the forums rant and get angry about bottom tier matches etc. so I was just curious. I play mostly tier VI up to X. For this experiment I took out tier X because there is no way to become bottom tier. I played for about a month and kept track of each tank, if I was top or bottom tier, and win or loss for each match. This is back in January before my great hiatus. I got the STRV and Fury so I did end up playing more tier 6 than normal to get use to these new tanks. At my stage of playing (over 32k games) I tend to mostly play tanks I like vs The dreaded tank grinding to get the next tank.  This lends to me playing way more premium tanks that other tanks.


My questions coming into this experiment.

1. What percentage of games am I top vs. bottom tier in a match?

2. Does my win rate (WR) vary at each tier based on being top vs. bottom?

3. Any variance between tiers for any of this Win/loss or top/bottom tiers?


Here is the data about games played as top vs bottom tiers...


  Games Top Tier Bottom Tier


Total 1003 538 465 .536
Tier 9 106 60 46


Tier 8 239 142 97 .594
Tier 7 358 172 186 .480
Tier 6 300 164 136 .546



Longest top tier streak - 12 games

Longest bottom tier streak - 14 games

Best day top tier - 26 of 36 games as top tier

Worst day top tier - 13 of 36 games as top tier


Observations and thoughts 

1. I was not overly surprised that tier 8 was the highest top tier rating as I tend to play almost exclusively premium tanks at tier 8 (IS-6, KV-5, T54 mod.1, O-47, Defender, etc.) to grind credits. They may not have complete top tier only status but it seems like they can pull top tier more often than not.

2. Tier 7 gets the lowest percentage due to overcrowded tier 8. I think tier 7 tanks get dragged into the very crowded tier 8 waiting list. There are a ton of great tanks at tier 7 but most players grind credits at tier 8 so you can get pulled up more often to keep clearing out the waiting times for tier 8 players.  Looking at the combined stats of tier 7 and 8 it appears to be true.


Here is some of the data I compiled over the month for win rate (WR)....


  WR Top WR Bottom WR Total

Total Games - 949 gms

.615 .647 .630

Tier 9 - 99 games

.596 .702


Tier 8 - 217 games .553 .632 ​.585
Tier 7 - 334 games .631 .649 ​.640
Tier 6 - 299 games .668 .647 .655


Observations and thoughts

1. I had a better win rate as bottom tier at almost EVERY tier.  I did not really see that coming.  I though as a good player I could carry more often as a top tier tank as I punish the weaker tanks.  Guess I was wrong.

2. Skills pay off more as bottom tier.  Making sure you are in a position to help the team as a bottom tier tank will help you win more games.  You can sway more pulling your weight as a bottom tier where lots of players struggle to contribute substantially in a match. Skillz pay da billz. 

3. I am a little surprised at my WR as top tier at tier 8.  I only win .553 percent of the time?  I might be overly aggressive as top tier in the beastly IS-6 and Defender so I actually die more than I realize. I may have to go back and look at that tier's stats and see what was going on.  I will update when I get a chance to look a little closer at that tier.

4. Looking at my daily logs I noticed I tended to stick to a tier when I got hot. Some days I did not even play a single game a one tier and have 15 to 25 games at another. That may be a secret to winning more. Having the ability to read the player base at the time and going to where the games are best suited for you to win.


If anyone wants individual tank info I can compile it but I won't bother listing each tank individually on here.  Please note these stats were done back in January before I took my break from the game so some of the newer premium tanks I either don't have or didn't play during this period of time. Hope you find this as interesting as I did.


What I learned

1. This experiment slowed down my game play.  I watched each match to the end. I think I learned a lot by watching and having that time to think about what I did wrong, if anything, when I died before the end of the match. I also stuck around trying to help teammates with comments.  I think I ended up helping several game become a win by calling out spots or advice during the ends of games.

2. MM is not rigged.  I expected to be a little over .500 due to honeymoon phases of new tanks and the like.  It was nice to see that it was pretty balanced over the long haul even if there were some really good and bad days mixed in.

3. Skills pay off more when you are bottom tier.  Side scraping, Making sure your shots pen and deal damage. Moving to advantageous spots. Map awareness.  These all come into play even more when bottom tier.  It is not a fluke when it happened at all the tiers I played at.  Fear a good player who is in a bottom tier tank.  They will influence the game more than being in the top tier. 

