So I had a question recently, and frankly I’m shocked that no one has asked this question before outside of a couple of Reddit posts that are more curious than actually serious (and can’t be taken as useful data in any case): what is the level-or overall hours played-of the average player in any given rank? It seems obvious that as player level or play time increases, so should their rank, but I don’t think that’s the case at all.
To begin with, I’m almost certain that Overwatch tries to place you in gold. If you get placed lower than gold, you’ve probably never played the game before. If you place higher than gold on a new account, you probably have played the game before. Gold is sort of the loose target, and I don’t quite know why.
So, I’m guessing this is happeningL the average player, with average experience in FPS games, gets placed in gold. Over time, they fall or climb to their actual rank and stay there until they lose competitive interest in the game, or have a mental epiphany and actually try to git gud. I would guess that at this point people start showing up on forums and complaining about ridiculous things that are holding their progress back.
As to what data I specifically want, I suppose it’s obvious, but player level is not at all indicative of the actual amount of time they spend in competitive. It is loosely correlated, yes, but with the number of quick play and arcade mode warriors, I don’t think you can use that as an absolute metric. For that reason, I want to make it clear that I’m more interested in the competitive play time of the average player at each rank, but I would be almost as interested to see the casual play time of the average player at each rank, and the ratio between the 2 play times. I realize that smurfs are going to screw up this data something fierce, but there’s not much anyone can do about it, past excluding bronze borders above plat. That could screw up the data even worse, though, so IDK what to do about that.
However, I need data that probably doesn’t exist to back this up. If the average/median player in bronze has more competitive playtime than then average/median gold player, that would be an obvious confirmation. Player level could be substituted if someone wants to use the Overbuff dataset, but what the accuracy of that figure is, I don’t think anyone could guess. It would be a decent starting guess, at least. I should do that, but IDK how at this point.
P.S. An obvious follow-up question to this one is this: if the placement target is actually gold, why is that? Is it because there is a smaller MMR range for QP (they use QP MMR in the absence of ranked MMR)? Or is it simply a mechanism by which Blizzard tries to reduce the effects of smurfing? My personal opinion is that the placement target should be bronze, or at best silver, with a stronger stat-based MMR gain should the smurfing feel terrible.
P.P.S. A bit off topic, but could the devs tweak the stat gain system to favor one-tricks less, at least in the long run? I watch streamers a lot, and it seems to me that OTPs are a little over-represented in masters because of the way stat-based MMR gains work.
