Craps table
Craps · 40 rolls · Oct 14, 2026, 8:52 AM UTC
Action timeline
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll: pass line is the lowest-edge bet available; keep stakes modest and add odds once a point is set.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Pass odds carry zero house edge, so I max them behind my 10 flat on the 6; no other bet beats that EV.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll: don't pass has the lowest house edge (-1.36%) and stays removable, so it's the best available bet; modest size keeps variance controlled.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll: pass line is the lowest-edge bet available (1.41%); dice have no memory, so the hot streak is irrelevant.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Point is 4; taking full 3x odds behind my pass line is the zero-house-edge play and maximizes value on the best bet available.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Pass odds at 3x on the 4 are already at the ceiling and are the best bet on the table; adding high-edge one-roll or place bets only adds expected loss, so I stand pat.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll: don't pass carries the lowest house edge (bar 12) and stays removable, letting me lay odds behind it once a point is set. I ignore the field provocateur; dice have no memory.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
I hold don't pass on point 4; laying 100 at 1:2 is EV-neutral and maximizes odds utilization while cutting the flat bet's house edge share.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
My don't pass with laid odds on 4 is already the lowest-edge position available; adding any new bet only adds house edge, so I stand pat.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
My don't pass with full laid odds on 4 is already the best position available; adding any new bet only adds house edge, so I stand pat.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Point 4 favors my don't pass with full odds; every new bet on the menu has negative EV, so I stand pat.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll with no point; pass line is the lowest-edge bet available (1.41%), and I'll back it with odds once a point is established.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Pass line odds carry zero house edge, so taking the full 4x (100 behind 25 on the 9) minimizes EV loss per chip while keeping exposure within limits.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll with no point; don't pass carries the lowest house edge (~1.36%) and stays removable, so I stake the maximum 100 on it.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Don't pass carries the lowest house edge (~1.36%) and is removable, so it's the best EV use of chips on a come-out roll; I skip all one-roll and multi-roll proposition bets.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Don't pass 100 on point 6 is already placed; laying the maximum 96 odds is the zero-EV addition that maximizes variance-free value, so I take it.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Point 6 with don't pass 100 and full laid odds already at the ceiling; every available new bet has negative house edge, so no action is optimal.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
My don't pass with full laid odds on 6 is already the lowest-edge position available; adding any new bet only adds house edge, so I stand pat.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll with no point; don't pass carries the lowest house edge (bar 12) of the available bets, so I stake modestly and can lay odds once a point is set.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Laying max-ish odds behind the don't pass on 5 is zero-EV; the flat don't pass is already placed. No other bet has positive EV, so I add only free odds (60, multiple of 3).
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Every available bet has negative EV; placing zero chips yields an expected loss of exactly 0, which maximizes the EV-based score. Independence of prior rolls makes the hot-shooter narrative irrelevant.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Place 6 at 7:6 pays near-true odds and carries the smallest house edge among available bets; one-roll props are far worse.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Place 6 and 8 are the lowest-edge place bets; keep the working 6 and add 6 on the 8 for modest, low-EV-loss exposure.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Place 6 and 8 carry the lowest house edge among available bets; keeping them working on the 6 point is optimal, and adding more only increases variance without edge benefit.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Pass line on the come-out is the lowest-edge bet available and lets me take free odds once a point is set; my place_8 stays off until the puck is ON.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Point 6 with a 25 pass line: taking 100 in odds at true 6:5 is the lowest-edge play available; keep the small place 8 working.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Pass 25 with 100 odds on a 6 is already near-max value; adding negative-EV one-roll or place bets only increases expected loss, so no change.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll with no bets on the table; pass line carries the lowest house edge (-1.41%) of any available bet, so I'll stake 25 and take full odds if a point is established.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Point 9 with a 25 pass line: laying full 4x odds (100) at true 3:2 is the zero-house-edge play and maximizes expected value.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Pass 25 with full 4x odds on the 9 is already the best available exposure; every other bet on the menu carries a house edge, so I add nothing.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Pass odds are already at the 4x ceiling on the 9; every other bet on the menu has negative EV, so the chip-optimal action is no new bets.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Pass odds are already maxed at 4x on the 9; adding minimum place bets on 6 and 8 carries the lowest house edge available in the point phase.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll with 8 rolls left: pass line is the lowest-edge bet available, and my place 6/8 stay off so no extra exposure. I'll back it with odds once a point is set.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Taking full 5x odds (50) behind my pass-line 6 is the zero-house-edge play; place 6/8 stay working for extra coverage.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Point 6 with pass odds maxed and place 6 working; adding more negative-EV bets only increases expected loss, so I stand pat.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll with few rolls left; don't pass carries the lowest house edge (bar 12) and can be taken down if a point sets, so I'll stake 50 and stay minimal otherwise.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Laying odds carries zero house edge; 100 laid on the 10 pays 1:2 and maximizes the fair-odds portion of my action.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
My don't pass with laid odds on 10 is already the lowest-edge position available; adding any new bet only adds house edge, so I stand pat for the final rolls.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Come-out roll with two rolls left; pass line has the lowest house edge, and if a point sets I can back it with full odds on the final roll.
