We do? 

http://vgchartz.com/hwcomps.php 

Nope, sorry but we don't. 

How exactly did you calculate correlation? I don't see any calculations in the spreadsheet, just the final numbers.

..........

.......fail....

Yes we do! Have you ever seen Wii market share this low? PS3 outsells wii everywhere nowadays, this is unprecedented.

Sure, but who cares?

I'm more interested in your "drastic decrease in wii sales [since the Slim]",  that was actually a 12k drop, less than it was between some average weeks earlier this year, despite the 500k new PS3 buyers. 

This gave me the info I needed.  You performed a correlation analysis based on the percentage market share for each month, instead of the raw numbers.  That does not tell us who is in competition.  Take July to August 08 for instance.  The Wii actually sold more in August, but lost market share because the 360 and PS3 were both up.  Doing the analysis by percentage like you did makes the Wii look like it lost sales to the 360 and PS3, but in actuality, this is not the case.

Doing the analysis with just the raw numbers gives a much different picture:

Wii vs 360 = 0.954
Wii vs PS3 = 0.880
360 vs PS3 = 0.862

What this tells me is that the sales numbers of these consoles are highly dependent on what time of the year it is.  They all rise in the holidays, and they all fall during the summer.  That is why the correlations are so close to 1.

On initial examination, we can also say that those consoles with the lowest correlations are in the most competition.  That would be the 360 vs PS3, and Wii vs PS3.  For whatever reason, the Wii vs 360 does not seem to be as in as much competition.

Given that these numbers are so close, a closer examination is probably in order.  Possibly going week by week.

EDIT:  I gave this a little more thought, and here is what I came up with.  The big problem is that the holidays skew the correlation towards time, instead of the competition.  One way to make this more relevant would be to just eliminate the holidays from the analysis.  Another way to do this would be to take smaller chunks of data.  An example could be using the weekly data over a three month period of interest (like around the time of a price cut, although you have to be careful the cut is not around the holidays).

Nowadays? 2 weeks hardly constitutes nowadays. Have you thought it may be something to do with the MASSiVE  price drop and the release of the slim?

So the Wii invaded Poland?

Wonder how Sony and Microsoft would feel about the idea of them joining forces? Some how I dont think they would be up for it, call it a hunch.

Shall I rally the troops???

No response?