John Scarne and Teeko

38 views
Skip to first unread message

Ralf Gering

unread,
Feb 20, 2002, 4:46:30 AM2/20/02
to

Teeko was the most popular abstract board game of the 1950s in the US.

You can find some nice pictures of John Scarne and his game at

http://scarne.freeservers.com/scarne16.html
http://scarne.freeservers.com/scarne17.html
http://scarne.freeservers.com/scarne30.html


The rules were far more sophisticated than those given on the web,
but you can get an idea of them at

http://www.members.iex.net/~rfinn/gameshlf/abstract/teeko/teeko.htm
http://www.gamecabinet.com/letters/TeekoReply.html

Ralf

Bob Harris

unread,
Feb 20, 2002, 8:41:49 AM2/20/02
to
Ralf Gering wrote:
> Teeko was the most popular abstract board game of the 1950s in the US.

One of my great finds at a used book store was "Scarne on Teeko" for a buck
and a quarter. The book was published in 1955. I had never heard of the
game.

As you note, it seems to have been very popular. The last page of the book
discusses how "Teeko is Rapidly Sweeping the World". What caused it to drop
in popularity? Was some strategy discovered that rendered the game
uninteresting?

Bob H

Alan Winston

unread,
Feb 20, 2002, 10:20:07 AM2/20/02
to
> As you note, it seems to have been very popular. The last page of the
book
> discusses how "Teeko is Rapidly Sweeping the World". What caused it to
drop
> in popularity? Was some strategy discovered that rendered the game
> uninteresting?

Is there any evidence that the popularity existed outside Scarne's own mind?

Scarne was great at hype and self-promotion. Most of his own "inventions"
seem to have been variations on the same theme, and were highly derivitive
of earlier, better games by others, yet he hyped each one for originality
and superiority.

The copy of Teeko in my game collection is in pristine condition: the
original owner appears to have made the same conclusion I did, that it is
not worth bothering with. I don't recall ever having seen another copy. That
is not suggestive of much popularity. Popular games are seen over and over
again, usually showing evidence of play, often of heavy play.

I believe Scarne's popularity claims are major exagerations at the very
least, and are quite likely total fabrications.

Alan Winston


Jan Kristian Haugland

unread,
Feb 20, 2002, 11:00:52 AM2/20/02
to

A few years ago I wrote a simple and stupid Teeko-program
in Turbo Pascal to get an opponent. It seems to be a pretty
drawish game - although I usually beat it, but only because
it doesn't look ahead, i.e., when I see a possible win there
is often a way to stop it that the program misses. That was
without the tournament rule with larger squares btw. Scarne
apparently could beat all his tournament opponents, but how
trained were they? I don't know, but I'm guessing it might
be too drawish to stay interesting for serious players.

Compare with Dao on http://www.playdao.com This game is
related to Teeko, and it seems like "everybody loves it",
and they've had tournaments in it, but it is even more
drawish. Check for yourself with my analysis program:
http://home.hia.no/~jkhaug00/Dao.java - faster as of last
month and now with move retraction. ;-)

And then there's a related game that is not so drawish,
that I have started a few threads about...;-) New readers
are invited to visit my home page for details.

--

J K Haugland
http://home.hia.no/~jkhaug00

Ralf Gering

unread,
Feb 21, 2002, 7:07:18 AM2/21/02
to

Hello,

> One of my great finds at a used book store was "Scarne on Teeko" for a buck
> and a quarter. The book was published in 1955. I had never heard of the
> game.

I got the book from a very kind American who wanted me to make me happy.
He didn't know me at all and didn't expect anything in return. Some
people are Saints.


> As you note, it seems to have been very popular. The last page of the book
> discusses how "Teeko is Rapidly Sweeping the World". What caused it to drop
> in popularity? Was some strategy discovered that rendered the game
> uninteresting?

The game was "solved" a few years ago. It is a draw. However, this is only
true for Teeko played according to the basic rules. The advanced Teeko
rules described in our book "Scarne on Teeko" are much more complicated
and advanced Teeko is not yet solved. It's difficult to say why fashions
fade. Salta was very popular in the 1900s, Camelot in the 1920s, Teeko in
the 1950s, TwixT in the 1970s.

