Tier 8 Post 1.0 Matchmaking Analysis

Introduction and Methods:

I noticed previously that tier 8 matchmaking was awful and I wanted to look at matchmaking in general.  I recoded data about my solo pub battles since patch 1.01 in an Excel spreadsheet.  After I started recording the game data, I learned that the community contributors that were at Tankfest were told that the system to make you more likely to be middle or top tier after being bottom tier was not implemented.

I played 1,013 solo pub battles since the release of patch 1.01 until this week.  For every battle, I recorded tank played, tank tier, battle tier, battle type(3/5/7, 5/10, or all the same tier), win/loss/draw, and other information.  Most of the battles were played during 7 to 11 PM East coast time on the US Central server.

Out of those battles, 484 of them were tier 8 tanks.  Of those tier 8 battles, 52 of tier 8 battles were pref matchmaking tier 8 tanks.

Results:

All tier 8 tanks including pref matchmaking tier 8 tanks:
Tier 8 total battles: 484
Tier 8 3/5/7 Top tier: 66 (13.64%)
Tier 8 3/5/7 Mid tier: 60 (12.40%)
Tier 8 3/5/7 Bottom tier: 186 (38.43%)
Tier 8 5/10 Top tier: 20 (4.13%)
Tier 8 5/10 Bottom tier: 119 (24.59%)
Tier 8 1(same) tier: 33 (6.82%)
Total Top tier battles: 86 (17.77%)
Total Top and 1(same) tier battles: 119 (24.59%)

Tier 8 Exclude Pref Matchmaking Tanks:
Tier 8 total battles: 431
Tier 8 3/5/7 Top tier: 52 (12.06%)
Tier 8 3/5/7 Mid tier: 53 (12.30%)
Tier 8 3/5/7 Bottom tier: 186 (43.16%)
Tier 8 5/10 Top tier: 17 (3.94%)
Tier 8 5/10 Bottom tier: 99 (22.97%)
Tier 8 1(same) tier: 24 (5.57%)
Total Top tier: 69 (16.01%)
Total Top and 1(same) tier: 93 (21.58%)

Tier 8 regular matchmaking 3/5/7 battles only:
Total tier 8 3/5/7 battles: 291
Tier 8 3/5/7 top tier battles: 52 (17.87%)
Tier 8 3/5/7 Middle tier battles: 53 (18.21%)
Tier 8 3/5/7 bottom tier battles: 186 (63.92%)

If 3/5/7 matchmaking were to follow the template after a large enough number of battles, the matchmaking should be like this:
Top tier: 3/15 = 20% of battles = Difference of 2.13%
Middle tier: 5/15 = 33.33% of battles = Difference of 15.12%
Bottom tier: 7/15 = 46.67% of battles = Difference of 17.25%

Tier 8 regular matchmaking 5/10 battles only:
Total tier 8 5/10 battles: 116
Tier 8 5/10 Top tier: 17 14.66%
Tier 8 5/10 Bottom tier: 99 85.34%

If 5/10 matchmaking were to follow the template after a large enough number of battles, the matchmaking should be like this:
Top tier: 5/15 = 33.33% of battles = Difference of 18.67%
Bottom Tier: 10/15 = 66.66% of battles = Difference of 18.68%

Conclusion:

It appears that matchmaking is completely broken for tier 8 tanks.  The statement by the community contributors that went to Tankfest appears to be true.  There is no system to make you more likely to be middle or top tier after being bottom tier.  Because of this all tier 8 premium tanks that people buy to make credits and train crews suffer.  Before the 9.18 matchmaking chance, regular matchmaking tanks were top tier between 35 and 42% of the time depending on who's data you look at.  Now, tier 8 tanks are bottom tier most of the time and have to spend more credits on consumables and premium ammo to successfully fight the higher tier tanks they face.