Hi! After a rocky string of matches in season 2, I decided to start logging all of my quick play matches results in a google docs, just to determine an average winrate during those matches (not good… we had an awful winrate of ~20-25% on Season 2, until they fixed the matchmaker).
After that, I just kept logging games out of habit and as something to do between matches.
Today I’ve reached 1,000 games and here’s the data, which can be surprising…
Some notes:
I’m a 40 years old veteran gamer with plenty of experience in FPS games. I would say my skill level has stayed more or less the same in the past years, not getting much better (or worse).
All of them are QuickPlay RQ matches, with Flex role selected. Backfills where the game ended in less than 1 minute after joining were not logged.
Most of the matches were played with friends, usually in a group of 3-4 people.
I barely play competitive that much, last time I played on OW1 I got to Diamond.
DPS: 258 → 554 (52,07% of the total).
Winrate: 50,00% → 51,8%
Support: 126 → 277 (26,03%)
Winrate: 53,17% → 48%
Tank: 117 → 233 (21,9%)
Winrate: 47,86% → 48,92%
Map modes:
Defense: 158 → 303 (29,82%)
WR: 50,00% → 48,84%
Push: 80 → 166 (16,34%)
WR: 38,75% → 43,3%
KOTH: 123 → 262 (25,79%)
WR: 50,41% → 50,76%
Attack: 140 → 285 (28,05%)
WR: 57,14% → 54,79%
Flashpoint (new): 48% (4,72%)
WR: 52,08%
I also have specific map and hero data, but that would take too much space and is more related to my playstyle, not all that useful to others.
What I get from this data (updated):
Overall the WR has gone slightly up for all roles but support. Not sure if related to my own skill or recent balancing changes.
Matchmaker is (still) surprisingly accurate, keeping an almost perfect 50% winrate over several matches. While some games were stomps, I would say on average the play experience was good enough and enjoyable (despite high skill differences on our party that sometimes could cause bad matchups).
For some reason I got double the DPS slots than support (tank can be understandable as there’s only one in the team), despite all players in our team picking all roles. My guess is that this is because the MM has an easier time finding proper skill matchups this way…
Last time I posted this, there were A LOT of win/lose streaks happening. While not completely gone yet (we still have >6 win/lose streaks from time to time), it looks a bit better this time.
Yea we want it close to 50 because then it would seem the matchmaker has made fair matches, but if it is actually close to 50 we just call it “forced 50/50”.
Oh, but it did. Season 2 MM was a trainwreck and they definitely changed things. For example, the way groups with different skills are matched.
As you can see, our winrate stabilized at 50% after that.
How? With a perfect matchmaker (not that it’s possible) we would also have 50% winrate on average. I don’t think my numbers can prove that it’s rigged. We’ve had our share of “unwinnable” matches, but more often than not we’re facing new smurf accounts (Endorse level of 1, some even with public profiles and <20 games on record) when that happens, so you can’t really blame the matchmaker for that.
I won’t deny we’ve been on the winning side in some stomps, but again, we’re sitting at 50% WR, so when I say the experience was good and enjoyable, I’m including also the matches where we lose.
People moan about blizzard being terrible, then claim the games rigged.
It’s either or. If they aren’t good enough to make a simple game, they certainly can’t come up with complete algorithms that miraculously predict human behaviour and determine the outcomes of matches like something out minority report.
Well I wasn’t recording my data but as i feel it’s really hard to win on your own (80% of my matches) now - winrate is around ~47% for me in QP
20% of my matches when i play with 2-4 friends it’s maybe around 51% winrate
in ow1 i only played solo or duo and my qp winrate was 52% (never played comp)
Do we take account whether we’re playing heroes we’re good with or not? I could play 500 games on Mercy and have a 39% win rate, but play 500 games with Lucio and have a 55% win rate
Maybe it’s just me, but this is formatted in a very confusing manner.
What is this supposed to represent?
Why did the number of matches played start at 501 and then go up to 1000? Is the first number meant to represent matches prior to being logged?
Also it would be very helpful if you made more clear splits between different sets of information, so that future readers wont have to struggle. Something like this:
Could instead be formatted like this:
[Party of 4: 148->288] [Winrate: 50,00% → 50,69%]
so that the amount of reoccurring symbols and punctuation is refined a bit more.
Overwatch is the only game where I felt like I personally had to apologize to the other team over how easy it was and how much they were getting destroyed effortlessly. Stomps are not fun on both sides of that. A lot of my games I knew I had no business playing at their competency and some games, I was invincible. I hate it