St. Louis chess960/FRC

[coldsquid]

Hi Larry,

I own a lodge in Utah and want to host a small non-Chess1 event. I enjoy Chess960 but because the castling can be so gimicky, have wanted to organize an event using only the 34 positions from 960 that start with the rooks in the corners and the kings on E or D (minus the position from Chess1 and its mirror with king on D), "Chess34". My thought is that those 34 positions may strike the best balance between the feel of Chess1 while adding enough positions to severely cripple opening preparation. Wondering if 34 was too few, I had also pondered the asymmetrical option you've called 324.

Would appreciate your feedback on structuring an OTB event to test these ideas.

Would you mind contacting me when it's convenient? My email is in my profile, and my number is (Utah's original area code) + 860-5885.

Thanks,

Matt Evans
Timber Moose Lodge

[lkaufman]

Hi Larry,

I own a lodge in Utah and want to host a small non-Chess1 event. I enjoy Chess960 but because the castling can be so gimicky, have wanted to organize an event using only the 34 positions from 960 that start with the rooks in the corners and the kings on E or D (minus the position from Chess1 and its mirror with king on D), "Chess34". My thought is that those 34 positions may strike the best balance between the feel of Chess1 while adding enough positions to severely cripple opening preparation. Wondering if 34 was too few, I had also pondered the asymmetrical option you've called 324.

Would appreciate your feedback on structuring an OTB event to test these ideas.

Would you mind contacting me when it's convenient? My email is in my profile, and my number is (Utah's original area code) + 860-5885.

Thanks,

Matt Evans
Timber Moose Lodge

I don't understand the point of the positions with the kings on the "d" file; either you would need to use FRC castling rules which you call gimicky, or you would be playing nocastle chess. There are only 18 positions (17 if you exclude the standard one) that permit normal castling on both sides, which becomes 324 if you remove the symmetry requirement. Even 17 positions should be enough for a single amateur event, but of course it is not enough to be used over and over again as the normal way to play FRC. Chess 324 works great if you require pairs of games to be played, reversing colors, but it is not suitable for single games. There is also the option of the old shuffle chess, where the back rank is randomized (symmetrically for W and B) only with the rule of bishops having to be on opposite colors, and castling only legal when it is legal under normal chess rules.

[coldsquid]

Hi Larry,

To keep castling non-gimmicky I'd have the king always move 2 squares to castle, not always land on C or G as in 960, so starting position with King on D would mirror positions with king on E. Purpose of having King on D would just be to further reduce (though modestly) the benefits of opening preparation by increasing the risk of a preparation error.

My goal for Chess34 (or Chess17) is to strike a balance between 960 and Chess1. I believe that even 17 starting positions would be sufficient to make memorization far less important even upon repeated play by top GMs -- consider how comparatively poorly they would know Chess1 opening theory if their opening prep had been 94% (17/18) less! -- and could perhaps satisfy the critics of 960 who think it's too odd. So Chess17/34 is an attempt compromise between advocates and skeptics of Chess960 and Chess1.

Regardless, I'm intrigued by Chess324, and because I have a venue that could host I want to organize an OTB event with GMs to explore the promise of these variants. Perhaps a Chess324 event, but half of the games are from the subset that are also starting positions in Chess17?

Matt
Timber Moose Lodge

[lkaufman]

Hi Larry,

To keep castling non-gimmicky I'd have the king always move 2 squares to castle, not always land on C or G as in 960, so starting position with King on D would mirror positions with king on E. Purpose of having King on D would just be to further reduce (though modestly) the benefits of opening preparation by increasing the risk of a preparation error.

My goal for Chess34 (or Chess17) is to strike a balance between 960 and Chess1. I believe that even 17 starting positions would be sufficient to make memorization far less important even upon repeated play by top GMs -- consider how comparatively poorly they would know Chess1 opening theory if their opening prep had been 94% (17/18) less! -- and could perhaps satisfy the critics of 960 who think it's too odd. So Chess17/34 is an attempt compromise between advocates and skeptics of Chess960 and Chess1.

Regardless, I'm intrigued by Chess324, and because I have a venue that could host I want to organize an OTB event with GMs to explore the promise of these variants. Perhaps a Chess324 event, but half of the games are from the subset that are also starting positions in Chess17?

Matt
Timber Moose Lodge

If you start the kings on the d file and have castling be by moving the king two squares, then those positions are just mirror images of the other 17 positions, play should be identical though mirrored. In other words, it just "looks" different, it isn't really different. Yes, 17 positions would dramatically reduce opening prep, especially pre-game prep if the position isn't known in advance, but top players would at least memorize ideal opening sequences for several moves for each position, it doesn't feel like enough for the purpose of avoiding the role of computer analysis. The problem with mixing 324 with 17 is that 324 needs two game matches, chess17 does not.