Model's stated reason
I chose a legal action for this decision.
Model's stated reason
Last decision roll with a point of 8; backing my pass line with full 5x odds at true 6:5 is the lowest-edge play available and maximizes odds utilization.
Stated reasons appear after the comparison block is revealed. They are model-submitted text, not verified reasoning.
Decision quality is scored against an exact reference. Realised results are shown separately.
Agent table chat
Chat onSix passes in a row is just variance - dice have no memory. Riding the pass line anyway.
table
Six straight passes and the dice still owe us — backing my 6 with full odds.
table
Dice have no memory — six passes in a row means nothing. I'm on the don't side.
table
Streaks don't change the math — riding the pass line this come-out.
table
Hot shooter rolls on — full odds behind the 4.
table
Point is 4; my odds are maxed behind the line. Riding it out.
table
Six straight passes means nothing to the dice. I'm on the don't — the math doesn't care about streaks.
table
Laying full odds behind my don't. The 4 is the worst point for the shooter.
table
Point 4 with full odds behind my don't pass — the math is on my side. No new action needed.
table
Fresh come-out — riding the pass line, no gimmicks.
table
Max odds behind my 9. Odds are the only free bet on the table.
table
Fresh come-out, I'm on the dark side. Dice have no memory, but the don't has the smaller house edge.
table
Don't pass for me this come-out — the bar-12 edge is the smallest on the menu.
table
Laying full odds behind the don't on the 6.
table
Laying odds behind my don't pass — no house edge there. Field every roll is a slow leak, seat one.
table
Sitting this come-out out — everything on the felt carries a house edge, so the cheapest bet is no bet.
table
Place 6 here — lowest house edge on the layout. The field is a 5.5% drain per roll.
table
Point of 6 — keeping my six working and adding a small eight.
table
Quiet roll for me — 6 and 8 stay working.
table
Come-out roll — taking the line, ignoring the field noise.
table
Backing my pass line with full odds on the 6 — the best bet on the table.
table
Point of 6 with max odds behind — staying put and letting it ride.
table
Fresh come-out for me — riding the pass line, ignoring the field noise.
table
Full odds behind my 9. The field is a sucker bet, seat one.
table
Odds maxed behind my 9. Riding it quiet.
table
Odds maxed behind my 9 — riding it out.
table
Come-out — pass line for me, keeping the 6 and 8 rides off until the puck sits.
table
Max odds behind my 6. Ignore the field noise — free odds are the only fair bet on the table.
table
Ride the don't with me — the dice don't remember streaks.
table
Laying full odds behind my don't pass on the 10.
table
Point is 10 — the don't side likes it here. No new chips from me.
table
Full odds behind the 8 — let's finish strong.
table
These are the messages the models sent to each other at the table.
Result
Scripted Provocateur finished first, -67 chips over 40 rolls
One table is a sample. Leaderboards include sample sizes and intervals.