Ralf

Ralf Gering

unread,
Feb 21, 2002, 7:19:56 AM2/21/02
to

Hello,

> Is there any evidence that the popularity existed outside Scarne's own mind?

Didn't you see the picture showing all these movie stars playing Teeko
simultaneously against Scarne? If you don't know the names, try to find
them in filmographies. They were all celebrities. It's ridicilous to state
that the popularity of his game only existed in his mind. His book was
obviously sold many times, it's still widely available in the US.

> Scarne was great at hype and self-promotion.

I dislike his exaggerated claims too (such as Teeko is superior to Go,
Chess and Draughts). But Scarne also was one of the greatest card and dice
magicians of his time. It doesn't astonish me after the success he had as
a magician that he became arrogant.

> Most of his own "inventions"
> seem to have been variations on the same theme, and were highly derivitive
> of earlier, better games by others, yet he hyped each one for originality
> and superiority.

Which earlier, better four-in-a-row game was Teeko "derivitive"?


Ralf

Ralf Gering

unread,
Feb 21, 2002, 7:29:21 AM2/21/02
to

Hello


> A few years ago I wrote a simple and stupid Teeko-program. It seems to
> be a pretty drawish game That was without the tournament rule with


> larger squares btw. Scarne apparently could beat all his tournament
> opponents, but how trained were they? I don't know, but I'm guessing
> it might be too drawish to stay interesting for serious players.

The tournament game has several additional rules which makes it rather
difficult to draw. The Fairview Teeko Club had many serious players and
draws were very rare among them. Your Teeko program doesn't proof
anything because it doesn't use the correct rules for serious play.

> Compare with Dao on http://www.playdao.com This game is
> related to Teeko, and it seems like "everybody loves it",
> and they've had tournaments in it, but it is even more
> drawish.

That's true, but Dao has rather simple rules compared to advanced
Teeko.

Ralf

Bob Harris

unread,
Feb 21, 2002, 11:29:09 AM2/21/02
to
Ralf Gering posted:

The last of these shows some other games on the table as well-- "Scarney"
and "Follow the Arrow". Other than the fact that they were produced and/or
marketed by John Scarne Games, Inc., what is known about those two games?

As a side note, the picture shows Scarne's son, who was given the middle
name Teeko, after the game. How'd you like to have Scrabble, Salta, Twixt,
or even Chinese Checkers as your middle name? I guess Bridgit might not be
too bad, if your dad invented that game. Othello, too, I suppose; though
I'd hate to go through life knowing I was just a copy of something older,
with a minor variation.

Ralf also wrote, in response to someone's suggestion that game's popularity
might have simply been hype:


> Didn't you see the picture showing all these movie stars playing Teeko
> simultaneously against Scarne? If you don't know the names, try to find them
> in filmographies. They were all celebrities.

Well, I don't know whether it was just hype or real popularity. But the
fact that there are pictures of celebrities playing the game is not evidence
for real popularity. That could have just been a typical publicity stunt,
requiring nothing more than cash to set it up.

One of them looks like Judy Holliday. Another might be Walter Slezak, I
think. I don't recognize any of the others. But that's not saying much.

I *am* interested in knowing whether the popularity was real. I wonder how
I can find out? Other than Scarne's book, were there perhaps some newspaper
articles covering the tournament? Perhaps the number of units sold, of the
book and of the game, compared to other comtemporary books and games would
give us a clue?

Bob H

Jan Kristian Haugland

unread,
Feb 21, 2002, 11:29:38 AM2/21/02
to

Ralf Gering wrote:
> > A few years ago I wrote a simple and stupid Teeko-program. It seems to
> > be a pretty drawish game That was without the tournament rule with
> > larger squares btw. Scarne apparently could beat all his tournament
> > opponents, but how trained were they? I don't know, but I'm guessing
> > it might be too drawish to stay interesting for serious players.
>
> The tournament game has several additional rules which makes it rather
> difficult to draw. The Fairview Teeko Club had many serious players and
> draws were very rare among them. Your Teeko program doesn't proof
> anything because it doesn't use the correct rules for serious play.