Your conclusions seem spot on.

Awake, aware and basics may overcome the reds and bad teammates. Or in my case my own mistakes and misses.

Gavidoc01 #25 Posted 23 March 2018 - 02:15 PM

    Platinum Card Wallet Warrior

  • Players
  • 57384 battles
  • 4,614
  • [III]
  • Member since:

View Postecpgieicg, on 06 September 2017 - 11:04 PM, said:


A minor objection: 



It totally is. I am top tier a lot more. And apparently, you are too.



WeepingAngel has 1003 battles tracked, Crusader6 has over 1000 battles tracked and I am 365 into 500 tracked that clearly show MM isn't rigged. Where is your analysis? 

Edited by Gavidoc01, 23 March 2018 - 02:16 PM.

To those who lose

Winrate is meaningless.




EricOtown #26 Posted 21 July 2019 - 06:28 AM

    Junior Sergeant

  • Players
  • 12516 battles
  • 212
  • [EX12]
  • Member since:
Just came across the old thread because I've been wondering about MM and percentages of being bottom tier/top tier.

A while back I kept stats similar to this, but not quite a comprehensive. I just wanted to track how often I was high tier vs low tier and if it varied by tier and by tank i.e. premium and tanks with higher tier guns.

I had very different numbers that this guy experienced but of course this post is from a few years ago.

I didn't save the numbers but I remember the general trends that I saw.

1. Definitely experienced a honeymoon period with new tanks where for my first dozen or so matches, I would be high tier about 80% of the time, if not 90 or 100%.

2. Premium tanks were high tier a significantly higher percentage of the time compared to tech tree tanks of the same tier.

3. When playing certain tiers, there was a significant difference in the percentage of time I was high tier vs. low tier. For example when playing tier 9 in tech tree tanks, I was low tier about 70% of the time. In tier 7 I was low tier about 60% of the time.
When playing Tiers 6 and 8 in tech tree tanks, I was high tier about 60% of the time. Tier 5 was close to 50/50.

4. All of these numbers were increased by about 10%, tilted towards being high tier when playing in premium tanks. So for example if I average 60% high tier in tier 6 and 8 tech tree tanks, I average closer to 70% high tier when playing premium tanks.

5. Regarding tanks of a given tier but playing with a higher tier gun, I didn't have enough numbers to be sure that it wasn't just random occurrence due to smaller numbers, but there was definitely a trend towards being lower tier more often when I was driving a tank with a higher tier gun compared to when I was driving tanks with the same tier gun as the tanks tier.

6. There were runs where I would be low tier for like 10 matches in a row, but I was never high tier for more than 5 matches in row and rarely more than a few in a row.

Overall across all tier combined, I was low tier about 57% of the time and high tier only about 43% of the time, but I think that may have been influenced to some degree because I played more games in odd tier tanks that in even tier tanks. I can't recall for sure if the split between the number of games odd and even tier was similar to 57% odd, 43% even, but think it was really close. I need to redo this little experiment playing the time number of games for each tier so that odd and even tiers are also equal and then see if my over all MM comes out closer to 50/50. But I would also need to account for premium tanks, tech tree tanks, and new tanks. Perhaps it would be best to do separate experiments for premium vs. tech tree and not use any new tanks to eliminate the honeymoon period from inflating my high tier numbers.

These numbers are exact. I had exact numbers at the time I did this experiment a couple weeks ago, but can't find where I saved the spreadsheet, so I'm just going off memory. While these numbers aren't exact, the overall trends are accurate.

Note: I only played tier 5-9 during my little test when I kept track of these numbers, but based on what I experienced in tiers 5-9, I was guess that tier 1 & 3 tanks are low tier more often then high tier. And tier 2 of & 4 tanks are high tier more often than low tier.

I assume that since we have +1 match making, tier 10 tanks other play against all tier 10's or they are high tier playing against some 9's. Tier 1 tanks the same but inverse. So this is what drives the rest of the tiers to be unbalanced. Odd tiers are low tier more often and even tiers are high tier more often.

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users