Training like kotov!

I've recently been reading the classic think like A Grandmaster by Kotov and have found the section on analysis very interesting. Kotov recommends playing through published games in the normal manner by moving the pieces on a board and when the positon begins to enter a more complex phase to stop.
he then recommends without moving the pieces, to calculate and visualise all the important lines and sub lines and then compare your findings with that of the author.
The technique certainly sounds a great way to improve. However, my game collections including my most treasured by Alekhine and Tal only rarely seem to go into the sort of deep analysis that Kotov is referring to.
i think this is the case with a lot of published games and it's obviously impractical to include all the various continuations that may have been considered by the player concerned.
Anyway, i was wondering whether anybody could recommend a good collection of games where at least once in every game there are reams of variations to comapre my analysis with.

I've looked at Igor Stohl's modern chess masterpieces on Amazon and it looks like it may fit the bill but without being able to browse it's difficult to tell.

I think Stohl’s book is a candidate. Also, Nunn’s “Understanding Chess Move by Move” and Shirov’s initial “Fire on Board” (I’ve not seen his 2nd book). Lots of variations in these. Maybe Kasparov’s “The Test of Time” too.

Have you read “Attack and Defence” (Dvoretsky/Yusupov)? This has some initial chapters on calculation that give an excellent follow up to the “raw” Kotov theory. Some other books are good too, but this is my favorite.

When practising “Kotov type” analysis, it’s better to complete your analysis before writing anything down. This means that you’re not using pen and paper as a crutch, and you have to try to get the lines clear in your mind so that you are able to write them down as a final step.

It is useful to choose games which you have annotated with variations in a book. But it’s also useful to check your analysis with a computer. Of course, a main factor is whether your analysis is correct or not. However, you also need to pay attention to the efficiency of your analysis. What order did you analyse the lines in? Did you waste time analysing a move before realising a much stronger candidate move was available? Did you analyse some lines too deep instead of evaluating sooner? Or maybe the opposite, not analysing deep enough? Did you jump between lines too much? etc.

Originally posted by Talisman
I've recently been reading the classic think like A Grandmaster by Kotov and have found the section on analysis very interesting. Kotov recommends playing through published games in the normal manner by moving the pieces on a board and when the positon begins to enter a more complex phase to stop.
he then recommends without moving the pieces, to calculate ...[text shortened]... it looks like it may fit the bill but without being able to browse it's difficult to tell.

Originally posted by Varenka
I think Stohl’s book is a candidate. Also, Nunn’s “Understanding Chess Move by Move” and Shirov’s initial “Fire on Board” (I’ve not seen his 2nd book). Lots of variations in these. Maybe Kasparov’s “The Test of Time” too.

Have you read “Attack and Defence” (Dvoretsky/Yusupov)? This has some initial chapters on calculation that give an excellent follow ...[text shortened]... r? Or maybe the opposite, not analysing deep enough? Did you jump between lines too much? etc.

Originally posted by Talisman
I'll certainly check out attack and defence and i'm also keen on looking at Tisdall's Improve Your Chess Now which is supposed to add a common sense edge to kotov's theory on calculation.
.