Soltis experiment

John Fairbairn · Post by **John Fairbairn** » Wed Nov 19, 2014 5:34 am

I like to look at chess books now and then because they have a lot to say about how to study the game, and I believe a large proportion of that can be applied to go.

The latest book I am looking at is "Studying Chess Made Easy" by Andrew Soltis. I am barely familiar with Soltis but he is a Grandmaster who seems well regarded as an author, and he writes for the New York Post.

Since I am not actually studying chess myself I can't vouch for the efficacy of his methods, but he does seem to make a reasonable case, and writes well.

His principles of good studying boil down to these:

1. Learning should be fun. His book identifies various ways to make it so.

2. It has to include hands-on learning. That is: theory + practice = success. Chess players have a big advantage in that they have easy access to good computer opponents, but in general playing real games is surely also part of the fun element in both go and chess.

3. It should be mainly independent. You can ask teachers and friends questions "but working alone works best". This may be contentious, but Soltis makes a strong case.

4. It's often subliminal. Just flipping through a magazine and looking at the diagrams "can be nearly as good as more intensive study". This may be contentious, especially to RJ, but it seems to be taken almost as a given in the chess world, judging by other books.

5. You have to be well motivated. In this context, this refers mainly to aspects of chess that may not appeal to you, e.g. the endgame. You have to find a reason to study those aspects. For example, learning the KBN vs K endgame may seem a waste of time because it's so rare, but look at it another way and you realise it teaches you valuable enclosure tactics that apply in other parts of the game. This is one area where a coach or mentor can help.

6. Avoid TMI: Too Much Information. Cramming doesn't work in the longer term. Ditto over-analysis.

7. Work below your consciousness. Soltis doesn't seem to differentiate this well from (4) but, as the rest of the book confirms, it shows he is very much an advocate of the sublimininal/subconscious school.

8. It's absorbing ideas. Again a case can be made to say this is a nexus with (4) and (7). Perhaps it's his way of making our subconsciouses absorb subliminally the idea that subliminal absorption by our subconscious is VERY important!

9. Be a rugged individual. Those who question and try things out are likely to improve most, though apparently not because they are thereby demonstrating greater intelligence. It seems more to do with the fact that they are more likely to be focused and tenacious in their study.

Anyway, he obviously goes into detail about all these things, and one point he makes strongly is the value of never letting a diagram go by without thinking about it. You don't have to get the right answer - there may not even be one right answer. The very act of thinking and learning to focus are what is beneficial. Of course it's a bonus if you can discuss it with someone else stronger as well, and it's also a bonus if you have a teacher who can point you at the most appropriate diagrams to look at. But these are bonuses, not a sine qua non.

That popped up in to my mind when I was looking at game that began as shown below.

Black 5 is nearly always in this lower left corner, and is nearly always high. I knew that already, but following Soltis I stopped to think about the whys and wherefores. I could come up with some reasons easily enough (e.g. high and facing the high nirensei shows consistency), but it dawned on me that it was much, much harder to say why other moves are inferior (indeed, whether they actually are inferior). For example, if I approached instead in the upper left (as a handful of pros have done), how would White demonstrate that was inferior?

So I thought I would toss this out for discussion. Even if we disagree or don't understand, the very act of discussing it will apparently be beneficial to our subconscious, if Soltis is right.

Aidoneus · Post by **Aidoneus** » Wed Nov 19, 2014 6:23 am

I'm a big fan of Andrew Soltis. When I joined Encyclopaedia Britannica I found it disgraceful that we had no signed article on chess. (Just a couple paragraphs.) While I could have written a decent short piece--I'm a lowly chess master--I wanted a grandmaster who could write a substantial article about all aspects of the game and its history. My first choice was Andrew, and he agreed. (Certain people at EB actually disparaged and resisted the inclusion of a signed article on a "mere" game.) Maybe I'm biased as the commissioner and editor of the EB chess article, but I highly recommend it for your reading pleasure!

Knotwilg · Post by **Knotwilg** » Wed Nov 19, 2014 6:54 am

My shot at it.

(Incidentally resisting the urge to discuss the proposition)

Uberdude · Post by **Uberdude** » Wed Nov 19, 2014 7:13 am

Bill, I know you were probably showing the disadvantages of playing normally after approaching the top left, but I think the d14 connection is rather uninspired. I would like to treat those exchanges lightly and approach at e4 now, or perhaps g14 for Takemiya style (and white probably g17 which does somewhat negate the nirensei influence, but black still aims at h16 press).

