Why is this position -2 points in favor of black?

Lol. He is the neo of the chess matrix.

My big question is why did the OP allow his opponent to do this to him?? Haha.

As others have already mentioned chess engines can (and often do) evaluate the same position by considerably different margins...sometimes by as much as .50 or more I've heard. And evaluations can change somewhat (or even drastically) depending on the depth you let them calculate at. I'm not sure what engine chess.com uses or to what depth it typically calculates out to, but these days, I believe Houdini is considered by many to be the "best" chess engine.

That said, I agree that black is clearly better in this position. Let's consider all of white's positional cons:
1.) Bad light square bishop.
2.) Correspondingly weak light squares (a consequence of the above point). Moving the white knight could potentially allow Qd3 at some point, for example.
3.) Passive knight with no obvious outposts/ways for it to improve. Unlike black's knight.
4.) Open a- file that black can make use of more easily than white. Black's rook, for example, could easily improve by going to a8 and potentially invading along the file. White can't do anything similar, particularly because black's queen is stopping Ra2.
5.) c3 pawn appears to be hanging in the current position and indeed, seems to be a big liability. The most "natural" ways of defending the pawn are Bd2 and Rc2. The former interrupts white's connection to the a-file and hence makes black rook invasions along it more attractive/possible. It also potentially allows ...e3 (deflecting the bishop from c3); Bxe3,Qxc3; Qxc3,Nxc3; Rc2 (hoping for ...Nd5; b5),Nb5! Black has now dissolved his e- pawn and now white's the one with a weak pawn on d4 (if not b4 as well). Which black can easily apply more attacking pressure to with Bb6 and Rd8. And black's control of d5 means white's not dissolving that weakness any time soon.

Rc2 prevents e3, but basically abandons any ideas white might have had of playing f5 and trying to go for kingside counterplay. In other words, it abandons any semblance of activity black might have hoped for.

Is all of this worth a -2 evaluation? I wouldn't think so. But I could definitely see a -1 or greater.

Also, keep in mind that I'm far from a chess authority by any means, and everything here has come purely from my ol' brain itself...no engines, in other words. So any mistakes in position evaluation are mine alone.

Maybe the idea of having the rook come to a8, then to a2, then pushing e3 forcing the bishop to take allowing rook takes rook, bishop takes rook then knight takes pawn, undermining and threatening the b and d pawns while also threatening Ne2+, winning the queen.

Well, since this whole thing came from a comp...we're still waiting for the OP to post those vars...

If you're curious, here's how it can get to that position:

The chess engine I used is Stockfish-231-64. Variations are:

-2.30 22. Bd2 g6 23. exf6 Rxf6 24. h4 Rf8 25. g5 hxg5 26. hxg6 b5 27. Ng2 Qd3 28. Qe1 Ra8 29. Qf1 Qxf1+

-2.50 22. f5 exf5 ...