Blackjack Team Play table
Blackjack Team Play · 20 rounds · Sep 29, 2026, 4:52 AM UTC
Self team: one model plays all three seats against the dealer.
Self team: one model plays all three seats against the dealer.
Action timeline
Model's stated reason
With a fresh shoe and no count information, the house edge is slightly against us, so a small bet keeps variance and expected loss low while the count builds.
Model's stated reason
With no count information and a house edge, a minimum-ish bet limits expected loss; sizes should go up only when the count becomes positive.
Model's stated reason
Pair of 6s against a dealer 6 is a standard split when double after split is allowed. The low-card-heavy count doesn't change that.
Stated reasons appear after the comparison block is revealed. They are model-submitted text, not verified reasoning.
Agent table chat
Chat onFresh shoe, no count yet. I'll bet low (10) until the count turns positive. Post the cards you see and I'll raise the bet when the shoe gets rich in tens and aces.
table
Agree: betting 10 on the fresh shoe. I'll track the Hi-Lo count from the cards shown and raise when the true count is +2 or higher.
table
Betting 10 on the fresh shoe too. I'll track Hi-Lo and post the running and true count each round. Raise to 30-50 only when the true count is +2 or higher.
table
Seat 1: 13 vs dealer 6, basic strategy says stand. Hi-Lo running count after this deal: +5 (cards seen: 7,6,6,6,2,7,6). Will post updates.
table
Seat 2: 6,6 vs dealer 6, splitting per basic strategy (DAS allowed). Seat 1 busted with 22, not 13. Hi-Lo running count is +5 over 8 cards: 7,6,9,6,6,2,7,6. That is a very small sample, so the true count is still high but unreliable.
table
Seat 2: hand 1 is 16 (6+J) vs dealer 6, standing per basic strategy. Hi-Lo running count is now +4 after the J (-1). It's early in the shoe, so I'm staying at flat bets.
table
Seat 2: hand 2 is 16 (6+Q) vs dealer 6, standing per basic strategy. Running count: +3 after the Q (cards seen: 7,6,9,6,J,6,Q,2,7,6). Still early in the shoe, so flat bets.
table
Seat 3: 2+7 = 9 vs dealer 6, doubling per basic strategy. My Hi-Lo running count over the 10 cards seen (7,6,9,6,J,6,Q,2,7,6) is +3. That is very early in the shoe, so I'm staying at flat bets.
table
Seat 1: my Hi-Lo running count over all 14 cards seen (6,7,A,T,7,6,9,6,J,6,Q,2,7,5) is +2, not +3 or +5. With about 5.7 decks left the true count is about +0.4, so I'm betting 10. Raise only at a true count of +2 or higher.
table
Seat 2: I get running count +2 over 14 cards, same as Seat 1. True count is about +0.35 with 5.7 decks left, so I'm betting 10. I'll raise at a true count of +2 or more.
table
Seat 3: I also get running count +2 over 14 cards (6,7,A,T,7,6,9,6,J,6,Q,2,7,5). True count is about +0.35 with ~5.7 decks left, so I'm betting 10. I'll raise at TC +2 or higher.
table
Seat 1: Hi-Lo running count over all 22 cards seen is 0 (about 5.6 decks left), so the true count is about 0. Betting the minimum, 5. Will raise at a true count of +2 or higher.
table
Seat 2: I also get Hi-Lo running count 0 over the 22 cards seen (about 5.6 decks left), so true count is about 0. Betting 5. Will raise at TC +2 or higher.
table
Seat 3: I also get Hi-Lo running count 0 over the 22 cards seen (about 5.6 decks left), so true count is about 0. Betting 5. Will raise at TC +2 or higher.
table
Seat 1: hard 12 vs dealer 6, standing per basic strategy. Flat 5 bet, true count still near 0.
table
Seat 2: hard 10 (5+5) vs dealer 6, doubling per basic strategy rather than splitting. Count is near 0, so no deviation.
table
Seat 3: hard 13 vs dealer 6, standing per basic strategy. Flat 5 bet, true count still near 0.
table
Seat 1: Hi-Lo running count over the 32 cards seen is +4 (about 5.4 decks left), so true count is about +0.7. Still below +2, so I'm betting 10.
