x
Chess - Play & Learn

Chess.com

FREE - In Google Play

FREE - in Win Phone Store

VIEW

Training to Instantly Solve Tactics Problems

Chessmo
Mar 17, 2014, 4:16 PM 18

"One of the primary goals of studying basic tactics is not just be able to solve simple and recurring problems, but to recognize both the position and the solution almost instantly."

--NM Dan Heisman from A Tactics Quiz

Over the past eight months, I've gotten much better at solving moderately challenging tactics problems and recognizing, and avoiding, my opponent's simple, 2-3 move tactical threats during games. I spent tremendous time and energy focusing on my basic tactical abilities from July to December of 2013.

So, it came as a shock that I still stink at instantly recognizing a basic tactical position and its solution, according to NM Dan Heisman's Tactics Quiz. I scored a "tactics rating" of 384, which is about 1200 points below my official USCF rating!

The tactics quiz presents the reader with 12 basic tactical positions, and the score is determined by both how many questions one gets correct and how quickly the quiz is completed. Though I got 9 of the 12 problems correct, I took a seemingly reasonable 14 minutes to complete the questions. It wasn't until I started calculating my score did I realize just how much of a time penalty was assessed for every 2 additional seconds one takes. Let's just say it is a lot!

After my anger at NM Heisman subsided for coming up with such an absurd scoring mechanism, I took a deep breath and thought about it. What he is telling me with this quiz is that I am still solving many basic tactics problems instead of instantly recognizing the solution. And more importantly, if I didn't address this once and for all, it would continue to hamper my chess progress long into the future.

You see, the theory is that if you can instantly recognize, for instance, a 2-3 move mate, or a 2-3 move knight fork, you will have a better chance of solving more complex tactics problems because you will be able to see the more basic version hidden within all that extra complexity. Once you recognize that basic tactical position, you can then just figure out how to rearrange the pieces into that known position.

I've been using Tactics Trainer for many years (over 3600 problems solved) but I've been measuring my success by my accuracy. I often will take 4-5 minutes to solve a problem that has an average time of only a minute. Another problematic function of Tactics Trainer is that the better one does, the harder the problems that one is fed. But in order to master basic tactical problems I needed basic problems, not ones that constantly adjust to my current skill level.

Then I remembered that one of my chess.com friends, Benedictine, told me a Tactics Trainer method that he used, where he sets it to "unrated" and then selects a rating range for the problems that is several hundred points below his current rating.

Since I like training based around systems and games, I like to record and measure my training progress, and I like to add scoring mechanisms--this is what keeps training fun. So, I took Benedictine's idea and added some more structure around it.

So, here is my new basic pattern training method that I am doing for 30 minutes each morning. I set Tactics Trainer to unrated and set the maximum range to 1000 (my current tactics rating is about 1550), so I am only served up problems rated 1000 and below--basic problems. Then, I solve as many as I can in 30 minutes, manually recording the overall start and end time, the number solved, and the number missed.

I plan to do this every morning until I can complete 90 problems with a 90% accuracy within my 30 minute time limit. Doing the math, this gives me an average of 20 seconds per problem.

Now for the gamification. When I accomplish my goal at the 1000 rating point level I get promoted to do problems at the 1100 level! Once I finish at the 1100 level I move up to the 1200 level, etc. I imagine it will take me several weeks or more to move up each level.

90 problems in 30 minutes at a 90% accuracy is very aggressive. But the problems at this level are also very easy. I've only been doing this for a few days now and I'll let you know how it goes.

Online Now