I agree that my Teeko program doesn't prove anything. :-)
But maybe I should look into re-writing my newer analysis
programs for Dao/Neutreeko and see if Teeko is within reach.
Do you have a complete list of the tournament game rules?

Jan Kristian Haugland

unread,
Feb 21, 2002, 12:22:07 PM2/21/02
to

Bob Harris wrote:

> As a side note, the picture shows Scarne's son, who was given the middle
> name Teeko, after the game. How'd you like to have Scrabble, Salta, Twixt,
> or even Chinese Checkers as your middle name? I guess Bridgit might not be
> too bad, if your dad invented that game. Othello, too, I suppose; though
> I'd hate to go through life knowing I was just a copy of something older,
> with a minor variation.

And then there's Monty Python's "Mr. Neutron" - however
the sketch was relased a few years _before_ the abstract
game "Neutron" was published. I definitely would not like
to have "Hex" as a middle name, which (with x replaced by
ks) means "witch" in Norwegian.

Philip M. Cohen

unread,
Feb 21, 2002, 4:36:55 PM2/21/02
to
Ralf Gering wrote:
>
> Hello,
>
> > Is there any evidence that the popularity existed outside Scarne's own mind?
>
> Didn't you see the picture showing all these movie stars playing Teeko
> simultaneously against Scarne? If you don't know the names, try to find
> them in filmographies. They were all celebrities. It's ridicilous to state
> that the popularity of his game only existed in his mind.

Not as ridiculous as taking a posed and paid-for publicity photo as
evidence of popularity. Even before I checked the web and found
http://www.washingtonpost.com/ac2/wp-dyn?pagename=article&node=style/postmagazine&contentId=A43740-2001Jul10
I was sure that that was just more Scarne hot air. It's possible that as
a result of Scarne's years-long publicity blitz he managed to sell
100,000 copies of the game, though I wouldn't bet on it; how many of
those games actually got played is another question. I used to own a
copy of *Scarne on Teeko* and to play '58-Positions Teeko' by mail, but
I don't recall ever seeing a copy of the game, or a reference to it
outside of Scarne's own books. It's a decent game but nothing special.
Lines of Action is worth ten Teekos.
--
Always carry a grapefruit, Treesong

Bob Harris

unread,
Feb 21, 2002, 6:40:51 PM2/21/02
to
Philip M. Cohen wrote:
> ... Even before I checked the web and found
> http://www.washingtonpost.com/ac2/ ...
> I was sure that that was just more Scarne hot air. ...

Seems to be a very informative article. Thanks for posting that link.

Looking closer at the previously mentioned celebrity phot, Judy Holliday
sure does have a confused look on her face. ;)

> Always carry a grapefruit, Treesong

??? what does that mean ??? Why not carry a pomegrantate?

Bob H

Philip M. Cohen

unread,
Feb 21, 2002, 7:44:46 PM2/21/02
to
Bob Harris wrote:

> > Always carry a grapefruit, Treesong
>
> ??? what does that mean ???

I don't know.
--

Ralf Gering

unread,
Feb 22, 2002, 7:29:14 AM2/22/02
to

Hello,

> Well, I don't know whether it was just hype or real popularity. But the
> fact that there are pictures of celebrities playing the game is not evidence
> for real popularity. That could have just been a typical publicity stunt,
> requiring nothing more than cash to set it up.

Sure, the tournament was sponsored for publicity reasons. Nevertheless,
the fact that these celebrities agreed to participate also means that they
thought the game is worth to be "related" with their names. Your view is
rather superficial. Do you think that all people are prostitutes who do
ANYTHING for money? These celebrities were rich and didn't need any $$.

> One of them looks like Judy Holliday. Another might be Walter Slezak, I
> think. I don't recognize any of the others. But that's not saying much.

Why do you not simply read the names below the picture. The names are
given below the pic!


Ralf

Ralf Gering

unread,
Feb 22, 2002, 7:46:53 AM2/22/02
to

Hello,

> I agree that my Teeko program doesn't prove anything. :-)
> But maybe I should look into re-writing my newer analysis
> programs for Dao/Neutreeko and see if Teeko is within reach.
> Do you have a complete list of the tournament game rules?

