My winrate with Shimakaze (let’s ignore the few games with Kagero till rank 15) is 53.4%. This means that any randomly chosen game has 53.4% chance to be a victory. For a game to be victory and the previous one being victory too, the chance is 53.4*53.4 = 28.5%. Let’s compare the theoretical chances to the real:
- Double-win: 31.4% instead of 28.5%
- Double-loss: 24.9% instead of 21.7%
- Triple loss: 14.0% instead of 10.1%
- Triple win: 16.8% instead of 15.2%
- Quad loss: 7.2% instead of 4.7%
- Quad win: 7.9% instead of 8.1%
As you can see, both wins and losses are more likely occur after a similar result. Especially losses tend to happen in clusters, quad losses happen 1.5x more often than they randomly should. The possible reasons for the not random losing streaks:
- me being tired, annoyed, distracted or other way unfit to play competitively
- if a game ends, many players queue again, so the next battle likely contain many of the people from the old game. If I manged to make them mad with my unique playstyle, they start the game mad when they see me on their team again.
- my win chances greatly depend on ship composition, for example with 4-4 destroyers in a game I likely lose, while with 1-1 I likely win. I also likely win more if bigger part of the cruisers has radar. Since most players play the same ships again and again, if there are lots of DDs in this game, the same thing will happen in the next
Other reasons can be assumed, but they don’t matter. What matters is to realize that I have much more losing streaks than I should. How to break this? By having a break after every defeat. Waiting half an hour after a lost game will likely fix the mentioned problems. Even if they persists (for example Zao lovers have a weekend rally), at least I play much less so lose less. On the other hand when the circumstances are favorable, I play again and again to utilize my luck.
A 53% winrate is about as close to a 1:1 ratio as one can get. It also implies that there’s a crap-ton of carrying poor players going on within matches. So either the match-maker is damn near perfect, or it’s rigged to keep the majority of players at around a 50% winrate to keep them playing(and possibly paying).
LikeLike
@NoGuff How does that imply that there is caaying poor players going on?
Also, regarding the match-maker: aside from other specific goals (like choosing players of the same rank), pairing players of the same strength is usually the job of a good match-maker.
Or in other words: With ELO ratings or similar internal ratings, a match-maker’s goal is to optimize to no difference in rating, as long as other consideration like queue-time etc. are met. So 53% is no surprise to me.
LikeLike
@Jean-Mira
I said it implies that carrying occurs. It would seem that from the other mechanics of the match-maker, such as the+2/-2 component, that under performing players tend to benefit from tier placement manipulation on any given team. Also, according to Gevlon’s own posts and data on this issue, it might even be more of a mechanic of the match maker having less players to match during low population times, but it still tends to support his experiences with cascading losses.
LikeLike