This would only make sense if PS3 and X360 games sold about half of what Wii games of the same age do... but they don't, they sell much much less.

A game with legs will sell in similar proportion to the new gamers buying the console... so it should come as no surprise that they fall when the console sales fall.

The difference is that PS360 games (most of them) at say, 40 weeks old have no where to fall from if the console sales decline, because they have already stopped selling.

As others have pointed out, the flaw of the theory is that the games are only purchased by the new owners.

Also, even if the legs for Wii games are better than games on HD console because Wii has a bigger install base, it doesn't change the fact that Wii games gave better legs. Because of the popularity of Wii, it enables the games released for the console to have such advantage. I don't see anything wrong with it.

I didn't know games like CoD:Waw were system sellers on the Wii...

When hardware sales fall, pretty much every software sales fall..

Case in point.

Week of April 11th :

Others : 132k Wii sold, 801k Wii software sales.

Week of April 25th :

Others : 96k Wii sold (-28%), 549k software sales (-32%).

No huge Wii release around either date either that could have artificially boosted early April Wii sales...

So how else would you explain the decline in software sales so similar to the one in hardware ?

 It looks that way because the HD console individually sell a lot less hardware than the Wii,

Case in point :

There are a tonn of PS3 game selling between 4.5k units and 6k units a week.

That's kinda pityfull sales. Thing is for every PS3 sold typically 3 Wii are sold on most weeks.

Multiply those 4.5-6k weekly sales by 3 and you get into the sales number of a lot of the games that people here call Wii games with legs...

This theory only takes account evergreen titles. Example if the Op theory is right madworld sells would have dropped by a similar percentage compared with hardware sells differential, this week compared to last week. But the drop in madworld sells do not correlate with percentage of wii hardware loss while titles like mario kart and wiifit do.

Can anyone could please verify with your amazing math skills mine are not so good.

Couldn't it just be that the factors which affect hardware sales are the very same factors which affect software sales?

I get your point about Playstation games like Little Big Planet and agree that if they are selling a fairly consistent percentage long after their release, then yes, they have legs too.  I agree that weekly sales of a few games like Mario Kart and Wii Fit are probably proportional to weekly hardware sales and can be taken as an indication that a good portion of new hardware buyers also pick up these games.  It could also be that weekly sales are by and large proportional to userbase size. I'd doubt however that the average person picks up 6 games on the day that they purchase their hardware so i think your explanation can't be used to explain the sales of all games.

The point i think you missed in your assessment is whether or not most or all of the Playstation and Xbox games with legs (by your own assessment) also had big, or relatively big, opening weeks.  My guess, i admittedly have nothing to back this up, is that they did.  They have big or biggish openings, due to a more informed userbase or hype and then continued to sell long after release due to their quality.  If this is in fact the case then it is still fair to say that HD games are front loaded and that early sales can be an indication of the potential for long term sales.  I would suggest that a slow start for a HD game often leads to less than stellar overall sales for the game

I'm not going to say all Wii games have legs and i concede the point about some people claiming "legs" without any consideration of the size of the userbase.  I would say a higher proportion of Wii games continue to sell relatively well (perhaps in proportion to weekly sales or userbase ratios or whatever) long after release DESPITE not having big openings, and i think this is the difference.  If a Wii game opens small we can't necessarily make any conclusions about it's long term sales potential.

@Ail
It seems to me that you are getting your wires crossed as to what you are actuallly stating.

Is this a reality check as to WHY wii games have legs or is this you saying that wii games do not actually have legs, with this is being due to the fact that legs for wii are purely based on consoles sold?

However, this does not really make that much sense, purely based on the fact that the wii consoles sold a week contributes to the legs wii games, logically.

So, Are you measuring legs PER console or legs of the population?

In reality developers only care about the latter.

So really... you might just be stating the obvious.

^ Bobacob. If he's stating the obvious then why are so many people arguing that his obseervations are wrong?

Madworld is not such a good example for what you are trying to illustrate. It is still at its front end, in that the weekly game sales are dropping from their opening peak, regardless of Wii sales numbers. When Madworld levels out and starts showing it's legs then you will be able to look at its sales and see if it goes up and down in similar % to Wii sales. At the moment though Madworld's sales graph is showing a linear drop with the line heading straight towards the x axis, If you look at the Madworld sales post peak (from week 3 to week 7 (current week) the line is basically straight and if one was to project the line to zero it would hit zero at week 10.

Compare this to a game that gets a lot of mention as having poor legs for the hype (KZ2) and the post peak graph covering the same period (weeks 3-7). KZ2's week 3-7 sales follows a more logarithmic curve so that if you project the KZ2 line to zero the line won't reach (practically) zero until about week 20.

I think for most games you can predict the sort of legs it will have by lookng at the sales graph between weeks 3 and 10. High and curvy = good, low and straight = bad, low and curvy = there is some hope, high and straight = freakish.

Madworld has 3 sales weeks left by which time we should be able to predict whether it will have good legs or bad legs. For it's week 8 sales if it is less than 10K this would be a bad sign, if more than 10K this would be hopeful news. For week 9 hopeful would be 8-10K, bad would be 5-6K. Week 10 hopeful would be anything north of 5K, bad would be anything south of 5K.

Interesting read, but it seems like you are confusing correlation with causation.

Yes, the analysis shows correlation. But there needs to be more sampling points in order to establish, or disprove, causation.