Komodo Dragon vs GM Wirig knight odds match

[lkaufman]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

[Chessqueen]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

My prediction is 4.5 to 3.5 in favor of GM Anthony Wirig, but i would have to say that his Blitz rating is mainly against other human NOT Engines, therefore that Blitz rating is incognito or insignificant when comparing against a top Engine like Komodo MCTS2, since other humans blunder and Engine do NOT under time pressure.

[lkaufman]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

My prediction is 4.5 to 3.5 in favor of GM Anthony Wirig, but i would have to say that his Blitz rating is mainly against other human NOT Engines, therefore that Blitz rating is incognito or insignificant when comparing against a top Engine like Komodo MCTS2, since other humans blunder and Engine do NOT under time pressure.

While I don't disagree with your prediction, I don't follow your logic. If his higher rating vs. humans is due to blundering less often, that is precisely the trait needed to do well with knight odds from Dragon. Judging from the match with Toth, a 2500 chess.com blitz rating is not high enough to avoid serious errors at 15' + 10" even with an extra knight, but a 2830 blitz rating may well be good enough.

[Uri Blass]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

My prediction is 4.5 to 3.5 in favor of GM Anthony Wirig, but i would have to say that his Blitz rating is mainly against other human NOT Engines, therefore that Blitz rating is incognito or insignificant when comparing against a top Engine like Komodo MCTS2, since other humans blunder and Engine do NOT under time pressure.

While I don't disagree with your prediction, I don't follow your logic. If his higher rating vs. humans is due to blundering less often, that is precisely the trait needed to do well with knight odds from Dragon. Judging from the match with Toth, a 2500 chess.com blitz rating is not high enough to avoid serious errors at 15' + 10" even with an extra knight, but a 2830 blitz rating may well be good enough.

1)15'+10'' is rapid and not blitz so I doubt if we can learn from blitz rating about result in rapid games.
2)I am not sure if the player who is better in normal chess is also better with odds.
opening knowledge of normal openings is important in normal chess when it does not give much in games with odds.

[lkaufman]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

My prediction is 4.5 to 3.5 in favor of GM Anthony Wirig, but i would have to say that his Blitz rating is mainly against other human NOT Engines, therefore that Blitz rating is incognito or insignificant when comparing against a top Engine like Komodo MCTS2, since other humans blunder and Engine do NOT under time pressure.

While I don't disagree with your prediction, I don't follow your logic. If his higher rating vs. humans is due to blundering less often, that is precisely the trait needed to do well with knight odds from Dragon. Judging from the match with Toth, a 2500 chess.com blitz rating is not high enough to avoid serious errors at 15' + 10" even with an extra knight, but a 2830 blitz rating may well be good enough.

1)15'+10'' is rapid and not blitz so I doubt if we can learn from blitz rating about result in rapid games.
2)I am not sure if the player who is better in normal chess is also better with odds.
opening knowledge of normal openings is important in normal chess when it does not give much in games with odds.

Experience with chess 960 (FRC) has shown that in general, the same players who are best in normal chess are also best in chess960, where knowledge of normal openings is useless. The correlation is not perfect, clearly some players benefit more than others from knowledge of opening theory, but it seems that the differences in this respect are small, maybe like 20 or 30 elo points. Regarding the blitz rating, if Wirig is stronger than Lenderman in online blitz, and Lenderman was about even at 6' +1" blitz with Dragon at knight odds Armageddon, then Wirig should be favored at that time control, and since longer time controls always favor the human he should be strongly favored at Rapid even with Armageddon scoring. My hope is that we have improved Dragon enough and learned which openings to avoid, but is that enough?

[Chessqueen]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

My prediction is 4.5 to 3.5 in favor of GM Anthony Wirig, but i would have to say that his Blitz rating is mainly against other human NOT Engines, therefore that Blitz rating is incognito or insignificant when comparing against a top Engine like Komodo MCTS2, since other humans blunder and Engine do NOT under time pressure.

While I don't disagree with your prediction, I don't follow your logic. If his higher rating vs. humans is due to blundering less often, that is precisely the trait needed to do well with knight odds from Dragon. Judging from the match with Toth, a 2500 chess.com blitz rating is not high enough to avoid serious errors at 15' + 10" even with an extra knight, but a 2830 blitz rating may well be good enough.

1)15'+10'' is rapid and not blitz so I doubt if we can learn from blitz rating about result in rapid games.
2)I am not sure if the player who is better in normal chess is also better with odds.
opening knowledge of normal openings is important in normal chess when it does not give much in games with odds.

