Blackjack Mentor | Blackjack Counter | Blackjack Expert | ||

Blackjack Audit | BJ Counter Databases | Blackjack Count Master e-Book | ||

Memo Safe | Balls! |

Experienced blackjack players know that casino blackjack rules vary enormously from place to place, and that subtle differences can have a significant impact on your earnings or loss rate. Common factors that alter your advantage include double after splits, hit or stand on soft 17, and blackjack payout. For card counters, additional factors such as bet spread, number of decks, number of players, and shoe penetration can also have a large impact on your earnings. In places such as Vegas where gambling establishments abound, the variety of blackjack game permutations is staggering. How do you choose the game that will give you the best advantage and earnings? Basic Strategy players and different rules The process of picking the best game for basic strategy players is fairly simple. Each rule variation has a specific advantage or disadvantage that affects your base expectation (the average amount of each bet you should expect to lose). Start with the standard "Vegas Strip" rules at an expectation of -0.02%: one deck, no double after splits, dealer stands on soft seventeen. Use the following table to determine your final expectation for a specific set of rules:
For example, a classic eight deck DAS, S17 game would have the following expectation for basic strategy players: -0.02 - 0.61 + 0.13 = -0.5%. This means that for each $10 you bet, you should expect to lose 5 cents on average. Note that this assumes the player is using Card counters and different rules For blackjack players who use a card counting system, using the above table as a guide is not the best way to maximize your earnings rate. A common mistake is to simply apply the above table of modifiers, and choose the best game accordingly. As an example, suppose you have a choice between two different casino games: **Game #1:**8 deck, DAS, S17, 75% penetration.**Game #2:**6 deck, DAS, H17, 75% penetration.
Which is the better game to play? In the previous article in our blackjack series, we already analyzed in detail the negative consequences of playing with an insufficient bankroll, so let's assume you will play with an appropriate bet spread and bankroll in both cases. The question is whether the decreased number of decks provides enough benefit to overcome the loss due to H17.
In the analysis above, we used Blackjack Audit, a blackjack simulator from DeepNet Technologies to collect all data. We ran 100 million rounds for each simulation using Stanford Wong's popular High-Low count system ("Professional Blackjack", Pi Yee Press). A standard multi deck/DAS basic strategy table was used for the simulations, as listed in our prior article: www.deepnettech.com/article1.html . The 'Fab18' indices are is an effective subset of 18 play indices that deliver 75% of the total gain in expectation, as first published by Don Schlesinger. Here are the detailed simulation settings we used: - 1 to 10 bet spread, defined for these true count values: TC 1: 3 units, TC 2: 5 units, TC 3: 10 units.
- Fab 18 indices: 9 vs 2 TC>=1:D, 9 vs 7 TC>=3:D, 10 vs 10 TC>=4:D, 10 vs A TC>=4:D, 11 vs A TC>=1:D, 12 vs 2 TC>=3:S, 12 vs. 3 TC>=2:S, 12 vs 4 TC<0:H, 12 vs 5: TC<-1:H, 12 vs 6 TC<-0:H, 13 vs 2 TC<0:H, 13 vs. 3 TC<-1:H, 15 vs 10 TC>=4:S, 16 vs 9 TC>=5:S, 16 vs 10 TC>=0:S, 1010 vs 5 TC>=5:P, 1010 vs 6 TC>=4:P, insure at TC >=3.
- 75% shoe penetration.
- no surrender, no double restrictions, up to three splits, no hitting split aces, can resplit aces.
- $5 unit bet size.
- Calculate true conservatively (number of decks rounded up).
- 100 million rounds per game simulation.
In the last row of table 2, we changed the Blackjack Audit simulation to spread to two hands at a true count of 3, and to three hands at a true count of 4 or greater. Although this increases the risk of ruin since more money is being wagered, it allows us to capitalize on the advantage at positive true counts delivered by the fewer number of decks. If you are playing with indices (even just the Fab18 indices), the games have almost identical expectations. The only case where you should avoid the six deck game is if you are playing with basic strategy and only using the count to adjust your bets. In all other cases, the six deck game is a bit better. If you play additional hands on high counts, then the six deck game has a significant 15% improvement over the eight deck game. Had we used the modifiers from table 1, we would have incorrectly picked the eight-deck game as the best option by a clear margin: **Game #1:**8 deck, DAS, S17. Expectation = -0.02 -0.61 +0.13 = -0.50%**Game #2:**6 deck, DAS, H17. Expectation = -0.02 -0.58 -0.20 +0.13 = -0.63%
The only way to reliably determine your advantage when card counting is to use a good blackjack simulation program to test the exact variables you are using. Make sure your choice of simulator program lets you easily and accurately specify not only the game conditions (such as DAS, H17, etc.), but also every subtle aspect of the your count system. For example, make sure you can model betting techniques you may employ such as hand spreading or 'wonging' (sitting out hands where the count is below a certain value). It is surprisingly easy to enter incorrect count system, betting, or index data into simulation programs, so make sure you can easily verify the settings! If the simulation does not exactly model the way you play, the results may be misleading. |