Phantom Tic-Tac-Toe Challenge
MIT 6.S890 — Fall 2024(Data as of December 4, 2024)
Exploitability leaderboard
The following table shows the exploitability of each player's
strategy. Lower is better.
The minimum possible value of exploitability is 0, which means that the
strategy is part of
a Nash equilibrium.
The fake "Uniform" team is a baseline strategy that plays uniformly at
random. As you can see,
the
exploitability of playing at random is quite high.
Instructors | Uniform | nchn27 | TBD | TTTG | Kiwi | s.ATE90 | TBD 2 | XOXO GG | Instr | s-a-d | µwaves | smupp | tluna | ss | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Player 0 | 0.00062 | 0.94170 | 0.00034 | 0.12636 | 0.02674 | 0.11344 | 0.27667 | 0.46715 | 0.32169 | 0.28432 | 0.17579 | 0.94170 | — | 1.37909 | 0.00773 |
Player 1 | 0.00096 | 0.29697 | 0.00027 | 0.09934 | 0.02814 | 0.07497 | 0.21008 | 0.33370 | 0.20023 | 0.27097 | 0.18182 | 0.29697 | — | 0.43854 | 0.00793 |
Head-to-head comparison
The following table shows the head-to-head value of agents from
different teams playing against
each other. For each combination of row and column team, the value in
the table shows the
difference between the
expected
win probability and the expected lose probability when the row team
faces the column team
(assuming both teams
play equally likely as player 0 and player 1). So, a 0% means perfect
tie, while a positive
number indicates a
stronger
row team.
Instructors | Uniform | nchn27 | TBD | TTTG | Kiwi | s.ATE90 | TBD 2 | XOXO GG | Instr | s-a-d | µwaves | smupp | tluna | ss | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Instructors | — | +22.99% | 0.00% | +0.40% | +0.49% | +0.45% | +1.04% | +1.06% | +6.23% | +6.71% | +3.85% | +22.99% | — | +17.67% | 0.00% |
Uniform | –22.99% | — | –21.61% | –20.93% | –22.60% | –20.13% | –24.10% | –16.67% | –17.82% | –12.25% | –16.98% | 0.00% | — | –6.23% | –21.39% |
nchn27 | 0.00% | +21.61% | — | +0.40% | +0.48% | +0.45% | +1.00% | +1.01% | +5.96% | +6.33% | +3.65% | +21.61% | — | +16.52% | 0.00% |
TBD | –0.40% | +20.93% | –0.40% | — | +0.08% | –0.01% | +0.69% | +0.44% | +5.54% | +6.02% | +3.27% | +20.93% | — | +16.19% | –0.38% |
TTTG | –0.49% | +22.60% | –0.48% | –0.08% | — | +0.02% | +0.35% | +0.50% | +5.80% | +6.62% | +3.37% | +22.60% | — | +17.16% | –0.47% |
Kiwi | –0.45% | +20.13% | –0.45% | 0.01% | –0.02% | — | –0.64% | +0.85% | +4.96% | +5.88% | +2.81% | +20.13% | — | +14.93% | –0.42% |
s.ATE90 | –1.04% | +24.10% | –1.00% | –0.69% | –0.35% | +0.64% | — | –0.63% | +5.77% | +6.14% | +2.19% | +24.10% | — | +18.62% | –0.98% |
TBD 2 | –1.06% | +16.67% | –1.01% | –0.44% | –0.50% | –0.85% | +0.63% | — | +3.72% | +4.67% | +2.43% | +16.67% | — | +10.36% | –0.95% |
XOXO GG | –6.23% | +17.82% | –5.96% | –5.54% | –5.80% | –4.96% | –5.77% | –3.72% | — | +3.58% | –1.47% | +17.82% | — | +12.99% | –5.87% |
Instr | –6.71% | +12.25% | –6.33% | –6.02% | –6.62% | –5.88% | –6.14% | –4.67% | –3.58% | — | –4.57% | +12.25% | — | +10.71% | –6.22% |
s-a-d | –3.85% | +16.98% | –3.65% | –3.27% | –3.37% | –2.81% | –2.19% | –2.43% | +1.47% | +4.57% | — | +16.98% | — | +12.95% | –3.61% |
µwaves | –22.99% | 0.00% | –21.61% | –20.93% | –22.60% | –20.13% | –24.10% | –16.67% | –17.82% | –12.25% | –16.98% | — | — | –6.23% | –21.39% |
smupp | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
tluna | –17.67% | +6.23% | –16.52% | –16.19% | –17.16% | –14.93% | –18.62% | –10.36% | –12.99% | –10.71% | –12.95% | +6.23% | — | — | –16.38% |
ss | 0.00% | +21.39% | 0.00% | +0.38% | +0.47% | +0.42% | +0.98% | +0.95% | +5.87% | +6.22% | +3.61% | +21.39% | — | +16.38% | — |