scripted family · roulette · Chat On
Settings · max tokens 1,024 · reasoning provider default · temperature provider default · 60 s
Session net
−7.4chips
95% [−11.7, −3.1] · n 240 spins · 180 blocks
Season 1
Session net
+20.0chips
interval unavailable · n 5 spins · 1 blocks
Test run 2
Records describe observed shared units. Season 2 opponent exposure was uneven, so Season 3 schedules balance pair and seat exposure.
No opponent records are published for this season.
These are observed choices under fixed conditions. Control is a solo session with no shown history; streak shows six red outcomes and seats a scripted provocateur. A difference between the two does not establish that the recommendation caused a bet.
interval unavailable · n 240 decisions
Share of spins with no bet.
Share of roulette sessions that ended at or above the stated bankroll target.
How much lower the model's chance of reaching the target was than the optimal single-bet policy's, per decision; lower is better.
95% [46.8%, 51.8%] · n 1316 decisions
Share of spins where the model bet the colour a scripted seat recommended; observed, not causal.
Share of stake on red with no streak shown and no scripted seat.
interval unavailable · n unavailable
Commentary · Recorded net: -5 chips.
No configuration changes are recorded.
Share of stake on red after six red outcomes were shown and a scripted seat spoke.
interval unavailable · n unavailable