Upgrade to Chess.com Premium!

How to get the most out of the Tactics Trainer!

It's been a while since my last blog update so here's a much longer (and more interesting!?) one than usual.

I've not been playing as much chess recently as I would've liked (a few rapidplays and a blitz event which I managed to win notwithstanding), partly because I've been busy coaching and partly due to a first holiday abroad (without playing chess) for years!

Anyway instead of boring you all with one of my recent games I thought I'd do something a bit different and talk about how to work at tactics. In particular the Chess.com Tactics Trainer is a really useful tool which allows players to keep sharp and work on improving that part of their game. 

Still, it got me thinking - will someone's tactics improve just from doing 10-15min tactics every day? I'm not sure, it can't be a bad idea and it will probably stop you getting worse/losing sharpness but will your tactical ability actually improve?

I have been trying something recently with my students where I set them puzzles (sometimes ones more difficult than they could normally manage) but I give them a set of rules to follow when working on the puzzle. So far (in obviously a limited sample) it seems to improve their calculation - at least at the time.  Anyway, time for the rules:

1. Find all of the Forcing Moves (Checks, captures and strong threats (of mate, to the Queen/Rook etc).

2. Examine them all, one by one, looking at the Most Forcing ones first (Checks are the most forcing as by definition they have to be stopped). Double-checks are more forcing still!

3. Make sure you examine all of your opponent's sensible (and in particular forcing) replies.

4. Don't finish analysing a line until there are no more forcing movesAt this point stop and Evaluate. If it's mate or you're clearly winning then you can stop there - if not then move on to your next forcing line.

Obviously this is easier said than done and if you have to calculate many moves ahead it's easy to lose sight of the position but I still think it's important to have these guidelines to work with! Anyway, let's see how it works in practice. Take today's Daily Puzzle (6/28/2012).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ok, that one was unusual as it was just a forced mate and it was solvable by just looking at the most forcing line. Let's have a look at another example, this time from a game:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 Ok, that was rather more difficult but I'm going to end with a very tough one. First let's see a position one move before the puzzle starts:

 

Ok, so now we're ready for the final puzzle! Good luck!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ok, please let me know if you find these rules useful - try them out on tactics trainer (although I am aware that there is a time penalty but I always think getting the answer is the main aim). I will be writing another post soon about how this can be applied to actual calculation during a game (rather than when doing tactics). That's all for now - see you soon!

Comments


  • 17 months ago

    adamstask

    Question for  Mischa: How can I go about rote learning forcing move sequences. I'm a beginner, I've been doing online tactics training for a few months, and I've gotten a bit better at it, and I think its helped my game. I would like to know how to study forcing sequences by rote. Is there a book or a website or a method to do this? Thank you. 

  • 17 months ago

    adamstask

    Question for  Mischa: How can I go about rote learning forcing move sequences. I'm a beginner, I've been doing online tactics training for a few months, and I've gotten a bit better at it, and I think its helped my game. I would like to know how to study forcing sequences by rote. Is there a book or a website or a method to do this? Thank you. 

  • 19 months ago

    chessatore

    Hey diogens, thanks a lot for your feedback. You may be on the right track with doing the tactics problems without timer to really calculate all variations as deep as possible. I have not tried that yet, as I considered the time somewhat a good simulation of a real game situation where you must make decisions under time pressure as well. But with my calculation speed I never manage to really fully calculate through a problem in time. So maybe that's why I dont get better.

    Another explaination could be that 48 hours of tactics training are just not enough to achieve a notable improvement. Maybe I gotta shoot all the way to 1000 hours to really see results. Sounds a bit scary tho, with one hour a day that would take like 3 years. :P

    Any other feedback of players? Maybe someone who did achieve a significant improvement willing to share his approach.

  • 19 months ago

    diogens

    chessatore Sure IM or NM can give you better advice about training methods.

    Looking at your graph the days you were not in top form the rating dropped dramatically. Then, round the middle of august you stabilized in the 2400-2500 range and probably you burnt out because it dropped again. The last week seem get stable again in the same range.

