ioi speaks out about ergh "VGC analysts"

[Torillian]

ioi said:

MaskedBandit2 said:

You cannot take such a tiny portion of a market and balloon it out to a precise and exact number to the nearest unit and rank them.

Let me put the ball in your court then. Tell me how you would do it. Here is some hypothetical raw data for three games:

Game 1

Week 1 - 984
Week 2 - 411
Week 3 - 219

Game 2

Week 1 - 477
Week 2 - 341
Week 3 - 356

 

Game 3

Week 1 - 147
Week 2 - 115
Week 3 - 99

Now assume that the data comes from a sample that represents 0.5% of the total population. Can you demonstrate to me how you would present the weekly chart for those 3 games for those 3 weeks in a way that you feel would make you happy.

Is there any standard way to represent the standard deviation for your model?  I guess my issue is right now what's being reported is only the peak of the curve that is being modeled based on previous comparisons but I personally at least would appreciate some kind of estimation as to how the curve that peak is part of looks in general.  Would reporting the standard deviation for the bell curve be a standard practice or is there something else that's often used?  Maybe reporting the 95% confidence range or something of that nature.