First of all let me say that I really love Chess Insights. It's a great feature especially for looking at which openings I should focus on.
When looking at average centipawn loss and filtering by material imbalance, I've found a bug where it seems to not be counting most of my moves, especially moves where material imbalance was positive.
I'm examining Average Centipawn Loss (ACPL) by Variant. With no restrictions or filters, this shows my Classical ACPL = 37.5, based on 2,872 moves.
However, if I filter by Material Imbalance, the numbers shown don't make sense. For example, if I check the four boxes for 0 to +1, +1 to +3, +3 to +6, and More than +6, it shows my Classical ACPL based on only 29 moves!
It is impossible that I've played 2872 moves and only 29 of them had positive material imbalance.
Also, when I filter to (equal, 0 to +1, +1 to +3, +3 to +6, or more than +6), it's looking at 53 moves. When I restrict to (less than -6, -3 to -6, -1 to -3, or 0 to -1) it's looking at 794 moves. But if I remove the filter it's looking at 2872 moves. Since those two sets of filters between them include every possibility, the sum of the number of moves for the two filters should equal 2872. However, 53 + 794 is a lot less than 2872.
When looking at average centipawn loss and filtering by material imbalance, I've found a bug where it seems to not be counting most of my moves, especially moves where material imbalance was positive.
I'm examining Average Centipawn Loss (ACPL) by Variant. With no restrictions or filters, this shows my Classical ACPL = 37.5, based on 2,872 moves.
However, if I filter by Material Imbalance, the numbers shown don't make sense. For example, if I check the four boxes for 0 to +1, +1 to +3, +3 to +6, and More than +6, it shows my Classical ACPL based on only 29 moves!
It is impossible that I've played 2872 moves and only 29 of them had positive material imbalance.
Also, when I filter to (equal, 0 to +1, +1 to +3, +3 to +6, or more than +6), it's looking at 53 moves. When I restrict to (less than -6, -3 to -6, -1 to -3, or 0 to -1) it's looking at 794 moves. But if I remove the filter it's looking at 2872 moves. Since those two sets of filters between them include every possibility, the sum of the number of moves for the two filters should equal 2872. However, 53 + 794 is a lot less than 2872.