The mission to teach casinos that they should not fear blackjack card counters is one that I have been on for a long time. This fear arises from two memetic misconceptions. First, that there are a lot of card counters. Second, that there is a lot of money to be made from card counting. Because of these misconceptions, the obsession with card counting by both players and casinos has lasted over 50 years. I doubt there is much I can do about it, but I will continue to try.

I posted two articles on this blog that address the second myth given above. These articles give an absolute upper-bound on the earning potential of a perfect high-low blackjack card counter. These articles hypothesized the world's greatest high-low card counter and determined his optimal win-rate. Here is the counter I considered:

  • The card counter uses the high-low card counting system.
  • The card counter knows every strategy index and plays perfectly.
  • The card counter plays exactly 100 hands.
  • Whenever the true count is above the index at which he has the edge, he wagers $100, otherwise he does not make a wager.
  • The casino doesn’t care about this counter and lets him play without “heat.”

In other words, we have a perfect card counter who is playing 100 hands, wagering $100 whenever he has the edge, otherwise he is sitting on his hands. Because he never wagers when the casino has the edge, and he wagers a flat amount of $100 when he has the edge, the earnings for this card counter are the maximum possible for any counter with a maximum bet of $100. Again, this card counter is winning at the maximum possible win-rate for a high-low card counter with a maximum bet of $100.

If you haven’t read my previous articles (or heard me talk about this topic), then this is an opportunity for you to take a quick quiz. Here are two quiz questions:

  1. If this card counter plays a six-deck shoe game with the rules H17, DOA (double on any two first cards), DAS (double after split), with the cut card placed 52 cards (1 deck) from the end, then on average how much do you think he will he earn per 100 hands?
  2. If this card counter plays a two-deck game with the rules H17, DOA (double on any two first cards), DAS (double after split), with the cut card placed 29 cards from the end, then on average how much do you think he will he earn per 100 hands?

After you have your guesses, divide them in half. Ready? And here are the answers:

  1. $33.58 (see this post).
  2. $66.30 (see this post).

The average blackjack card counter earns far less. The following table gives some win rates for real-world blackjack play:

6d doa das win per 100 hands $100 max bet


For more information, read my blog post, "The Win-Rate of the Average Blackjack Card Counter."

There is genuine disbelief in these numbers from both sides of the table. In my seminars, most in attendance guess $200 or more. I had one trainee guess $8,000. The magnitude of these overestimates is not limited to the casino-side. Most card counters falsely believe they are making a lot more as well. Hopefully this FAQ will clear up some misunderstandings.

The World's Greatest Blackjack Card Counter FAQ

Question 1. You must have made a mistake in your computer simulations. How can I trust that your results are right?

The easy answer is that I didn’t conduct the simulations. Instead, I contacted the premier blackjack programmer on the planet, Norm Wattenberger, who did these simulations for me. Norm has the best card counting simulation software available, Casino Verite. This software has been around for almost 20 years. Norm also hosts an Internet message board, www.blackjacktheforum.com, so if you have any other doubts about the data I was given, ask Norm!

Question 2.  I am winning at a much higher pace than your numbers suggest is possible. How can you be right when I am winning so much more than you predict?

You’ve been lucky. Play enough and your win-rate will go down.

Question 3. What point are you trying to prove beyond telling casinos that they should basically ignore red chippers who might be counting?

To put all forms of advantage play on an equal footing so that they can be compared apples-to-apples. In this context, blackjack card counting is near the bottom. Just about every other method of advantage play is superior.

Question 4. In your six-deck results, you say that the card counter should play at +1 and above. But doesn’t the house have an edge at a +1 true count?

According to the results of Norm’s simulation, the player has a 0.199% edge over the house when the true count is +1 in the six-deck game.

Question 5. You have the player play whenever the count is +1 or higher. Can’t he make more money by wagering when the true count is higher so that he has a bigger edge?

The player maximizes his earnings by wagering whenever he has the edge and at no other time. If he waited until the true count was +3, then he would be giving up all the earning potential of those +1 and +2 true counts.

Here is the equation that gives the card counter's earning potential:

Win-per-100-hands =

($100)x(100 hands)x(average edge)x(bet frequency)

A player who played beginning at a higher true count would have a higher average edge, but a much lower bet frequency. Overall, his win-per-100-hands would be lower.

Question 6. I bet a lot more than $100 as my maximum bet, so I will make a lot more than your numbers suggest. How can you say your numbers are valid for me?