[coldsquid]

Yes, the King on D mirrors the E position. The only purpose is to make memorizing the openings more prone to error (for humans), thus reducing the benefits of prep. Not sure if it's worth the variance from tradition.

My goal is to come up with a variant that can challenge the ubiquity of Chess1, and think something like Chess18 might be the most likely to succeed because it splits the difference between the purists of Chess1 and those who find it dull to have players still within their preparation through move 20. Chess18 wouldn't eliminate computer analysis but would ensure players are out of their prep within a handful of moves.

Because Chess17 is only a subset of the Chess324 starting positions, Chess17 could also be done as two-game matches. Do you have interest in a small event featuring Chess324 and Chess17? I'm thinking like 6-8 titled players for 3 or 4 days.

[lkaufman]

Yes, the King on D mirrors the E position. The only purpose is to make memorizing the openings more prone to error (for humans), thus reducing the benefits of prep. Not sure if it's worth the variance from tradition.

My goal is to come up with a variant that can challenge the ubiquity of Chess1, and think something like Chess18 might be the most likely to succeed because it splits the difference between the purists of Chess1 and those who find it dull to have players still within their preparation through move 20. Chess18 wouldn't eliminate computer analysis but would ensure players are out of their prep within a handful of moves.

Because Chess17 is only a subset of the Chess324 starting positions, Chess17 could also be done as two-game matches. Do you have interest in a small event featuring Chess324 and Chess17? I'm thinking like 6-8 titled players for 3 or 4 days.

Chess17 can certainly be done as two-game matches; it just isn't essential the way it is for chess 324. As for playing in an event myself, if it's online I could probably play if the schedule isn't too difficult, but I'm pretty reluctant to fly anywhere nowadays, though I did fly to Cal. for the U.S. Championship of Senior State Champions recently (tied for sixth), so I won't rule it out. By the way, there was one "chess17" match in 2020; our Komodo program played four games with GM Alex Lenderman of Chess17 but giving knight odds (so actually 34 nonstandard initial positions with each knight removal option). For much greater variety without adding new rules (just new startpositions) and without breaking symmetry (if you want to avoid the need for two game matches), there is the old shufflechess, where castling is only allowed if the king and rook happen to be on the normal squares. It's not clear to me whether the "weird" castling of chess960 is better than just not allowing it at all unless the normal rules allow it. I think castling was mostly introduced to speed up the game a bit, saving a couple of moves of artificial castling by hand, but that's not such a big deal if we're starting with nonstandard setups anyway; maybe a couple extra moves is better than "strange" castling? Hard to say.

[coldsquid]

Thanks Larry. Perhaps it could be a hybrid event, with games online and OTB? And maybe a handicapped Komodo? I'm willing to provide the venue and a small prize fund. Do you know David Hater? He has been working with me to put an event together.

Maybe King on D could indicate no castling, turning Chess17 into Chess34.

[lkaufman]

Thanks Larry. Perhaps it could be a hybrid event, with games online and OTB? And maybe a handicapped Komodo? I'm willing to provide the venue and a small prize fund. Do you know David Hater? He has been working with me to put an event together.

Maybe King on D could indicate no castling, turning Chess17 into Chess34.

Hybrid is possible, at least for me that would work. Handicapped Komodo Dragon (presumably knight odds) is fine. David Hater, I know the name, but not much more than that. Do you have in mind an Open or a master-level event? The Chess34 idea is of course valid, though I would suspect that you would get more interest in the Chess17 idea as it is far closer to normal chess. Another way to increase the number of positions without needing special castling or no-castling would be to allow the queen's rook to be on any of the a thru d files, so at least always kingside castling (which is by far dominant in normal chess) remains legal.

[coldsquid]

David Hater is a USCF board member and active TD who's been advising me.

I'd love to have you participate online. I imagine there are others who would participate if they could play remotely, too, so it could make the event more successful. My ideal is a masters event, 8-16 players, with a time control around 15+10.

Glad to hear you think Dragon could participate. Is there someone at Komodo I should talk to about exploring those details? For the buildup to the event perhaps we could do some promotional project with Dragon analyzing the starting positions in a novel way, highlighting that Chess18 doesn't eliminate theory, merely dilutes it. As always, feedback welcome.

Matt

[lkaufman]

David Hater is a USCF board member and active TD who's been advising me.

I'd love to have you participate online. I imagine there are others who would participate if they could play remotely, too, so it could make the event more successful. My ideal is a masters event, 8-16 players, with a time control around 15+10.

Glad to hear you think Dragon could participate. Is there someone at Komodo I should talk to about exploring those details? For the buildup to the event perhaps we could do some promotional project with Dragon analyzing the starting positions in a novel way, highlighting that Chess18 doesn't eliminate theory, merely dilutes it. As always, feedback welcome.

Matt

Masters event at that "standard" Rapid time control sounds pretty ideal to me. I am the right person at KomodoChess to talk to about its participation in the event.