Hey guys, this is how percentages work...

Each week, percentages get nudged in the direction of that week's sales. You'll see what I mean in a second.

Let's take the last week of September, worldwide sales:

Xbox 360 - 205k
Wii - 194k
PS3 - 76k
Total - 475k

So the Xbox 360 had 43% of sales, the Wii had 41% and the PS3 had 16%.

This means that if, from today, sales kept going that way for an infinite amount of time, the amount of sales before today would be irrelevant, and the final market share would be 43%, 41% and 16%.

Since console generations don't last for an infinite amount of time, what happens in reality is that market share tends to get closer and closer to the average weekly sales of each console.

Since the 3 consoles didn't launch at the same time, there is a fixed amount of Xbox 360s counting for market share, which is making the market share percentages be different from what we consider now to be a typical week of sales. Get it?

Now, let's take a more typical week, since the last week of September was heavily affected by Halo 3's launch. Week ending the 2nd of September:

Wii - 256k
X360 - 129k
PS3 - 93k
Total - 478k

The Wii makes for 53% of sales, the Xbox 360 for 27%, and the PS3 for 19%. If you consider this a typical week until the end of this generation, this means that final market share will be getting closer and closer to 53% for the Wii (increasing from now), 27% for the Xbox 360 (decreasing from now) and 19% for the PS3 (slowly increasing from now).

The bottom line is that rising market share doesn't imply good sales performance, it just means that market share is getting closer and closer to what the typical (or average) sales week is.

I hope that was understandable.

@NJ5 However, in an alternate universe where all three consoles had launched at the same time, market share increasing would imply increased weekly market share, which would imply good sales. Due to the year head-start by the 360, however, increasing market share doesn't necessarily mean increasing weekly sales, as per your formulation.

Mathematically, you can consider that as a sub-case of different launch times, and vice-versa too, by assuming the launch time as being the launch time of the first console to be launched. This means there's really no difference between different or equal launch times.

Just imagine that the Wii and the PS3 sold 0 consoles per week in the first year, and then started selling. So increasing market share still doesn't imply increasing weekly sales. For example, they sell a lot at launch (but not enough to overcome the top console) and sales then decrease, but market share is always increasing as the consoles find their space in the market. It's all about average sales.

Or looking at it another way.  The Wii and PS3 are stealing some of the marketshare the 360 had of the next generation seeing as the 360 had 100% before the launch of the Wii and PS3. 

I agree with what NJ5 has to say, but, there is no way that we can gauge this average and marketshare this early on in the next generation.

I reckon by the end of next year, we should be able to see how all three consoles are doing marketshare-wise.

2008 is the make or break years for all consoles IMHO.  Although the Wii is gaining momentum at the moment and the PS3 is by far at the back of the class.  2008 should be the defining year for 1st, 2nd and 3rd places.  Now, I;m not stating that if the Wii is first, 360 2nd and PS3 is 3rd, that this is the way it is going to go...what I am stating is that if the PS3 can crawl up from 17% marketshare to a more reasonable level - 25-30% and the 360 drops from the current 40% to closer to 30%, then this will be an indication of what will happen.

On the other hand, we could end up seeing the PS3 with 20%, the 360 with 30% and the Wii with 50%...who knows. 

@NJ5 - I disagree - what it mathematically means, is that, at this point in time, the % of units sold of a given console is greater than the average of all previous weeks of this generation. If you want to include a bunch of 0%s then, yes, that is a skewed and somewhat useless figure. But if market share is increasing then, relative to the other consoles, sales of a given console are increasing.

Gah, that wasn't particularly clear, but mathematically it's true. I have to do work now.

KBG is expecting a lot from a price drop in one region (admitedly Eu is Sonys best region) i expect WW weekly to average around 120-130k for the 4 weeks after the 40G is launched (up 40k, more than doubleing Eu sales)....If it launches next week then thats what i expect, however i am not sure when it comes out so it might be less than that. remember the 120k is an average, opening week in Eu i think tripling Eu is possible.
Id work out marketshares but i dunno the 40g release and i dont want to predict Wii sales (i have no idea how low they will go before xmas)

It doesn't work just with zeros, that was just an easy extreme example to prove my point. I could give you an example with more realistic numbers but people are probably sick of my rambling by now :P

When you have the time, try simulating some scenarios with differently successful launches and you'll see what I mean.

It's launching tomorrow or Thursday depending on the countries, so we'll get 4/5 days worth of data with the new pricing structure. I'm predicting 80k weekly sales at most (even if we scale those 4/5 days of sales to 7), for the reason explained in my sig.

ben did a similar accusation/purposefully misinterpretation  of some of my comments before, which is pretty annoying to be around.