Plz explain the 50% win rate to me

General Discussion
184

Can someone explain to me then why all this talk about win rates with heroes?

For example: for years threads about X hero being 58% win rate or 43% win rate and this is a soft indicator that this hero may be buffed or nerfed. But with blizz reply, if the game matches you with similar players and if you’re on a win streak and versing harder players which is fair enough, then why all this talk about nerfing or buffing because it’s not around 50%?

I truly believe there is other things the blizz reply is not telling us and the 50% has other factors other than what the reply state.

185

Those are not the same thing.

Heroes should have good and bad matchups. They counter some heroes and are countered by some heroes.

When a hero wins 80% of its games against everything else ( As was the case with release Samuro) then there is a problem because its obviously too strong.

Same thing when a hero loses the majority of its games (Like Currently Cassia) no matter who plays them. It’s an indicator.

186

This HERE is the Fing problem… the Matchmaker should NEVER make me be the worst person on the team or the best. This is the #1 complaint we have! But what seems to happen, is that your teammates do not get better, but the enemy does as you win more. MMR only works in 1v1 games. As soon as you have a team of 5 people, MMR doesn’t work anymore to calculate skill. You can be the best person in the game and still lose and you get the same amount of points lost. This is why we wanted Performance Based MMR, but your “engineers” could not even figure out how to properly implement it even though other games have some form of it working as intended.

2 Likes
187

Assuming we are talking about ranked, because no one cares about win rate of other game modes, this is the key point.

Does everyone always give their best to win? In my experience they don’t!

The problem is there are four roles and several heroes for each role, a player should have a different MMR for each role, maybe for each hero. But we draft so the MM does not know what you are going to pick.

Say a high MMR tank is bored and wants to play assassin or a main assassin is last pick and tank is needed. Etc etc.

Those players will under perform their MMR screwing the MM over.

Then there is the other issue, people not playing ranked with their MMR not decaying. Example, players consistently playing barely 10 ranked games per season still in plat.

Blizzard is afraid to hurt their feelings, as any supplier is rightly afraid to offend their customers but people not playing ranked should never be above silver. 10 games over 4 months means they do not play ranked.

In the current system MMR should decay faster than it is now, for everyone.

They could rework the system into making everyone start from bronze/silver every season with a much higher mobility at low ranks (until diamond). With mobility decreasing the higher you get.

1 Like
188

Ah the Trump approach.

1 Like
189

50 win minimum by end of season to get rewards should also be used to trigger some sort of MMR decay, if you did not play those 50 wins, you get decay from the amount of games you did not play to get 50 wins, so If you played 10 games, won 9 and lost 1 and you end up Plat 5, you should decay to SIlver 1 or Gold 5 next season (pre-placement).

190

The only reason not everyone has exactly 50%, is because Blizzard’s matchmaking is as bad as their report system.

2 Likes
191

Every Grand Master I saw so far has north of 55-60%+ win rate in all seasons, they for sure aren’t being forced to win only 50%.

Can’t be a pure coincidence they are also the highest ranked people in the game.

Also seen plenty of Bronzes with less than 40%, surely not a coincidence.

2 Likes
192

No coincidence. Grand Masters just have no life, and players like me that don’t care about 20 minutes of bs, just to have a comeback, just to have your stats look better, will always have a lower winrate, because the matchmaking just doesn’t account for those things.

1 Like
193

Hardly an explanation - I frequently see GM’s who have played 4+ years but have 500-1000 matches played. Lifetime.

I played 10k matches and have never broken D3 in solo ranked play.

Games played doesn’t translate to rank.

1 Like
194

I’ve had (haven’t seen him online for 2 years) a friend with 15k+ games (well over 5k in ranked) who never broke out of gold (usually hovering around bronze). Some ppl are just bad at the game and have no intention (or need) of improving. They get evenly matched games and they’re happy with that. That’s the whole idea of a working ranking system (and I do agree it is working just fine (always has been) [sure, an injection of 10 million new players would be awesome for an even better matchmaking experience, but that’s a daydream]).

1 Like
195

Exactly. The “GM’s are just players who nolifed” argument makes no sense. It’s not how much you play, it’s how you play. There are eternal bronzes with 20k games played. There are Grand Masters who joined HotS 2 years ago and played 400 total games.