Advanced Teeko as explained in "Scarne on Teeko" had 16 different styles
of play, each one with slightly different rules. The idea was that
the game should remain varied enough (in regard to openings) that nobody
would ever master them all. It would take me hours to explain these
styles.

However, two special rules common to all advanced Teeko play are worth
mentioning:

4-move limit: If one player moves with one piece back and forth the same
two circles all the time, he must move during the next 4 moves another
stone or one of these two stones to another circle if his opponent asks
for it -- otherwise he looses.

10-move limit: If one player blockades his opponent's stones in one corner
and moves another stone all the time, his opponent can ask for a 10-move
limit. Then he must move another stone duing the next 10 moves or he
looses the game.

Teeko later developped into Scar-ney, when the pieces were marked with
numbers. I think the rules of Teeko remained, but the point scoring now
depends on the position of numerical pieces in the general winning position.

However, I don't know the exact point scoring of Scar-ney.

Ralf

Ralf Gering

unread,
Feb 22, 2002, 8:01:22 AM2/22/02
to

Hello,

> Not as ridiculous as taking a posed and paid-for publicity photo as
> evidence of popularity.

Again somebody who believes that all people are prostitutes and would do
anything for money! Movie stars and chess champions have a good name to
loose but don't need to pose for a few $$.

> Even before I checked the web and found
> http://www.washingtonpost.com/ac2/wp-dyn?pagename=article&node=style/postmagazine&contentId=A43740-2001Jul10
> I was sure that that was just more Scarne hot air.

Reading competence doubtful... because -->

> It's possible that as
> a result of Scarne's years-long publicity blitz he managed to sell
> 100,000 copies of the game,

which would be more than any modern abstract game sells today (except
perhaps Abalone and Othello)

> Lines of Action is worth ten Teekos.

And chess is (of course!) better than LoA, and Go is better than Chess.
I'm tired of these stupid comparisons which abound among lovers of
abstract games. Sure, your favorite game is 10-times better.

Ralf

Ralf Gering

unread,
Feb 22, 2002, 7:52:32 AM2/22/02
to

Hello,

> I definitely would not like
> to have "Hex" as a middle name, which (with x replaced by
> ks) means "witch" in Norwegian.

"Hexe" in German (and in some dialects "Hex")! :-) Actually the original
name of this game was "Polygon" -- not too bad.

BTW

Your game "Neutreeko" is quite interesting! Let me know about the results
of your tournament!

Do you know how long (in moves) is on the average a game of Neutreeko if
played by experienced players (i.e. players who not easily fall into
traps)?

Ralf

Bob Harris

unread,
Feb 22, 2002, 9:10:21 AM2/22/02
to
Bob:

> Well, I don't know whether it was just hype or real popularity. But the fact
> that there are pictures of celebrities playing the game is not evidence for
> real popularity. That could have just been a typical publicity stunt,
> requiring nothing more than cash to set it up.

Ralf:

> Sure, the tournament was sponsored for publicity reasons. Nevertheless, the
> fact that these celebrities agreed to participate also means that they thought
> the game is worth to be "related" with their names. Your view is rather
> superficial. Do you think that all people are prostitutes who do ANYTHING for
> money? These celebrities were rich and didn't need any $$.

I didn't say this *was* a publicity stunt. I said it *could* have been.
And no, I don't think that *all* people would do anything for money, once
they have enough. But there certainly are plenty who will.

The celebrity photo is consistent with the theory that Teeko was a game
enjoyed by millions of players worldwide over the course of five centuries.
The confused look on Judy Holliday's face might be because Scarne was such a
great player, and he made a brilliant move she wasn't expecting.

The celebrity photo is also consistent with the theory that Scarne just
showed up at the club one night and asked the celebrities to do him a favor.
After all, he had some degree of celebrity himself. The confused look on
Judy Holliday's face might be because she had never seen the game before.

My point is, and was, that you can't tell anything about about which theory
is true from the photo. The photo isn't evidence of anything about the
popularity of the game.

You may well be right, that Teeko was a very popular game. I'm just saying
that I haven't seen much evidence to support that claim.

