Why is my team full of potatoes?

You are ahead–by far–on hero damage, kda and siege damage, it appears that whenever your team wins a team fight it’s thanks to you but you lose the game anyway. Why does this happen, is it that Blizzard wants to impose a 50% win-rate onto us by giving us potatoes instead of human team mates?

-HOTS uses “account mmr” to balance games.
Hero mmr is a number which represents your skill on any hero, say Diablo. Account mmr is the average of all your hero mmrs. Your mmr increases if you win a game and decreases if you lose.

Every player has a small pool of heroes we are very good at and we are, comparatively, much worse with all the others. Because we are best with the characters we have practiced the most, whenever we try something new, we are going to be matched against enemies worthy of our best characters, not of the one we are trying, therefor we are likely to lose that game simply because of how the system works. It is literally impossible to get a fair game if you are trying something new.

How does this explain me having potatoes on my team even when im on my main?

-When we are on our main or playing one of our best characters, there is a high probability that we get an ally or many of them who arent on one of their best characters. When this happens, they look like a potato to us simply because the system put them in an unfair match that they are very likely to lose, the actual system doesn’t know better. This is the origin of potatoes.

If this is a big problem to you, the most direct solution is this: create one account for every hero you play, this way your hero mmr and your account mmr will be the same and you get fair games every time regardless of the hero.

If HOTS used hero mmr instead of account mmr to balance games, it would be impossible for the potato situation to exist (if we apply some additional simple systems not discussed in this writing for the sake of brevity).

-Salty unreworked Whitemane.

4 Likes

Because there are more inexperienced players than decent ones.

So if you know what you’re doing, odds are you will be with at least 3 players who don’t, and your priorities will be at odds.

5 Likes

This. There is no conspiracy.

Two thirds of the players (nearly 70%) are between bronze to gold.

I don’t know OP’s MMR but if he/she is considerably high, the game just can’t find similar allies. It will throw that player in whatever match is available and make the average MMR of both teams as close as possible.

That chart is for Ranked, but the distribution for QM probably isn’t too far off.

5 Likes

Hm, is this a result of having the MMR directly translate to rank?

I´m quite convinced i read at some time that bronze was the most popolous rank in SL (or maybe HL back then).

1 Like

I don’t know, there are no fresher stats than from 2019, when they translated MMR directly to a person’s rank.

The huge overrepresentation at Gold might also be because Gold 3 was the starting point for new players from 2015-2018 (also 1700 MMR, or mid-gold, on Hotslogs).

Many of those old players probably still remain somewhere in Gold if they still play HotS, as long as they do their placements.

Its not like enemy team have none of those. Both teams can end up having at least one bad player not just yours.

And because they don’t decay, they could have stopped playing for 4 years and come back right back in gold.

i decayed a bit after not playing a year, from plat 1 to plat 4 despite winning three placements, but it’s not a very meaningful decay

maybe if i didnt play 5 years i would have dropped very low

but someone who was gold 1 some years ago will probably still place in gold

Someone on the forum recently complained they had decayed from diamond all the way to bronze, but upon further digging they admitted they had not logged in for a very long time.

I don’t have the topic bookmarked because I assumed it was fairly irrelevant, but I think the poster missed several years worth of seasons.

That’s about 1 division per season. I would call that a very slight decay.

1 Like

Are you claiming they use two forms of MMR?

:rofl:

It would be impossible for it’s matchmaking to be decent.

Because it would take a minimum a few dozen games to get an accurate hero MMR. Hundreds Is probably more accurate.

I’m sure there is a better system them “we adjust your effective MMR down for low level heroes, and up for high level heroes”. But it certainly isn’t per hero MMR.

Now, per role MMR… eh, that might be viable.

I don’t know exactly how they do it now, in regards to where they pull MMR from, but it used to be “skipping” a season of ranked would have the game pull from your URD MMR as baseline (or QM if you didn’t have an URD MMR)

They publicly changed that, but for very long term breaks, I wouldn’t be shocked if they did the same sort of thing.

Nah, the way that the rank distribution works is that they target which % of players should be in each rank.

And if there is a large change in reality, they adjust what MMR applies to what rank.

Here is a site that seems to have gathered info on the matter accurately:

The problem with this is that your macro knowledge is the same for every accounts, even for parts of your micro skills like stutter stepping.

I don’t want to have to play 100 games of the same hero to be placed at its right MMR. Even at that point, it gets worse. Both of my accounts have around 3k QM MMR. My main is lvl 2671 and I face an average team of gold/plat, meaning that my team mates are around silver/gold. My smurf is at lvl 235 and I face and average team of bronze1/silver, and I’m with bronze 3/4. I find my games way more fair on my main account.

2 Likes

it would be less accurate than hero mmr matchmaking. Nevertheless i can see this becoming a reality simply because it sounds less risky.