I would like to see Komodo Dragon crush IM Rozman, in Knight Odds

[Chessqueen]

I know that Komodo Dragon could possibly win 4.5 out of 6 games against IM Rozman, Levy, but it would be more exciting than to see Komodo Dragon losing to a player like GM Ben Finegold, it is time to see Komodo Dragon winning even if the offer is $500.00 for each win and $100.00 for a draw. Who knows he might get 3 draws out of 6 games
https://ratings.fide.com/profile/2039877/chart

What is the point of playing knight odds with a lower rated IM than Andras Toth, whom we beat 4.5 to 1.5? Just that he has a bigger internet following?

In that case pick up a player a little bit higher than Andras, but not as goos as GM Smerdon, or GM Finegold, with your suggested rules: Dragon must always open 1.e4, alternating knights, but the opponent can never play the same first move twice with a given knight missing. Since there are at least six very sensible good replies to 1.e4 with either knight missing, up to 12 games can be played this way with no repetitions

[lkaufman]

I know that Komodo Dragon could possibly win 4.5 out of 6 games against IM Rozman, Levy, but it would be more exciting than to see Komodo Dragon losing to a player like GM Ben Finegold, it is time to see Komodo Dragon winning even if the offer is $500.00 for each win and $100.00 for a draw. Who knows he might get 3 draws out of 6 games
https://ratings.fide.com/profile/2039877/chart

What is the point of playing knight odds with a lower rated IM than Andras Toth, whom we beat 4.5 to 1.5? Just that he has a bigger internet following?

In that case pick up a player a little bit higher than Andras, but not as goos as GM Smerdon, or GM Finegold, with your suggested rules: Dragon must always open 1.e4, alternating knights, but the opponent can never play the same first move twice with a given knight missing. Since there are at least six very sensible good replies to 1.e4 with either knight missing, up to 12 games can be played this way with no repetitions

I think we need to choose a GM for it to have any good publicity value if we win the match, pretty much any active, popular, well-known GM in the 2400s FIDE would be a good candidate.

[Chessqueen]

I know that Komodo Dragon could possibly win 4.5 out of 6 games against IM Rozman, Levy, but it would be more exciting than to see Komodo Dragon losing to a player like GM Ben Finegold, it is time to see Komodo Dragon winning even if the offer is $500.00 for each win and $100.00 for a draw. Who knows he might get 3 draws out of 6 games
https://ratings.fide.com/profile/2039877/chart

What is the point of playing knight odds with a lower rated IM than Andras Toth, whom we beat 4.5 to 1.5? Just that he has a bigger internet following?

In that case pick up a player a little bit higher than Andras, but not as goos as GM Smerdon, or GM Finegold, with your suggested rules: Dragon must always open 1.e4, alternating knights, but the opponent can never play the same first move twice with a given knight missing. Since there are at least six very sensible good replies to 1.e4 with either knight missing, up to 12 games can be played this way with no repetitions

I think we need to choose a GM for it to have any good publicity value if we win the match, pretty much any active, popular, well-known GM in the 2400s FIDE would be a good candidate.

There are lot of females between 2400 to 2430 that fall under that category
Like Krush, Irina and many others https://ratings.fide.com/profile/2012782
https://www.2700chess.com/women

[Chessqueen]

I know that Komodo Dragon could possibly win 4.5 out of 6 games against IM Rozman, Levy, but it would be more exciting than to see Komodo Dragon losing to a player like GM Ben Finegold, it is time to see Komodo Dragon winning even if the offer is $500.00 for each win and $100.00 for a draw. Who knows he might get 3 draws out of 6 games
https://ratings.fide.com/profile/2039877/chart

What is the point of playing knight odds with a lower rated IM than Andras Toth, whom we beat 4.5 to 1.5? Just that he has a bigger internet following?

In that case pick up a player a little bit higher than Andras, but not as goos as GM Smerdon, or GM Finegold, with your suggested rules: Dragon must always open 1.e4, alternating knights, but the opponent can never play the same first move twice with a given knight missing. Since there are at least six very sensible good replies to 1.e4 with either knight missing, up to 12 games can be played this way with no repetitions

I think we need to choose a GM for it to have any good publicity value if we win the match, pretty much any active, popular, well-known GM in the 2400s FIDE would be a good candidate.

