Latest Updates:
Page Index Toggle Pages: 1 [2] 3 4 
Topic Tools
Very Hot Topic (More than 25 Replies) Smith and Tikkanen's Woodpecker Method (Read 17309 times)
Jupp53
God Member
*****
Offline


be

Posts: 873
Location: Frankfurt/Main
Joined: 01/04/09
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #36 - 08/14/18 at 01:29:24
Post Tools
an ordinary chessplayer wrote on 08/13/18 at 02:23:54:
Jupp53 wrote on 08/12/18 at 19:44:40:
So I would recommend to see the book as YATB yet another tactic book. My question: How is it under this point of view?

... I'm pretty sure the authors would dispute your characterization as YATB. But reading between the lines it seems to me that you don't think there is any benefit to repetition. ...
Quote:
The acquisition of expertise in complex tasks such as high-level chess, professional basketball, or firefighting is intricate and slow because expertise in a domain is not a single skill but rather a large collection of miniskills.
I picked that one because I think pattern recognition, while significant, is just one skill among many in chess.


There is a benefit of repetition. My comment was only about the expectation. I just read a sentence, which I think is very true imo: The master has failed more often, than the apprentice has ever tried. So repetition helps you to see, what you did and did not unterstand. Skills need practise, repetition is a good way to practise.

My point is: Don't expect to hold the level reached by studying too long. The best effect for a tournament will be finishing a circle two or three days before it begins. Then there will be a time of three or four weeks and the effect on concentration and short-term-memory will be faded. The effect to long-term-memory will still be useful.

Thank you to all answers to my question. Especially the short review of ReneDescartes was helpful.
  

Medical textbooks say I should be dead since April 2002.
Dum spiro spero. Smiley
Back to top
 
IP Logged
 
dfan
God Member
*****
Offline


"When you see a bad move,
look for a better one"

Posts: 720
Location: Boston
Joined: 10/04/05
Re: Smith and Tikkanen's Woodpecker Method
Reply #35 - 08/13/18 at 14:51:30
Post Tools
Stigma wrote on 08/13/18 at 14:25:46:
I like it too, but one possible criticism is won't there necessarily be individual differences here? Someone like Tikkanen may be able to trust his intuition that "this must be winning" way earlier in a line than I could or should. This can depend on differences in both strength, playing style and "thinking style" (more calculating vs more intuitive).

I think of it as actively training your intuition; you learn what sorts of positions you should be able to evaluate rather than calculate. Of course there is no harm (scoring-wise) in calculating further if you feel you need to.
  
Back to top
 
IP Logged
 
Stigma
God Member
*****
Offline


There is a crack in everything.

Posts: 3043
Joined: 11/07/06
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #34 - 08/13/18 at 14:25:46
Post Tools
dfan wrote on 08/13/18 at 12:54:36:
ReneDescartes wrote on 08/13/18 at 12:07:54:
One very nice feature is a little check the authors place before the point you have to see in each variation, even if the variation goes on further. "This is as far as you have to see. It is obvious that Black's king cannot survive." This is great. I know of no other material that trains you where to  stop calculating.

The scoring in the Yusupov series often gives you full credit for a variation well before the end of it. Like you, I appreciate the feature.

I like it too, but one possible criticism is won't there necessarily be individual differences here? Someone like Tikkanen may be able to trust his intuition that "this must be winning" way earlier in a line than I could or should. This can depend on differences in both strength, playing style and "thinking style" (more calculating vs more intuitive).
  

Improvement begins at the edge of your comfort zone. -Jonathan Rowson
Back to top
 
IP Logged
 
dfan
God Member
*****
Offline


"When you see a bad move,
look for a better one"

Posts: 720
Location: Boston
Joined: 10/04/05
Re: Smith and Tikkanen's Woodpecker Method
Reply #33 - 08/13/18 at 12:54:36
Post Tools
ReneDescartes wrote on 08/13/18 at 12:07:54:
One very nice feature is a little check the authors place before the point you have to see in each variation, even if the variation goes on further. "This is as far as you have to see. It is obvious that Black's king cannot survive." This is great. I know of no other material that trains you where to  stop calculating.

