- Solve tactics
- Play at a pace that doesn't leave you with a big time deficit at the end
- During the game doubled check your candidate moves for errors
- A usual way to catch errors is calculating forcing moves (checks, captures, threats). Try to punish your intended move as if you were your opponent.
This is an example of a game I just played where I was up a knight in the middlegame
A pawn on f3 is a great attacking aid, and your pieces were uncoordinated, so the blunder didn't happen in a vacuum. Notice you never once had your rooks connected on the back rank. A move like 24.Bb2 or 31.Bb2 will be pretty automatic once you get more experienced. Moves like Re3 and b4 will be avoided.
I lose like you do, from winning positions, but for me it is always because of blunders. 9 out of 10 games I lose in rapid is simple because of big blunders. Even in the games I win, I do blunders, but they are just not enough for losing.
Your case seems different, if I take account the game you posted. In that game, it seemed like your opponent just played better to be honest. He sacrificed his knight, for a strong attack to a player rated lower than him. Of course, he risked a little, but he wasn't playing vs a titled player, nor a very long time limit, which could prevent him from succeeding with his attack. Maybe, it wasn't a good example the game you posted. Even with a piece up, I would be really scared with the position you got and for sure, wouldn't think that I was winning, except maybe in a daily game (were u can analyze a lot). Your king's position, your opponents pawn on f3, your missing bishop... looks like a horror movie to me. I admire your confidence, thinking you were winning.
Why 19. g3? You could have taken the Knight with the bishop. g3 also weakened your kings defense. (I am a beginner, maybe what I'm saying is nosense)
Why 19. g3? You could have taken the Knight with the bishop. g3 also weakened your kings defense. (I am a beginner, maybe what I'm saying is nosense)
It is 
Because the knight was defended by the queen. The way you say it, he would just exchange his Bishop for opponents Knight, but the way he did it, he took the Knight for nothing 
. It worked, because the knight was pinned. Of course, he weakened his position in the process, but with a long time limit and accurate play, he could hold on and win.
Thanks for the explanation dk-Ltd
In addition to the excellent advice from godsofhell, keep in mind this principle when you're up a piece in the middlegame: try to simplify to an endgame by exchanging pieces (but not pawns). The fewer the pieces left, the more your extra piece counts, and eventually when you have just one piece and pawns left vs. your opponent's pawns, the win is easy as little counter-play is possible.
In practice this means that when you are a piece up in the middlegame, you can afford to be more aggressive and go for an attack, because the best defence against an attack (sometimes the only defence) is to exchange off the attackers. And that falls right into your plan of exchanging to an endgame.
This is why, at a higher level, players resign immediately after losing a piece with no compensation. The player with the extra piece can use it to their advantage in any sort of attack (or defence), obviously. Now combine that with how the other player can't defend against an attack without exchanging to a hopeless endgame. It's literally a lose-lose situation, so there's no point in continuing the game. But until you reach a level where your opponents would resign in such a lost position, you will have to practise how to exploit the extra piece...
I lose like you do, from winning positions, but for me it is always because of blunders. 9 out of 10 games I lose in rapid is simple because of big blunders. Even in the games I win, I do blunders, but they are just not enough for losing.
Your case seems different, if I take account the game you posted. In that game, it seemed like your opponent just played better to be honest. He sacrificed his knight, for a strong attack to a player rated lower than him. Of course, he risked a little, but he wasn't playing vs a titled player, nor a very long time limit, which could prevent him from succeeding with his attack. Maybe, it wasn't a good example the game you posted. Even with a piece up, I would be really scared with the position you got and for sure, wouldn't think that I was winning, except maybe in a daily game (were u can analyze a lot). Your king's position, your opponents pawn on f3, your missing bishop... looks like a horror movie to me. I admire your confidence, thinking you were winning.
I felt that I was winning. It was a rapid game (20 min). during the engine analysis, the engine gave me a +3 advantage, so i guess i was winning. but you're right, i was kinda scared of hus queen's position and as @godsofhell1235 pointed out, that pawn of f3 was deadly. I tried to get rid of that pawn, but it defended by the rook on f8 most of the time, so i couldnt do much about it. The engine said I was winning the game until I played 34. R3b2 which made go from winning to losing
Why 19. g3? You could have taken the Knight with the bishop. g3 also weakened your kings defense. (I am a beginner, maybe what I'm saying is nosense)
well, cuz I felt that I was winning a piece and also he was threatening checkmate on g2 and I didnt want to trade my bishop for a knight
To maintain your advantage deeper into each game is progress lose or draw.
In addition to the excellent advice from godsofhell, keep in mind this principle when you're up a piece in the middlegame: try to simplify to an endgame by exchanging pieces (but not pawns). The fewer the pieces left, the more your extra piece counts, and eventually when you have just one piece and pawns left vs. your opponent's pawns, the win is easy as little counter-play is possible.
In practice this means that when you are a piece up in the middlegame, you can afford to be more aggressive and go for an attack, because the best defence against an attack (sometimes the only defence) is to exchange off the attackers. And that falls right into your plan of exchanging to an endgame.
This is why, at a higher level, players resign immediately after losing a piece with no compensation. The player with the extra piece can use it to their advantage in any sort of attack (or defence), obviously. Now combine that with how the other player can't defend against an attack without exchanging to a hopeless endgame. It's literally a lose-lose situation, so there's no point in continuing the game. But until you reach a level where your opponents would resign in such a lost position, you will have to practise how to exploit the extra piece...
yes, I'm actually aware of that principle. it just wasnt that easy to exchange pieces in this game
It's starting to get very annoying. Lately, in most of the games I play, I get a winning position (being up a piece in the middlegame). after that, I make these big time blunders to quote the engine "from winning to losing". Its very frustrating. It's like, in the start of the game, I play fairly well and get an advantage and somewhere near the end, I make the losing move all of a sudden. I'm not able to stay stable throughout the game and have lost many many games because of this problem. How do you overcome this problem?