No because you could create emulators that work well with some games but don't work with other games. The N64 emulator on the PSP for example, it runs Super Mario 64 okay but most games run poorly on it.

This might be an interesting read for you.
http://www.tested.com/tech/gaming/2712-why-perfect-hardware-snes-emulation-requires-a-3ghz-cpu/

In short: emulators don't use real emulation but rather use a lot of shortcuts to increase performance. It's really dependent on the game and how well it's optimized for emulation how it runs.

It's not a theory, it's a fact and an answer to the question : because the instruction emulation time is totally different depending on the instruction, you can't use emulation to determine which game have high cpu requirements. Some people talk about 3 or 4 instructions, but it can be a lot higher than that, especially if you avoid custom tricks for different games (the fact you have custom development to improve different game proves alone how game behavior can be different). Also, some of the gpu instructions could be (and probably are) emulated on the cpu, leading potentially to an higher cpu use for more gpu intensive games, again depending on the instruction.

Depending of the hardware source and target, emulator implementation, you could find situations where cpu usage are matching very well, but still, by essence, it's not deterministic.

Even disregarding that the answer is no, none of those games are all that CPU intensive anyway.
Games like Call of Duty World at War/MW1/BO/MW3 and Boom Blox 1 & 2 are probably the best stress tests of the Wii CPU.

Proof ? 

As far as I'm aware there aren't any benchmark test results, but the devs of MW1 Wii talked about how hard it pushed the CPU, and other Wii COD games offer similar or even more intensive situations in terms of number of AI characters at a time, etc.

And you can set up some insane physics demonstrations in Boom Blox.

Do you have a source for that claim ? 

Rigid body simulation isn't all that impressive but that's just my opinion. 

Took some digging back to 2009, but: 

Other than memory restrictions, CPU time was the other main bottleneck. Even running the absolute lowest quality assets, a huge amount of CPU time is tied up in level scripting, AI logic, line-of-sight traces, and everything else central to a first person shooter. When everything first fit in memory, the game ran at about 2 frames per second, and hundreds of engine optimizations had to go in to bring that up to 30. Cod4 runs with 32 AI at times (safehouse, heat, and the bog are the worst cases), and each one is calculating animation matrices, doing geometry traces, and running thousands of lines of script - every single frame. We weren't able to reduce the number of AI without significantly changing the gameplay, and so other elements of the engine had to be re-engineered to maintain an acceptable framerate.

http://www.neogaf.com/forum/showthread.php?t=371684&page=28%5D

Yes, "lol neogaf", but the guy had enough inside info that turned out to be correct that it can be safely assumed he was indeed who he said he was.

So I guess some GAF users can be trusted ... I wasn't going to be snide about where the source came from but do you have some track record of his statements ? 

I still trust Pete Dodd as a leaker even if he let his emotions screw him up that one time. 

In the thread he discloses some control customization options that turned out to be true, as well as some other stuff. To be honest though I can't be bothered sifting through all 36 pages, it took me long enough to find the CPU statement haha. XD