So you dont belive only 4% of people is actually willing to play tank?
For me sounds about right. If you look at formations of LFG you notice tanks are last roles to be filled and you wait for them most. And there is issue of what Tank you play with.
Most of “tanks” are good guys who wanna boost their chances by filling for team needs. At least from my experiance.
no, it is speculation.
You don’t have any numbers, statistics, or data. You have no clue how many players insta lock dps. You have no clue how many compositions consist of 2/2/2 or 1/2/3 variations across all ranks.
You have nothing to back up your ‘logic’.
You are speculating.
I do.
If a lot more people want to play DPS than healer or tank, then they’ll have to stand on line that much longer.
This is accurate. Tanks are usually the last filled role in any game. It’s that one guy/gal who waits till everyone else picks and then is like,
sigh 
Click Reinhardt.
Practice what you preach.
You have no idea of the playerbase numbers or exactly how the matchmaker functions. They said they worked on the matchmaker for a solid year(don’t remember the exact number but it was very long), that dictates that it’s not going to simply function off a rudimentary equation of players.
There’s much more that’s going to factor in besides total number of players, such as number of players per region, average ping, etc. It’s fine to have speculation about it, to outright say that the calculations are off is pretty foolish and shows you’re not using logic at all.
And really? Does this snarky reply add anything to your discussion?
Oh the woes of flex mains that just played main tank role 20 games straight
Stop trying to pretend you understand things when clearly you’re not.
Here is my data, derived from attached video (I even have sources):
Here are some calculations:
Assuming the average queue time is 1:30 (the average of 1-2 minutes), this means that it takes the matchmaker 1:30 to find 12 players that are appropriate to match in the same game, ie they have similar SR.
I realize that this is different for different SRs, but according to Jeff’s statement, this is the average queue time, and this is what these calculations reflect.
This means that for a specific SR (one that reflects average queue times), about one player per 7.5 seconds presses “Play Ranked” in average.
These are hard facts.
Now, using these facts, and assuming no behavioral change, which is EXACTLY what Jeff Kaplan is doing, if you would weigh in hero category preference, the matchmaker would instead require 13-26 minutes of queue time. Since we have NO behavioral change, one player per 7.5 seconds for the same SR will press “Play Ranked”. But the matchmaker, who now will require not only 12 players to be able to create a game but 4 DPS, 4 Healers and 4 Tanks as well, will - according to Jeff - have to wait for 13-26 minutes instead of 1:30. During 13-26 minutes, 104-208 people of this SR will press “Play Ranked” (because we have no behavioral change), and for it to be true that the matchmaker cannot find a game sooner than this, only 4 of these 104-208 players have queued as tanks.
If you don’t get it now, please get off my topic.
This is your data?
This…
Yeah I am getting out of this discussion, because it is clearly going to go nowhere.
Have fun with your ‘logic’.
You have to take into account that the current matchmaker doesnt take role preference into account at all.
Lets say that we have just as many players for roles as many heroes for the said roles exist. We have total 30 heroes, 7 heals, 7 tanks and the rest 16 is dps.
Lets say it takes 1 min to create a team, in our case this means in 5 min all players are in the game.
If we go with 2/2/2 rolequene that means we can create 3 teams in 3 minute but will leave out 12 people who are mostly dps. To get them in the previous players have to return and rotated in again what could mean an additional 5-6 minutes in quene if on top of the playtime of the teams.
The sad reality of role quene is that it generates horrific wait-times and if you dont believe it ask any wow player how long it takes to get into a popular endgame dungeon as dps.
I found it a questionable number but since he didn’t provide background so we can’t really judge.
The only game I can compare it too directly is FF Online… it has this sort of queue system for dungeons, and you know what people do there if queueing for DPS, they either
wait it out… sometimes a long time…
breakdown and decide to tank or heal and get in games quickly…
give up waiting and do something else and check back.
I am not sure how you claim that a flex queue option is a fact… he kind of indicated this wouldn’t work because the same people would still end up going tank healer while selfish dps one tricks just stand their ground …
I wouldn’t take too much from it, he also said they’re paying attention to console balance, which has to be a joke.
You have asserted no meaningful numerical evidence. You are speculating.
If you had say, a simulation, as the devs have potentially ran, then I would be more inclined to call your argument logic-based.
I think this is a duplicate
The problem is that if you get to choose, people will choose dps,
If you queue, you are queued and could have 6 mercy mains on ur team. u flex to make comps but if u say u can queue as what u want, me and 13 of my friends agree we would stop maining support and tank and just play dps since its more fun and what we want to play.
Thats the reason, once u tell people u can queue dps and well have healers and tanks so dont worry, people will play dps knowing theyll get that.
He really sounded more like he was talking about it in Character specific rather than Role Specific… I think he mentioned if you want to play 76 specifically… it would take that long if all DPS where split by either character or sub role (Hitscan, Sniper, etc)
If it was character specific I can’t even imagine how long Genji would take.,… probably around 2-3 Hours
If it worked like LFG 2/2/2 Groups do… DPS just gets matched without any pre-determination to WHAT DPS is played… it’d surely be much faster.
Right now, that sounds pretty accurate to me.
Once upon a time, tanks (at least, certain ones) were more popular in my experience, but over time and various balance changes, the role has become more and more miserable and some tanks (like Hog and D.Va to a degree) have lost their former charm for many players. Pair that with the limited number of choices and even more limited tank parings synergy-wise and yeah…
Tank is pretty much always last picked in my games, if its picked at all, and especially, no one wants to pick main tank. Whereas Support? Usually at least one is insta-locked and a second usually follows quickly.
That said, there’s ways tanks could be made more appealing and it would be a lot easier to do so if triple+ tank comps were taken out of the picture.
Its not a matter of being onetrick or not, the whole problem stems from the fact that we have 7 support, 7 tanks and 16 dps.
Just by looking at the numbers its obvious that dps will have more waittimes.
It also doesnt help that tanking and healing requies different set of skills to get maximum gains from while dps only needs the standards with good aim. We only have like 2 healers and 2 tanks who can play as dps the rest requies much more to be valuable (not talking about moira).
People claiming that I’m speculating or that I have no data, please refrain from following this topic any further. You clearly have a Trump-like view on what facts and calculations actually mean.
I share your feeling that tanks are very impopular. But not that impopular. There just can’t be as little as one tank player in every 50 players.
No, that wasn’t the case. He mentioned 76 as an example, but he has never indicated that the discussion was about breaking it down any further than the three roles Damage, Tank and Healer.
I’m not sure what he meant by claiming that, but if so, he hasn’t thought it through. If there is one queue for each hero role, there can’t be a problem with the option to queue up as every role (or two of them). When you get a game, of course you will be assigned one of the roles in that specific game, but the queuing specifically can be done as any role.
I did. And I understand that it’s easier to try and come up with your own way of reasoning than understanding mine. But what I said makes sense. Reading it again might help.
Not really in my experience. Even in games where the sheer number of tanks and healers equals DPS, DPS player inevitably tend to be the worst one-tricks of them all and least willing to heal or tank. I have seen this tendency across multiple games, including WOW and FF… and even Team Fortress for instance.
FF even allows you to have one character who literally can learn all classes… and even then its the same problem, people want to DPS and have to wait… despite no SR system and being able to play whatever they want to learn…
I read it then and now i read it again but my point still stands. Sheer numerical statistics shows an approx 500-600% quene time increase and in the games where they have role quene the dps waittimes are huge.
But if you think otherwise i suggest a little experiment. Go and make a post where you ask what is their favorite hero and why. Count each answer, put them into roles and see for yourself about what are the avarage percentages of the roles people like to choose.