Experience with chess 960 (FRC) has shown that in general, the same players who are best in normal chess are also best in chess960, where knowledge of normal openings is useless. The correlation is not perfect, clearly some players benefit more than others from knowledge of opening theory, but it seems that the differences in this respect are small, maybe like 20 or 30 elo points. Regarding the blitz rating, if Wirig is stronger than Lenderman in online blitz, and Lenderman was about even at 6' +1" blitz with Dragon at knight odds Armageddon, then Wirig should be favored at that time control, and since longer time controls always favor the human he should be strongly favored at Rapid even with Armageddon scoring. My hope is that we have improved Dragon enough and learned which openings to avoid, but is that enough?

And the same apply to No castling chess GM Anand is in better shape than GM Kramnik and it is showing in NC Chess as well

[Chessqueen]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

My prediction is 4.5 to 3.5 in favor of GM Anthony Wirig, but i would have to say that his Blitz rating is mainly against other human NOT Engines, therefore that Blitz rating is incognito or insignificant when comparing against a top Engine like Komodo MCTS2, since other humans blunder and Engine do NOT under time pressure.

While I don't disagree with your prediction, I don't follow your logic. If his higher rating vs. humans is due to blundering less often, that is precisely the trait needed to do well with knight odds from Dragon. Judging from the match with Toth, a 2500 chess.com blitz rating is not high enough to avoid serious errors at 15' + 10" even with an extra knight, but a 2830 blitz rating may well be good enough.

1)15'+10'' is rapid and not blitz so I doubt if we can learn from blitz rating about result in rapid games.
2)I am not sure if the player who is better in normal chess is also better with odds.
opening knowledge of normal openings is important in normal chess when it does not give much in games with odds.

Experience with chess 960 (FRC) has shown that in general, the same players who are best in normal chess are also best in chess960, where knowledge of normal openings is useless. The correlation is not perfect, clearly some players benefit more than others from knowledge of opening theory, but it seems that the differences in this respect are small, maybe like 20 or 30 elo points. Regarding the blitz rating, if Wirig is stronger than Lenderman in online blitz, and Lenderman was about even at 6' +1" blitz with Dragon at knight odds Armageddon, then Wirig should be favored at that time control, and since longer time controls always favor the human he should be strongly favored at Rapid even with Armageddon scoring. My hope is that we have improved Dragon enough and learned which openings to avoid, but is that enough?

[Chessqueen]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

My prediction is 4.5 to 3.5 in favor of GM Anthony Wirig, but i would have to say that his Blitz rating is mainly against other human NOT Engines, therefore that Blitz rating is incognito or insignificant when comparing against a top Engine like Komodo MCTS2, since other humans blunder and Engine do NOT under time pressure.

While I don't disagree with your prediction, I don't follow your logic. If his higher rating vs. humans is due to blundering less often, that is precisely the trait needed to do well with knight odds from Dragon. Judging from the match with Toth, a 2500 chess.com blitz rating is not high enough to avoid serious errors at 15' + 10" even with an extra knight, but a 2830 blitz rating may well be good enough.

1)15'+10'' is rapid and not blitz so I doubt if we can learn from blitz rating about result in rapid games.
2)I am not sure if the player who is better in normal chess is also better with odds.
opening knowledge of normal openings is important in normal chess when it does not give much in games with odds.

Experience with chess 960 (FRC) has shown that in general, the same players who are best in normal chess are also best in chess960, where knowledge of normal openings is useless. The correlation is not perfect, clearly some players benefit more than others from knowledge of opening theory, but it seems that the differences in this respect are small, maybe like 20 or 30 elo points. Regarding the blitz rating, if Wirig is stronger than Lenderman in online blitz, and Lenderman was about even at 6' +1" blitz with Dragon at knight odds Armageddon, then Wirig should be favored at that time control, and since longer time controls always favor the human he should be strongly favored at Rapid even with Armageddon scoring. My hope is that we have improved Dragon enough and learned which openings to avoid, but is that enough?

Mr. Kaufman, are you referring to the small improvement from Komodo MCTS1 to Komodo MCTS2, I thought that you have to feed the correct Knight Openings and which one to avoid such as the 1. f4 ! which favor Komodo MCTS etc... or Komodo MCTS2 plays thousand of openings by itself and learned which one to avoid?

