23 minutes ago, Ginkapo said:Undeadguy, statistics dictate that outliers should be checked for variance against population averages. You do not compare two opposite outliers, you compare the outlier to the overall population to see if there is a significant variation to determine if it a contributing factor or not.
With all due respect, Geek was correct in his approach.
Edit: Take a group of people, line them up in height order. Now with a standard group of people the tallest person is a bit taller than average but with the tolerance of the variation that is present in the overall population. However, comparing just the shortest and tallest person you can see a massive difference! This is to be expected.
Now add an NBA player to your group. Now the tallest person is abnormally tall compared to the overall variance and average of the population.
What the regional data shows is that the winners are not abusing activations significantly more than anyone else. Many players have views of whether a 4.5 average ship count is healthy for the game, however no amount of statistics will give any judgement on this. Where the line between helpful activations ends and needless activation padding begins is a subjective choice which varies from person to person. Thats why these discussions are so vitrolic, because it never seems to be acknowledge that its a subjective discussion.
Are you saying the winners/top 4 and the bottom quarter are outliers? Because that's what I was comparing. Now if there was a 1 ship fleet in the top 4, I'd agree that it may be an outlier if it's not consistently repeated. We could say 8 ships is an outlier, but I think that has been repeated before by Tokra.
As it is, there is nothing in the data set that looks like an outlier. And if there was, how do you dismiss it without thorough investigation? Geek's response was the bottom quarter doesn't matter because those players were just looking to show up and get prizes, to which there is no data to support this claim or dismissal of data.









