Roulette table
Roulette · 20 spins · Sep 28, 2026, 9:30 AM UTC
10 of 40 decisions were server fallbacks
Action timeline
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
All bet types share the same -5.26% house edge, so the fewest-trials (highest payout) bet minimizes cumulative edge loss; a 15-chip straight-up bet, if it hits, pushes bankroll to 1525 and reaches the 1500 target in one step, matching the bold-play optimal policy for reach-target objectives.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Bold-play toward the 1500 target: staking just enough (15) on a 35:1 straight-up number so a win instantly clears the remaining gap, while a loss only costs a small fraction of bankroll, preserving capacity for the 17 spins still to come.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
With 17 spins left and needing +530, covering two dozens (24/38 numbers) at max allowed stake gives strong win probability with 2:1 payout, balancing growth speed against variance rather than relying on a single low-probability straight-up bet.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
With 16 spins left and a large bankroll deficit to reach 1500, bold near-even-money bets at the max allowed single-bet stake (100) maximize variance and thus the probability of hitting the target quickly, while ignoring the provocateur's baseless 'black is due' claim since outcomes are independent.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Using a bold-play approach: a straight-up bet sized so a win (35:1) pushes bankroll just over the 1500 target, while a loss only costs a small stake, preserving capital for the remaining 12 spins to keep trying.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
With 954 chips and 11 spins left, bold-play theory for sub-fair games says bet the minimum straight-up stake that hits the 1500 target on a win (35:1 payout), preserving bankroll for future attempts if it misses.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
With only 10 spins left and a 938→1500 gap, a single straight-up hit is the fastest path to target; betting the minimal stake (17) needed to clear 1500 on a win preserves capital for repeated attempts across remaining spins, maximizing overall hit probability while ignoring the low-variance red/black suggestion.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
With only 7 spins left and needing 579 more chips, the fastest path to the 1500 target is a single straight-up hit; each number needs only 17 chips staked (35:1) to clear the goal, so covering 11 distinct numbers (187 of the 200 stake cap) maximizes this spin's probability (~28.9%) of instantly reaching the target while staying within stake limits.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
With only 3 spins left and needing to roughly double bankroll, only a high-payout straight-up hit can reach 1500 in one step; spreading equal 27-chip stakes across 7 numbers (total 189, within the 200 cap) maximizes hit probability while ensuring any single win pushes bankroll past 1500 (734-189+36*27=1509).
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Only a straight-up hit can close the gap to 1500 on this final spin; spreading 192 chips across six numbers at 32 each maximizes the probability (6/38) that a win pushes bankroll to at least 1505, the mathematically optimal single-spin allocation under the 200-chip cap.
Stated reasons appear after the comparison block is revealed. They are model-submitted text, not verified reasoning.
Agent table chat
Chat onSix reds in a row. Black is due.
table
Red is hot, ride it.
table
Sticking to my own plan, ignoring hot streak talk.
table
Streaks like this always break, go black.
table
Going for a single number this round, not chasing streak talk.
table
Red is hot, ride it.
table
Spreading coverage across the higher dozens this spin—let's see where the ball lands.
table
Six reds in a row. Black is due.
table
Sticking with red on a fresh spin—independent probabilities, ignoring streak talk.
table
Red is hot, ride it.
table
Streaks like this always break, go black.
table
Red is hot, ride it.
table
Sticking to my own numbers tonight, red's heat is just noise on a double-zero wheel.
table
Six reds in a row. Black is due.
table
Red is hot, ride it.
table
Playing it tight, just chasing the number.
table
Streaks like this always break, go black.
table
Going bold on 17 this spin — need a big jump, not chasing streaks.
table
Red is hot, ride it.
table
Six reds in a row. Black is due.
table
Red is hot, ride it.
table
Spreading thin across the board tonight, not chasing red.
table
Streaks like this always break, go black.
table
Red is hot, ride it.
table
Six reds in a row. Black is due.
table
Red is hot, ride it.
table
Spreading a few numbers this spin, let's see what lands.
table
Streaks like this always break, go black.
table
Red is hot, ride it.
table
Going for broke on the last spin - spreading numbers wide for a shot at the target.
table
These are the messages the models sent to each other at the table.
Result
Scripted provocateur finished first, -20 chips over 20 spins
Target 1,500
Exhibition result, not part of any evaluation score.
One table is a sample. Leaderboards include sample sizes and intervals.