A very close friend of mine and fellow gaming consultant, Bob Del Rossi, passed along an interesting story about a situation that occurred at the Desert Inn Hotel/Casino back in the 1990s. Del Rossi started in gaming in Atlantic City in surveillance, and because of his desire to master card counting, he advanced to become one of the members of Harrah’s “card counting” team. During that time, Atlantic City casinos used some of their more knowledgeable employees to detect and confirm whether a player was counting cards. In addition, Bob engaged in card counting on a more personal level after moving to Las Vegas in the late ‘80s. In other words, when it comes to counting cards, Bob “knows his stuff.”
During the time period of this story, Bob worked as a pit manager on grave shift in the main pit at the old “DI” hotel. On this occasion, Bob had just been introduced to his new shift manager and new member of the DI team, Tom. He and Tom discussed a number of issues involving the shift and DI standard procedures, but like a good floor supervisor, Bob always positioned himself so that he could talk and watch the games at the same time. Tom thanked Bob for his time, shook his hand and as he walked way, Bob asked Tom for a favor. During their short conversation, Bob had noticed some characteristics about a specific blackjack customer he had been casually watching. “Could you please call surveillance and let them know there is a player sitting last position on BJ #2 who shows a strong indication he might be counting cards.” Tom agreed, but then hesitated. He turned back to Bob and said, “Don’t tell me you were counting down that six deck shoe while we were talking shop?”
Bob told Tom there was no way that he could manage to keep a running count in his head and still carry on a conversation with Tom; however, Bob had noticed certain indicators that led him to suspect the player was a serious, if not professional-level card counter. A call to surveillance was placed, and 30 minutes later surveillance had confirmed that the player was, in fact, a serious card counter. From that point on, Tom thought that Bob had to be one of the sharpest floor supervisors in gaming. Bob was golden.
How did Bob snap to the card counter without counting down the shoe, or watching the game for a lengthy period of time? Bob used a common sense technique known as “counting short cuts.” He noticed that the player in question appeared to exhibit a correlation between his wager size and the strategy he used to play his hands. Based on Bob’s experience in card counting, he recognized a pattern that only a person counting cards would use, and based on this pattern recognition, Bob felt the player was worthy of a more intensified investigation by surveillance.
A Suspected Card Counter Must Have an Adequate Bet Spread
One of the biggest misconceptions involving the threat of card counting is the degree a counter must spread his wagering to achieve a reasonable advantage over the game of blackjack. When sitting at a table and playing through most of the shoe, the card counter must use a sufficient minimum-to-maximum bet spread. If the suspected player does not meet this bet spread requirement, he or she will not be able to win enough money to make his or her time at the blackjack table worthwhile. The professional-level card counter—the only counter that is a true threat to the casino—does not count cards for fun.
The question is: What size of minimum bet spread does the professional-level card counter need to use in order to make his time at the table beneficial? Following is a chart I use in my “Finer Points of Card Counting” session of my Table Games Tactics seminar (Table 1).
Note: Single deck betting units is based on a 3:2 game. In a single deck game where BJs are paid 6:5 the spreads are 1 to 14 units (0.8 percent), and 1 to 16 units (1 percent).
The professional-level card counter strives for an average player advantage of 0.8 percent as a minimum level, with 1 percent or greater as their desired average player advantage. If you suspect a player is counting on a six-deck game with a shuffle point of 1.5 decks (75 percent penetration), he must use a minimum spread of 1 to 12 units (example: $25 minimum, $300 upper). If the minimum size of spread is not utilized, or the player does not spread the upper level on a majority of the player advantageous situations (true count of > +1), the counter will not achieve a large enough gain to make his time at the table worthwhile.
When observing suspected count play, first ask yourself whether or not the suspected counter has an adequate bet spread to win an acceptable amount of money from the casino. If the answer is “yes,” then you can do what Bob did to analyze the suspected counter’s play: Use the card counter detection short cuts. Remember, a bet spread is not the sole indictor that a person is counting cards. A player wagering within the parameters of the listed spreads, but wagering based on outcomes such as progressive betting systems is not an indication of card counting. The professional-level card counter must spread to a higher wager only when composition of the remaining cards in the shoe or deck give him a mathematical advantage over the house and bet as little as possible when the game is neutral or the house has the advantage.
Look For a Correlation Between the Size of the Bet and Hand Playing Strategy
Let’s assume for the moment that an observed player is strongly suspected of counting cards. If he is wagering with many units or few units of his bet spread, what does that indicate? Is the count plus (high in tens/aces), or is it minus (high in small cards)? If the person is wagering on the higher end of the bet spread, then this indicates the count must be plus. If the person is wagering near the bet spread minimum, then this indicates the count is neutral or minus. The reason the shortcut system is effective is through the observed correlation between the size of wager indicating the count and the strategy decisions the player uses to deviate from basic strategy. If after a period of observation the player is noted making deviations from basic strategy that are in direct correlation with the size of the bet, the floor supervisor or surveillance operator has evidence that the person is likely counting. Note: This method is used for detecting a card counter, not for confirmation. Accurate confirmation is conducted through a different set of observed factors.
In general, hand strategies fall into three categories: (1) hit and stand decisions, (2) double down and hand splitting decisions and (3) whether or not to take insurance against a dealer’s ace up card. In approximately four out of every five hands, the professional card counter will use basic strategy to play his hand. However, one hand out of five, the counter will deviate from basic strategy based on the composition of the cards remaining in the deck or shoe. Following are hand strategies one would look for when observing situations of minimum or upper wagering levels.