    Why you don't rest? i.ex. train each other week. The week you are resting tactically go for some theoretical endgame study, videos, openings but nothing very stressing.

    They are many who think that 3 intenssive months of hard work will push them to another platform. But then you burn out, the rating drops some hundred points of your peak and you think you can't do better, therefore, disappointment. I believe if you try the each other week method during 1 year you will see real improvement. Anyway, 2500 is quite high and quite a few of titled players hold with this tactical ability. Another reason to work other aspects of your game.

    I am much lower rated than you and this summer I almost didn't puzzled at all. But I have tried to train calculation because the CC TT is quite pattern recognition oriented. So I just looked at certain positions without time bounds and try to calculate variations as wide and far as I could (and write them down). This hasn't reflected in any rating but I find now that I am capable to understand better and find some interesting lines that wouldn't have occurred before.

  • 19 months ago

    chessatore

    After a long break from chess, I decided to make an effort to improve my game. I figured that sharpening my tactical vision would have the greatest impact. So three month ago, I started using the tactics trainer regularly but unfotunately there is no visible development. It is basically stagnating around the same level.

    Here is the chart:

    Now, I am asking myself what could be the reasons for this rather disappointing result and especially wonder if there are other training methods that are more to the point. I red that the russian chess school has special training methods to improve board visualization. Any comments and suggestions are highly apreciated as I dont feel like giving up just yet.

  • 20 months ago

    forrie

    I would also like to add that the one rule that work very good for me is to look individually at both your own and your opponents pieces to see where they can move.

    If you see where they can move and where not you will see the threats. Then combine the movements of the pieces to calculate tactics - pay particular attention to move order in your calculation.

  • 21 months ago

    nyLsel

    Nice blog!

  • 21 months ago

    NM ozzie_c_cobblepot

    I only think rules are useful for TT to the extent that they can be used in OTB chess. Personally I play better when I have less process to go through per move - so a checklist of "ok now I'm supposed to find the forcing moves, okay now look at it, ok am I done, now evaluate, ok now what was the next forcing move, okay look at it, etc." involves too much context switching and not enough chess playing. What if your process fails to include asking the question "what would my opponent's next move be?", or you just forget to ask that question.

    Instead, I prefer a method where you try to learn patterns over process. The only process thing I do at the beginning of every tactic is just to count material. So if I'm down an entire rook, I'll avoid tactics which win an exchange :-).

    One example of trying to learn patterns is asking the question "are any of my opponent's pieces unguarded?". I can recall a TT problem, rated fairly high, where there is a rook on a3 unguarded. And sure enough, the tactic involves <blah blah blah> ...Qc1+ picking up the loose rook.

  • 21 months ago

    IM Trendle

    I have now written the follow-up blog and it can be read @ http://blog.chess.com/view/anand-gelfand-amp-what-is-a-critical-position

  • 22 months ago

    IM Trendle

    I'm glad people like the article and thanks for the feedback in general about the Tactics Trainer.

    diogens I am going to cover things like the rule you suggest in a follow-up blog/article which will be much more about applying this idea to real games. Watch this space!

  • 22 months ago

    Mate91

    I can do only 3 TT a day, but I always do them and I enjoy it. I will probably become a member later. I like the time limit a lot, I think mate in 3 situations should be recognized without calculation, this teaches players to think fast and recognize obvious patterns based on position. I did not have much problems with the time limits so far, it is a good feature.

    Nice article, I much appreciate it! I love the 3rd puzzle it was the most challenging.

  • 22 months ago

    diogens

    About the TT feature in this site, I heared many criticism about the clock and penalty points received for being to "slow". I don't agree.

