Skip to content

Update NN challenge results and future plans #196

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Naphthalin merged 1 commit into from
Mar 24, 2026
Merged

Update NN challenge results and future plans #196

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion content/blog/2026/Odds_Blitz_Challenges.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,9 +32,15 @@ Note that unlike in the first challenge, 2'+1" will not be treated as an equival

While theoretically one could win the prize in ten games, I don't expect any human to do that, and it likely will again take a "super" GM (one who has exceeded 2700 FIDE) to win the prize, or at least someone at that level in blitz (2900 Lichess blitz), but I could be wrong. In general I think games at two knights odds tend to be more exciting than at queen for knight, as the bot is more likely to take risks trying to win when it has a queen.

##### UPDATE (2026-03-24): We have a winner!

The NN challenge went live last Friday night. Several strong players attempted it, including streamer "Chess-Network" who scored +9=11-58, IM Kacper Drozdowski ("Kacparov") who scored +5=4-12, GM Keith Arkell ("AtomRod") who scored +1=2-22, GM Jeffery Xiong ("FeegLood") who scored +8=8-12, and finally the winner, GM Andy Woodward ("mraquariyaz67") who scored +10=3-2. I was exactly right in saying that the winner would need a 2900 blitz rating (both Kacper and Andy have that). Andy's games in particular showed that a bit more opening variety is needed (though he would have won anyway); a recent temperature change already implemented for the single piece odds bots should help once implemented for the multipiece bots. Based on the players who also played at queen for knight odds at 3'0" or the similar 2'1", it appears that NN is a slightly larger handicap, perhaps about a half pawn larger. It is also clear that rook odds is closer to knight odds than to two knights odds; the standard 1-3-3-5-9 values are misleading here.

Copilot AI Mar 24, 2026

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The time control notation "2'1"" looks like a typo/inconsistent with the earlier "2'+1"" format used in this post. If this is meant to indicate 2 minutes + 1 second increment, consider changing it to "2'+1"" for clarity.

Suggested change
The NN challenge went live last Friday night. Several strong players attempted it, including streamer "Chess-Network" who scored +9=11-58, IM Kacper Drozdowski ("Kacparov") who scored +5=4-12, GM Keith Arkell ("AtomRod") who scored +1=2-22, GM Jeffery Xiong ("FeegLood") who scored +8=8-12, and finally the winner, GM Andy Woodward ("mraquariyaz67") who scored +10=3-2. I was exactly right in saying that the winner would need a 2900 blitz rating (both Kacper and Andy have that). Andy's games in particular showed that a bit more opening variety is needed (though he would have won anyway); a recent temperature change already implemented for the single piece odds bots should help once implemented for the multipiece bots. Based on the players who also played at queen for knight odds at 3'0" or the similar 2'1", it appears that NN is a slightly larger handicap, perhaps about a half pawn larger. It is also clear that rook odds is closer to knight odds than to two knights odds; the standard 1-3-3-5-9 values are misleading here.
The NN challenge went live last Friday night. Several strong players attempted it, including streamer "Chess-Network" who scored +9=11-58, IM Kacper Drozdowski ("Kacparov") who scored +5=4-12, GM Keith Arkell ("AtomRod") who scored +1=2-22, GM Jeffery Xiong ("FeegLood") who scored +8=8-12, and finally the winner, GM Andy Woodward ("mraquariyaz67") who scored +10=3-2. I was exactly right in saying that the winner would need a 2900 blitz rating (both Kacper and Andy have that). Andy's games in particular showed that a bit more opening variety is needed (though he would have won anyway); a recent temperature change already implemented for the single piece odds bots should help once implemented for the multipiece bots. Based on the players who also played at queen for knight odds at 3'0" or the similar 2'+1", it appears that NN is a slightly larger handicap, perhaps about a half pawn larger. It is also clear that rook odds is closer to knight odds than to two knights odds; the standard 1-3-3-5-9 values are misleading here.

Copilot uses AI. Check for mistakes.

For future challenges we are considering a more flexible approach to the time limit, with players playing until someone has an overall winning record including at least ten won games, and giving prizes to the players who win matches at the shortest time controls for the specified odds.

##### Summary of the challenge conditions:

- [LeelaPieceOdds](<https://lichess.org/@/LeelaPieceOdds>) plays white with both of its knights removed [lc0.org/NN](<https://lc0.org/NN>)
- time control is 3'+0"
- challenge is open to all titled players FM and above (GM, IM, FM) with a 2500+ Lichess blitz rating (50+ games) and their identity revealed, with possible extensions after personal request (see above)
- the first participant who makes an overall plus score which includes at least ten wins (starting now, i.e. 2026-03-21 00:00 UTC) will win the challenge and the $300 prize
- the first participant who makes an overall plus score which includes at least ten wins (starting now, i.e. 2026-03-21 00:00 UTC) will win the challenge and the $300 prize
Loading