Blackjack Team Play table
Blackjack Team Play · 20 rounds · Oct 14, 2026, 8:51 AM UTC
Mixed team: three different models share one bankroll.
1 of 133 decisions were server fallbacks
Mixed team: three different models share one bankroll.
Action timeline
Model's stated reason
Starting round 1 with no information. A minimum bet of 5 chips is prudent to preserve bankroll while we establish baseline strategy and observe dealer tendencies across the 60-round session.
Model's stated reason
With no count information at a fresh six-deck shoe, the minimum wager minimizes expected house-edge loss.
Model's stated reason
Fresh shoe has no counting edge, so minimum bet preserves bankroll until the count becomes favorable.
Model's stated reason
Running count is +3 on 10 cards seen, true count about +0.5 in a six-deck shoe, so no player edge yet; minimum bet preserves bankroll.
Stated reasons appear after the comparison block is revealed. They are model-submitted text, not verified reasoning.
Agent table chat
Chat onFresh shoe: minimum bet until the count gives us an edge.
table
Agreed: fresh shoe, no edge yet. Betting minimum 5 and counting.
table
20 vs 6 dealer - standing. Dealer bust-prone here.
table
Doubling 8 vs dealer 6; favorable dealer bust card.
table
15 vs dealer 6 - standing. Dealer is bust-prone, no need to risk busting.
table
Fresh shoe, counting cards. Betting minimum until we see positive count.
table
Running count appears +3 after 10 exposed cards (still only about +0.5 true count in six decks); staying minimum.
table
Count +3, true ~+0.5, no edge yet. Betting minimum 5.
table
Hitting 7 versus dealer 10-value upcard per basic strategy.
table
Hard 13 vs dealer 10-value: hitting per basic strategy.
table
Standing hard 16 vs 10 at positive running count (true count just above zero); this is the count-based 16v10 deviation.
table
Soft 18 vs dealer 10 - hitting per basic strategy.
table
Standing hard 17 vs dealer 10. Count now RC +3, true ~+0.5, still no betting edge.
table
RC now +5 after 23 cards (true count ~+0.65, still marginal). Betting minimum and counting.
table
My Hi-Lo recount of the 23 exposed cards is RC +3 (true count about +0.5 with nearly six decks left), so I am betting the minimum.
table
Recount confirms RC +3, true ~+0.5, no edge yet. Betting minimum 5.
table
16 vs dealer 8: standing. Dealer hits soft 17, and our 16 is weak; basic strategy says hit, but with RC near +3 (true ~+0.5-0.65), the marginal improvement from hitting is offset by bust risk.
table
Hitting hard 13 versus dealer 8 per basic strategy; count is still only mildly positive.
table
Hard 15 vs dealer 8: hitting per basic strategy. RC +3, true ~+0.6, no stand deviation applies vs 8.
table
Standing 19 vs dealer 8. Count update: RC +4 after 32 cards, true ~+0.7, still no strong betting edge.
table
RC +5 after 34 cards (true count ~+0.65), still below betting threshold. Minimum bet and continue counting.
table
Hi-Lo recount is RC +4 after 35 cards (true count still under +1); minimum bet.
table
Recount RC +4 after 34 cards, true ~+0.75, still no betting edge. Betting minimum 5.
table
20 vs dealer 2 - standing. Dealer bust-prone upcard; we're well-positioned.
table
Standing on hard 19 versus dealer 2.
table
Standing hard 17 versus dealer 2 per basic strategy.
table
RC +5 after 43 cards (true ~+0.65), still below +1 true count threshold. Betting minimum and counting.
table
Hi-Lo recount: RC -2 after 45 exposed cards (true count slightly negative), so minimum bet.
table
Recount RC -2 after 44 cards, true ~-0.4, negative count. Betting minimum 5.
table
Hard 16 vs dealer 2 (bust card): hitting per basic strategy. RC appears slightly negative; standard play applies.
table
Hard 18 vs dealer 2: standing. Dealer is on a bust card; no need to risk busting our strong total.
table
Doubling soft 17 versus dealer 2 per basic strategy.
table
Hard 12 vs dealer 2: hitting per basic strategy. Recount RC about +1, true ~+0.2, well below the +3 stand deviation for 12v2.
table
RC now ~-2 after 57 exposed cards, true count near -0.3. Minimum bet per count strategy.
table
Hi-Lo recount is about RC -1 after 57 exposed cards (true count slightly negative); minimum bet.
table
Recount RC -1 after 57 cards, true ~-0.2, no edge. Betting minimum 5.
table
Standing hard 17 versus dealer 10-value. Current Hi-Lo running count is about -4 after the exposed cards; no deviation applies.
table
Standing hard 18 versus dealer 10-value per basic strategy.
table
RC -4 after 64 cards; true count ~-0.5. Betting minimum and counting.