There are lot of females between 2400 to 2430 that fall under that category
Like Krush, Irina and many others https://ratings.fide.com/profile/2012782
https://www.chess.com/game/live/3083158 ... =irochka83
https://www.2700chess.com/women

Another great candidate would be this Young person from Israel
https://ratings.fide.com/profile/2813408
https://www.chess.com/game/live/3737279 ... airparkhov
https://www.365chess.com/players/Yair_Parkhov

[Uri Blass]

The question is if the komodo Dragon team is interested to have a match like it(when Komodo Dragon play at the same time against 6 different players with fide rating 2000-2400) and the only missing thing for it is some sponsor to pay for the humans to play.

We would be willing to do it, but not to sponsor it. Playing six games at once is of course no problem for Komodo when giving knight odds, it won't hurt the results measurably, if at all. The difference between Rapid and Standard chess for humans is probably in the range of 100 to 200 elo, call it 150. If Komodo Dragon is performing at 2425 FIDE giving knight odds in Rapid, it should perform about 2275 at Classical, so maybe a better elo range for the opponents is 2150 to 2400. The concern for this sort of event is possible cheating, especially if there are any cash prizes. We've been careful so far only to play matches with well-known players who would be very unlikely to cheat.

I wonder if you are interested only in knight odd matches or maybe with possible bigger odds when the choice of the odds is of the human opponents.

I thought about something like the following:
Time control is 90+30 when the 6 human opponents may have a choice between
1)white knight odds(smallest odd)
win=6 points,draw=4 points
2)black knight odds
win=5 points draw=3 points
3)a1 rook odds
win=4 points draw=2 points
4)a8 rook odds
win=3 points draw=1 point.

Additional 2 points for every player so even losers get 2 points
but a winner with white knight odd may get 8 points.

The prize money for the human opponents is going to be proportional to the number of points they get with fixed total prize money for the humans.

Fide elo range for the opponents 1700-2200 and
games OTB with an arbiter to prevent cheating.

The reason that I think about lowering the fide elo is because I consider to donate some money for young females in israel
and the list of best famales that are born 1996 or later in israel based on fide rating is the following when only 7 out of them are below 20 years old:

1)Michal Lahav 2178 (birth year 1999)
2)Michelle Katkov 2144 (birth year 2003)
3)Ronit Levitan 2128(birth year 2004)
4)Limor Altshul 2087 (birth year 1998)
5)Adi Federovski 1979 (birth year 1996)
6)Noam Gadasi 1918(birth year 2007)
7)Or Shatil 1883(birth year 2010)
8)Liad Shaul Seidner 1818(birth year 2004)
9)Noga Orian 1791(birth year 2007)
10)Maya Tadmor 1776(birth year 2001)
11)Dana Kochavi 1726(birth year 2007)

This leaves me only with 7 players that are not above 20 year olds.
I know one of them does not live in israel today and I am not sure if the rest of them are going to want to play with odds against engines with the prizes that I may be ready to offer(did not ask them at this point of time).

[lkaufman]

The question is if the komodo Dragon team is interested to have a match like it(when Komodo Dragon play at the same time against 6 different players with fide rating 2000-2400) and the only missing thing for it is some sponsor to pay for the humans to play.

We would be willing to do it, but not to sponsor it. Playing six games at once is of course no problem for Komodo when giving knight odds, it won't hurt the results measurably, if at all. The difference between Rapid and Standard chess for humans is probably in the range of 100 to 200 elo, call it 150. If Komodo Dragon is performing at 2425 FIDE giving knight odds in Rapid, it should perform about 2275 at Classical, so maybe a better elo range for the opponents is 2150 to 2400. The concern for this sort of event is possible cheating, especially if there are any cash prizes. We've been careful so far only to play matches with well-known players who would be very unlikely to cheat.

I wonder if you are interested only in knight odd matches or maybe with possible bigger odds when the choice of the odds is of the human opponents.

I thought about something like the following:
Time control is 90+30 when the 6 human opponents may have a choice between
1)white knight odds(smallest odd)
win=6 points,draw=4 points
2)black knight odds
win=5 points draw=3 points
3)a1 rook odds
win=4 points draw=2 points
4)a8 rook odds
win=3 points draw=1 point.

Additional 2 points for every player so even losers get 2 points
but a winner with white knight odd may get 8 points.

The prize money for the human opponents is going to be proportional to the number of points they get with fixed total prize money for the humans.

Fide elo range for the opponents 1700-2200 and
games OTB with an arbiter to prevent cheating.

