It's not pretty obvious at all. In my house back in college we had a Wii which as far as I'm aware never had Wii Sports played at all. There may be a significant fraction of the core that never sticks the game into their Wii for the simple reason that they're not interested in it. That might amount to a very small percentage of Wii owners, but it may also amount to a larger percentage of the users who submit Nintendo channel data, due to that being self selecting.

Well if we're relaying tales of experience, all of the traditional gamer Wii owners I know have all played Wii Sports a fair bit. I don't know any Wii owner who has never played it, in fact.

MadWorld - TH=53,699 Av. pp=5h 15m (in 4 plays - like me, it seems most people didn't really rate this game [ball park# ~100,000)

This is more than my total time i took to beat this game ... in case you are wondering i finished mad word in 3:55 H.

The point is that anecdotes are not enough. We simply don't know how many people don't put Wii Sports into the console for whatever reason, and your near 0% figure is just your guess based on anecdotal evidence.

It is, but it's a guess that seems more reasonable than one that puts the figure higher. I don't know that everyone in the UK who owns and watches TV has ever turned it on to BBC1, but I would guess that the figure that haven't would be close to 0%. True, I can't prove it, but it's a reasonable working assumption, and if anyone disputes it then I think the burden of proof lies with them.

Well unless it is anywhere near your example then it doesn't really make much difference. But even then the assumption that all games can therefore be multiplied by the same figure is wrong.

Zack and Wiki for instance I doubt the multiplier is much more than 3 or 4... in fact it can't be 4 because then it would have gained a silver award from ELSPA (though sometimes I am unsure if they count 90 or 100% of the market)

Endless Ocean also doesn't seem to have gained a silver award... which I do find a little odd, but it does mean that it's multiplier must only be about 3 at best.

Not quite sure what you mean by the multiplier - does the channel read how many different files are on each game, and feedback literally a per person set of data rather than, as I presumed, a per system set? If that is so, then the Wii Sports data would suggest that far less than 10% of Wii owners are submitting data, and if that's the case, the raw number of players indicated by the data for other games may also be less than 10% of the actual figure (though acknowledging that such a figure is likely to vary depending on how much the game lends itself to a gamer who uses the Wii's online functionality - one would expect, for example, data for VC/WiiWare games to be based on a much bigger percentage sample than that for Wii Sports [i.e. well above 10%).

No... the multiplier is just whatever number the users of a game on the Channel need to be multiplied by to get their actual UK sales.

The multiplier for Wii Sports is about 10 (if you look on the channel, divide the total sessions by the average sessions you get the number of owners of that game that also use the Nintendo Channel.... well roughly anyhow ie ignoring game sharing, though some of that should be nullified by those that have entered the Nintendo Channel and never returned and then gone on to buy their games)

Wii Sports total sessions divided by average sessions works out to ~610,000 (i'd do it exactly if you want but not at the moment) and of course you multiply that by about 10 to get the UK sales.
That "multiplier" is what I am guessing you used to work out the UK sales of those other games.... making the assumption that all games Nintendo Channel "sales" equate to a similar percentage of actual sales. You did say that it might not be accurate mentioning Call of Duty as an example... but I don't think you realised the extent of the possible error. As I said earlier the multiplier for games can range from just 3 (or perhaps even less in some cases.... though I would say that is limited to things like Wiiware that require an online connection just to get the game) all the way up to at least 20. Factors that affect this are mostly things like "casualness" so games like Fire Emblem and Okami are at the very low end.... with a multiplier of maybe 3 or 4 (ie the "hardcore" who buy these games are also more likely to have the Wii online and send their Nintendo Channel data) while some games like the cheap 3rd party Wii Play knock-offs have a much larger multiplier.

Oh I see - yes, what you're calling the multiplier is just what I calculated from Wii Sports.

What surprised me about the data was more just how many systems in the UK had data for certain games on them - now like you say, the multipliers for each game will vary from the 10 suggested by the Wii Sports data, but I don't see why such a multiplier would be dramatically lower for a game such as Z&W. True, users doesn't equate to (first-hand) sales, but it raises an interesting point: suppose an original IP is something of a commercial failure due to low first-hand sales, but Nintendo Channel data suggests that quite a large number of people ended up playing the game anyway (perhaps through second-hand sales, renting or borrowing), would a publisher consider such data when considering whether to release a sequel? I mean, Z&W may well have sold nowhere near 200,000 copies in the UK, but suppose (hypothetically) that the multiplier from Wii Sports is similar to that for Z&W. That would mean that, despite low sales, around 200,000 UK Wii owners have experienced the game. As such (given the game's critical acclaim), the number interested in a sequel might be higher than sales of the original suggested.