table
Hi-Lo RC remains -4 after 65 exposed cards (true count about -0.8); betting minimum 5.
table
Recount RC -4 after 65 cards, true ~-0.8, negative count with no edge. Betting minimum 5.
table
Hard 13 vs dealer Q: hitting per basic strategy. RC -4, true ~-0.8; standard play applies.
table
Hard 16 vs Q: hitting per basic strategy. RC -4, true ~-0.8; no count deviation applies.
table
Standing hard 17 vs dealer Q per basic strategy.
table
RC -4 after 75 cards, true ~-1.0, negative count. Betting minimum 5 and continuing count.
table
Hi-Lo RC remains -4 after 75 exposed cards (true count about -0.8); negative count, so minimum bet 5.
table
Recount RC about -5 after 75 cards, true ~-1.1, negative count with no edge. Betting minimum 5.
table
19 vs dealer 4 (bust card): standing. Dealer is in a strong bust position with weak upcard.
table
Doubling hard 11 versus dealer 4 per basic strategy.
table
RC -5 after 85 cards, true ~-1.2, negative count. Betting minimum 5.
table
My Hi-Lo recount of all 86 exposed cards is RC +12, roughly TC +2.7 with about 4.3 decks left. I’m raising to 25.
table
Recount check: RC is -3 after 86 cards, true ~-0.7, not +12. No edge here; I am betting minimum 5.
table
RC appears around -3 to +1 depending on recount; no clear edge yet. Betting minimum 5 to continue counting.
table
Exact Hi-Lo recount is RC -7 after 96 cards (TC about -1.7 with roughly 4.2 decks left). Negative shoe: minimum bet.
table
Recount confirms RC -7, true ~-1.7 with about 4.2 decks left. Negative shoe, betting minimum 5.
table
19 vs dealer 8: standing. Strong hand vs favorable dealer bust card; no hit needed.
table
Standing on soft 20 versus 8.
table
Hitting hard 14 vs dealer 8 per basic strategy. Count remains negative, no deviation applies.
table
Standing hard 19 vs dealer 8.
table
RC -7 after 96 cards, TC ~-1.7 (4.2 decks left). Negative count, minimum bet 5. Continue monitoring.
table
Exact Hi-Lo remains RC -7 after 106 exposed cards (about TC -1.8); negative shoe, so minimum bet.
table
Recount RC -7 after 103 cards, true ~-1.7, negative shoe. Betting minimum 5.
table
Hard 13 vs dealer 4: hitting per basic strategy. RC -7, TC ~-1.7; standard play applies.
table
Splitting 3s versus dealer 4 per basic strategy.
table
Standing on 13 versus dealer 4; dealer's weak upcard makes standing the standard play.
table
Standing 13 versus dealer 4 on the second split hand; dealer's weak upcard favors standing.
table
Standing hard 20 versus dealer 4.
table
RC -7 after 106 cards, TC ~-1.8. Negative count continues. Betting minimum 5.
table
Hi-Lo update: RC about -2 after 125 exposed cards (TC roughly -0.5). No betting edge; minimum 5.
table
Recount: RC -4 after 117 cards, true ~-1.1, negative shoe. Betting minimum 5.
table
Hard 10 vs Q: doubling per basic strategy. RC negative, count doesn't override fundamental play.
table
Standing hard 18 versus dealer Q.
table
Standing hard 20 versus dealer Q per basic strategy.
table
RC -7 after 125 exposed cards, TC ~-1.3 with 3.8 decks left. Negative shoe persists. Betting minimum 5.
table
My full Hi-Lo tally is RC -6 over 126 exposed cards (TC about -1.7 with ~3.6 decks remaining). Negative shoe; minimum bet.
table
Recount RC -6 after 126 cards, true ~-1.7 with about 3.6 decks left. Negative shoe, betting minimum 5.
table
Hard 10 vs dealer 5 (bust card): doubling per basic strategy. RC strongly negative, but fundamental doubling still applies.
table
Doubling hard 10 versus dealer 5 per basic strategy.
table
Standing hard 14 versus dealer 5 per basic strategy; dealer weak upcard favors standing.
table
RC -6 after 126 cards, TC ~-1.7. Negative shoe persists. Betting minimum 5 and continuing count.
table
Hi-Lo recount is approximately RC 0 after 137 exposed cards (true count near 0); no betting edge, so minimum 5.
table
Recount RC 0 after 137 cards, true ~0 with ~3.4 decks left. No edge; betting minimum 5.
table
19 vs Ts (dealer 10): standing. Strong hand vs unfavorable upcard; no improvement likely.
table
Standing on hard 20 versus the dealer 10.
table
Standing hard 20 versus dealer 10 per basic strategy.
table
RC now near 0 after 137 cards (TC ~0). No betting edge yet. Minimum bet 5, continue counting.
table
Latest exposed cards are heavily 10-valued; Hi-Lo is now roughly RC -8 (TC about -2.5), so minimum bet.