I choose $100 and 100 hands so that my results are scalable and can be used by any player to analyze his play. So, consider a counter who plays 150 hands with a maximum bet of $200. This counter has a maximum theoretical earning potential of (150/100)x(200/100)x($33.58) = $100.74 in the six-deck shoe game I described.

Another reason for the “100 hands at $100” scheme is that I use the same parameters for all of my card counting analysis. This allows various card counting opportunities to be compared and ranked against each other.

Question 7. I use Kelly betting, so I win at the maximum possible rate, how can you say that you can win more without using Kelly betting?

If you use some sort of Kelly betting for a session of play, then you have a maximum bet for your session determined by this type of betting. Whatever your maximum bet is, use that as your bet when you scale my results for your session. You will earn less than that. You simply cannot win at a faster rate than by either betting $0 or making a maximum bet when you have the edge.

For most professional card counters, Kelly betting is a non-issue. Sufficiently well-funded players and teams have more practical limitations. These include avoiding CTRs as well as low table maximums and heat avoidance.

Question 8. I use a much better card counting system, how much more will I make?

In fact, you will probably make a lot less in practice, especially on the double-deck game. First of all, you don’t know all the indices that the perfect card counter knows. Perhaps you only know the Illustrious 18, for example, which gives up a lot of value on double-deck games. Second, you are getting heat, so you may use some cover plays (like not splitting 10s). You probably use a bet ramp of some sort, so you are placing less than your maximum bet in some situations when you have the edge.

The win-rates I give are unattainable in practice. Your best practical win-rate with a stronger system is almost certainly lower than these optimal theoretical results.

Question 9. I play a game with much better rules and cut-card placement than those you wrote about. Aren’t I making a lot more because of having a better game?

Again, more research can certainly be done here by someone who is willing to do it. In an ideal world, every possible combination of card counting system, cut card placement and game rules would be available for you to reference. The point of my work is not to answer every question; it is to give a sense of the magnitude of the misconception.

Question 10. What about a single-deck game where blackjack pays 6:5 and the penetration is down to last 12 cards and using a count which includes side counting values such as 7,8,9,5 and Ace for more accurate play variations?

You're trying to count a 6/5 game and I'm trying to cure cancer.

Question 11. In your article, a person with a $ 5000 max bet will win  $3314 per 100 hands. So at approx 30 hands per hour back counting, is he looking at $ 1100 per hour?

Your numbers refer to my double-deck results. These results are not based on the hands you actually play, they are based on the hands you observe in total. Win-per-100 hands refers to 100 hands, whether played or not. So this player will win $3314 per 100 hands on average, even if he only plays 30% of them.

The terms to read about to fully understand this are "average edge" and "bet frequency." I write about them in this post.

Question 12. What software should I buy if I want to double check your results or run my own simulations?

Casino Verite by Norm Wattenberger. Buy it here. The module that does these simulations is called CVCX. You can download a trial copy for free.

Question 13. I wouldn't be surprised if you work for the casino industry with your propaganda talk; someone interested in counting could find this article and actually believe counting isn't worth it.

Both are true.

Question 14. I have to say that I disagree with you and think your article is pure Bull****. I believe that card counting does give the player a huge edge and it is not a waste of time.

You have obviously been drinking.

Question 15. You are one of the biggest sociopathic, self-hating imbeciles in gambling history. But yes, your math is right.

Your point?

Question 16. You are wrong.

That’s not a question, but my dog agrees with you sometimes.

Question 17. I think you couldn't cut it as a card counter so your ego has the only possible explanation that counting doesn't work no matter what the computer data says.

I see your logic. It may be too deep for some readers, so let me simplify it: I suck therefore I make stuff up. See question #1.

Question 18. You say you used the numbers you did so the results can be scaled. So,the Greatest DD player with a $500 Max Bet will make $330/hr. And, with a Max Bet of $1,000 she is making $660/hr. Do you make significantly more than $300 - $600/hr at your job? Do most people?

Why stop there. A person who makes $1M bets will make $330k per hour! Card counting at the $1k level (or even the $50 level) isn't a 40 hour per week job that you just show up and do at your local casino.

Question 19. Everything you do is self-serving and this was no different, so you used a low max bet to emphasize your point, that card counter are no threat and casinos should focus on other methods of AP and more precisely your services.

Wait, someone's calling, be right back ... maybe ... nope, not a casino. Wait, oh wait, another call ...

About the Author
By

received his Ph.D. in Mathematics from the University of Arizona in 1983. Eliot has been a Professor of both Mathematics and Computer Science. Eliot retired from academia in 2009. Eliot Jacobson