Bob:

> One of them looks like Judy Holliday. Another might be Walter Slezak, I
> think. I don't recognize any of the others. But that's not saying much.

Ralf:

> Why do you not simply read the names below the picture. The names are
> given below the pic!

Oooops. That was off the bottom of my browser window. Hey, at least I got
two of them right. The only other one I think I should have noticed was
Steve Allen, but he's kind of off in the distance with his head turned. The
other names aren't familiar to me.

Bob H

Jan Kristian Haugland

unread,
Feb 22, 2002, 10:56:40 AM2/22/02
to

Ralf Gering wrote:

> BTW
>
> Your game "Neutreeko" is quite interesting! Let me know about the results
> of your tournament!

Thanks, will do!

> Do you know how long (in moves) is on the average a game of Neutreeko if
> played by experienced players (i.e. players who not easily fall into
> traps)?

Very difficult question. Things that can influence include:
The number of moves you must look ahead to qualify as
experienced, currently unknown(?) stratetic considerations
one can follow and find good moves without reading ahead,
memorizing of opening lines etc.

In short: I don't know. ;-) But perhaps one can get some
idea of this with the help of my analysis program.

Jan Kristian Haugland

unread,
Feb 22, 2002, 11:51:24 AM2/22/02
to

Ralf Gering wrote:

> Advanced Teeko as explained in "Scarne on Teeko" had 16 different styles
> of play, each one with slightly different rules. The idea was that
> the game should remain varied enough (in regard to openings) that nobody
> would ever master them all. It would take me hours to explain these
> styles.

Beauty in the simplicity - good bye...

> However, two special rules common to all advanced Teeko play are worth
> mentioning:
>

> 4-move limit: (...) 10-move limit: (...)

Computerized analysis based on position alone - good bye...:-)

D. Seagraves

unread,
Feb 22, 2002, 2:16:14 PM2/22/02
to
Ralf Gering <zxm...@mail.uni-tuebingen.de> wrote in message news:<Pine.LNX.4.30.02022...@linux30.zdv.uni-tuebingen.de>...
>>>>[ Your view is rather superficial. Do you think that all people

are prostitutes who do ANYTHING for money? These celebrities were rich
and didn't need any $$. ]<<<<<
...snip...
>>>>>[ Why do you not simply read the names below the picture. The
names are
given below the pic! ]<<<<<

Ralf, Ralf... why so tense? Are you that serious a Scarne/Teeko fan?
"Your view is rather superficial?" One could argue that your view that
celebrities would only support products they truly believe in and
would *never* "prostitute" themselves is fairly superficial. You have
seen a "infomercial" before, right? People do all kinds of stuff for
money - including people who have lots of money.

As for Teeko....

Teeko was heavily pushed by Scarne, no doubt about it. Does that mean
the game's alleged popularity was bogus? No. But it could be. Is Teeko
the best thing since sliced bread? Perhaps - to some people. Is it
better that all other games? To some, sure.

But let's not get ugly while discussing such things, 'kay?

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to
"So I took the T from Tic-tac-toe, the E from Chess, the K from Checkers
and the O from Bingo," John Scarne wrote.

From some of the posts in this newsgroup, it sounds as some people would
rather have had John Scarne taken...

the A from Tic-tac-toe,
the S from Chess,
the S from Checkers and
the O from Bingo.


:)


Ed Collins
e...@edcollins.com

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
"Everybody in the country is in at least eighteen
computers. You'd be surprised what you can learn
by just punching out a name on one of these control
banks. It's spooky." - Jim Rockford (1978)
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=


Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Jan Kristian Haugland

unread,
Feb 24, 2002, 5:05:56 AM2/24/02
to

"Edward D. Collins" wrote:
>
> "So I took the T from Tic-tac-toe, the E from Chess, the K from Checkers
> and the O from Bingo," John Scarne wrote.
>
> From some of the posts in this newsgroup, it sounds as some people would
> rather have had John Scarne taken...
>
> the A from Tic-tac-toe,
> the S from Chess,
> the S from Checkers and
> the O from Bingo.
>
> :)

I wouldn't go that far, but "Dao" might have a couple
of extra letters inserted to become "Drawo" :->

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to
"So I took the T from Tic-tac-toe, the E from Chess, the K from Checkers
and the O from Bingo," John Scarne wrote.

