Every major card counting system in one place — Hi-Lo, KO, Hi-Opt I/II, Omega II, Zen, Wong Halves, Red 7, and Uston APC. Card values, betting correlation, playing efficiency. Sourced from the Blackjack Review Encyclopedia and QFIT. Always free.
Every counting system can be measured on three dimensions. Higher is better in all three. The trade-offs between them determine which system fits your game.
How accurately the count predicts when to bet bigger. Highest priority in shoe games (4–8 decks) where most edge comes from bet sizing.
How accurately the count tells you to deviate from basic strategy on individual hands. Highest priority in single and double-deck games.
How accurately the count tells you to take insurance. The single most-impactful index play across all game types.
Card values for every position, plus efficiency metrics. Sourced from the Blackjack Review encyclopedia, with values cross-checked against QFIT and Stanford Wong's published tables.
| System | A | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | BC | PE | IC | Type | Lvl |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Hi-Lo | −1 | +1 | +1 | +1 | +1 | +1 | 0 | 0 | 0 | −1 | .97 | .51 | .76 | Balanced | 1 |
| KO (Knock-Out) | −1 | +1 | +1 | +1 | +1 | +1 | +1 | 0 | 0 | −1 | .98 | .55 | .78 | Unbalanced | 1 |
| Red 7 | −1 | +1 | +1 | +1 | +1 | +1 | +½ | 0 | 0 | −1 | .98 | .54 | .78 | Unbalanced | 1 |
| Hi-Opt I | 0 | 0 | +1 | +1 | +1 | +1 | 0 | 0 | 0 | −1 | .88 | .61 | .85 | Balanced | 1 |
| Hi-Opt II | 0 | +1 | +1 | +2 | +2 | +1 | +1 | 0 | 0 | −2 | .91 | .67 | .91 | Balanced | 2 |
| Omega II | 0 | +1 | +1 | +2 | +2 | +2 | +1 | 0 | −1 | −2 | .92 | .67 | .85 | Balanced | 2 |
| Zen Count | −1 | +1 | +1 | +2 | +2 | +2 | +1 | 0 | 0 | −2 | .96 | .63 | .85 | Balanced | 2 |
| Wong Halves | −1 | +½ | +1 | +1 | +1½ | +1 | +½ | 0 | −½ | −1 | .99 | .57 | .72 | Balanced | 3 |
| Uston APC | 0 | +1 | +2 | +2 | +3 | +2 | +2 | +1 | −1 | −3 | .90 | .69 | .90 | Balanced | 3 |
A note on Hi-Opt I, Hi-Opt II, and Omega II: these systems treat the ace as 0 in the main count, which lowers betting correlation. The published BC values assume no ace side count. With a proper ace side count, Hi-Opt II and Omega II reach BC of ~0.99 — at the cost of significantly more mental load.
Detailed view of each system, with its strengths and the kind of player it fits.
The standard. Taught by Blackjack Apprenticeship, used by the MIT team, recommended for almost every shoe-game player.
Best for: Shoe games. First system to learn. Easy enough that you can talk through it. Lower error rate than complex systems wins back the playing-efficiency gap in real-world play.
Hi-Lo with the 7 added in. Unbalanced — no true count conversion needed. Bet directly off the running count once you hit the pivot.
Best for: Beginners. Casino floor work where mental load matters. Skips the true-count conversion that trips people up. Slightly weaker than Hi-Lo for advanced index play, but the simplicity wins.
Snyder's design. Counts only red 7s as +1 — black 7s are 0. The only system where card color matters. Optimized to peak betting strength right at the pivot.
Best for: 6-deck and 8-deck shoe games where it actually outperforms KO and Hi-Lo at typical penetration. Niche but real. Requires you to track card color, which doesn't suit everyone.
Ace-neutral. Higher playing efficiency than Hi-Lo, but lower betting correlation. Designed for single-deck play with an ace side count.
Best for: Single-deck and double-deck (pitch) games. Pair with an ace side count for serious play. Use the high PE for index plays; use the side count for bet sizing.
Significant power upgrade. Among the highest playing efficiencies of any practical system. Ace side count strongly recommended.
Best for: Serious single-deck and double-deck players. Top-tier PE makes index plays significantly more profitable. Heavier mental load — only worth it if you can run it without errors.
Bryce Carlson's design from "Blackjack for Blood." Counts the 9 (where Hi-Opt II doesn't), giving slightly different trade-offs.
Best for: Pitch games where Hi-Opt II's edge would also apply. Slight betting-correlation advantage over Hi-Opt II. Choice often comes down to which textbook you learned from.
Snyder's design. Counts the ace (unlike Hi-Opt II / Omega II), giving strong BC without an ace side count. The most balanced level-2 system.
Best for: Players who want level-2 power without juggling an ace side count. Strong all-around. Solid PE, near-Hi-Lo BC. The best "single number" level-2 system.
Stanford Wong's high-precision count. Fractional values (½, 1½) make running totals harder. Highest BC of any common practical system.
Best for: Counters who want maximum BC and don't mind fractional running totals. Many doublers convert to integers (×2 mentally), turning it into a level-3 integer count. The most demanding practical system.
Ken Uston's Advanced Point Count. Highest playing efficiency of any common system. Ace-neutral — requires ace side count.
Best for: Elite single-deck specialists. The PE of 0.69 is essentially the ceiling for human-playable systems. Total mental load is the highest on this page — every card has a different weight.