The scoring in the Yusupov series often gives you full credit for a variation well before the end of it. Like you, I appreciate the feature.
  
Back to top
 
IP Logged
 
ReneDescartes
God Member
*****
Offline


Qu'est-ce donc que je
suis? Une chose qui pense.

Posts: 1046
Joined: 05/17/10
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #32 - 08/13/18 at 12:07:54
Post Tools
Well, the sample is up on Forward Chess, and except for the introduction, it seems to be just a book of tactics from world champions, arranged in three levels of difficulty and then chronologically by champion. Pono, Khalifman, Topalov, etc. are included--not my taste, but ok.

One very nice feature is a little check the authors place before the point you have to see in each variation, even if the variation goes on further. "This is as far as you have to see. It is obvious that Black's king cannot survive." This is great. I know of no other material that trains you where to  stop calculating.

Another nice feature is that not all the positions have a tactic that works! Until you have them memorized, you'll have to verify everything using "System 2," as the saying goes (slow, careful logical reasoning). This is a nice feature--but only if the book is viewed as in YATB. If the unproven theoretical idea behind the Woodpecker method is, as An Ordinary Chessplayer avers, to train System 1 (fast, intuitive, "blink" thinking),
then the red herings are an idiotic,throw-in-the-kitchen-sink feature. The authors more or less admit this.

There is an old world-champions tactics book by Tangborn, and a large Russian multi-volume set of complicated WC combinations, but nothing like this book, which fills a nice niche--as Yet Another Tactics Book.


  
Back to top
 
IP Logged
 
an ordinary chessplayer
God Member
*****
Offline


I used to be not bad.

Posts: 599
Location: Columbus, OH (USA)
Joined: 01/02/15
Re: Smith and Tikkanen's Woodpecker Method
Reply #31 - 08/13/18 at 02:23:54
Post Tools
Jupp53 wrote on 08/12/18 at 19:44:40:
So I would recommend to see the book as YATB yet another tactic book. My question: How is it under this point of view?

I don't have the book, maybe someone else can answer your specific question. From what I have seen online, I'm pretty sure the authors would dispute your characterization as YATB. But reading between the lines it seems to me that you don't think there is any benefit to repetition. What would a trial of the Woodpecker Method look like? I suppose, have the experimental group train for a tournament solely using the Woodpecker Method, and have the control group train for the same tournament but using only new problems. It seems like this could be set up at a school or camp.

I finished the Thinking, Fast and Slow book. He actually mentions chess in a couple of places. Once he briefly mentions Kasparov vs Deep Blue (not interesting), then on pages 238-240 he uses chess as an example of skill acquired by practice. It's somewhat interesting, although I think most experienced chessplayers already know most of what he says there. I'll just give a brief quote:
Quote:
The acquisition of expertise in complex tasks such as high-level chess, professional basketball, or firefighting is intricate and slow because expertise in a domain is not a single skill but rather a large collection of miniskills.
I picked that one because I think pattern recognition, while significant, is just one skill among many in chess.
  
Back to top
 
IP Logged
 
Jupp53
God Member
*****
Offline


be

Posts: 873
Location: Frankfurt/Main
Joined: 01/04/09
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #30 - 08/12/18 at 19:44:40
Post Tools
A well researched psychological fact:

Training of sustained attention (I use "concentration" further) is only possible with short time effects. Is there someone doubting that concentration is one part of the process leading to the performance in chess?

You can see the "Woodpecker method" partly as a concentration training. So doing it before a tournament you will get: Better understanding of some positions you could not solve at first or later. Activating your chess memory in the area the tasks the book gives you. If done till the last week before the tournament some specific chess concentration training.

The second to last and the last effect will vanish after a month, at least if its about concentration. I do not remember any specific research to the memory. But there is surely someone who knows more.

So I would recommend to see the book as YATB yet another tactic book. My question: How is it under this point of view?
  

