It does take home advantage into account. You can see that in the list of upcoming matches and the likelihood
of a win here - so it gives Tahiti a better chance of beating New Caledonia tomorrow despite being lower-rated - reason being they're at home.
First-move advantage in chess is very similar to home advantage in football; it's statistically proven to heavily influence the outcome of a game (white scores about 55% and black scores 45%).
I'm not dismissing his comments because he doesn't agree with the rating system; I'm dismissing them because he doesn't attempt to engage in it and still says that Kenny needs a result against France and stuff like that or that we drew in 2009 so we can draw again (a bizarre comparison seeing as our squad is much worse than 2009), or tries to say that there's much of a muchness between third and fifth; that sort of stuff. Elo is pretty well-respected as a measuring system and has even been incorporated into the FIFA rankings.
Elo doesn't predict individual results. What it does do is say that when you play a team 220 points lower at home (Ireland v Luxembourg), you should be winning 80% of the time (or whatever it is). And it can happen that the lower-rated team win too. If those results start exceeding chance, then something's going wrong (and indeed by the time of that Luxembourg result, we had dropped 100 points in Kenny's first ten games, which is pretty stark)
So it's not about saying France are favourites - we know that. It's about showing by how much they're favourites, and even why the comparison with the Portugal draw (even if they were up for the game in Dublin, which they weren't) isn't all that relevant because our chances of drawing with France are 40% lower.
So yeah, we can make it difficult for them even if we lose - that's fine. But when Boomers says "If we had a competent manager I'd be very confident of a draw at least", for example - well then it really helps to put the task ahead of us in perspective. That's the point of Elo.