I went by number of accounts.
What were the number of accounts?
40 million, iirc
And shot for a 95% confidence interval with a 5% margin, due to it being a forum poll.
Your poll had 283 respondents, not over 400?
Oh oops. I linked the wrong poll. That one is invalid because it didnât hit the target number. Let me go find the first one.
60-40 split looks a very strong reason to go with separate OQ mode.
And a very bad reason to ditch OQ for RQ as an only option.
2 Likes
Absolutely. I am in favor of it. I believe the polling and the population numbers are valid and prove that role queue is preferred by the majority. But that the majority is quite slim, and thus speaks to the idea that there should be a separate mode for those who prefer open queue.
As the majority does like it, I think that ditching RQ in favor of OQ would be the wrong solution. But OQ alongside RQ makes the most sense.
2 Likes
Okay so now Iâve done the calculations:
Your pollâs 432 respondents are 0.0011% of the 40,000,000 playerbase.
My 136 respondents are 0.0033% of 4,000,000 - I assume forum participants make up less than 10%, i.e. 4 million participants. Thatâs of course being extremely generous.
My polls are 3 times more representative than yours, relative to total size.
Therefore my numbers are in fact sufficient.
Thatâs not how that works. Your poll is of the same population group as mine, but has less respondents. Therefore, itâs not as accurate.
Furthermore, your poll asks what role a person mains rather than looking at which heroes are most played. Many people do not play what they feel is there main more than other roles for a variety of reasons.
Ah, but your poll tries to answer a question about the playerbase - or thatâs what youâre using it as evidence of. Mine tries to answer a question about the forum.
My poll specifically asked: Which is your most played role?
My poll specifically (not NixonIsSadâs) didnât even include the word âmainâ at all.
Okay, yours is definitely better than Nixonâs.
Sorry I mixed the two.
I went to the sample size calculator to check and it said that you need 287 for 4,000,000 people. Which would be 10% of the population of players also being forum users.
Can you give the link to your sample size calculator?
Edit
I see, thanks.
Yeah it looks like youâre right. My sample size is insufficient.
Maybe I should do another one and try to bump/promote it more?
I still believe weâd get the same results with Damage not being the majority, just like we did with Reddit.
Perhaps. It would definitely give context to future polls. Iâm not convinced it would matter to this one as play data also aligned with the polling. Which means itâs highly likely that enough people who play support also play dps and so it didnât strongly impact their opinion.
However, that doesnât mean wouldnât be good to know for future polling. Itâs always good to keep possible sample bias in mind.
I donât quite understand this point. You said before:
The numbers about how we play were in regards to which Arcade modes were popular as I recall. What would that have to do with role preferences?
Iâll probably make another poll shortly.
If you compare the population of role queue QP to Open QP, you can see that the role queue is more popular by a slim majority. The devs have also said in the past that QP had the most users but not the most playtime/engagement, the latter would be ranked.
If people have the same opportunities of playing QP over QPC, then that likely means that they are playing QP for the role queue.
The same is true of open comp vs role queue comp, but Open comp wasnât around long enough to draw that conclusion at the time of posting.
Thatâs a very strong indication that in terms of RQ the forum poll is representative of the general population. As both indicate a small majority preferring RQ.
I donât know about that.
Specifically on the question of Role Queue Quick Play being preferred, the chart shows that in Korea and China, non-restricted modes i.e. Quick Play Classic and Arcade - No Limits were more popular than Role Queue Quick Play.
China: 12.1+17.3 = 29.4
29.4 > 21.2
Korea:
12.3+2.6 = 14.9
14.9 > 13.5
Also, polls even on this forum on that question didnât seem to be so definitive, for example:
From: Overwatch survey
https://us.forums.blizzard.com/en/overwatch/t/very-large-overwatch-survey-finally-the-results/367034
They show they are more played but in terms of preference they also indicated a slim majority for RQ. Remember rather than comparing population sizes across regions you should be comparing similar roles to the same region.
13.5% played RQ QP vs 12.3% played QPC. Which shows a very small preference for RQ over QPC.
That poll is from before RQ was implemented and also didnât reach the number of respondents needed.