Supranormal Acitivity in Chess

     The Universal Attack 1...g6/2...Bg7/3...e6/4...Ne7 is ideal against the English for a number of reasons. First of all it is very sharp (Just look at the game Duliba-Moody. He, by the way, is the latest American Postal Grandmaster. I got a good game against him.)

     Here I take on Fritz 10 with the Universal Attack. Not surprisingly the computer won, but I had excellent counterplay. The computer wandered aimlessly trying to find a plan and eventually, through extensive maneuvering,

 can up with a winning strategy. Against a human opponent, there were excellent swindling chances but against a World Class computer these are impossible.

Rybka casts Dispel Magic.  Moody withers.  Game over.

There's a medium for ppl who like to make shit up, and then dress shit up with a secondary education in an effort to scam idiots out of their ducats.  It's called AM radio.  This is a chess forum.  I think you'll find that the percentage of gullible listeners is far lower here than it is on Coast to Coast with Art Bell.   

Fritz has a tough time with Universal Positions. Witness this game.

It's no mistery computers dont like closed positions and they dont like to be taken out of book either but even without opening books computers are just waay too powerful for your average player.

I think maybe White should win this one.

Here's another Sicilian where White is at least equal in the endgame. Anyone above about expert level could win or draw this ending.

     Within one week of getting my Fritz 10 engine, I managed to get two favorable middlegame positions and two equal middlegames at the rate of 120'40, 60'20, '30. In this letter I will describe how Universal theory can be used in the Caro-Kann. It is a totally new way to play against the Caro-Kann and I encourage others, based on the game here, to consider trying this strategy as well. The main idea of Universal Positions is to acquire controlled space which I define as time. World Champion Anatoly Karpov was so successful with Universal Positions with his  "boa constrictor"  approach which I used against Fritz 10 in the Closed Sicilian, The Wilkes-Barre/Traxler, and here, in the Caro-Kann.

     Universal theory is useful against computers because they would have to become unimaginably complex to play Universal Positions and Classical chess simultaneously. To get them to play Universal positions successfully, then it will be necessary to curb their aggressive tendencies. Here, for example, Fritz presses from the start but

 gets nowhere. Clearly if a 1600 player can hold his own against a 2750 computer, this has practical significance.

Here's a rematch with Fritz 10 in the Universal Attack aka the Anti-English.

This appears to be a potential cook of one of the main lines of the Scandinavian Defense. Critics claim that Fritz 10 played poorly, but it played book on move 2 (but it does look wrong).

For those of you who would like to try something incredibly tactical, you'll like game number on in the French/Sicilian and those of you who prefer a quieter positional game will like game # two. Both were played at 120'50 60'20 '30 against Fritz 10.

I think White missed a win here somehow but Fritz claimed the middlegame offered equal chances.

This reminds me of one of my favourite Fischer losses, a very nice strategic win for Kovacevic with black.

Great Game. Thanks---it was fun to see the tactics/strategy. Here's a gambit in the Alekhine no one says is sound. Never mind that Fritz 10 barely got out of the opening alive, but the gambit is "unsound" and any class player could walk away with a free pawn. I was told, in effect that the move order 1.e4 Nf6 2.d4 Nxe4 is not as good as 1.e4 Nf6 2.d4 d6 3.c4! Nxe4. No one seems to notice that if White gets in 3.c4/

4.Nc3 before Black gets in dxe4 that Black is busted.

By the way, I believe that the reference to "Judit Polgar vs Vladimir Kovacevic"  is one of the quite frequent errors in chessgames.com. There is no player in the FIDE ratings list called "Vladimir Kovacevic". If he was a GM, it was very likely Vlatko Kovacevic, the player who won a game against Fischer 17 years earlier.

Looking at the Fischer-Kovacevic game again, I would reassess it as my very favourite Fischer loss. Remarkable is that Kovacevic was an IM, and the game was won in a year when the only other normal time control game Fischer lost with white was against Bent Larsen, who was Fischer's closest competitor in tournaments at that time.

When you get these Fritz engines on the ropes, you've got to finish them off before they get back into the game; here a simple error converts a clearly favorable middlegame into and endgame loss. Fritz nailed me with a combination I never saw coming.

This may qualify more as "supernatural" than supranormal. While playing a game against Fritz 10 in the Two Knights' Defense, it checked my King. When I looked at the board, my King had disappreared! Somehow the computer thought it could capture my King with check. When I left the position on the board to see if my King would reappear it didn't even though I waited about 10 minutes. It was only after I stripped back several moves did my King reappear!

Any system of sufficient complexity has the potential at times to appear magical, or "supranormal" to casual observers.  Imagine what people of just 20 years ago would make of Touch technology (present in the iPod and numerous cell phones), which is now ubiquitous to us, but was unheard of in Bobby Fischer's day.  The modern world gives us more and more wonders as we explore the bounds of our complex systems.

Chess certainly fits the bill, as there are plenty of positions that probably have never been reached in history.  Many of these positions will likely never occur because they are "bad" in one way or another, but one way they might occur is through deep analysis by computer engines, as in some of the examples you gave.  This said, I don't think any serious student of mathematics or game theory would attribute these occurances to paranormal activity.  There is plenty of opportunity for even just plain chance to account for these variances.

Despite the belief that computers play "flawless" Chess, it can do us a lot of good to understand that Chess engines are programmed, and are only as good as the collective theory behind their programming.  Computers don't actually know how to play Chess; someone has to tell the Chess engine what a good move is.  So despite the historic, "supranormal" peaks during the careers of some of the famous GM's, all of these occurances are well within normal playing ability.

I highly recommend Josh Waitzkin's The Art of Learning for more insight into how advanced disciplines (such as Chess and Martial Arts) can take on the characteristics of the paranormal by virtue of their complexity.  Josh points out that "moving in smaller circles" can streamline processes to the point where they certainly appear magical to observers that aren't in-the-know about the underlying mechanics.

I think you'll find that if you look far enough back this statement doesn't hold true for "never ever in the course of history":

http://www.chessville.com/BillWall/EarlyComputerChessPrograms.htm

That is a really great article; thanks for posting it! As for chance---in an early posting I referred to a computer program, supposedly a 1700 computer, was seeing ahead farther than even a Grandmaster (based on a critical concept that required the ability to see ahead at least 8-10 moves in a matter of seconds).

For this to be due to chance, for example, we can assign a 50% chance the computer would choose the right move for the wrong reasons. This took place over at least 120 half moves or 2 raised to the power of 120. This is odds of roughly 1X10 to the 36th power. Chance doesn't seem like a good bet.

Who is reading only the first sentence sloughterchess spouts and then ignores the rest if he doesn't bring up anything interesting (read: ignore whole posts if there is a hint of: I beat fritz, a GM and i was tutored by one, invented 20 openings, scientific nonsense. Oh and grammatical nonsense like his own name, words like supranormal)? I'm surprised he didn't spout the thread: Sicilian Dragon REFUTED!