When the Suspected Player is Wagering Toward the Upper Level of Their Bet Spread
A higher wager indicates the deck is plus, and when the deck is plus, it contains a higher composition of 10 value cards and aces. The player who is counting cards will exhibit the following strategy changes:
1) Hit and Stand in a more “passive” manner. He will stand on hands such as 16 v. T, 16 v. 9, 15 v. T, and possibly 12 v. 2 or 3.
2) Double down and split more “aggressively.” The player will double on hands such as 10 v. T, 10 v. A, 9 v. 2, and 9 v. 7. The player may also split 10 value cards; TT v. 5, and TT v. 6.
3) The player will insure any hand total, including 12s through 16s.
In situations where surrender is offered, the higher unit spreads will also be subject to more aggressive surrender options including 16 v. A, and 15 v. 9 or T.
Regarding hand splitting; many executives question the wisdom of splitting 10 value cards in any situation; however, splitting TT v. 5 & 6 falls under the top 18 hand deviations a professional card counter will regularly use. Don’t turn your back on a suspected player just because he was observed splitting TT when wagering a large bet. Also, don’t rule out a suspected player because he didn’t split 10s. Some professional-level counters are reluctant to split 10 value cards because they feel it attracts too much attention. In most cases, the professional structures his entire play in order to attract the least attention possible.
The most important deviation for the counter, which I refer to as the “million dollar indicator,” is the taking of insurance. Most “everyday” players do not take insurance unless they are offered “even money” when holding a two-card blackjack on a 3:2 game. In order to maximize his strategy deviation return, the card counter has to take insurance on his hand anytime the count reaches “+3 true count” or higher (hi/lo count system). Watch for hands where the dealer offers an ace up-card. If you see at least a 90 percent correlation between a larger bet and the player taking insurance on any hand, you’re probably watching a card counter in action (or an advantage player getting the dealer’s hole-card).
When the Suspected Player is Wagering the Minimum Level of Their Bet Spread
A lower or minimum wager indicated the deck is minus or is composed of a larger number of lower-value cards such as twos through sixes. Lower wagering will correspond with following strategy changes:
1) The player will be more “aggressive” in hitting busting hands. He will hit on hands such as 16 v. T, 16 v. 9, 15 v. T, and possibly 12 v. 4, 5, or 6. This includes hitting some surrender hands such as 88 v. A, and 17 v. A.
2) The player will double down and split more “passively.” The player will hit on hands such as 10 v. 9, and 11 v. A.
3) The player will not insure any hand total; the most apparent being the refusal of taking “even money” on 3:2 blackjack hands.
It’s important that the executive using the “shortcut” method focuses on the contrast between key hand strategy plays during both higher end wagers and minimum wagers. For example: standing on 16 v. T and 15 v. T with a larger wager, and then hitting the same hands with a minimum wager. Another common example would be doubling down on 11 v. A and 10 v. T while wagering a larger bet, and hitting both hands when wagering near minimum. Of course, insurance is “the” key play and needs to receive serious scrutiny.
Pass the Information Along to a Person Who Can Confirm the Suspect is Counting
In the late 1980s, while working for Bally’s Las Vegas, I was involved in training floor supervisors to watch our newly implemented double deck blackjack games. The three areas I stress for protecting the double deck games from card counting attack were (1) understanding basic strategy, (2) knowing the minimum bet spread requirements for the double deck game dealt at Bally’s and (3) utilizing the detection short cuts as described previously. I always emphasized the value of detection. One has to detect the presence of a card counter before the suspected counter can be counted down and confirmed. Using the shortcuts also helps side step time wasted due to “false positives.” False positives occur when the casino focuses its efforts on the wrong player. For example, when a player is winning a large sum of money; however, he does not have an adequate bet spread, and/or does not show a correlation to the strategy of play dictated by the shortcuts.
One situation that occurred at Bally’s is a fine example of how using the shortcuts works to detect possible professional-level card counters. While working as a pit manager in a double deck pit, I was approached by one of my floor supervisors. The following is the information she passed along to me:
• The player in question was wagering from $25 to $125. Bally’s double deck game allowed the professional-level counter a reasonable gain if he bet from one to five units.
• The player had been seen taking insurance with a 14 v. A while wagering $125.
• The floor supervisor noted that on a number of hands, the player stood on 15s and 16s against a T while wagering $100 to $125. At other times, the player was observed hitting those same hands while wagering $25.
• The player split TT v. 6 while wagering $125.
After listening to her explanation of these observations, I contacted surveillance and tipped them to the player and the noted play/wager observation. Less than 30 minutes later, the shift manager arrived in our pit and proceeded to back the player off the game for counting cards.
A blackjack player suspected of counting cards must have an adequate bet spread, based on the number of decks and game rules, in which to overcome the basic mathematical advantage of the game and gain a worthwhile return for his time on the table. Don’t waste time and resources observing and counting down players who do not utilize an adequate bet spread. Just because they are winning doesn’t mean they are counting cards. Note: Don’t let your focus on card counting cloud your mind to other advantage play and cheating possibilities. I can think of numerous situations where floor operations and surveillance were unsuccessfully focused on card counting while the player was actually employing another advantage play technique.
If you observe a player spreading an adequate bet spread, look for a correlation between the size of his bet and the hand decision strategy he employs. Remember to watch for deviations from basic strategy in areas of hitting/standing mostly with 15s and 16s versus the dealer’s 10 value card, double down on 9s and 10s on borderline plays and splitting two 10 value cards against a dealer’s up-card of five and six. The most important player decision to look for is insurance. Watch for a player taking insurance with any hand value when he is wagering a higher limit bet.
Using this shortcut method is only good for detecting card counters. Once a player has been detected using the shortcut method, the information must be passed along to someone capable of confirming the count play. Confirmation needs to be accomplished by someone with access to counter catcher software, or has at the very least, a semi-professional knowledge of how to count cards.