If you don't solve a tactics puzzle is it re-introduced later? Maybe you are simply seeing people solve it correctly with a second or third chance later.
Nah I don't think this ever happens, unless you go through all 60,000+ tactics problems, which very few people have done.
To give an example of the kind of problem I'm talking about: I found this puzzle extremely difficult and it took me a long time to solve, yet the success rate is 50%! Like is it really true that 50% of people are able to solve this?!
Have you ever seen Hikaru do tactics? He gets 2300 puzzles in just a few seconds. It's ridiculous. I assume this puzzle isn't given to beginners... so yeah, if the tactic's rating is correct, and it's only given to people e.g. 200 +o r - that rating, then exactly 50% of people should solve it.
Calculating for 15 minutes means you don't understand the pattern... which is great! It means the puzzle can teach you stuff, but someone who knows more of the relevant patterns should be able to solve it pretty quick. It took me ~5 minutes to be sure the solution was right, but had the first move as a candidate pretty quickly. It'd probably take Hikaru 5 seconds lol.
IMO it's good to give yourself a time limit. I'd say 15 minutes should be the maximum UNLESS you want to train calculation/visualization too. If you want to train that, then sure, spend 30 or 60 minutes on a single position, that's fine... but if this becomes a habit it may screw up your time management in real games. For that reason, and because tactic training is about learning new patterns, I like a time limit of around 10 minutes. Spend the extra time reviewing the solution and try it again a few days later.
And part of a low clear rate is tempting alternatives. For example there was no 3 move line to win the rook for your knight, there are no (good) checks, and the rook is the only loose piece... so the ideas for the puzzle are pretty limited. People who get stuck might be able to calculate their way to a solution even though as you found out that's not easy!
To give an example of the kind of problem I'm talking about: I found this puzzle extremely difficult and it took me a long time to solve, yet the success rate is 50%! Like is it really true that 50% of people are able to solve this?!
Yes, sometimes I'm also surprised by high success rates.
In this case, I made the correct moves so in the statistics I would count as "solved". But to be honest, I didn't calculate to the end. Bf3 looked just obvious to allow Ne4. But that was all of my thinking.
So, you can "solve" those puzzles without understanding them from the start, which increases the success rate.
Hi Stiggling, good stuff.
Well, if you know most of the usual tactic patterns, it might strike you as a surprise when Chess.com tells you to mate him out of a seemingly equal position. One of the reasons why even masters miss tactics is because they’re not looking for it - the position doesn’t really look like you’re going to Smother-Mate him anytime soon.
Try solving the crazy, Mikhail-Tal-ish tactics that promote creativity. It’s not all that easy - there’s a reason he was world-famous for tactics. See above.
(Note: in Game 1 Tal’s opponent resigned instead of getting mated.
Eg, I got it in 110 seconds, and Tal would’ve got it in about 10 seconds (probably less).
Because only the very best even get to attempt very difficult problems and since those attempting the problems are very good, they solve the problem half the attempts.
exf6 is an easy candidate to spot, but Nxe6 2 moves later is amazing to me. I could probably learn some good stuff by going over that game carefully. The 2 bishops + e7 pawn for the queen looks totally lost for white.
Ok the engine says maybe it's just equal, but still, that's a lot better than it looks at first (to me anyway)
harder puzzles have higher elo which means its given generally to higher elo players, so if this puzzle had a difficulty of 2300, then its more likely to be given to a 2300 players to do then a lower player, so its reasonable for those groups with the appropriate skill level to get the answer correct at a reasonable frequency while a generally lower rated player might find it extremely hard.
harder puzzles have higher elo which means its given generally to higher elo players, so if this puzzle had a difficulty of 2300, then its more likely to be given to a 2300 players to do then a lower player, so its reasonable for those groups with the appropriate skill level to get the answer correct at a reasonable frequency while a generally lower rated player might find it extremely hard.
Agreed.
Sometimes I'll spend 15+ minutes solving an extremely difficult tactics puzzle. I'll eventually figure out this epic 5 move combination, and I'll be super proud of myself for getting it right, only to then realise that the puzzle has a 50% success rate.
Is everyone on this site a genius or something? Like I'd think that the success rate on some puzzles should be less than 5% because they're so incredibly complex, yet people always seem to know the right answer.