I had similar results, and man 40% is a really bad number for a change as severe as this.
1 Like
This is where OP or someone else chimes in with a “it’s 20% now bc we put it on a more popular echochamber where it is obviously going to be in favor as well”
1 Like
I found the original results of that with the questions. Here’s the link.
An important consideration: It’s obvious that repeated polls will give similar results if the sampling happened in a similar fashion. That doesn’t mean that the results are reliable though. The sampling is still non-random, more likely to attract certain demographics (just look at the reddit poll that ±1400 competitive players and only ±70 ‘casual’ players answered) and most important of all: these polls completely ignore that part of the playerbase that already left because of role queue, how small or large that portion may be.
Like I said above in an earlier post, these are called survivor effects and they have caused many a research to have yielded misleading information because the researchers didn’t take them into account.
2 Likes
Well, polls done on day 1 of role queue at least had the opportunity to capture this metric. The polls are from various slices in time since Role Queue was implemented until now. And they have stayed consistently about the same in the end results.
So, yes, there is a sampling error in the sense that the sample is not random but self selected. That is true. But even accounting for that, every single poll (all of them) have showed the same trend of about 2/3 - 3/4 of the player base being in support of role queue.
In the meantime, the people who claim that the majority do not like role queue have no data of any kind. The best argument is that our data is flawed, but without producing any actual evidence that their position is supported. All evidence so far, regardless of the objections to it, all points in the same direction regardless of when or where the evidence was gathered.
There’s also no true indication about how many people have left because they didn’t support role queue. For all we know, that number could be insignificant. We don’t know. But there’s no empirical reason that I know of to believe that number is meaningful.
Well, 60/40 split is not that great. That’s still a lot of people.
Current results are more 75/25 than 60/40
Ty…was Interested in seeing that questions weren’t leading and such which I’m not seeing…so that’s at least good
but now that I can actually read the questions and see the graphs I do have one gripe with the poll…and it’s that (if I’m reading correctly) 93% of the people polled were from people in “competitive” overwatch subreddits…the vast majority of which barely play it based on one of the other questions…which lead to wonder if there are inherent biases given that you would expect those people to WANT to see changes to comp…
And there’s very little if anything with regard to the QP which was also affected by the change…so a more globalized poll would be better
(On top of timing thing which I’ve mentioned a few times on this thread)
Then there’s another important confounding factor known as the ‘honeymoon phase’. LFG is maybe one of the best examples to date of this. At its release, it was being praised as the best thing since sliced bread and now it’s pretty much dead.
That’s often the difficulty with interpreting poll results, one confounder often gets replaced by another, or multiple confounders are interacting with eachother.
Well yeah, I never claimed that the majority doesn’t like role queue. I think it’s more around 50/50, which is still entirely in the area of possibility considering the variance on the existing data.
That being said, I wouldn’t call it a good system if only half of the playerbase or a slight majority is in favor of it. Just look at political decisions where the opinion is more or less 50/50 divided, those only create tension among the population.
But that’s now precisely the biggest issue. All these polls have one main thing in common: the casual playerbase is underrepresented simply because of the nature of the polls (casuals are always a minority on forums, subreddits, etc…) yet they’re also a substantial chunk of the playerbase. If you wouldn’t get similar results, then it would be really concerning.
To give a simple example: imagine that you want to know the height of the average human. You measure the height of a whole bunch of people, but the vast majority are men. Then several other people also measure the height of a big number of people, but every time there are much more men than women among the measured individuals. Everyone will have similar results, but they all will overestimate the actual height of the average person.
Such a thing will probably also not be seen in a sudden decrease in player numbers, but more likely in a stronger decline over longer timespans. It might even still be too early to assess how many have/are/will be leaving over role queue. It’s also not a change that stands in a vacuum, but there also several long term consequences tied to it. To give an actual example, just look at the double shield meta: if it will last too long, you can expect a good amount of people to leave because of that in the coming months. You could argue that people then don’t leave for role queue specifically, but it’s a consequence of it nonetheless.
