For NA, games are at its peak around 22-23h EST and are 5.5 times more popular than around sunrise.
Same goes for EU at Central European Summer Time, but is 22.5 times more popular.
On the contrary to popular belief, weekend games (Saturday) are only 27% more popular than week games (Wednesday). This means weekend games quality aren’t sufficiently altered by casual weekend players. Only toxicity (alcohol, drug) could alter it sufficiently, but it’s not measurable.
This is some interesting information. For your conclusion, actually I think we miss an information about the population itself. Is it really the same population that plays during the week days and the week end? Maybe it can be that the 27% more population on the week end is actually -50% of the population of the week days and +77% of the “casual” players. This would also (beyond alcohool 
) alter the games while still showing +27% more popularity during the week end.
Edit: and actually, this is the number of games. So, it could also be that people have more time during the week end and the number of games has not such a strong relation with the population itself.
So HotS is still very playable and match quality is high if you’re an evening player in EU.
Que times should be short and skill gaps between players not too high during that time of the day (early night).
Always roll my eyes at the people who blame weekenders ruining their games as if only the bad players are allowed to be on your team.
At the moment 5 stacks are getting near instant queues in ANZ between like 10pm and 12am. It’s so wierd that when you are stacked you get quicker queues IMO lol.
That can happen on any region. Some 5 stacks deliberately pick off peak hours as the MM has no choice but to match them vs solo’s as the player population is too low during those hours. Nothing can really be done about this, unless you give the 5 stack hours long queues for something called Quick Match. I’m not defending this practice, just noting the catch 22.
I was thinking that it is possible to obtain some minimal size population with these data. If we take the EU number of games (~30k in an hour), we can have an estimation of the population
30k/3 * 10 = 10k
Divide by 3: taking the length of a game as 20 minutes
Multiply by 10: 10 players in each game
So this is a minimal number of players. Need to take into account that all the games are not uploaded and that it surely is only QM and SL (maybe ARAM also?).
Oh I was talking about SL :stuck_out_tongue so even after writing that my friends were queued as 4 and it said “long wait expected” I join and instant queue. Got 250 points for the win as well. Next games after midnight was 10+ min then less than 5min waits.
I just thought it was weird it would queue a 4 stack longer than a 5. My guess is they relax rules around 5 stacks a bit more but still try to balance with a 4 stack.
pretty poor form if they are specifically off peak to play as 5 vs 5 solos. I only like stacking with friends cause I’m less likely to get a potatoe and/or have a Laugh while waiting for the next game though some of my friends don’t even have mics.
I think you would need to include ARAM. For me I’ve noticed when I play on NA it feels like most people play ARAM. Since ARAM has been made official I’ve noticed the queue times for QM increase and ARAM almost always instant
Well, I just “played” a little with the filters to see what it can do. And it should also take into account the ARAM games that are uploaded and that is the case. Well, at least the numbers are consistent.
But… there is something really strange with that filters. Having less games when selecting all regions than only EU region is a little bit of weird. Another thing is that it seems to be a cumulative number of games and not an average one (you can see it when selecting 7 days or 15 days and the number is doubled). So if these numbers are OK, actually, the minimal population should be 10k/60. That is really not a lot. But again, it is a minimal population.
Well at the moment, I am really not sure that these numbers should be trusted since they are not consistent, even not talking about correctness… And actually, I think they have something like the player ID so it actually would be easier to directly look at the population.
Around 10-15% of games are uploaded, I think.
I wouldn’t look at All Region, as it’s not important, buggy or not.
The thing that could influence individual region numbers from the day of the week significantly are players that actually upload their games. Maybe the casual weekend players don’t upload as much as daily consistent players.