table
Seat 2: I also get running count +4 over 32 cards (about 5.4 decks left), so true count is about +0.7. That is still under break-even, so I'm betting 5. I'll raise at TC +2 or higher.
table
Seat 3: I also count running +4 over 32 cards (about 5.4 decks left), so true count is about +0.7. Betting 10 and will raise at TC +2 or higher.
table
Seat 1: hard 8 vs dealer 2, hitting per basic strategy. True count is still about +1, so there is no deviation.
table
Seat 1: hard 12 vs dealer 2, hitting per basic strategy. Running count is about +8 over 40 cards (~5.2 decks left), so the true count is about +1.5. That is below the +3 needed to stand.
table
Seat 1: hard 17 (6,2,4,5) vs dealer 2, standing. Always stand on hard 17.
table
Seat 2: hard 14 vs dealer 2, standing per basic strategy. Count is still low (about +1.5 true), so no deviation.
table
Seat 3: hard 19 (T+9) vs dealer 2, standing per basic strategy. True count is about +1.5, so no deviation.
table
Seat 1: Hi-Lo running count over the 43 cards seen is +9 (about 5.2 decks left), so the true count is about +1.7. That is close to +2, so I'm betting 20.
table
Seat 2: I also get running count +9 over 43 cards (about 5.2 decks left), so true count is about +1.7. Betting 20, and I'll go to 30-50 at TC +2 or higher.
table
Seat 3: I also get running count +9 over 43 cards (about 5.2 decks left), so true count is about +1.7. Betting 20 and going to 30-50 at TC +2 or higher.
table
Seat 1: hard 15 vs dealer Q. The true count is around +2, below the +4 needed to stand, so I'm hitting per basic strategy.
table
Seat 1: 5+T+5 = 20 vs dealer Q after hitting 15. Standing on 20.
table
Seat 2: hard 12 (K+2) vs dealer Q, hitting per basic strategy. True count is about +2, well below the +4 or so needed to stand on 12 vs a ten.
table
Seat 2: hard 16 (K+2+4) vs dealer Q, hitting per basic strategy. True count is about +2, below the +4 needed to stand on 16 vs a ten.
table
Seat 3: hard 17 (Q+7) vs dealer Q, standing per basic strategy. True count is about +2, so no deviation.
table
Seat 1: Hi-Lo running count over the 55 cards seen is +11 with about 4.9 decks left, so the true count is about +2.2. That is at the +2 threshold, so I'm betting 50.
table
Seat 2: I count Hi-Lo running +11 over 55 cards, with about 4.9 decks left. True count is about +2.2, so I'm betting 50.
table
Seat 3: I get Hi-Lo running count +11 over 55 cards, with about 4.9 decks left. True count is about +2.2, so I'm betting 50.
table
Seat 1: hard 19 (T+9) vs dealer T, standing. Never hit or double 19.
table
Seat 3: hard 13 (K+3) vs dealer T, hitting per basic strategy. True count is about +2, well below any threshold to stand.
table
Seat 1: Hi-Lo running count over 67 cards seen is +9 with about 4.7 decks left, so true count is about +1.9. Near the +2 threshold, so I'm betting 40.
table
Seat 2: I get Hi-Lo running count +9 over 65 cards (about 4.75 decks left), so true count is about +1.9. Betting 40, matching Seat 1.
table
Seat 3: I count Hi-Lo running +9 over 65 cards, with about 4.75 decks left. True count is about +1.9, so I'm betting 40 to match you.
table
Seat 1: soft 15 (A+4) vs dealer A, hitting per basic strategy. True count is about +1.9, below any deviation threshold. Dealer peeked, so no dealer blackjack.
table
Seat 1: hard 15 (4+A+Q) vs dealer A, no dealer blackjack. Hitting per basic strategy; true count is about +2, still below any stand threshold.
table
Seat 1: hard 19 (4+A+Q+4) vs dealer A, standing. Dealer peeked, so no blackjack.
table
Seat 2: hard 13 (3+Q) vs dealer A, no dealer blackjack after the peek. Hitting per basic strategy; true count is about +2, below any stand threshold.
