Up two pawns, couldn't win

I'm not talking about forcing wins where they don't exist so much as creating positions where the endgame conversion into a win is relatively simple. That is, try to have a clear endgame path before you convert into an ending, since if one outplays the opponent in the opening and middle-game, it doesn't make sense to grab a small advantage and rush into an ending in which the expertise needed is far more than should be necessary, especially if the opponent happens to be a brilliant endgame player. There's such a player on our club team. He's careful with his openings, mediocre in middle-games, and a dominant force in endings. He was an accountant all his working life, which might explain it. So it's more important to develop control over a game such that the nature of the endgame achieved is characterised by simple, winning plans. Perhaps you're relatively strong at endings yourself, I don't know. I'm relatively strong in complex situations that can't easily be calculated and which demand positional insight and also at planning and executing aggressive surprise attacks by masking them as routine moves. I'm pretty crap at endings much of the time, although I have played difficult endings perfectly on my day because I have good calculative ability if I can be bothered. I think it just reflects what we enjoy most in chess.