If black were to low approach the bottom left, the patient kosumi feels like a plausible answer, making miai of left and lower sides, though I don't feel black is inferior.

Bill Spight · Post by **Bill Spight** » Wed Nov 19, 2014 7:47 am

Uberdude wrote:Bill, I know you were probably showing the disadvantages of playing normally after approaching the top left, but I think the d14 connection is rather uninspired.

My sentiments exactly.

Bill Spight · Post by **Bill Spight** » Wed Nov 19, 2014 7:51 am

Aidoneus wrote:I'm a big fan of Andrew Soltis. When I joined Encyclopaedia Britannica I found it disgraceful that we had no signed article on chess.

Britannica still has signed articles?

Quick! Tell Wikipedia. Tell Sensei's Library.

RBerenguel · Post by **RBerenguel** » Wed Nov 19, 2014 8:48 am

Very interesting John. I have also been reading the chess literature lately (well, I have stopped a little after a while) to find interesting stuff on "improvement." I'll check Soltis book when I can.

But, what strikes me is your last paragraph,

John Fairbairn wrote: I knew that already, but following Soltis I stopped to think about the whys and wherefores. I could come up with some reasons easily enough (e.g. high and facing the high nirensei shows consistency), but it dawned on me that it was much, much harder to say why other moves are inferior (indeed, whether they actually are inferior.)

I started playing backgammon a few months ago, and I'm still at the beginner-intermediate level, where I have read some, played some games but still need to double-check everything. And yesterday the backgammon subreddit (reddit is an online community site with smaller subgroups, like for go, backgammon, and mostly any theme you fancy) complained about its lack of activity. And one of its moderators posed a problem suitable for anyone:

: bg.jpg (17.08 KiB) Viewed 15384 times

White to play, moving counterclockwise (home is on the right.) Since I'm still at a relatively new level, I decided to give my proposed solution in relatively long prose, arguing why some moves were inferior to my proposed move. And it was actually pretty enlightening for my understanding of the position. Giving a full account of whys, a better player can easily tell me "here, look, this assumption you make is wrong because of this." And I'll learn from it, easily.

I put my proposed move between spoiler tags in case you play backgammon and want to give a shot (if you do, please do likewise.)

Bill Spight · Post by **Bill Spight** » Wed Nov 19, 2014 9:22 am

I think that the main reasons for the high approach have to do with the long distance relationships between stones, an appreciation of which goes back at least as far as Murase Shuho in the 19th century. The main proof of the value of these relationships is empirical. Playing the whole board wins games.

As for which corner to approach, let me use my heuristic of comparing (non-sente) candidate plays as though they were miai. If Black plays one of them, let White play the other.

Click Here To Show Diagram Code: [go]$$Bcm5 Black approaches bottom left $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . X . . . | $$ | . . . 2 . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . O , 1 . . . . , . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

Click Here To Show Diagram Code: [go]$$Bcm5 Black approaches top left $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . , . a . . . , . . . . . X . . . | $$ | . . . 1 . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . O , 2 . . . . , . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

In the second diagram, White threatens a pincer in the top left, which will work with the enclosure in the bottom left. In the first diagram, White threatens to approach the top right corner from the top side, but that threat is not as good as the threat of the pincer in the second diagram. Also, in the first diagram the Black approach stone has a good relationship to the Black stone in the bottom right, while in the second diagram the Black approach stone is almost isolated. (I would probably play

at "a" to remedy that.) I prefer the first diagram for Black.

Click Here To Show Diagram Code: [go]$$Bcm5 White pincer $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . X . . . | $$ | . . 3 . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . O , . . . . . , . . . . . X . . . | $$ | . . . . 1 . . 2 . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

Given the White D-17 stone, the low approach to the bottom left corner looks better than it would if White had played at D-16. If White pincers, for example, Black can approach the top left corner.

Click Here To Show Diagram Code: [go]$$Bcm5 White kosumi $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . 3 . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . 2 . . . . . . . . . . . . . . . | $$ | . . O , . . . . . , . . . . . X . . . | $$ | . . . . 1 . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

Likewise, if White plays kosumi or other response on the left side, Black can still approach the top left corner.

Click Here To Show Diagram Code: [go]$$Bcm5 White encloses top left corner $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . X . . . | $$ | . . . 2 . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . 3 . . . . . . . . . . . . . . . | $$ | . . O , . . . . . , . . . . . X . . . | $$ | . . . . 1 . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

If White encloses the top left corner, Black can play the kake, aiming at a framework on the bottom side.

