How to practise middlegames

Sort:
jackie71

I am wondering if practising middlegames resulting from certain openings against a computer is sensible. I enjoy chaotic and complicated positions, so I don't think that playing against engines will give me the outlook that I need, and since I play openings that branch out quite a bit, I don't really have many opportunities to practise different middlegames from every single line by just playing games online. I heard there was an engine on Lichess ("Maia" if I recall correctly) that plays very "human like". Should I practise against Maia? Otherwise, is anyone down to practise some lines with me? I'm looking for D4 and sicilian players, preferably around 1700 rapid.

zix3i7im_V3
So for middlegame practice I’d maybe recommend considering lessons from chess dot com like silman’s “roots of positional understanding” collection from 1998. It’s on chess dot com and I’ve been enjoying it and think it’s very instructive. Daniel Naroditsky’s videos are pretty instructive too but they’re not strictly about middlegames and making a plan.

About Maia, I have played her before and she does play very much like a human compared to most chess playing computers but compared to most humans I don’t know. Not that I think the creators failed in creating what they set out to do. I actually think they did a very good job but I still think it’s not quite what you might be expecting. At least for some moves as it will play maybe 75% human moves but 25% of the time some computer or inexplicable move that kinda ruins the facade. And that can heavily influence the rest of the game in chess. Uhm definitely give her a try and play with her though. I’m just saying she’s not exactly like playing a human yet in my experience. I guess. Uhm 🤷🏻‍♂️
jackie71

Thanks for the answer, but I want to practise middle games that come from specific openings. I want to setup the position up to move 16, and play the game from there on. I'm not really looking for general principles of the middlegame, I'm looking for familiarity with middlegames that result from specific openings that I play.

tygxc

Sparring against a strong engine is good for progress, but bad for your ego.
Another way to practice is take a grandmaster game, take one side and cover up the moves. You make the move of grandmaster A. Now you think about your reply, preferably with a real chess set and a real chess clock and a score sheet. Think about your move, play it, note in on the scoresheet, now look what the grandmaster played and play that instead. Continue like that and at the end compare your moves to the grandmaster moves.

zix3i7im_V3
Yeah I mean in that case playing the bots is probably fine. I mean the bots will play the most critical lines and if you can win against those then you can win against weaker moves. Also bots or strong computers will know the theory for the lines/openings you want to practice. I suppose you want to familiarize yourself with all the options in that opening and how to respond to them not just the best lines. Yeah just playing bots of different strengths starting from set positions of your choosing should work fine afaik.
Kevin0157
tygxc wrote:

Sparring against a strong engine is good for progress, but bad for your ego.
Another way to practice is take a grandmaster game, take one side and cover up the moves. You make the move of grandmaster A. Now you think about your reply, preferably with a real chess set and a real chess clock and a score sheet. Think about your move, play it, note in on the scoresheet, now look what the grandmaster played and play that instead. Continue like that and at the end compare your moves to the grandmaster moves.

I think this method might be the best I've  seen. Think I'll try it myself. I also see, when I analyze my games, that I do make a lot of questionable moves in the middle game. I thought tournament play might help, and it has a bit, but I still deviate from the best moves all too often. The difficult decision to make is to begin with "which" opening. There are 500 FIDE numbered openings, and almost an unlimited number of variations. But in order to understand the nuances of the middle game, the method you suggest above, would likely be the best.

tygxc

#6
The opening does not really matter. Proof: play the good side of a bad opening against an engine and watch how the engine destroys you with middle game tactics despite its bad opening.