These are things that people say. But the evidence carefully sampled from actual games is really compelling and the examples are purposefully chosen. Excellent work!
Huh, this is very cool - and I can think of examples of these patterns being both found and missed in my own games (pretty sure I blew a chance to remove the defender of a bishop on g5 with tempo sometime in the past week). Your point about cultivating that attitude of always being ready to find your opponent's mistake is a good one.
Very nice! Another thing is that in most of the puzzles you show, the targets are very limited, so when you know it's a puzzle, it's easy to focus in on the right idea.
The one that seems odd to me is #5, in the bottom center. The target seems less obvious to me, so I would expect it to be harder as a puzzle. And the correct move seems to be pretty in flow: the previous move wasn't something that needed to be responded to, and black's move is at least somewhat in-line with his plan. I wonder why this scored so highly?
Yes, I agree that #5 is the odd one out. I think what's going on here is we know White's last move was h3, and Black's last move was probably Be7 or a6, meaning the same tactic was possible on the previous move! So when the position in occurred in a real game, Black already missed it once, and probably didn't find it on the next chance either.
When you looked for the position in games, did you limit it to cases where the last move was the same as in the puzzle? Looking at the four most recent games for that position in the opening explorer, three of them had white's last move as Be3, so the tactic was new. One had white's last move as a3, and the tactic was available last move.
Although this being true even a decent percentage of the time could be enough to tip the scales in favor of this position.
I suppose this is probably inevitable. As you say, you should be focusing on what changed with the last move. Particularly in blitz/rapid, you're pulling your focus away from things which didn't change, and this is normally going to be a good tradeoff, but not always.
Hmm, no. For the games I just looked at the position, but the puzzle includes the previous move, so that is a potential source of error or at least confusion.
These are things that people say. But the evidence carefully sampled from actual games is really compelling and the examples are purposefully chosen. Excellent work!
Very helpful and well researched!
This is an excellent Article!
Huh, this is very cool - and I can think of examples of these patterns being both found and missed in my own games (pretty sure I blew a chance to remove the defender of a bishop on g5 with tempo sometime in the past week). Your point about cultivating that attitude of always being ready to find your opponent's mistake is a good one.
Very nice! Another thing is that in most of the puzzles you show, the targets are very limited, so when you know it's a puzzle, it's easy to focus in on the right idea.
The one that seems odd to me is #5, in the bottom center. The target seems less obvious to me, so I would expect it to be harder as a puzzle. And the correct move seems to be pretty in flow: the previous move wasn't something that needed to be responded to, and black's move is at least somewhat in-line with his plan. I wonder why this scored so highly?
Yes, I agree that #5 is the odd one out. I think what's going on here is we know White's last move was h3, and Black's last move was probably Be7 or a6, meaning the same tactic was possible on the previous move! So when the position in occurred in a real game, Black already missed it once, and probably didn't find it on the next chance either.
When you looked for the position in games, did you limit it to cases where the last move was the same as in the puzzle? Looking at the four most recent games for that position in the opening explorer, three of them had white's last move as Be3, so the tactic was new. One had white's last move as a3, and the tactic was available last move.
Although this being true even a decent percentage of the time could be enough to tip the scales in favor of this position.
I suppose this is probably inevitable. As you say, you should be focusing on what changed with the last move. Particularly in blitz/rapid, you're pulling your focus away from things which didn't change, and this is normally going to be a good tradeoff, but not always.
Hmm, no. For the games I just looked at the position, but the puzzle includes the previous move, so that is a potential source of error or at least confusion.
A good statistical proof of the puzzle bias! Puzzles are solved better, when players know that they are puzzles.