counting · 9 min read

Card Counting Systems Compared: Hi-Lo, KO, Omega II, Zen

Four counting systems, ranked by what actually matters: edge per round, variance, learning curve, and N0. Spoiler — for most players, Hi-Lo still wins.

There is no shortage of card-counting systems. Hi-Lo, KO, Hi-Opt I, Hi-Opt II, Omega II, Zen Count, Halves, Mentor, REKO, Uston APC — all of them claim some advantage over the others. Most beginners pick one based on what their first book recommended. Most never compare.

This article compares the four systems that actually matter for working advantage players: Hi-Lo, KO, Omega II, and Zen. We'll rank them on edge, variance, complexity, and N0 — the only metric that combines edge and variance into a single ranking.

What a counting system actually does

Every count system assigns a value to each card. As cards leave the shoe, you add their values to a running count. When the count is positive, the remaining shoe is rich in tens and aces — the player has an edge. When negative, the casino has more edge than usual.

True count = running count divided by decks remaining. The bet ramps up with the true count. That's the entire game.

Hi-Lo: the standard

Tags: 2-6 = +1, 7-9 = 0, 10-A = -1. Balanced (the deck sums to zero), so you must convert running count to true count by dividing by decks remaining.

Hi-Lo is the world's most-used count for one reason: it's complex enough to extract most of the available edge, simple enough that a smart amateur can learn it in a week and execute under pressure. Every other system on this list trades simplicity for marginal edge.

KO: the unbalanced shortcut

Tags identical to Hi-Lo except the 7 is +1 instead of 0. Unbalanced — the deck doesn't sum to zero, which means you skip the true-count conversion entirely. Just bet based on running count.

KO is what we'd recommend over Hi-Lo if not for one thing: every Hi-Lo book, lesson, and online resource is more comprehensive than the KO equivalents. The minor edge gain isn't worth the smaller community.

Omega II: the high-end balanced

Tags: 2 = +1, 3-4-6 = +2, 5 = +2, 7 = +1, 8 = 0, 9 = -1, 10-T-J-Q-K = -2, A = 0. Combined with a separate ace side count for betting decisions.

Omega II's payoff is the playing efficiency: it tells you when to deviate from basic strategy more accurately than Hi-Lo. For an experienced AP grinding many hands, that 0.16 efficiency advantage adds up. For a part-time player making 200 hands a trip, it's noise.

Zen Count: the middle ground

Tags: 2-3 = +1, 4-5-6 = +2, 7 = +1, 8-9 = 0, 10-J-Q-K = -2, A = -1. Balanced. Designed by Arnold Snyder as a Hi-Lo / Hi-Opt II compromise.

Zen is the sweet spot for players who want better than Hi-Lo without an ace side count. The catch: fewer published resources than Hi-Lo or Omega II, and the multi-level tags are harder to learn than KO.

Which one should you pick?

Honestly, the system you'll actually drill is better than the system that's optimal in theory. The N0 differences between Hi-Lo (13,000) and Omega II (10,000) are small relative to the variance of casino play. If you'll learn Hi-Lo and play 1,000 hours over your career, you'll out-earn the player who half-learned Omega II and gave up.

Pick Hi-Lo unless you have a specific reason not to. Switch to KO if you struggle with the true-count conversion. Move to Omega II only after Hi-Lo is muscle memory and you want the playing-efficiency upgrade.

What we left out, and why

Hi-Opt I, Hi-Opt II, Halves, Uston APC, Mentor, REKO — all real systems, all with their adherents. They mostly fall between Hi-Lo and Omega II on the complexity-vs-edge curve. None is dramatically better than the four above. If you find yourself in a forum debate about whether REKO beats Hi-Lo by 0.03% N0, you've left the productive part of the discussion.

Practice the chosen system

Reading about counting systems is not the same as drilling them. Use the TableSharp counting trainer to get reps in — running count, true count conversion, deviations against the Illustrious 18. Pick a system, drill 30 minutes a day, ship two months later.

Drill in the trainer

Card Counting Trainer

Published 2026-05-06. Last updated 2026-05-06. Spot an error?