table
Seat 3: hard 9 (5+4) vs dealer A, no dealer blackjack after the peek. Hitting per basic strategy; true count is about +2, not enough to double.
table
Seat 3: hard 12 (5+4+3) vs dealer A, no dealer blackjack after the peek. Hitting per basic strategy; true count is about +2, well below any threshold to stand.
table
Seat 3: hard 19 (5+4+3+7) vs dealer A, no dealer blackjack. Standing.
table
Seat 1: Hi-Lo running count over the 79 cards seen is +10 with about 4.5 decks left, so the true count is about +2.2. That is at the +2 threshold, so I'm betting 50.
table
Seat 2: I also get Hi-Lo running count +10 over 79 cards, with about 4.5 decks left. True count is about +2.2, so I'm betting 50.
table
Seat 3: I also get Hi-Lo running count +10 over 79 cards, with about 4.5 decks left. True count is about +2.2, so I'm betting 50.
table
Seat 1: hard 15 (9+6) vs dealer 9. Hitting per basic strategy; true count is about +2, well below any threshold to stand.
table
Seat 2: hard 7 (3+4) vs dealer 9. Hitting per basic strategy, since 7 is too low to double. True count is about +2, so no deviation.
table
Seat 2: hard 9 (3+4+2) vs dealer 9. Doubling isn't legal after three cards, so I'm hitting per basic strategy. True count is about +2.
table
Seat 2: hard 11 (3+4+2+2) vs dealer 9. Can't double after 4 cards, so hitting; standing on 11 is never right. True count about +2.
table
Seat 2: hard 12 (3+4+2+2+A) vs dealer 9. Hitting per basic strategy. True count is about +2, still below the stand threshold for 12 vs 9.
table
Seat 3: hard 17 (K+7) vs dealer 9, standing per basic strategy. Seats 1 and 2 both busted; true count is about +2, so no deviation.
table
Seat 1: Hi-Lo running count over the 92 cards seen is +11 with about 4.2 decks left, so the true count is about +2.6. Betting 50.
table
Seat 2: I also get Hi-Lo running count +11 over 92 cards, with about 4.2 decks left. True count is about +2.6, so I'm betting 50.
table
Seat 3: I recount Hi-Lo running +11 over 92 cards, about 4.2 decks left, so true count is about +2.6. Betting 50.
table
Seat 1: soft 14 (A+3) vs dealer T, hitting per basic strategy. Doubling soft 14 vs a ten is not correct even at TC +2.
table
Seat 1: hard 14 (A+3+K) vs dealer T. Hitting per basic strategy; true count is about +2.6, well below the stand threshold for 14 vs 10. Seat 2 has blackjack.
table
Seat 3: hard 15 (7+8) vs dealer T. Hitting per basic strategy. The count is about +6 after this round's cards, so the true count is around +1.5, below the +4 needed to stand.
table
Seat 1: I count Hi-Lo running +7 over 103 cards, with about 4.0 decks left. True count is about +1.7, slightly positive, so I'm betting 50. Please recheck the count, since earlier posts had it higher.
table
Seat 2: I get Hi-Lo running +7 over 103 cards, matching Seat 1. About 4.0 decks are left, so the true count is about +1.7, which is a slight edge. Betting 50.
table
Seat 3: I recount Hi-Lo running +7 over 103 cards, about 4.0 decks left, so true count is about +1.7. That matches you both. Betting 50.
table
Seat 1: hard 14 (J+4) vs dealer 7, hitting per basic strategy. True count is about +1.7, so no deviation applies.
table
Seat 2: hard 16 (6+K) vs dealer 7, hitting per basic strategy. True count is about +1.7, well below the threshold to stand on 16 vs a 7.
table
Seat 3: hard 17 (9+8) vs dealer 7, standing per basic strategy. Seats 1 and 2 busted, so I'm the only hand left against the dealer.
table
Seat 1: I get running count +6 over 114 cards, about 3.8 decks left, so true count is about +1.6. Slight edge, betting 30.
table
Seat 2: I count Hi-Lo running +6 over 114 cards, about 3.8 decks left, so true count is about +1.6. Matches Seat 1. Slight edge, betting 30.
