I was thinking about how to rate the "legs" of the sales for a game. A simple  idea would be to do something like this:

total sales-sales for first X weeks

------------------------------------------

                 total sales

which would of course give the percentage of sales after the inital release weeks.

But do you think that absolute numbers should be included or any other criteria used? And if you agree with the first formula, what value would you give to X?

And holidays are a big complication as well.

What's your thoughts?

well once you built up enough non holiday data you would be able to "adjust" for the holiday boost and even see how much the holiday boost was by other numbers.

However that would probably take a program like SPSS. I miss SPSS and SAS.

I was thinking a really long time about trying to make a "leg" formula and I came to the conclusion that there were to many variables that determines legs...

BUT, one thing I did decided on was it would not be just the first week, but the first 2-3 weeks...

The problem I ran into was games like Halo and Pokemon, Which each sold over 4 million launch week world wide so their % will be skewed even though they still enjoy solid sales week in and week out. They have legs even though their % would be jacked up... then there is games like Cooking Mama which actually chart lower than halo or pokemon but would have a MUCH Better %

Japan Only. The numbers are a few months old. 

I think it´s dificult to make a leg formula because there are too many factors to each launch like time of the year, etc...