[Chessqueen]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

Next Komodo Dragon2 Knight Odds will be against GM Anthony Wirig https://ratings.fide.com/profile/615013, to be scheduled. Here are some of the Black Opening of GM Anthony Wirig, he has not played against 1. f4

https://www.chess.com/analysis/game/live/18273701861

https://www.chess.com/analysis/game/live/18227515499

https://www.chess.com/game/daily/338277639

[lkaufman]

GM Anthony Wirig (FIDE 2500 but higher on next list, chess.com blitz 2830, born 1983) will take on the latest Komodo Dragon in an eight game Rapid (15' + 10") match on July 21 and July 22, starting 1 pm Eastern Daylight Time, 7 pm Central European Daylight Time each day. Komodo is giving knight odds, alternating between b1 and g1 knight removed. Games will be played on chess.com, while the event can be viewed on twitch. Fifteen minute break between the games to allow time for commentary/discussion. Armageddon scoring, meaning that draws count for Komodo, although there is also a smaller prize for match victory by normal scoring in case it becomes clear that a win with Armageddon scoring is not likely for the GM, as well as $100 per win and $25 per draw so he will still has incentive to hold the draw if it becomes clear that a win is unlikely. In the event that the score is tied 4 to 4 with Armageddon scoring, a ninth tiebreaker game will be played.
The format is essentially the same as in the match which Dragon won from IM Andras Toth (2377 FIDE, 2500 chess.com blitz) by 4.5 to 1.5 recently, except 8 games instead of 6. Based on the FIDE rating difference of 123 (someewhat more with live ratings) Komodo would still be favored, but based on the huge 330 elo gap in chess.com blitz ratings, Wirig should win by 5.5 to 2.5 by normal scoring. Rapid being midway between blitz and standard, the best estimate is that Wirig is about 230 elo stronger at Rapid, so he should be a slight favorite by normal scoring, perhaps a slight underdog by Armageddon scoring. But he has the advantage of having seen the six Toth games, and of having experience playing blitz games with Stockfish top level on lichess at knight for pawn (f7 or g7) odds (2 to 2 score). Knight odds is much larger than knight for f7 or g7 pawn, so I'm not too optimistic, but I hope it will be a competitive match. I think Komodo could probably beat some GMs who are not so strong in blitz or rapid at knight odds, but 2830 chess.com blitz is a very high rating for a 2500 GM, more than 100 above GM Lenderman who was about even with Dragon at Armageddon knight odds at time limits around 6' + 1", so this could mean trouble for the Dragon. We'll see!

My prediction is 4.5 to 3.5 in favor of GM Anthony Wirig, but i would have to say that his Blitz rating is mainly against other human NOT Engines, therefore that Blitz rating is incognito or insignificant when comparing against a top Engine like Komodo MCTS2, since other humans blunder and Engine do NOT under time pressure.

While I don't disagree with your prediction, I don't follow your logic. If his higher rating vs. humans is due to blundering less often, that is precisely the trait needed to do well with knight odds from Dragon. Judging from the match with Toth, a 2500 chess.com blitz rating is not high enough to avoid serious errors at 15' + 10" even with an extra knight, but a 2830 blitz rating may well be good enough.

1)15'+10'' is rapid and not blitz so I doubt if we can learn from blitz rating about result in rapid games.
2)I am not sure if the player who is better in normal chess is also better with odds.
opening knowledge of normal openings is important in normal chess when it does not give much in games with odds.

Experience with chess 960 (FRC) has shown that in general, the same players who are best in normal chess are also best in chess960, where knowledge of normal openings is useless. The correlation is not perfect, clearly some players benefit more than others from knowledge of opening theory, but it seems that the differences in this respect are small, maybe like 20 or 30 elo points. Regarding the blitz rating, if Wirig is stronger than Lenderman in online blitz, and Lenderman was about even at 6' +1" blitz with Dragon at knight odds Armageddon, then Wirig should be favored at that time control, and since longer time controls always favor the human he should be strongly favored at Rapid even with Armageddon scoring. My hope is that we have improved Dragon enough and learned which openings to avoid, but is that enough?

Mr. Kaufman, are you referring to the small improvement from Komodo MCTS1 to Komodo MCTS2, I thought that you have to feed the correct Knight Openings and which one to avoid such as the 1. f4 ! which favor Komodo MCTS etc... or Komodo MCTS2 plays thousand of openings by itself and learned which one to avoid?

We have improved Dragon, especially MCTS, since Dragon 2, but there is no automated editing of the small handicap opening book. All I can do is to remove or downgrade lines that seem to work out poorly in games with humans (mostly myself). All openings are bad for White at knight odds, but some are worse than others and should be avoided.