In any event these differences are subtle. I doubt if they affect the win rate at the pro level by as much as 3%. Which is why the long history is important. It takes years or decades to see significant results.

gowan · Post by **gowan** » Wed Nov 19, 2014 10:23 am

Re #4 in John Fairbairn's post, I have some supporting evidence from a different area. Some of my friends are classical musicians and they emphatically say that they get some "practice" value from just reading a score, not actually playing their instrument(s)

Re Bill Spight's posting about the Britannica and Wikipedia concerning signed articles. Just because an authority has written a signed article does not guarantee that the article is a good one. On Wikipedia, as everyone surely knows, articles can be edited by people who did not author the original version. In effect there is a community interested in a particular article and that community is likely to contain several authorities on the topic of the article. Because of that the correctness of the article might well be better than an article written by a single expert not subject to editing by other authorities. This works on Wikipedia because the editing community is very large. On SL the community is probably much smaller and there are fewer authoritative writers. In the SL case good articles often get spoiled by subsequent editing by people who don't know what they are writing about. Such edits are more likely to notice and fixed on WP.

Finally, some years ago there was talk about a study comparing WP articles with articles in EB. I don't have a reference to this and I suppose it could be an urban legend but the study supposedly found that there were comparably many mistakes and types of errors in the EB articles as in the WP articles.

RBerenguel · Post by **RBerenguel** » Wed Nov 19, 2014 10:34 am

gowan wrote:Re #4 in John Fairbairn's post, I have some supporting evidence from a different area. Some of my friends are classical musicians and they emphatically say that they get some "practice" value from just reading a score, not actually playing their instrument(s)

I'm not a musician, but based on musician's recommendation and general praise, on "musical learning" I recommend Effortless Mastery.

Bill Spight · Post by **Bill Spight** » Wed Nov 19, 2014 11:37 am

gowan wrote: Re Bill Spight's posting about the Britannica and Wikipedia concerning signed articles. Just because an authority has written a signed article does not guarantee that the article is a good one.

Britannica may want a signed article to indicate an authoritative source. For Sensei's Library the opposite applies. Relatively little on SL is authoritative. IMO, not signing contributions makes it appear that articles are more authoritative than they are. SL started out with signed contributions, but switched to Wikipedia style some years ago. I argued against the change, but lost that debate.

I'll say no more now, so as not to hijack this thread.

gowan · Post by **gowan** » Wed Nov 19, 2014 12:07 pm

Here is a link to the USA State of Delaware website with information about how professional librarians evaluate WP vs EB

http://library.blogs.delaware.gov/2013/ ... le-source/

Aidoneus · Post by **Aidoneus** » Wed Nov 19, 2014 12:33 pm

I say, too many cooks "may" spoil the broth. Let me add, however, that I like Wikipedia, open-souce software, and collaboration in general.

A rebuttal: http://corporate.britannica.com/britann ... sponse.pdf

Aidoneus · Post by **Aidoneus** » Wed Nov 19, 2014 3:09 pm

Hi John,

You have private message receipt disabled, so I'm responding to your query here.

I retired from EB in 2009. I have a 1st edition facsimile and an older 13th edition set, but I suspect you might want the 11th or 12th edition. Both the 9th/10th edition (1902: http://www.1902encyclopedia.com/) and the famous 11th edition (1911: http://encyclopedia.jrank.org/) are in the public domain and online but I don't believe that either contained an article on Go.

Having said all of this, it sounds like the reference to Go might be in an article about India. In any case, if your friend can relate the name of the article, the printing, and the author, I might be able to find out something directly or through one of my former colleagues at EB. EB keeps a copy of every printing in its reference library. Searching for the word "go" is, of course, nigh on impossible!

Regards,
Bill

mhlepore · Post by **mhlepore** » Wed Nov 19, 2014 6:26 pm

John Fairbairn wrote:...but it dawned on me that it was much, much harder to say why other moves are inferior (indeed, whether they actually are inferior). For example, if I approached instead in the upper left (as a handful of pros have done), how would White demonstrate that was inferior?

...

I think this is an important point, but it applies mainly to strategy (fuseki) and not tactics (especially life and death).

Herbert Simon won the Nobel Prize for his work on bounded rationality and satisficing. Human's don't make decisions because we've eliminated all other options as suboptimal, but because we found one that is good enough for the moment.

Life In 19x19

Soltis experiment

Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment

Re: Soltis experiment