Agent 0 Fortune, on 07 March 2014 - 12:42 PM, said:
Your conclusion that when players cannot play in 5-11 man groups they instead drop solo doesn’t make sense to me. You are suggesting that 7 players who cannot drop as a seven man group have instead been dropping as seven individuals and not as a 3 man and 4 man team?
No, not at all
What I'm suggesting is that when we have a 4man and a 5th jumps on they HAVE to drop solo if there's noone else to drop with or there's noone on they WANT to drop with. You're discounting this aspect which is specifically what I'm referring to. I could have been a bit clearer there though. It also completely discounts that if you have the 7 you mentioned that those two groups cannot play with one another. That means they're dropping groups but still can't play with their buddies.
I'm not disputing the data they used. I'm saying they have seriously misinterpreted the data. There's a huge difference.
4 out of 5 dentists recommend x gum
That's a pretty common statistic thrown around. Is that accurate though? Well.........
What if the question posed to the dentists was
"Which gum would you suggest?"
or
"Would you suggest a gum?"
There's a huge difference in the bias of the two questions I just proposed. Then you have to see WHY those 4 dentists chose that gum. It's about understanding the data. PGI just isn't in my opinion.
The statistical analysis is skewed because they don't take into consideration the correlation between group limitations and more solo players now than before the group limitations.
84% is NOT an accurate representation of solo drops. I drop solo lots of times when I'm just wanting to try out builds, just have time for a few quick games, etc. That does NOT mean that I'm a solo player lone wolf type. The same player dropping into games could be counted multiple times in different sections. They might drop solo, then group up, then do a 12 man as more players from their unit join.
I can see all of these factors even with the minimal data they've released. I can see it because I DO have an understanding of statistics beyond simple percentages. I can also understand why 84% is completely wrong. By wrong I mean it's being misunderstood. There's very rarely a hard number when dealing with stats. You can't simply say
"Well we're taking this item off the menu because even though it was the most popular and most sold product for a year but then we limited the number we were going to sell. Now a year later it doesn't sell as many units because we limited the number sold per day."
See how that is a fallacy and would be taking an item off of a menu due to lower sales after limiting the number you're going to sell? That's very similar to what we have with PGI at the moment. They're making a decision based off of a bad interpretation of the data, not because the data itself is bad.
This is a very simplistic and basic idea of how stats and statisticians work when it comes to data sets and how even good data can be biased, misunderstood, misrepresented, etc.
Agent 0 Fortune, on 07 March 2014 - 02:21 PM, said:
The problem I have with this "what if" scenarios and conspiracy theories is that they don't address the point you are trying to make. If you want 5-11 matches petition for that, don't try to attack the data, recognize that you are in the minority, but that even as a minority withing the community, your desires and ideas for a better game are just as valid as anyone's.
This is where you're misunderstanding. We're not attacking the data, we're pointing out that the data is being interpreted very poorly.