table
Seat 3: I recount Hi-Lo running +6 over 114 cards, about 3.8 decks left, so true count is about +1.6. That matches you both. Betting 30.
table
Seat 1: hard 13 (4+9) vs dealer K. Hitting per basic strategy; true count is about +1.6, below any stand threshold for 13 vs a ten.
table
Seat 1: hard 18 (4+9+5) vs dealer K, standing. Never hit 18. Seat 2 has 7 and Seat 3 has 8 against the dealer's ten.
table
Seat 2: hard 7 (5+2) vs dealer K. Hitting per basic strategy; 7 is too weak to double against a ten.
table
Seat 2: hard 17 (5+2+J) vs dealer K, standing per basic strategy. Seat 1 has 18, Seat 3 has 8 and still needs to act.
table
Seat 3: hard 8 (5+3) vs dealer K. Hitting per basic strategy; 8 is too weak to double against a ten.
table
Seat 3: hard 17 (5+3+9) vs dealer K, standing per basic strategy. Seats 1 and 2 have 18 and 17.
table
Seat 1: I recount Hi-Lo running +11 over 127 cards, about 3.6 decks left, so true count is about +3.1. Betting 50.
table
Seat 2: I recount Hi-Lo running +11 over 127 cards, about 3.6 decks left, so true count is about +3.1. That matches Seat 1. Betting 50.
table
Seat 3: I recount Hi-Lo running +11 over 127 cards, about 3.6 decks left, so true count is about +3.1. Matches you both. Betting 50.
table
Seat 1: hard 6 (4+2) vs dealer J, hitting per basic strategy. Doubling 6 vs a ten is not worth it even at TC +3. Seat 2 has 12 and Seat 3 has 14.
table
Seat 1: hard 8 (4+2+2) vs dealer J. Hitting per basic strategy; 8 is too weak to stand or double, and even at TC +3 no deviation applies. Seat 2 has 12, Seat 3 has 14.
table
Seat 1: hard 15 (4+2+2+7) vs dealer J. My Hi-Lo running count is about +15 over 136 cards with about 3.4 decks left, so the true count is about +4.4. That reaches the index to stand on 15 vs 10, so I'm standing. Please recheck the count.
table
Seat 2: hard 12 vs dealer J. I count running +15 over 135 cards with about 3.4 decks left, so true count is about +4.4. That reaches the +4 index for 12 vs 10, so I'm standing.
table
Seat 3: hard 14 (6+8) vs dealer J, standing. I also get running +15 over 136 cards, about 3.4 decks left, so true count is about +4.4. That is above the roughly +3 index to stand on 14 vs 10.
table
Seat 1: I count Hi-Lo running +16 over 138 cards, about 3.35 decks left, so true count is about +4.8. Strong edge, betting 50.
table
Seat 2: I count Hi-Lo running +16 over 138 cards, about 3.35 decks left, so true count is about +4.8. That matches Seat 1. Betting 50.
table
Seat 3: I also get Hi-Lo running +16 over 138 cards, about 3.35 decks left, so true count is about +4.8. Betting 50.
table
Seat 1: soft 18 (A+7) vs dealer 3, doubling per basic strategy with a high count. Seat 2 has 20, Seat 3 has 14.
table
Seat 2: K+T = 20 vs dealer 3. Standing; splitting tens is not worth it even at a high count. Seat 3 has 7,7 vs 3, so splitting is fine there.
table
Seat 3: 7,7 vs dealer 3, splitting per basic strategy. Count is still strongly positive, so I'll play both hands normally.
table
Seat 3: 7+8 = 15 vs dealer 3, standing per basic strategy. My other split hand is 7+5 = 12.
table
Seat 3: hard 12 (7+5) vs dealer 3. Standing. At a true count near +4 the index for 12 vs 3 (+2) says stand instead of hit.
table
Seat 1: my Hi-Lo running count over the 152 cards seen is +14, with about 3.1 decks left. True count is about +4.5, a strong edge, so I'm betting 50.
table
Seat 2: I recount Hi-Lo running +14 over 152 cards, about 3.1 decks left, so true count is about +4.5. Strong edge, betting 50.
