by dugman
Mike is right: there is a huge selection bias in the population of users rating PL. Over 95% of them have played Pandemic. A vast majority of those users rank P as 7+.
.
Keep in mind that a very, very high percentage of BGG users in general have played Pandemic. Saying a lot of the people who rate Legacy have played Pandemic is like saying a lot of people who rate games here have played Monopoly.
this is looking at it backwards.
it isn't the fact that the people who rate PL have played pandemic. It is the fact that the people who like pandemic aren't playing or rating PL.
He and I said "rate" not "like".
I'm equating "Rate as 7+" as "like". is that an unacceptable assumption?
You can assume what you like, neither of us said "like".
Are you kidding me?
let me rephrase...
it isn't the fact that the people who rate PL have played pandemic. It is the fact that the people who rated pandemic low aren't playing or rating PL.I got your point, I just don't like your passive aggressive tone.
Also I was responding to his post above, not your posts, where he was using the fact that the bulk of people who rated the game played Pandemic and that Pandemic is popular with them skews the results.
Allow me to give a really similar argument to what he posted:
95% of players of players who rate the game are still alive. Of those people many of them played Pandemic and gave it a 7 rating. Therefore the results are skewed because the players who aren't still alive never got the chance to play the game and that kept them from giving it a bad rating.
That was his argument in that post and it's logically faulty. The fact that the majority of people who rated the game played Pandemic and gave it a good rating could be in part due to the simple fact many people played Pandemic and that it is a good game. Even if we were talking about a completely separate game from Pandemic you would expect a lot of the people rating that other game to have played Pandemic and liked it because that is what the overall rating pool is like.
Here's another hypothetical example. Say 1000 people are in the rating pool that could have rated Legacy. 999 of them played Pandemic and liked it, the other one hated Pandemic. When Legacy comes out the 999 all rate the game and give it high marks, the one outlier doesn't rate the game. Legacy's rating is techincally "skewed" because 100% of the people who hate the game didn't rate it. But the overall effect of that bias is trivial because that negative population is so small. Therefore in order to show this sort of bias to have a major impact the poster above would have to demonstrate that the potentially missing negative votes were numerous enough to skew the overall ratings. Not to mention there could also be a lot of missing positive votes from people who missed out on the first print run, for example, or who didn't have time to get into a big campaign but otherwise would like the game. Maybe it's biased in a positive direction, maybe it's not, but he didn't argue any of that in the post I replied to.
So I'm not disputing that it's possible the ratings are skewed one way or another, I'm saying his logical argument was faulty and that to show the results are really biased you need to show the missing votes would actually have been enough to turn the tide of the positive votes.