I unearthed new patterns, not only usable for double attacks, but for other themes as well. Which is logical, because the problems are only 2 moves deep. Double attack + 1 other theme, or 1 other theme + double attack, idealiter.
I take my time to ingrain the new patterns. To grow a feel for it. Even after training for 26 days in a row, the new feel is still a bit wobbly. It is still easy to be distracted by old habits.
Taking one month per theme seems to be about right. But I expect I can't do without some exercises to maintain the newly learned patterns in the future.
So far, it seems to work. I won al my games at the club by tactical means, and the new patterns play a role in every game. So far, the competition hasn't been very stiff, though.
 |
| Black to move |
[solution]
New pattern: get a feel for connecting the dots between:
- target 1
- target 2
- attacking square
- attacker
Today I'm going to start with a new theme: exposed king.
Which is a mate in the middle of the board, in 90% of the cases.
An interesting problem, even if rather simple. Two possible variations immediately appear to be viable.
ReplyDeleteThe first (and most obvious) variation is a fork of the two Rooks with 1. ... Qg4. Unfortunately, that is followed by an immediate realization that White can defend both Rooks with 2. R(h5)d5, removing that Rook (WRh5) from attack and protecting the other Rook (WRd1). There is no Black followup that produces a "killer" on the next move.
The second variation took a few seconds longer to "see". Why? Because one of the "targets" is an empty square AND it requires combining two tactical themes. The two motifs in the position are (1) an unprotected piece [WRh5; WRd1] with a "clue" provided in the "fork/double attack" tactical theme under study, and (2) the possibility of a back-rank mate [the "+1 other theme"]. Combining the two themes, it becomes "easy" to find 1. ... Qf7 threatening the WRh5 and checkmate on the back rank on f1. White will either lose a Rook or be checkmated.
Great observations about the distraction of old habits, and the continuing need to "refresh" the training.
Congratulations on finding a simple way to improve!
Took 1:42 on it, but probably no one cares because it's simple pattern training. Obvious back-rank problem.
ReplyDeleteSo you spilled a minute. What did you do during that minute?
DeleteGood question. I first looked at 1...Qg4, then 1...Qe4 and 1...Qc4, just trying keep the rooks from connecting, or exploiting that they weren't connected. Then, after the wasted minute and a half realizing that it was a simple double-attack (rook and back-rank threat). When I realized the other moves didn't work, I figured there must be a simpler, more direct move.
DeleteTo find an idea is often easy, then ( the speed for ) to find the refutations of the false ideas are critical. It is psychological the more problematic task
DeleteI remember this puzzle, did we had it before?
ReplyDeleteLike so often there are some weaknesses wich needed to be refuted, for example the Queenfork at g4.
To find an idea is one problem, to find the right idea is an other problem and to find the refutations for the wrong ideas is a third problem and this.. quick is problem nr 4
"I remember this puzzle, did we had it before?"
DeleteI thought so too, at one moment during writing this post. But I checked the past 3 years, and I didn't use the diagram as far as I could see.
Very good, those exposed king exercises. Seeing these patterns is like meeting old acquaintances which you thought to have passed away long ago.
ReplyDeleteI said "Taking one month per theme seems to be about right."
ReplyDeleteI just calculated how much time I spent on the theme fork/double attacks: 10 hours (source: solving time Chess Tempo)
The most valuable new patterns are discovered in the 1800 - 2000 rating range. That applies for both the fork/double attack theme and for the exposed king theme. Below 1800 is good for refreshing already known patterns. Above 2000 doesn't make sense when you don't know the 1800 - 2000 patterns. Above 2000 are the same patterns as 1800 - 2000 basically, but more intertwined and with more distracting themes.
ReplyDelete