^ Yeah, that's what I said myself, and what I tried to do in the second part of my previous post.
The difference being that I cut it to the "core" games.

I know, I know... it's a foggy distinction.
But I _expect_ Rayman Raving Rabbids and Game Party and LEGO games to have legs on the Wii as well as similar games had on the PS2, because they sell sort of proportionally to the install base.
The reason being that they are not bought by people who know when they are coming out or exactly what they are ( the front load ), but mostly by less informed people who buy them based on the genre and name only, when wandering among the shelves. A "casual" sales model, if not "casual" gameplay if you want.

Out of those you named, I assessed Sonic and the secret rings and Umbrella Chronicles in my previous one yet. That leaves RE4:

Resident Evil 4 - Wii edition
Total: 1660k - 150k 84k 50k 38k 35k 43k 34k 29k 30k 34k ...
About 530k i.e. one third of total in the first 10 weeks and then I have no data. Seems to have more legs than most, but as with Okami this could just be the port effect: gamers waiting for the retail price to go down, especially since many of them could keep playing their GC version. I know I was one of them :)

I still don't see statistical proof that the Wii is different from the other consoles when third-party new core games are considered. Only that its market has very little of them and relies very much on casual shopping and system selling first party products.

Edit:

The day I have a bit more time I'll try to fit most sales curve with a function like this:

S(w) = F *exp(-w^2/W^2) + L * I(w) + N * ( I(w)-I(w-1) )

S(w) are the sales for a given week and they are the sum of

- a gaussian "front-load" contribute with scaling factor F and width in weeks W

- a "legs" contribute that is proprtional to the installed base I(w)

- a "new user" contribute that takes into account the people who have just acquired a console ( I(w) -I(w-1) ) and are catching up with the library content

I expect "core games" to have big F and some N, "casual audience" games to have big L and moderately good N, system sellers to have big N, moderate F and even smaller L

Wait

How do the legs on the GC and PS2 versions of RE4 compare after eighteen months, exactly?

Look for yourself at the comparison chart

It's exactly as I expected it: more front loaded on the PS2 (more Wii owners than PS2 owners already had it for the GC), then the Wii sales catch up after price goes down, then after 60 weeks the "legs" contribute is practically the same (the curves are parallel).

Very interesting! I notice that its legs are much, much, much better than the Cube version.

^Well, in my magic formula legs are proportional to install base :)

...Even though the Wii version has better legs than the PS2 version?

Lost you there, sorry. I thought the chart showed same legs after 60th week?

Oh, sorry.

...Even though the Wii version's legs appear roughly equivalent to those of the PS2 version?

The legs contribute is L * I(w). The L factor is still different for each game and each version.
Higher L means higher share of release-time-indipendent shoppers that are likely to buy a given game. It could be that L is approximately the same for GC and Wii, though I actually think that it was bigger on GC (more "core" gamers also means higher L).
The PS2 actually looks like it has a smaller L indeed having a bigger I. That lower L accounts for install base demographics ( at that point in its life I guess the PS2 was even more "casualized" than the Wii) and for other factors ( the carry-over crowd from the GC to the Wii is more likely to buy a RE4 copy than the average PS2 owner, basically the F and N factor for them are so flat to contribute to L )

Basically too many parameters to draw a resonably subtle conclusion, but the general shape of the curves is not unexpected.

So you're saying that we have insufficient data to make a meaningful comparison as of right now?

EDIT: I apologize sincerely for dragging you into this conversation about one game, I didn't mean to nitpick like that.