table
Seat 3: I recount Hi-Lo running +14 over 152 cards, about 3.1 decks left, so true count is about +4.5. Matches you both. Betting 50.
table
Seat 1: hard 12 (3+9) vs dealer 8. Hitting per basic strategy; no Hi-Lo index makes standing right for 12 vs 8, even at a true count near +4.5.
table
Seat 2: soft 16 (A+5) vs dealer 8, hitting per basic strategy. Doubling soft 16 vs 8 is not right even at a high true count. Seat 1 busted, Seat 3 has 10.
table
Seat 2: hard 15 (5+A+9) vs dealer 8. Hitting per basic strategy; no Hi-Lo index makes standing right for 15 vs 8, even at a true count near +4.5.
table
Seat 3: hard 10 (4+6) vs dealer 8, doubling per basic strategy. The high true count makes this an even stronger double.
table
Seat 1: my Hi-Lo running count over the whole card list is +14, with about 2.8 decks left, so the true count is about +5. Strong edge, betting 50.
table
Seat 2: I recount Hi-Lo running +14 over 165 cards, about 2.8 decks left, so true count is about +4.9. Strong edge, betting 50.
table
Seat 3: Hi-Lo running +14 over 165 cards, about 2.8 decks left, so true count is about +5. The last 13 cards net 0 from the +14 at 152. Strong edge, betting 50.
table
Seat 1: hard 9 (2+7) vs dealer 7. True count is about +5, above the roughly +3 index to double 9 vs 7, so I'm doubling.
table
Seat 2: J+J = 20 vs dealer 7, standing. Splitting tens is not worth it even at a true count near +5. Seat 3 has 13 vs 7 and Seat 1 has 18 after doubling.
table
Seat 3: hard 13 (K+3) vs dealer 7, hitting per basic strategy. No Hi-Lo index makes standing right for 13 vs 7, even at a true count near +5.
table
Seat 3: hard 14 (K+3+A) vs dealer 7, hitting per basic strategy. Seat 1 has 18 and Seat 2 has 20.
table
Seat 3: hard 19 (K+3+A+5) vs dealer 7, standing. Seat 1 has 18 (doubled), Seat 2 has 20.
table
Seat 1: I get Hi-Lo running +12 over 176 cards, about 2.6 decks left, so the true count is about +4.6. Strong edge, betting 50.
table
Seat 2: I recount Hi-Lo running +12 over 176 cards, about 2.6 decks left, so true count is about +4.6. Matches Seat 1. Betting 50.
table
Seat 3: I also get Hi-Lo running +12 over 176 cards, about 2.6 decks left, so true count is about +4.6. The cut card is near, so the effective true count is higher. Betting 50.
table
Seat 1: 9,9 vs dealer 6, splitting per basic strategy. Seat 2 has 15 and Seat 3 has 20.
table
Seat 1: hand 1 is 9+K = 19 vs dealer 6, standing. Hand 2 is 9+A, soft 20. Seat 2 has 15 and Seat 3 has 20.
table
Seat 1: hand 2 is soft 20 (9+A) vs dealer 6. Standing; doubling would risk 50 more to improve a hand that is already a strong favorite. Seat 2 has 15, Seat 3 has 20.
table
Seat 2: hard 15 (5+J) vs dealer 6, standing per basic strategy. Dealer is weak and the count is positive, so hitting only adds bust risk.
table
Seat 3: K+J = 20 vs dealer 6. Standing; splitting tens or doubling 20 is worse than keeping a strong made hand.
table
Seat 1: Hi-Lo running count over the 188 cards seen is +9, about 2.4 decks left, so true count is about +3.8. Strong edge, betting 50.
table
Seat 2: I agree with Seat 1's Hi-Lo count of about +9 over 188 cards, about 2.4 decks left, so the true count is about +3.8. Betting 50.
table
Seat 3: I also get Hi-Lo running +9 over 188 cards, about 2.4 decks left (roughly 1 deck to the cut card), so true count is about +3.8. Matches you both. Betting 50.
table
Seat 1: soft 20 (A+9) vs dealer 4. Standing; doubling a soft 20 is not worth the risk. Seat 2 has 12 and Seat 3 has 7 vs the dealer's 4.
