Don't understand idea behind 10. Bb3

10. Ba2 is classed as an inaccuracy by the engine, with an eval of 0.16. It suggests 10. Bb3 as the best move, with an eval of 0.71, but I don't understand how Ba2 can be so inferior. The only thing I can come up with is that if the knight and rook move to the centre it would leave the bishop and the a-pawn undefended briefly which could lead to some kind of combination or just cost me a tempo having to play Ba2-b3 at a later stage anyway. Am I thinking along the right lines here or am I just missing something obvious?

It's just low depth bias in a early middlegame position, not very significant. After some time both Leela and Stockfish rate both moves about the same. There is an important difference though : if White wants to play the a4-a5 break, then the bishop should stand at b3, not a2. This way it will be easier to recapture on a5 if bxa5 and the b2 pawn will protected if Black plays something like Rb8 at some point.

You're right Nerwal, I wasn't even thinking about that pawn break. Thanks! I just used the depth 30 chess.com stockfish, I thought that would be deep enough the provide an accurate eval.