The reason that I think about lowering the fide elo is because I consider to donate some money for young females in israel
and the list of best famales that are born 1996 or later in israel based on fide rating is the following when only 7 out of them are below 20 years old:

1)Michal Lahav 2178 (birth year 1999)
2)Michelle Katkov 2144 (birth year 2003)
3)Ronit Levitan 2128(birth year 2004)
4)Limor Altshul 2087 (birth year 1998)
5)Adi Federovski 1979 (birth year 1996)
6)Noam Gadasi 1918(birth year 2007)
7)Or Shatil 1883(birth year 2010)
8)Liad Shaul Seidner 1818(birth year 2004)
9)Noga Orian 1791(birth year 2007)
10)Maya Tadmor 1776(birth year 2001)
11)Dana Kochavi 1726(birth year 2007)

This leaves me only with 7 players that are not above 20 year olds.
I know one of them does not live in israel today and I am not sure if the rest of them are going to want to play with odds against engines with the prizes that I may be ready to offer(did not ask them at this point of time).

[lkaufman]

The question is if the komodo Dragon team is interested to have a match like it(when Komodo Dragon play at the same time against 6 different players with fide rating 2000-2400) and the only missing thing for it is some sponsor to pay for the humans to play.

We would be willing to do it, but not to sponsor it. Playing six games at once is of course no problem for Komodo when giving knight odds, it won't hurt the results measurably, if at all. The difference between Rapid and Standard chess for humans is probably in the range of 100 to 200 elo, call it 150. If Komodo Dragon is performing at 2425 FIDE giving knight odds in Rapid, it should perform about 2275 at Classical, so maybe a better elo range for the opponents is 2150 to 2400. The concern for this sort of event is possible cheating, especially if there are any cash prizes. We've been careful so far only to play matches with well-known players who would be very unlikely to cheat.

I wonder if you are interested only in knight odd matches or maybe with possible bigger odds when the choice of the odds is of the human opponents.

I thought about something like the following:
Time control is 90+30 when the 6 human opponents may have a choice between
1)white knight odds(smallest odd)
win=6 points,draw=4 points
2)black knight odds
win=5 points draw=3 points
3)a1 rook odds
win=4 points draw=2 points
4)a8 rook odds
win=3 points draw=1 point.

Additional 2 points for every player so even losers get 2 points
but a winner with white knight odd may get 8 points.

The prize money for the human opponents is going to be proportional to the number of points they get with fixed total prize money for the humans.

Fide elo range for the opponents 1700-2200 and
games OTB with an arbiter to prevent cheating.

The reason that I think about lowering the fide elo is because I consider to donate some money for young females in israel
and the list of best famales that are born 1996 or later in israel based on fide rating is the following when only 7 out of them are below 20 years old:

1)Michal Lahav 2178 (birth year 1999)
2)Michelle Katkov 2144 (birth year 2003)
3)Ronit Levitan 2128(birth year 2004)
4)Limor Altshul 2087 (birth year 1998)
5)Adi Federovski 1979 (birth year 1996)
6)Noam Gadasi 1918(birth year 2007)
7)Or Shatil 1883(birth year 2010)
8)Liad Shaul Seidner 1818(birth year 2004)
9)Noga Orian 1791(birth year 2007)
10)Maya Tadmor 1776(birth year 2001)
11)Dana Kochavi 1726(birth year 2007)

This leaves me only with 7 players that are not above 20 year olds.
I know one of them does not live in israel today and I am not sure if the rest of them are going to want to play with odds against engines with the prizes that I may be ready to offer(did not ask them at this point of time).

Interesting idea, those are indeed the four most relevant handicaps for engine vs human games (except for strong GMs; games beyond rook handicap aren't very useful as the engine may essentially just give up). They are roughly 100 elo apart, maybe a bit more going from Black knight to White rook. So estimated perf. ratings at classical time control might be White knight 2275, Black knight 2175, White rook 2050, Black rook 1950. So if you want to include some female players in the 1700s that's not too unreasonable at the largest handicap. One thing I would change: it's important for all handicap games to discourage draws, since with any plausible contempt setting the engine will always take an early repetition when down a piece or more; it's just too easy to make a draw. So maybe scoring should be such that draws are equivalent to 1/3 of a win; wins could be 3,6,9, and 12 points for the four handicaps in descending order, with draws getting 1/3 of those numbers.