The Slow Chess League database is ~6000 games and most of the games are normal openings. The Larsen is only 1% of the games. The average rating as I recall is around 1450.
Larsen Opening Why can't I find any articles about it.
The Slow Chess League games are formal. You negotiate for a few days to arrange a time to play. When you do play, it's 45/45 or 90/30 time controls which are roughly equivalent to regular tournaments.
Which is to say, Slow Chess players are fairly serious about their games. These are not throwaway blitz games or even thirty minute games.
Except for the fact that I play in the comfort of my own home, I find these games as draining as rated USCF games.
In the Slow Chess world on chess.com 1.b3 scores impressively:
Games 65 White 63% Draw 6 % Black 31%
"Slow Chess" and "Slow Chess on Chess.com" are two completely different things! Two problems with "Slow Chess on Chess.com".
1) Significantly weaker competition because of all the low rated players here, and because the same level of effort is not applied here as it is over the board, and
2) 30 minutes each on here is considered "Slow Chess". Slow Chess in FIDE and USCF is more like 40 moves in 2 hours followed by Sudden Death in 1 hour, or a more rapid version of "slow" might be Game in 90 minutes with a 30 second increment.
Slow Chess Over The Board is where you should be looking for your stats. White's numbers are not very good at all.
Also, a lot of people are using the basis of "traps" as their validity for the opening. If you are playing for traps, you have no clue how to play chess. Chess is not about traps. Sure, if there is a trap involved in what is a "productive move against correct play", then great, but all of the 1-trick ponies that amateurs try to play in what is otherwise a useless move is one of the biggest bad habits that many players below the expert level have.
When deciding how good an opening is, ASSUME BEST PLAY FROM THE OPPONENT! Now how good is it? The Queen's Gambit is still a slight advantage for White, as is the Ruy Lopez. Larsen's opening ain't worth jack.
ThrillerFan: Geez, cool your jets. You don't know what you are talking about.
There is a group that began on chess.com known as the Slow Chess League in which all games have time controls of 45 min + 45 sec increment or 90 min + 30 sec increment. All games are played over the chess.com live server. These games generally last 1-4 hours.
The Slow Chess League makes available all games under its auspices in a pgn file which is regularly updated. I've written software which does database things to pgn archives.
I'm interested in what happens at the class level of play and this is a convenient database for my investigation.
I don't particularly care if this matters to you.
I consider your "Larsen's opening ain't worth jack" animus silly. The Larsen is just an opening among several dozen. It was largely pioneered by a super GM of his time and is still played by GMs on occasion.
You are welcome not to play it.
If anyone knows of a good database of OTB games by class players, pass it along. I haven't found one, though I haven't looked hard either.
The trouble with OTB class play is that the games are not saved systematically.
I sometimes use chessgames.com where you can download big chunks of the database by position in the opening tree. Those include class players too, but not many and they are selected, presumably, for theoretical significance rather than to reflect typical class games.
Here's a listing of players of who have played 1.b3 a dozen times or more from that database:
93 Vladimir Bagirov
47 Bent Larsen
34 David Ian Calvert
33 Baadur Jobava
31 Pavel Blatny
30 Gerard Welling
23 Tatiana Grabuzova
23 Hikaru Nakamura
23 Elisabeth Paehtz
22 Ljubomir Ljubojevic
20 Juan Manuel Bellon Lopez
19 Artashes Minasian
18 Richard Rapport
18 Bill Wall
17 Lars Karlsson
17 Albin Planinc
15 Yelizaveta Orlova
14 Vladislav Artemiev
13 Nathan Goodhue
13 Branko Lovric
12 Tamaz Gelashvili
12 Simon Kim Williams
956 players from this dataset played 1.b3 at least once.
Hmm, a database with OTB games by players U2000.
I guess you could find some opening traps?
Not sure what else you would use it for.
Considering how so many commenters here blather on and on about how class players should study openings and play openings, one might think there would be some interest in actual data of class players' opening choices and results.
I'm interested personally because I don't think the opening choices of 2300+ apply that well to the pool in which I play. So I'm curious about what works at my level.
I suspect within ten-twenty years Chessbase or Everyman or some school of chess teachers will bring large statistical resources to bear on class players and their openings.
There will be a market for this information.
Publishers can target their opening books better. Class players may find more effective opening choices beyond aping GMs or following their fancies. Chess teachers may find more effective ways to teach openings.
That's sort of my point. If a class player blunders a rook on move 30 and loses, it's not helping you know how good the opening was. If their opponent was out of book on move 5 and started guessing, the game wont be representative of the opening's effectiveness.
bb_gum234: The same is true if a GM blunders a pawn on move 30 and loses. But if you deal with large enough samples this stuff cancels out.
In any event I'm not seeking the eternal truth of a chess opening's goodness, but its effectiveness in a particular pool of players during competition.
Also I'm curious about looking for patterns in class player data.
You seem to miss the point of statistics.
It's not to find perfect truth from perfect data.
It's to use mathematical techniques to squeeze information out of imperfect, noisy data.
Agreed.
I'd have to take a closer look at those games and maybe read some literature on 1.b3.
The King's Gambit does remarkably well too on Slow Chess. Not sure about that either.