No I’m going to give you 10 consecutive replay codes. I will screenshot the list. I will not be selective whatsoever. Will you accept this proof?
Actually how about this:
That’s around 6 hours of open queue in total. The 4-hour video was live-streamed. Perfectly reasonable compositions in all of them.
Sigh.
This is an important subject but it always ends the same…
I swear I should just keep a txt file somewhere just to paste it on these forums.
So…
No, we can’t just play open-queue. It is not the same for various reasons. First, because role-queue exists and takes half the players away, the remaining being people sick of flexing and just willing to go DPS for once. Second because open-queue doesn’t have the same visibility as role-queue.
The main mode of the game, the first thing a new player is going to see makes it the most popular. The main mode is quick play. So, role-lock.
Do an experiment and change this. Most players will not go to the arcade to keep their precious role-lock, they will play the game as served.
Make open-queue the main mode and suddenly, it’s the most popular. It’s not rocket science.
Next issue: I am a flex player. Well, was a flex player. Now I am tank.
What happened is pretty self-explanatory. During the golden age of open-queue, people were willing to flex. Now they can’t.
If our DPS is struggling, we can’t switch. I can’t pick up the pace for him, I can only stay confined in my role.
Which brings me to: role-queue or the creation of a toxic bomb.
Since everyone is locked to their own role, thus role-rivalry appeared. Role-lock created a lot of frustration and anger. For everybody.
There are no benefit to role-queue. The only reason we got it was to counter goats as fast as possible.
So you want to force people to play the mode you prefer by taking away the mode they prefer? 
Open Q seems to be much more popular in Korea than in other regions and they don’t seem to have any issue whatsoever due to the “visibility”.
You can “flex” in Open Q as much as you want.
That this means in practice that you will play Main Tank in an overwhelming amount of your games has always been that way. Role Q might have made the issue worse, but there is no way to actually prove this.
This is purely anecdotal, but my personal experience is that toxicity went down by a lot.
Many games pre Role Q where lost in the hero select screen, simply because 4-5 people were arguing about who gets to play DPS and who has to fill.
And whoever ended up “having” to fill often used toxicity as a vent for their frustration that came from not playing the role they prefer - which often also got increased by the problematic that they heavily underperformed on that role because it wasn’t their main role.
Factually wrong.
According to the devs they started to work on Role Q before goats became a thing, so that it fixed the goats issue appears to be more of an lucky accident…
As one of the most important reasons Jeff named being able to play the role you want within your appropriate rank for that role.
Not really. Just with a quick scan of the first video I found matches where at least one of the teams had 3 dps, there was even one where there were 4 dps. And this is considering that you are tanking, so if you were a dps player then almost all games would’ve had 3+ dps players since dps players almost never swap, nor should they.
for sake of argument, I will accept that for now.
not at all
the quote includes several very specific elements from this dataset
If you wish to believe it is sourced elsewhere, so be it…you may believe what you wish to believe, same as every forum member
it might, and yet, it doesnt seem to be what was done in this case
given the use of specific elements of the publicly-available dataset in the statement, I dont see it as speculation. As above, one can believe what they wish to believe
I dont see how this relates to the topic at hand, but I do agree on the basis that I have no reason not to believe that the devs do not have access to all game usage
Thank you for providing a reference
The statement in question that I am in turn referring to is publicly available
Jeff said in the very quote I provided: “right now” and not “4 weeks ago”.
The most logical explanation would be that player behavior simply hasn’t significantly changed over the time of four weeks and therefore looks very similar.
The values could even have changed by several percent without rendering either of the statements in the quote incorrect, so they hardly suffice to be used as a “finger print”.
Not at all.
See above.
Well, you jumped to the conclusion that it is usage data, which isn’t necessarily the case. Especially since Jeff himself is talking about “popularity” in the quote and most certainly had access to the required data to make a statement about it.
That’s kind of a… weird response?
Anyways… I didn’t want to try to convince you any further and considering you saying this:
…that is probably the right move.
Have fun doing whatever you are trying to do here.
I would wish you luck, but I doubt it would help.