Well, let’s see what we can infer from your examples:
The 2017 is faster 5-60, 30-50, and 50-70 and only 1/10th of a second slower in the 1/4 mile.
The 2018 adds a more complex 10 speed transmission and, of course, the turbo.
Car and Driver’s tests show identical fuel mileage (Fuelly is great, but only when the sample size is sufficient. You’re looking at a sample size of 7 here, which when you take into account the fact that these samples were gathered by non-experts who may very well be reading the mpg estimator in the MID, isn’t very good data).
So, we got more complex in the engine, and the transmission, and get identical, or nearly so if we’re being generous, mileage and better performance only by the very slightest of margins and only in a drag race, in a car that’s 120 pounds lighter and, having added an underbody aerocover has a 3% lower coefficient of drag than the 2017 (source: Honda press kit).
Huh. Doesn’t look like the turbo really does much of anything, does it… Oh wait! Yes it does, because the official mpg numbers from fueleconomy.gov show the 2018 getting 2mpg better combined than the 2017, just like I said it would in the official tests that the turbo was meant to game.