There is no such thing as a "best" card counting system unless you are a robot, in which case you would be keeping track of each and every card. Humans, on the other hand, have limitations. For shoe dealt games, all card counting systems perform within a tenth of a percent (or so) of each other. For single deck games, a balanced multi-level count with an ace side-count can show a significant theoretical improvement over unbalanced and single-level counts, however, the player runs the risk of mental fatigue and errors. The bottom line for most players is that "simple is best!" I recommend the Hi-Lo, Red Seven, K-O or Zen for shoe games and the Hi-Opt I, Zen or Omega II for single-deck. What is the best all around count? I like the Hi-Lo because it allows me to concentrate on more important things... like convincing the pit boss that I am a loser!

Obviously, the answer to this question is not as easy as it appears. Several approaches have been used in the past to evaluate card counting systems. One analytical approach is the calculation of several performance parameters (e.g., playing, betting, and insurance efficiencies). The results are then used to approximate the potential of one system over another. Another approach that is used is to simulate each system against typical game conditions on a high speed computer. Simulations can provide an accurate real-world estimate of the advantages and win-rates that are possible in playing a particular system.

However, the problem with coming up with a ‘best’ card counting system is that you can always come up with a ‘better’ card counting system. Instead of a single-level* ‘unbalanced’ count you could assign more accurate point values to each card and determine true counts by the exact number of decks or cards remaining. You could improve ‘playing’ efficiency by assigning a ‘zero’ to the Ace and side counting each of them. You could also side count other cards such as 7s, 8s, and 9s thus improving your play against specific hands. You could also incorporate play variations (changes to basic strategy) based on specific counts by remembering ‘every’ index number for ‘every’ play possible. To improve the accuracy of your insurance decisions you could also keep a separate count of all the tens in the deck or shoe. Of course, you don’t won’t to forget all the ‘practical’ advice each system offers in regard to betting, playing, camouflage, and other tips and tricks of the trade.

Peter Griffin, author of the classic text The Theory of Blackjack, wrote "If one’s ambition is to raise overall strategic efficiency beyond the 70% level, perhaps as high as 90%, it is imperative that the primary system be quite simple and hence allow great flexibility for incorporating several auxiliary, independent sources of information."

I believe the above comment was one of the most important suggestions ever made about card counting. Griffin suggested that it may be better to keep your base count simple to allow your brain the ability to perform other tasks and to utilize other sources of information. These other sources of information can often improve the potential of a single-level count over an advanced 2- or 3-level count that doesn’t use this information. This information includes side counts, shuffle tracking, ace location strategies, key card techniques, and dealer errors. My own experience at card counting has shown that Griffin was probably right.

Human error is another reason to keep it simple. The most advanced card counting system in the world is not going to do you any good if you can’t play it accurately. Thus, the best card counting system may be one that perfectly balances theoretical power and your human ability to execute it accurately.

*Footnote:  A single-level count assigns point values in such a manner that the non-zero point values are the same in absolute value, namely +1 or -1. The single-level Hi-Lo count, for example, assigns 2 - 6 as +1, 7 - 9 as 0, and Tens and Aces as -1.