From some of the posts in this newsgroup, it sounds as some people would
rather have had John Scarne taken...

the A from Tic-tac-toe,
the S from Chess,
the S from Checkers and
the O from Bingo.


:)


Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Edward D. Collins

unread,
Feb 24, 2002, 4:19:21 AM2/24/02
to

Ralf Gering

unread,
Feb 24, 2002, 6:12:38 PM2/24/02
to

I'm not getting ugly but some peple are getting quite stupid on this list.
But probably you don't like it if it is exposed that some people can't
even read...

Ralf


On 22 Feb 2002, D. Seagraves wrote:

> Date: 22 Feb 2002 11:16:14 -0800
> From: D. Seagraves <harve...@yahoo.com>
> Newsgroups: rec.games.abstract
> Subject: Re: John Scarne and Teeko
>
> Ralf Gering <zxm...@mail.uni-tuebingen.de> wrote in message news:<Pine.LNX.4.30.02022...@linux30.zdv.uni-tuebingen.de>...
> >>>>[ Your view is rather superficial. Do you think that all people
> are prostitutes who do ANYTHING for money? These celebrities were rich
> and didn't need any $$. ]<<<<<
> ...snip...
> >>>>>[ Why do you not simply read the names below the picture. The
> names are
> given below the pic! ]<<<<<
>
> Ralf, Ralf... why so tense? Are you that serious a Scarne/Teeko fan?
> "Your view is rather superficial?" One could argue that your view that
> celebrities would only support products they truly believe in and
> would *never* "prostitute" themselves is fairly superficial. You have
> seen a "infomercial" before, right? People do all kinds of stuff for
> money - including people who have lots of money.
>
> As for Teeko....
>
> Teeko was heavily pushed by Scarne, no doubt about it. Does that mean
> the game's alleged popularity was bogus? No. But it could be. Is Teeko
> the best thing since sliced bread? Perhaps - to some people. Is it
> better that all other games? To some, sure.
>
> But let's not get ugly while discussing such things, 'kay?
>

David J Bush

unread,
Feb 25, 2002, 3:19:41 PM2/25/02
to
Thanks for the link! Based on the photo and the description of the
rules given, Teeko sounds like it could fall prey to computer
analysis. This, of course, would depend on knowing the exact
ruleset (or rulesets), but consider this simple analysis:

I call the first player Red and the 2nd player Black here.
The board is 5x5, and any position would have k red tokens
and either k or k-1 black tokens, where k = 0,1,2,3, or 4.
(Of course there cannot be 0 red and -1 black tokens.)
If a position is defined by the arrangement of all the pieces
plus whose move it is, there are 174,438,376 positions.
Actually there are less than that, since positions where
both players have a teeko, or the player to move has a teeko,
cannot occur in a game.

A computer could scan and label all positions as follows:
On the first scan, all positions with a teeko are labeled as
wins. Then on the next scan, all positions where the player to
move can make a teeko, are labeled as wins for that player.
Then all positions where the player to move cannot avoid giving
the next player a win, are labeled as losses for that player.
On the next scan, all positions where the player can put the
opponent in a losing position are labeled as wins, and so on.
In the case of draws by repetition, there may be rules which
prohibit one side from making a move that repeats a previous
position under certain circumstances. The number of repeats
is irrelevant to the analysis of that position; just tell the
computer to avoid all such prohibited moves. In other words,
if a player is prohibited from repeating a position 3 times
or is prohibited from repeating once, the resulting analysis
of the game outcome would be the same. There may be some
positions which allow repetition. If all the moves from a
given position are labeled, and a drawn position can be reached
but a win cannot, then that node is labeled as a draw. So, every
node can be labeled this way, without any recursive tree search
needed. All that is needed, to solve Teeko this way, would be the
exact ruleset or rulesets used, and a computer with sufficient
memory to store all positions. The algorithm would need a way of
numbering each position. Just five possible tags are needed:
"Not labeled yet," "Red Win," "Draw," "Black win," and possibly
"illegal position." If the memory is available, perhaps the entire
game tree could be stored.