Medical textbooks say I should be dead since April 2002.
Dum spiro spero. Smiley
Back to top
 
IP Logged
 
an ordinary chessplayer
God Member
*****
Offline


I used to be not bad.

Posts: 599
Location: Columbus, OH (USA)
Joined: 01/02/15
Re: Smith and Tikkanen's Woodpecker Method
Reply #29 - 08/10/18 at 21:34:49
Post Tools
Stigma wrote on 07/21/18 at 21:43:00:
But the crucial combination that the psychological pattern recognition literature puts the spotlight on is "thinking" that's nonverbal, subconscious and automatic. It's this sense of pattern recognition that enables strong players to play a decent game even in bullet, or 2800 players to perform at 2600 level even in blitz and simuls (I didn't bother to look up the actual figures now, but there is a well-known article on this that I can dig up). It's this form of pattern recognition that time and again in the psychology literature has been thought to be the main distinguishing factor between players of of different strengths, to the extent that some researchers even started doubting that calculation skill ("search") had anything to do with it (though it turned out it had).
Great insight there.

I am reading Kahneman (2011) Thinking, Fast and Slow, and the nonverbal, subconscious and automatic thinking is what he calls System 1. The reason why I highlighted pattern recognition is because I strongly suspect there is more than one inference system involved in System 1 chess thinking. If I am correct about that, then to call it all pattern recognition is to misidentify at least some of that thinking, even though pattern recognition is, I believe, certainly in there.

Does the Woodpecker Method work? I.e., does it improve (some) System 1 chess thinking? I don't see why it wouldn't. In fact, all the logical reasons given by the authors for why it works could be total bollocks, and yet following the method could still work. What chess psychologists call pattern recognition might turn out to be something else entirely, and yet the method could still work.

Will the Woodpecker Method lead to the claimed rating gains? In most cases, no. The reason is that our rating is determined by the weakest part of our game. If whatever the Woodpecker Method improves is not our most glaring weakness, any rating gain for us would be modest if any. But for some players it might be the ticket to success.

I agree with ReneDescartes that whether it's better to repeat known positions versus move on to new ones is not yet determined. I can see a theoretical argument that if the idea is to train System 1 then speed is important and therefore keeping System 2 disengaged is necessary. But at this time that's just a story. It needs testing.
  
Back to top
 
IP Logged
 
Uhohspaghettio
God Member
*****
Offline


I Love ChessPublishing!

Posts: 510
Joined: 02/23/11
Re: Smith and Tikkanen's Woodpecker Method
Reply #28 - 08/10/18 at 00:15:47
Post Tools
Man, this stuff by Vik-Hansen is such garbage. Pattern just means something similar, like recognizing a face by facial pattern (chess positions have been shown to activate similar regions of the brain activated when trying to recognize faces). The word and idea of "pattern" in chess makes sense and is useful, what doesn't make any sense and is not useful is questioning its existence as he is doing. An example of academic catastrophe, how the process could have gone so wrong that such a thing got published, suitable only as an April Fool's.
  

"I don't recall saying good luck."
Back to top
 
IP Logged
 
ReneDescartes
God Member
*****
Offline


Qu'est-ce donc que je
suis? Une chose qui pense.

Posts: 1046
Joined: 05/17/10
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #27 - 07/31/18 at 00:37:32
Post Tools
Well, in the absence of evidence it seems as likely to work as any other method, and the training we believe in is, within reason, often the one that will work best  for us, if only because we work best for it--just as with openings (but don't ask me for my evidence on that!). I also agree it would be nice if it were true.
  
Back to top
 
IP Logged
 
Stigma
God Member
*****
Offline


There is a crack in everything.

Posts: 3043
Joined: 11/07/06
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #26 - 07/27/18 at 15:24:20
Post Tools
ReneDescartes wrote on 07/27/18 at 14:18:11:
But where is the evidence that going through the same book again later is preferable to doing another tactics book (leaving aside for the moment review of the problems you got wrong, which has so much common sense in its favor that, even without evidence, I can't imagine it not being very worthwhile)?