Blizzard will also never ever tell us if their playerbase starts shrinking faster (Just look at WoW. As soon as the playerbase was taking a nosedive in WoD, they just stopped reporting any subscriber numbers. Since then they haven’t released any of the sort except of those super misleading ‘40 million copies sold!’). Hell, they’re even reluctant to tell us that the playerbase is declining at all. People have to derive that from their financial reports.
ahaha 220 votes? i think i have 220 friends who leave OW because of shields, goats, auto aim characters, bad MM and long queue time … only my friends broke all your statistics
Where are the stats?
There is no way that poll is accurate, it is a small sample and a potentially biased one at that. Only people on the forums responded. People on a video game forum tend to care more about the game then those who don’t, and also invest more time into it. Not scientific, but general reasoning. By that logic, people on the forums are more likely to be supportive of any changes than say, Jo Shmo the dps main, who also plays other games. Jo Shmo is more likely to move on once 2-2-2 goes live, and takes his voice with him. Multiply this hundreds of times over, and the major voice of dissent is gone, meaning that those left are more likely to like role lock, and thus is appears that the majority of the community approves it.
The other issue is only a small group of people are given voices. People who don’t use the forums have no voice, meaning they are left out. If it was a forum survey, thats one thing, but even then, it is only people that saw that particular thread. There needs to be an in-game survey to truly determine if the player base wants 2-2-2.
Here is one of the excerpts from thw official dev response I was talking about:
And even more importantly, here’s the point where the overall very positive opinion of the majority of the game’s player base is touched upon:
at this point its 553 votes.
And you can get your friends to vote, its the best were gonna get here.
Also.
Goats is gone.
Only 1 auto aim character, which is mercy. If you still think moira has auto aim you just need to get better at the game and kill her.
Bad MM is often what people used when role queue wasnt a thing because of 5 dps comps, and long queue times are when role queue became a thing.
Half of your “reasons” why your friends left mean nothing.
At least he’s trying… where’s your effort?
they are not play in OW anymore, why they should vote? Why do you think peope will ask or wait something? Customers will pay someone with a better product.
double shield not better
moira, winston, brigite, RH and another set of spam hero like sigma, junk etc almost without aim skill requirements
5 dps is bad only because of disbalance/ Any healer or tank bring more impact to the game, if all heroes have the same impact 5DD will not be a problem and anyone can play what ever he want. And 2-2-2 changes nothing, we just have DD moiras, DD zarya, DD Diva, DD hamod and DD dzen, still has 5 or 6 DD, and win is not about who is more skilled, but about who peak more shields and who peak more skill independent heroes.
If blizzard will not fix the issue, OW will die because most of DD(who is the biggest part of comunity) will leave to another project. Already now most of battle royals has bigger comunities, CS has bigger comunity.
Are you really think OW is outlive TF 3 in 2020 with so bad gameplay characters like sigma and orisa?
The People that dislike role Q are here in the forums whining and complaining daily.
They majority of people that play this game are not in the forums posting daily.
And the majority like role Q.
People can vote even if they dont play the game anymore 
Its still not goats.
Moira isnt autoaim, its a beam, winston has been in the game since day 1, if they hated him so much they wouldve left day 1. Junk and sigma are not auto aim.
Its not that the heroes arent more balanced, people just prefer to play dps heroes because its fun to kill the enemy, which you can do less of with diff roles.
I think OW will outlive a game that isnt even a real thing
I said this about 2-2-2 and gave it until Thanksgiving for the energy to change, and let’s say I see Christmas coming a bit early
I’m more interested in “What role do you usually queue for?”
It broke down how the poll is actually split and the lowest population was DPS? While ingame, the DPS outnumbered the other roles several times over.
That puts into question the validity of the OP’s poll, especially considering the forums tend to be mostly supports and tanks.