GM’s likely just have better micro, more efficient brains etc., they are just overall more valuable teammates.

196

I just realised that my injection of players notion could give the devs bad ideas. I rescind that notion. I never said that.

(Reference to terrible examples of Fortnite and PubG padding playerbase problems with adding bots to games).

197

Heh. I legitimately and unironically thought all the Bnetplayers and MagicPants were bots when I joined HotS. Took me about 6 months to realize they’re human players.

There’s thousands of them and they mostly never speak.

1 Like
198

I played a game of PubG with 95 bots in it (so only me and 4 other players). So far I haven’t launched PubG after that…
Great way to kill, murder and butcher a game.

I really really hope that terrible idea never comes to Hots. Even bad players don’t want to play vs bots that “pretend” to be players.

1 Like
199

This is impossible. Someone is always going to be the best on the team, and someone will always be the worse.

1 Like
200

There are first class people and other plebs. First class should get extra treatment. /s

201

Ok, when i say “GMs have no life” i am talking about… you know… “the 99%”, not your ONE friend that plays ranked.

202

But how does playing more game exclude you from a supposed 50% algorithm?

Logically, the more games you play, the easier it is for the game to steer you because it has more stats on you.

1 Like
203

More games you play, the more you can get screwed over by bad luck. In an ideal world, all games would be even, but as soon as you get a loss due to afk, feeder, troll, etc. You fall behind and can never get back until the luck swings back in your favor (system still working against you). It’s like gambling, the odds are against you, if you win big… GO HOME (Which is what people did back then, they made an account, got lucky/carried in placement, and only played their 10 games a season to maintain rank).

What I mean is, that the system should try to find 5 equally skilled teammates and opponents. YES, naturally one of those 5 is going to be the worst or best. But as I win/lose more, I should NOT become the worst on team or best naturally, it should not give me a team of potatoes and expect me to carry them until I win enough to become the potato, the system needs to always give me equally skilled teammates. It seems the more I win, the higher chance I have to be the one banning (highest MMR) and when I am banning, more often then not, the teammates I get are inferior (sometimes by a HUGE margin).

My whole complaint about the forced 50% is not that it “makes” me lose, it’s that it does not give me quality matches. Most of my games are 1 way stomps with MAYBE 5-10% being “even” matches.

Forced 50% is the feeling that there is nothing you can do to win some matches (or lose some matches). I do not care about being higher ranked, I just want to have consistent even matches with MAYBE 5-10% being stomps, not the other way around.

There is this huge misconception that some players believe the Forced 50% is preventing them from ranking up. (most think that by ranking up, they will get higher quality matches, which is not true (I used to think that, so I started getting frustrated when I lost because it set me back)), I have played in every MMR range (Except Masters/GM) and all of them have the same issue of inconsistency in teammate/enemy quality.

I know this is a team game, but if you want to use MMR as a system, I think it’s time we start talking about changing the game to allow for more individual performance. Right now, the way it feels is that whoever gets unlucky and gets the worst player out of 10, loses 99% of the time.

“A chain is only as strong as it’s weakest link”, so why do the other 4 get punished just as much for losing because of the worst player?

PBMMR needs to happen, even if it screws Bronze and Master/GM players over. It worked for Silver to Diamond.

Maybe we should just find the correct algorithm to determine who was the worst player on BOTH the friendly team and enemy team and those players lose more or win less points to help fix the ladder. So like, Worst player would lose 300 points instead of 200 and the rest of team only loses 175. Same with Winning team. The guy who was carried, gets only 100 Points and rest of team gets 225. This kind of system should speed up the proper ranking of individual players in a team based game.

We also need to bring back SOLO queue only and Team League to make such a system work, because it would screw everything over if there is a huge spread of MMR in games. UNLESS, groups are always matched based on the Highest group MMR. SOo if a D5 player with 1000MMR queues with a Silver with 200 MMR(Numbers For illustrative purposes only), both are counted as 1000MMR and both go up and down by the same amount of MMR. So if you lose 30MMR, both go down by 30, they would NOT be adjusted to prevent boosting yet allowing groups to team up together regardless of rank.

Thoughts?

1 Like