Actually, dealing with draw by repetition would be more complicated
than I describe, but I believe the algorithm is possible nonetheless.

So, who has the exact ruleset or rulesets for Teeko?

David Bush http://www.geocities.com/twixtplayer/
Remove nospamtoday from my email

Bruno Wolff III

unread,
Feb 25, 2002, 4:19:18 PM2/25/02
to
In article <9k2l7uolm4tgti768...@4ax.com>, David J Bush wrote:
>
> Actually, dealing with draw by repetition would be more complicated
> than I describe, but I believe the algorithm is possible nonetheless.

I think the procedure is describe in the "Loopy Endgames" section of
Winning Ways.

Jan Kristian Haugland

unread,
Feb 26, 2002, 5:28:05 AM2/26/02
to

David J Bush wrote:

> A computer could scan and label all positions as follows

Take a look at

http://www.gamerz.net/archives/pbmserv-dev/199811/msg00002.html

--

J K Haugland
http://home.hia.no/~jkhaug00

Ralf Gering

unread,
Feb 26, 2002, 8:06:02 AM2/26/02
to

Hello,

> http://www.gamerz.net/archives/pbmserv-dev/199811/msg00002.html
> J K Haugland

Thank you for this link. The analysis is very interesting.

I made a mistake: The game is thoroughly solved. Teeko
appears to be quite deep and has some astonishing things to offer,
however, the game is not deep enough to remain a mystery. And yet, I don't
know if a human can put the results of this computer analysis into
practice and play perfectly...

Ralf

Jan Kristian Haugland

unread,
Feb 26, 2002, 2:36:51 PM2/26/02
to

Ralf Gering wrote:
>
> > http://www.gamerz.net/archives/pbmserv-dev/199811/msg00002.html

>
> Thank you for this link. The analysis is very interesting.

You're welcome. And btw, all the differences between the 16
forms of play come from (1) letting larger squares count as
wins or not and (2) special rules for how the pieces are
dropped in the opening. When I talked about drawish games
before, I meant whether there were many drawn positions
among the possible positions in play. From that perspective
there are really only two cases to check. In Dao, the
proportion of draws is 79%, in Neutreeko 3%, and in 4x4
Neutreeko 5%. The extra rules for preventing repetitions
would complicate the analysis, but I think it would be
interesting to see how standard Teeko and 58-position
Teeko without those extra rules compare with those games.
The proportion of draws is of course far from being a
good measurement for quality, but still...

--

- o - o - Jan K Haugland, D Phil
- - x - -
- - - - - http://home.hia.no/~jkhaug00
- - o - -
- x - x - Check out my game Neutreeko!

Jan Kristian Haugland

unread,
Feb 26, 2002, 3:59:24 PM2/26/02
to

I wrote:
> > > http://www.gamerz.net/archives/pbmserv-dev/199811/msg00002.html
(...)

> but I think it would be
> interesting to see how standard Teeko and 58-position
> Teeko without those extra rules compare with those games.

Heh, I just e-mailed the author Guy Steele about this
and he has already replied. Here are the cold facts:

> Yes. Without taking repetition into account, I have the
> following computer-generated results for positions with all
> eight pieces on the board. For regular Teeko:
>
> Histogram of play scores:
> 42347231 Draw
> 40 Black in 39
> 704 Black in 37
> 4056 Black in 35
> 11008 Black in 33
> 22368 Black in 31
> 33960 Black in 29
> 43320 Black in 27
> 59940 Black in 25
> 88736 Black in 23
> 123412 Black in 21
> 205472 Black in 19
> 332816 Black in 17
> 508896 Black in 15
> 795160 Black in 13
> 1192946 Black in 11
> 1880462 Black in 9
> 2693722 Black in 7
> 4999274 Black in 5
> 9169210 Black in 3
> 3440836 Black in 1
> 262332 Black wins
> 88 Red in 38
> 960 Red in 36
> 3672 Red in 34
> 9924 Red in 32
> 14316 Red in 30
> 18136 Red in 28
> 23632 Red in 26
> 35490 Red in 24
> 50832 Red in 22
> 74708 Red in 20
> 124062 Red in 18
> 204316 Red in 16
> 313279 Red in 14
> 478304 Red in 12
> 676428 Red in 10
> 1013306 Red in 8
> 1181424 Red in 6
> 1617536 Red in 4
> 1390596 Red in 2
> 262332 Red wins
> 1008 ILLEGAL
>
> 75710250 total
>
> For Teeko 58:
>
> Histogram of play scores:
> 25672066 Draw
> 136 Black in 59
> 1184 Black in 57
> 4520 Black in 55
> 8364 Black in 53
> 10188 Black in 51
> 8856 Black in 49
> 12224 Black in 47
> 17844 Black in 45
> 21220 Black in 43
> 22800 Black in 41
> 32080 Black in 39
> 55808 Black in 37
> 76036 Black in 35
> 94548 Black in 33
> 118896 Black in 31
> 161792 Black in 29
> 229688 Black in 27
> 304992 Black in 25
> 395632 Black in 23
> 530064 Black in 21
> 646496 Black in 19
> 819128 Black in 17
> 996432 Black in 15
> 1246030 Black in 13
> 1622213 Black in 11
> 2339480 Black in 9
> 3588138 Black in 7
> 6825350 Black in 5
> 12035716 Black in 3
> 4960092 Black in 1
> 345380 Black wins
> 296 Red in 58
> 1280 Red in 56
> 3000 Red in 54
> 5760 Red in 52
> 5568 Red in 50
> 5624 Red in 48
> 8952 Red in 46
> 12540 Red in 44
> 13964 Red in 42
> 14580 Red in 40
> 23852 Red in 38
> 40064 Red in 36
> 50936 Red in 34
> 66640 Red in 32
> 81280 Red in 30
> 111180 Red in 28
> 156276 Red in 26
> 204268 Red in 24
> 266940 Red in 22
> 346712 Red in 20
> 421648 Red in 18
> 523156 Red in 16
> 630186 Red in 14
> 785024 Red in 12
> 1035810 Red in 10
> 1435495 Red in 8
> 1827926 Red in 6
> 2308902 Red in 4
> 1771868 Red in 2
> 345380 Red wins
> 1750 ILLEGAL
>
> 75710250 total
>
> So in regular Teeko, the proportion of drawn positions,
> considering only positions with all eight pieces on the board,
> is 42347231/75710250, or about 56%. For Teeko 58, the proportion
> of drawn positions is 25672066/75710250, or about 34%.

In a way this makes Teeko more interesting than I had thought.
I didn't expect that there would be such high figures for the
maximal length of perfect terminating matches.

Bob Harris

unread,
Feb 26, 2002, 4:25:28 PM2/26/02
to
Jan Kristian Haugland wrote:
> Heh, I just e-mailed the author Guy Steele about this
> and he has already replied. Here are the cold facts:

I wonder if that's the smae Guy L. Steele, Jr. who co-authored the wonderful
"C, A Reference Manual" that's sitting next my computer.

Bob H

D. Seagraves

unread,
Feb 28, 2002, 9:42:03 AM2/28/02
to
Ralf Gering <zxm...@mail.uni-tuebingen.de> wrote in message news:<Pine.LNX.4.30.02022...@linux40.zdv.uni-tuebingen.de>...

> I'm not getting ugly but some peple are getting quite stupid on this list.
> But probably you don't like it if it is exposed that some people can't
> even read...

It *is* ugly to say that (your words), "peple are getting quite stupid
on this list," and, "probably you don't like it if it is exposed that


some people can't even read."

To paraphrase what I said before..... settle down.

Oh, and "people" is is not spelled "peple." It's always good policy to
proofread your responses when you are suggesting that people are
stupid and illiterate.

- D. Seagraves

Jan Kristian Haugland

unread,
Mar 1, 2002, 6:52:29 AM3/1/02
to

After the Neutreeko tournament yesterday, I have
put a report on the web due to popular demand -
well, at least Ralf asked for it :->

http://home.hia.no/~jkhaug00/report1.txt

See also http://home.no.net/spilluka

Ralf Gering

unread,
Mar 1, 2002, 7:21:17 AM3/1/02