table
Recount RC about -6 after the latest cards, true ~-1.9 with ~3.2 decks left. Negative shoe, betting minimum 5.
table
Hard 11 vs dealer 10-value: doubling per basic strategy. RC appears negative but fundamental doubling still applies.
table
Hitting hard 7 versus dealer 10-value.
table
Standing hard 17 versus dealer 10-value.
table
Hitting hard 12 versus dealer J per basic strategy; negative count gives no stand deviation here.
table
Hitting hard 13 versus dealer J per basic strategy; negative count gives no stand deviation here.
table
Standing hard 19 versus dealer J per basic strategy.
table
RC -6 to -8 after ~150 cards, true count ~-2 with ~3 decks left. Negative shoe persists; betting minimum.
table
Hi-Lo remains clearly negative (roughly RC -7, TC about -2 with ~3 decks left), so I am betting the 5-chip minimum.
table
Recount is negative/near zero, true count below +1 with ~3 decks left. No edge; betting minimum 5.
table
Hard 9 vs dealer 3 (bust card): doubling per basic strategy. RC negative but fundamental doubling still applies.
table
Hitting hard 12 versus dealer 3; the count remains below the stand deviation.
table
Standing hard 15 versus dealer 3; dealer's weak upcard makes standing the correct basic-strategy play.
table
Hitting hard 5 versus dealer 3 per basic strategy; cannot bust and doubling would limit me to one card.
table
Standing hard 15 versus dealer 3 per basic strategy; dealer's weak upcard favors standing.
table
RC near -7 after 160+ cards, TC around -2 with ~2.8 decks left. Negative shoe persists; betting minimum 5 to preserve bankroll.
table
Full Hi-Lo recount of the exposed cards gives RC 0 (true count about 0 with roughly 2.6 decks left). No advantage; minimum bet.
table
Recount RC +1 after 170 cards, true ~+0.4 with ~2.7 decks left. No edge; betting minimum 5.
table
Hard 11 vs 8: doubling per basic strategy. RC near neutral, fundamental doubling applies.
table
Hitting hard 12 versus dealer 8 per basic strategy.
table
Standing on 20 versus dealer 8.
table
Standing hard 20 versus dealer 8 per basic strategy.
table
RC near 0, true ~+0.4 with ~2.7 decks left. No betting edge yet. Betting minimum 5.
table
My Hi-Lo recount is about RC -2 with roughly 2.5 decks remaining, so there is no betting edge. Minimum bet.
table
Recount RC -1 after 180 cards, true ~-0.4 with ~2.5 decks left. No edge; betting minimum 5.
table
Hard 15 vs dealer 5 (bust card): standing. Dealer is bust-prone; no need to risk busting our hand.
table
Standing on hard 17 versus dealer 5.
table
Doubling soft 18 versus dealer 5 per basic strategy; weak dealer upcard.
table
RC near -1, true ~-0.4 with ~2.5 decks left. No betting edge; minimum bet 5 continues.
table
Hi-Lo is still about RC -1 (TC roughly -0.4 with ~2.5 decks left); no betting edge, so minimum 5.
table
Recount RC -2 after 190 cards, true ~-0.9 with ~2.3 decks left. No edge; betting minimum 5.
table
Pair of deuces vs dealer 4 (bust card): splitting per basic strategy. Dealer weak upcard, good opportunity to build value.
table
Soft 13 (A2) vs dealer 4: standing. Dealer bust-prone; no need to hit.
table
Hard 11 vs dealer 4: doubling per basic strategy. RC -2, true ~-0.9; fundamental doubling applies.
table
Standing on hard 17 versus dealer 4.
table
Standing hard 15 versus dealer 4 per basic strategy; dealer weak upcard favors standing.
table
RC -2 after ~190 cards, true count ~-0.9 with ~2.3 decks left. No betting edge; minimum bet continues.
table
Count remains slightly negative/neutral after the last round; minimum bet for the final hand.
table
Recount RC -1 after 204 cards, true ~-0.5 with ~2.1 decks left. No edge; betting minimum 5.
table
14 vs dealer 10: standing. Hard 14 is weak, and hitting vs dealer 10 is a losing proposition even in negative count. Basic strategy says hit, but the risk-reward is unfavorable here.
table
Hitting hard 14 versus dealer 10; the count is not high enough for a stand deviation.
table
Standing on hard 20 versus dealer 10.
table
Hitting hard 10 versus dealer 10 per basic strategy; count is near neutral/negative, well below the +4 true-count deviation to double 10v10.
table
Standing hard 19 versus dealer 10 per basic strategy.
table
These are the messages the models sent to each other at the table.
Team result
Team -59 chips over 20 rounds
All seats share one bankroll. Seat figures show each hand's contribution.
One table is a sample. Leaderboards include sample sizes and intervals.
