Think you get unfairly treated by the matchmaker? You may be right.

Like many others, I've felt the matchmaker is biased against players in lower tiers. With the 3/5/7 matchmaking, you should in theory be top tier 3 games out of 15 and mid-tier 5 games out of 15, if you look at the games with +2 matchmaking. However, it feels like you're bottom tier more often than that. But is that really true?

Well, the numbers don't lie. To prove or disprove this theory, I wrote a little python script that parses the replay files into a simple format to see what the stats actually have to say…

Let's start with Tier X. I haven't really played it much, so results are hardly statistically significant, but it gives you an idea. Also, this only looks at the standard capture-the-flag game mode; results from other game modes are either not relevant (frontline, halloween events) or very few.

     76 ctf,10,all-same-tier,
     18 ctf,10,plus-one-top-tier,
     22 ctf,10,plus-two-top-tier,

Out of 116 games, almost 2/3rds are same-tier games. Sounds decent enough I guess, although I personallly think Tier X should only play against tier X.

Tier 9:

     34 ctf,9,all-same-tier,
    173 ctf,9,plus-one-bottom-tier,
    196 ctf,9,plus-one-top-tier,
    175 ctf,9,plus-two-mid-tier,
     47 ctf,9,plus-two-top-tier,

…seems slightly skewed. I should be bottom tier more often than top tier in plus-one games, but this is likely skewed by the abundance of people playing tier 8 premiums. On plus-two I'm mid-tier more than three times as often as top tier, while it should only be around 50% higher if it did an unbiased match. At 34 games, the same-tier gameplay is a rare unicorn indeed – but stay tuned, because it's not the rarest!

Tier 8:

   1047 ctf,8,all-same-tier,
    415 ctf,8,plus-one-bottom-tier,
    190 ctf,8,plus-one-top-tier,
    459 ctf,8,plus-two-bottom-tier,
    152 ctf,8,plus-two-mid-tier,
    174 ctf,8,plus-two-top-tier,

The tier with the most same-tier games. Nice! Sadly, many of the "Tier 8" premiums are in reality on par with many tier 9's, so that's hardly a relevant metric. But I'm doing this to prove the match maker is biased, so let's ignore that.

In +1, more than twice the games are in bottom tier than in top tier, which is definitely skewed towards providing tier 9 players a nice set of XP piñatas. In +2, it's even more skewed – you can expect to play almost 60 % of your games as a bottom tier when you get +2 matchmaking.

Tier 7:

    334 ctf,7,plus-one-bottom-tier,
    167 ctf,7,plus-one-top-tier,
    215 ctf,7,plus-two-bottom-tier,
    229 ctf,7,plus-two-mid-tier,
     89 ctf,7,plus-two-top-tier,

Wait, did I forget to copy a line? No, you're seeing it right: Out of about a thousand games, not a SINGLE ONE was same-tier gameplay, making that the rarest game of all. You're skewed towards bottom-tier gameplay in +1 MM,

What's even more interesting is that you'll spend more than half your time as the bottom tier when playing tier 7.

Tier 6:

    364 ctf,6,all-same-tier,
    321 ctf,6,plus-one-bottom-tier,
    182 ctf,6,plus-one-top-tier,
    397 ctf,6,plus-two-bottom-tier,
    124 ctf,6,plus-two-mid-tier,
     36 ctf,6,plus-two-top-tier,

Again, more of the same. Out of 1424 games, just over half is played at bottom tier.

I have so few games at lower tiers, so I'll won't bother posting them.

Conclusions

Yes, the matchmaker is skewed. Yes, you're spending an unfair amount of time as bottom tier.

If you want to be nice about it, it is likely a result of trying to reduce wait times. If you have a conspiratorial mind, it is likely because they need to provide the top players some satisfaction of "winning" by providing a chance to club seals (though these seals do have teeth… if they press the 2 key and ruin their credit gain).

Source: https://www.reddit.com/r/WorldofTanks/comments/r4shpe/think_you_get_unfairly_treated_by_the_matchmaker/

leave a comment

Your email address will not be published. Required fields are marked *