Card counting is a technique that lets a blackjack player track the changing composition of the remaining deck without memorizing individual cards. The principle is simple: when more low cards (2-6) have been dealt out, the remaining deck is rich in 10s and aces, which favors the player. When more high cards have been dealt out, the deck favors the dealer.
Every counting system assigns each card a small positive, negative, or zero value. As cards come out, you mentally add or subtract those values, keeping a running total. When the count is high, you bet more (the deck favors you) and sometimes deviate from basic strategy. When the count is low, you bet the table minimum.
The running count is what you maintain as cards are dealt. For balanced systems (Hi-Lo, Hi-Opt II, etc.), you have to convert the running count to a "true count" by dividing by the number of decks remaining. A running count of +6 with 3 decks left is a true count of +2 — and that's the number you make decisions on.
Unbalanced systems (KO, Red 7) skip this step. They start at a non-zero number based on the number of decks, and use a "pivot" point to indicate when the deck favors the player. Less mental work, slightly lower precision.
The classic temptation is to assume a level-2 or level-3 system will make more money than Hi-Lo. The math says yes — on paper. The reality is that humans make more errors with more complex systems. A level-3 system with a 3% error rate often performs worse in practice than a level-1 system with a 0.5% error rate.
This is why the MIT team and most professional counters use Hi-Lo for shoe games. The simplest tool that captures most of the gain is usually the right tool.
The point values, betting correlation, playing efficiency, and insurance correlation values for each system are computed from running every system through millions of simulated hands. The numbers on this page are sourced from Michael Dalton's Blackjack Review Encyclopedia and Norm Wattenberger's QFIT simulator data — the two most widely-cited sources in the counting community. Cross-checked against Stanford Wong's Professional Blackjack and Don Schlesinger's Blackjack Attack where the systems are covered there.
Counting cards isn't worth anything if you don't have basic strategy nailed first. Start there with the interactive basic strategy chart. Then come back here, pick a system that fits your game, and practice it until you can run it without losing track of conversations. After that, the card counting 101 guide walks through actual table technique. Index play (the Illustrious 18) is the final layer — coming soon.
The questions counters and aspiring counters ask most often.
Hi-Lo for most players. It is the industry standard, taught by every major blackjack training program (including the MIT team's training), and despite being a level-1 system it has betting correlation and insurance correlation in the same range as more complex systems. For absolute beginners who want to skip the true-count math, KO is the easier alternative — it is unbalanced, so you bet directly off the running count without converting to true count.
A balanced count's card values sum to zero across the entire deck. Hi-Lo, Hi-Opt I/II, Omega II, Zen, and Wong Halves are all balanced. To use a balanced count for betting, you must convert running count to true count by dividing by decks remaining.
An unbalanced count has values that sum to a non-zero number; KO and Red 7 are the common ones. Unbalanced counts use a 'pivot' point and let you bet directly off the running count, eliminating the conversion step but at the cost of some playing efficiency.
Betting Correlation is how well a counting system tracks the player's true expected value as cards are dealt. A BC of 1.00 would be perfect. Hi-Lo's BC of 0.97 means its bet-sizing decisions correlate 97% with the actual mathematically optimal bet sizing. BC matters most in shoe games (4-8 decks) where most of your edge comes from betting larger when the deck favors you.
Playing Efficiency measures how well a counting system informs you about deviating from basic strategy on individual hands — like standing on 16 vs. 10 at high counts, or taking insurance. PE matters most in single and double-deck games where individual hand decisions have a larger impact on overall expected value. Hi-Lo's PE of 0.51 is the trade-off for its simplicity; Hi-Opt II's 0.67 is one of the highest among practical systems.
Insurance Correlation tracks how well a count predicts when the insurance side bet becomes profitable. Insurance pays 2 to 1 if the dealer has blackjack, and it becomes correctly +EV when the deck has a high concentration of 10-value cards. Most counting systems have IC in the 0.76-0.91 range. The MIT team and most modern counters use the count to make insurance decisions, since insurance is the single most-impactful index play.
Probably not, unless you play single-deck games. The conventional wisdom from Stanford Wong, Don Schlesinger, and most professional counters is: in shoe games, Hi-Lo's lower playing efficiency is essentially compensated by its lower error rate. Higher-level systems require more mental work, which means more mistakes per hour. The expected-value gain on paper is often wiped out by real-world execution errors.
Use Hi-Lo for shoe games. Use Hi-Opt II or Omega II for serious single-deck play, with an ace side count.
A side count of aces is a separate running tally of aces seen, kept independently from your main count. Some systems (Hi-Opt I, Hi-Opt II, Omega II) treat aces as 0 in the main count to maximize playing efficiency, then add a side count of aces for betting decisions. The ace is the most valuable card for betting — getting blackjack pays 3:2 — so knowing aces are still in the deck is worth tracking. Side-counting aces is mentally taxing and is usually only done for high-edge situations or in pitch games.
The Illustrious 18 is a list compiled by Don Schlesinger of the 18 most-impactful basic-strategy deviations based on the count. Things like "stand on 16 vs. 10 if true count is 0 or higher" or "take insurance if true count is +3 or higher." These 18 plays capture roughly 80–85% of the gain available from index play. Memorizing them is the standard next step after mastering basic strategy and a counting system. We're building a dedicated Illustrious 18 reference tool — coming soon.