I don't want an interminable discussion, I just am curious why you believe this is supported by what we know about pattern recognition?

There's no direct, "scientific" evidence that I'm aware of. That was just an afterthought anyway, the rest of the paragraph was the actual point I wanted to make there.

But OK, if that's what you want to focus on, I have had some success with going through the same tactics sets again personally. Especially as a warm-up just before tournaments. And I could repeat my earlier point that compared to new material there may be less need to calculate, so more of the time is spent effectively drilling the patterns into the brain.

And wouldn't it be nice if it's true that once you find a really good tactics book, you can get that much out of it - instead of having to seek out new material all the time?

There could also be a more general point for chess study here (taking the same point beyond just tactics), one I've seen claimed by various people: It's better to study a few good books well, until you really know the material, than to study many books superficially. And examples of good ways to ensure you know some material well are of course spaced repetition and (maybe to a lesser extent) the woodpecker method.
  

Improvement begins at the edge of your comfort zone. -Jonathan Rowson
Back to top
 
IP Logged
 
ReneDescartes
God Member
*****
Offline


Qu'est-ce donc que je
suis? Une chose qui pense.

Posts: 1046
Joined: 05/17/10
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #25 - 07/27/18 at 14:18:11
Post Tools
Stigma wrote on 07/26/18 at 21:41:12:
@ReneDescartes:
Well, I think the research and our knowledge of pattern recognition is prescriptive.

Thanks to that knowledge, I know not to work on a big tactics book by thinking long and hard on each problem and then, if I finally get through the entire book, put it away and consider myself finished with it - a mistake I think many adult improvers still make. Instead I will look up the solutions after several minutes, and repeat the positions several times, especially the ones I struggle with. And go through the same book again later (if it's a good one) whenever I feel my tactical level starting to drop.


But where is the evidence that going through the same book again later is preferable to doing another tactics book (leaving aside for the moment review of the problems you got wrong, which has so much common sense in its favor that, even without evidence, I can't imagine it not being very worthwhile)?

I don't want an interminable discussion, I just am curious why you believe this is supported by what we know about pattern recognition?
  
Back to top
 
IP Logged
 
Stigma
God Member
*****
Offline


There is a crack in everything.

Posts: 3043
Joined: 11/07/06
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #24 - 07/26/18 at 22:02:22
Post Tools
@Jupp53:
Very interesting observations on the value of the "subconscious" vs automaticity as models of nonconscious behavior, thank you. I have to think about this a bit more and get back to the thread over the weekend.

I am trying to take a break from ChessPub even if it doesn't look that way... I have a weekend tournament coming up and still quite a bit of work left this week.
  

Improvement begins at the edge of your comfort zone. -Jonathan Rowson
Back to top
 
IP Logged
 
Stigma
God Member
*****
Offline


There is a crack in everything.

Posts: 3043
Joined: 11/07/06
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #23 - 07/26/18 at 21:41:12
Post Tools
@ReneDescartes:
Well, I think the research and our knowledge of pattern recognition is prescriptive.

Thanks to that knowledge, I know not to work on a big tactics book by thinking long and hard on each problem and then, if I finally get through the entire book, put it away and consider myself finished with it - a mistake I think many adult improvers still make. Instead I will look up the solutions after several minutes, and repeat the positions several times, especially the ones I struggle with. And go through the same book again later (if it's a good one) whenever I feel my tactical level starting to drop.

Apart from the "memorizing games" technique (which is probably overkill, though there are some anectodes about strong players having used it), that tactical 1900 player you're mentioning could have been me, except I was only rated 1500 when I went through Reassess Your Chess and some of Dvoretsky's writings on prophylaxis. The latter is obviously a mindset/thinking technique, but in the Silman case I don't think his "Silman Thinking Technique" ever did much for me - I now see it as largely a motivational trick to get people to study all those great strategic games. Pattern recognition deriving from those games plus Silman's excellent, engaging explanations likely did the heavy lifting.

