"The higher compression rate the engine has, the higher octane number is needed" is definitely part of the story, the broader part of the story is that the more prone to detonation an engine is, the higher the octane of fuel you want to put in it. A relatively low performance, low compression engine with a poor combustion chamber design might well need higher octane fuel than a much higher performance engine with better chamber design and cooling.
In some engines higher octane fuel does give more power and in some engines it might just reduce the chance of it going boom!
Well, if the thread starter's MW3 does not suffer from pre-ignition/detonation and has a modern engine with knock sensor set up for 96, I cannot see any benefit in putting in anything higher - why try to fix a problem you do not have. But then there is the ethanol issue. As far as I know 96 octane has 5% ethanol mixed in, whereas 98 has zero. True, ethanol can act as a solvent and be detrimental to plastics, rubber, gaskets and so forth. But I do not think this applies to a modern engine. I suppose the engineers are very well aware of what is put into the tanks. On older cars or veterans it might be an issue though. And ethanol is hygroscopic, true. To this day I have never heard of a problem related to (these obviously small amounts of) water in the fuel, but thinking about it, I am considering filling the tank with 98 for winter storage but stick to the specified 96 for everyday use.
