COOS & NSXT hit most of the points although I could nitpick the analogy to plug wires etc. However, every unbiased scientific dyno test I've seen over the years (just a few) have been unable to confirm a meaningful HP gain on a normally aspirated engine. Granted, that's in a shop for short runs and in some cases on an engine dyno where the engine is out in the air, but the last test actually showed a net loss throughout the range. (They attributed that to the fact that the header was properly tuned to start with and the inside coating slightly reduced the diameter.) And that was with a mild steel header! Though not nearly as good as coatings, SS headers are far better than plain steel at holding in the heat.
As for intake temp, you shouldn't be drawing air directly from the engine compartment so that 100 degrees is not meaningful. It may translate into a few degrees of IAT at most as the air races through the intake. Mind you, lower temps under there are sill a good thing, but probably not for that reason.
The most likely benefit is on a turbo motor where retaining heat in the exhaust helps to spin the turbo up faster, but I've never seen a proper study on that. Bottom line, theories are great and people with lot's of book learning will gladly tell you all kinds of things, but the real world invariably has factors not accounted for by textbook theory.