Silman himself has been known to recommend obviously pattern-based training techniques like rushing through lots of GM games from the database, "chess movie style". He claims he developed his thinking technique and the more general idea of imbalances because lower-rated players thought he was crazy when he told them looking at vast amounts of grandmaster games was a good training technique, and so he needed an alternative they would accept!

If the poster you're quoting is wrong, it's not because trying to increase one's storage of patterns wouldn't work. But maybe because it would be too boring and/or don't feel like actual training, so many people wouldn't actually stick with it.

And I don't believe de Groot's original subjects, as enormously talented as they were, had as good training methods as we do today. If they did, you'd be hard pressed to explain why world class players are taking less and less time to reach the elite level (though with some well-known exceptions like Fischer). The chess information explosion and internet play must have a lot to do with that.
  

Improvement begins at the edge of your comfort zone. -Jonathan Rowson
Back to top
 
IP Logged
 
ReneDescartes
God Member
*****
Offline


Qu'est-ce donc que je
suis? Une chose qui pense.

Posts: 1046
Joined: 05/17/10
Gender: Male
Re: Smith and Tikkanen's Woodpecker Method
Reply #22 - 07/25/18 at 23:08:49
Post Tools
What I'm mostly complaining about is sloppy thinking as people cite deGroot's research or its terminology when advocating particular kinds of study, as if that research were prescriptive. In our forum, for example, you get this sort of thing, as advice to a player asking for general help [the poster does not speak English as a first language, and I have no wish to mock either his ideas or his expression of them]:

"What you need for thinking is as many patterns as possible stored in your brain (which has to make use of this storage in his own enigmatic way. You only can wish that this knowledge may [come to mind when it's needed] - and it does cf. smothered mate)...But one thing is clear to me: You need much more patterns. I would say, Top players have at least some 50,000 in store...I definitely feel that pattern knowledge is the key. As it can be used for strategical patterns too. And too for Endgames."

Now, what that poster advises is merely (1)studying large books of tactics and (2)making file cards or Chessbase files of interesting positions one encounters and practicing them--both of which I myself do and would recommend doing, but not because of a psychological theory. (I would conjecture that the benefit of making your own file-card index and drilling it has more to do with active learning and with identifying and repairing your deficits than with the theory of pattern-recognition, which, in its ability to find benefit in almost any method of study with the exception of visualization exercises, is nearly a universal solvent). Strictly speaking, I would say that we just don't know what works best and why. Scientific studies would help (not that they're forthcoming).

Yes, of course, I'm also thinking of Heisman, de la Maza, and Smith/Tikkanen advocating repeating the same exact exercises, but they don't actually say that the theory indicates that their methods are best. Interestingly, I know personally a private devotee of this method who started using it in the early 1990s by memorizing games, citing pattern recognition.  He was a 1900-rated tactically-minded player. He only took a leap, however, when he read Silman's explanatory positional book Reassess Your Chess--without going out of his way to repeat anything.

I'm also thinking of the vogue for what you once called "a la carte" strategic and positional study manuals--van de Oudeweetering, Soltis's 100 Grandmaster Secrets, Guliev, and Broznik/Terekhin. I own those books, and I think them of decent quality; the last one I think very good; for that matter, My System is not so far from their approach. But given the kind of statements I've heard, and the way people now like to throw around the term "pattern recognition," I suspect some purchasers might think that the method of these books is scientifically supported by deGroot, etc., when for all we know the best way to acquire patterns is the way deGroot's original subjects presumably did it--playing a lot, interacting with  strong players, reading and annotating master games, preparing openings, examining studies, etc.



 
« Last Edit: 07/26/18 at 19:49:18 by ReneDescartes »  
Back to top
 
IP Logged
 
Page Index Toggle Pages: 1 [2] 3 4 
Topic Tools
Bookmarks: del.icio.us Digg Facebook Google Google+ Linked in reddit StumbleUpon Twitter Yahoo