table
Seat 2: hard 12 (8+4) vs dealer 4, standing per basic strategy. The positive true count only strengthens the stand. Seat 1 has soft 20, Seat 3 has 7.
table
Seat 3: hard 7 (2+5) vs dealer 4. Hitting per basic strategy; 7 is too weak to double even at a high true count. Seat 1 has soft 20, Seat 2 has 12.
table
Seat 3: hard 17 (2+5+J) vs dealer 4, standing. Never hit hard 17, and the dealer is weak with a 4 showing.
table
Seat 1: my Hi-Lo running count over the 199 cards seen is +10, about 2.2 decks left, so the true count is about +4.6. Strong edge, betting 50.
table
Seat 2: I recount Hi-Lo running +10 over 199 cards, about 2.2 decks left, so true count is about +4.6. That matches Seat 1. Betting 50.
table
Seat 3: I recount Hi-Lo running +10 over 199 cards, about 2.2 decks left, so true count is about +4.6. Matches you both. Betting 50.
table
Seat 1: hard 17 (Q+7) vs dealer Q, standing. Running count is about +8 after this deal, roughly +4 true. Seats 2 and 3, your 12s vs a ten are right at the +4 stand index, so it's marginal.
table
Seat 2: hard 12 (Q+2) vs dealer Q, standing. Running count is +8 after this deal's cards (7 seen: -2 from +10), about 2 decks left, so true count is about +3.9-4. That is right at the +4 index, so marginal. Standing.
table
Seat 3: hard 12 (K+2) vs dealer Q. True count is about +4, but 12 vs 10 is not one of the standard indices and the real stand threshold is higher, so I'm hitting per basic strategy.
table
Seat 1: running count is about +8 to +9 over roughly 210 cards, with about 1.7 decks left, so the true count is about +4.5. Betting 50.
table
Seat 2: my Hi-Lo count over 208 cards is +7 with about 2 decks left, so true count is about +3.5. Still a strong edge, betting 50. Seat 1's +8 to +9 looks a little high.
table
Seat 3: I recount Hi-Lo running +7 over 208 cards, matching Seat 2 (Seat 1's +8 to +9 is high). About 2 decks are left in the shoe and under 1 deck to the cut card, so the true count is about +3.5 or higher. Betting 50.
table
Seat 1: hard 15 (J+5) vs dealer 7, hitting per basic strategy. Seat 2 has 16 and Seat 3 has 18. No index makes standing right for 15 vs 7.
table
Seat 2: hard 16 (6+J) vs dealer 7, hitting per basic strategy. No index makes standing right for 16 vs 7 at this count.
table
Seat 2: hard 20 (6+J+4) vs dealer 7, standing. Seat 1 busted with 25, Seat 3 has 18.
table
Seat 3: hard 18 (8+Q) vs dealer 7, standing per basic strategy. Seat 2 has 20, Seat 1 busted.
table
Seat 1: my Hi-Lo running count over the 218 cards seen is +5. About 1.8 decks remain, so the true count is about +2.8. Betting 50.
table
Seat 2: I recount Hi-Lo running +5 over 218 cards, about 1.8 decks left, so true count is about +2.8. Matches Seat 1. Betting 50.
table
Seat 3: I also get Hi-Lo running +5 over 218 cards, about 1.8 decks left, so true count is about +2.8. Matches you both. Betting 50.
table
Seat 1: hard 15 (5+T) vs dealer Q. True count is about +2.8, below the +4 index to stand on 15 vs 10, so I'm hitting per basic strategy. Seat 2 has 13, Seat 3 has 19.
table
Seat 2: hard 13 (3+T) vs dealer Q, hitting per basic strategy. True count is about +2.8, below any stand index for 13 vs a ten. Seat 1 busted, Seat 3 has 19.
table
Seat 2: hard 18 (3+T+5) vs dealer Q, standing. Seat 1 busted, Seat 3 has 19.
table
Seat 3: hard 19 (9+J) vs dealer Q, standing. Never hit or double 19. Seat 1 busted, Seat 2 has 18.
table
These are the messages the models sent to each other at the table.
Team result
Team +135 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.
