Don't forget that what is easily refutable by a super GM (or better) rated computer engine probably won't be by your human opponents. Engines make us all feel inadequate, but luckily that's not who we play.
depressing annalysis
for example, there was an article on this site about how some of Tal's (a great chess player back in the 60's) great chess sacrifices were easily refuted by the computer, but that it would be practically impossible to find the defense to them in an actual game by a human. So what the computer found easily would be very hard for Tal's opponents to find. So don't get too down on urself.
I wouldn't say "practically impossible", I'm sure Carlsen or Kasparov would have often found the refutations, which were typically later found after the game (and sometimes on the board, I imagine those were left out of the Tal's Glorious Sacrifices book).
However I agree with the main point. The computer doesn't understand ideas or systems. If the computer can calculate that some obscure move now would be +0.3 later due to some 30-ply calculation, and your opponent clearly doesn't see this freakish move combination, then really what worth is it for the computer to come up with that. You are better off not knowing about that unrealistic combination because it could limit your move options.

http://www.chess.com/livechess/game?id=1082653503 the game where stockfish said i made two inaccuracies and a blunder
well lets face it , i am no gm but i have been shocked lately when i started putting my best game in stockfish well there are things i can improve but basically , this gets deppressing like how can i get my courage to play back and improve after knowing my best play was easily refutable ?