    I more often train at CT (a site focused on puzzling). There are two modes:

    Standard: you receive the same +/- points no matter the time. Some people play this mode rather quickly but I take my time, no matter if I need 10/15 min., I'm training caculation

    Blitz: with time. Must play really fast. Here you train pattern recognition

    I would say chesscom TT is somewhere in the middle. There is more time than in blitz but you need to recognize patterns, if you relay just in calculation from scratch you'll probably get penalized. Has happened to me many times but I rather play something planned than a random move. If you loose too many points for being too slow, you must improve your pattern recognition, thus, do more puzzles. As your rating decreases you should get easier puzzles until the point that you aren't "too slow" anymore. I would say that chesscom TT is focused on playing rapid games, 20 to 45 min per game.

  • 22 months ago

    diogens

    IM, I would add a first rule:

    1. See if your partner move has created any weakness (the most common, a hanging piece or pawn)

    About tactics, I read a recent article from GM McShane where he assesses that the saying "chess is 99%" tactics is 95% true. In my short experience OTB playing against a 1700-2000 pool, I would say that more or less, we have a similar tactical ability. But if you want to improve to the +2200 pool a big step forward must be taken in this matter. I see that here in the TT rating, FIDE players +2100 have TT rating of +2500.

    Tactics = Pattern Recognition + Calculation

    If you spend the whole game calculating non stop engine like, on move 20 you'll probably burn out. When we face a puzzle "we know" that there is something there to be solved. But in a real game there's no wendy to advert you "start calculating, there is a tactical shot right now". And that's why pattern recognition is so useful. It tells you when to start caculating as deep as you can and when to play more swiftly.

    To improve your pattern recognition I would say, do as much puzzles as you can. 100 weekly puzzles is a good target (I'm lazy and do 1/3 of that). But not only puzzles, go through games, openings, endgames, all of them have recognizable patterns and will save you a lot of time during the game.

    For caculation I would apply your method. Few more difficult puzzles and dive deep in the position. But not too many because you'll get tired and start messing up your thoughts. All this obviously depending on the capacity of the pupil (age, physical shape, energy, personal skills...)

  • 22 months ago

    Ironknight777

    Thankz for the insight. 

  • 22 months ago

    P_G_M

    Beachdude67 

    I agree 100% with you. The time feature of the tactics trainer should be removed, it discourage people from using it. I have only use the tactics trainer once because I did not like the time feature, tactics should be graded on how precise you can solve the problem, not how fast you can solve the problem, in a tournament game you have plenty of time to solve a tactical mating combination most probably involving a sacrifice, if you calculate wrong then you will lose the game, there is no second chance in real games.

  • 22 months ago

    Czechman

    My problem isn't so much solving the problem, it's doing it within the time constraints. It's frustrating to find a mate in 4 and only get one point because I didn't do it fast enough. Frown

  • 22 months ago

    Mischa

    Personnaly, I think it's much more productive to learn say 100 forcing sequences by heart than solving thousands of tactical puzzles.  The real art in chess regarding calculation, is in checking forcing moves in ' normal ' positions.  Mating combinations between 2 - 4 moves should be counted as a fail in study, if you have to put your calculating hat on.  A little bit of rote learning may not sound sexy but it is the only way.

  • 22 months ago

    kludvig6

    Regular members like me can only do 3 tactics a day. How is that going to help?

  • 22 months ago

    Beachdude67

    Actually Trendle I went straight back to the tactics trainer to test them out and yeah, I did find them useful! One of my weaknesses is that I will sometimes see a line that appears to go nowhere and dismiss it. Often these are the most forcing lines. So placing an emphasis on forcing lines is good.

    One drawback I do see with the tactics trainer on chess.com is the time feature. In a tournament game it would be normal for a player to have 5 or 10 minutes to examine and analyze lines. The chess.com tactics trainer will actually punish you if you spend the time making an analysis. I don't particularly think that this is a good thing. I spent several minutes last night analyzing a position in the tactics trainer, found the elegant solution, and wind up with -6 points because I took too long to do it.

    Please don't interpret my earlier posts as being critical of your blog. It was really, really excellent :)

  • 22 months ago

    IM Trendle

    @Beachdude67 Yes, but the question is do/would you find the rules useful in your work with Tactics Trainer?

Back to Top

Post your reply: