Did "Rapid Chess Improvement" by Michael de la Maza work for you?

The three basic approaches :

• De La Maza : repeat a set 7 times, faster and faster
• Woodpecker method : repeat a set until you get it 100% right
• Spaced Repetition : repeat a set, with priority on failed puzzles, and then spacing the interval between two repetitions further and further for puzzles you've done right

The three methods make you work on your tactics and help long term retention. Provided the puzzle set is adequate (ie. reasonably easy), they should help in some way.

Are these approaches meant to be mixed, hicetnunc?

My latest thinking is that once you do a problem set in a single day (like finishing 7-circles) you then need to repeat this (i.e. solve the entire set in a day) with the same problem set at intervals, whilst at the same time starting a new set. Hopefully then you can keep the old positions in your head whilst you try to shovel new positions in.

There's a good blog on this here, this bloke has really thought about this stuff:

http://empiricalrabbit.blogspot.co.uk/2010/11/introducing-expanding-repetitions.html

I'm certainly an advocate of spaced repetitions. 

My own method is quite flexible, but more or less I do something like this:

- puzzles 1-100 (the easiest)

- puzzles 1-200

- puzzles 1-300

- puzzles 101-400

- puzzless 201-500

etc. (comming back to the easiest ones from time to time)

To be fair, hard work is about the only way to achieve rapid improvement in just about anything, not just chess. I do admit that's often not much fun, though.

Well, the usual SRS methods (the third approach) are supposed to take into account how human memory works. Typically, the interval between two repetitions will increase reasonably fast, and you'll have to wait at least a couple of months before tackling the 5th or 6th repetition (depends on the exact parameters of your SRS set/tool you're using, but if I'm not mistaken that's the idea of the science underlying the method).

Compared to this approach, MDLM's 7 circles looks more brutal, but I have no idea if the final circles help with long-term-memory or not ?! So, yes in this case, maybe it makes sense to repeat the set a bit later.

These methods are good for learning particular positions by heart or even very similar ones where the candidate moves and solutions are going to be much the same.  What about similar positions where the candidate moves and solutions are substantially different?  To be able to do that you have to do more than just memorise.  You have to understand the position.  That is not hard when the position is not complex.  You understand the position without really trying.  But complex positions are a different story.

petrip, I disagree.  It is not hard to see potential blows as you put it.  But making them happen in complex positions requires something more than the takes, takes, takes approach.  With that approach, you are going to get into a lot of time trouble.  Anybody can learn simple geometry.  The best tacticians in fact have a feel for the position more than just seeing two pieces lined up for a potential fork, etc.

This discussion is about how to train to see relatively simple tactics very quickly, nothing more.

I have "read" it: To save your time and to sum up the message of the book: Work 8h a day on tactics and you'll become better. Well, I knew that before ...

petrip, my point is that there is training and training.  There are players who have been doing these tactical puzzles for many years and got only marginally better.  There are others who learn a lot more from the same tactical puzzles.  Mechanically doing these puzzles only gives you so much.  As to Mr Tikkanen, I am sure he did a lot more than the woodpecker method to get his GM title.

petrip, agreed.

Silman recently wrote that he got to expert level purely on tactics himself.  Yet he continues to push the idea that you need the stuff in his strategy books to get there.  I don't know how it is with most adults, but I know that with most strong kids, tactics are definitely enough to get them to the expert level.

what is expert level?

And what does it take to get the not-strong kids to the expert level?

Expert is apparently 2000 to 2200.  Tactics are no use if you don't get reasonable positions out of the opening or you face an opponent who is very solid positionally.

Yes. Great book!

Don't know.  They usually quit chess before they get close to expert level.

That book is full of brilliancies.

"Each player should choose an opening that attracts him. Some players are looking for a gambit as White, others for Black gambits. Many players that are starting out (or have bad memories) want to avoid mainstream systems, others want dynamic openings, and others want calm positional pathways. It’s all about personal taste and personal need.

For example, if you feel you’re poor at tactics you can choose a quiet positional opening (trying to hide from your weakness and just play chess), or seek more dynamic openings that engender lots of tactics and sacrifices (this might lead to more losses but, over time, will improve your tactical skills and make you stronger)." - IM Jeremy Silman (January 28, 2016)