Casual · just starting
REKO
One tag set, no true-count division, every index at +2. Real advantage for minimum effort — the right first system. Step up to KO when you want a little more without learning division.
Train with REKO →The Library
Every system in the room, explained like a human: what the tags mean, which games each count actually wins at, and what it costs you in effort. Hover any ? for plain-English definitions.
Find Your System
Every system exists for a reason. Match the reason to your life, your game, and your error tolerance. In the current public 6-deck H17 fixture, the simmed gap between the easiest real count and the industry standard is about $33/hr.
Casual · just starting
One tag set, no true-count division, every index at +2. Real advantage for minimum effort — the right first system. Step up to KO when you want a little more without learning division.
Train with REKO →Committed · teams
The tried-and-true. Every book, index table, and blackjack team speaks Hi-Lo. Learn it if you want the shared language — not because it's always the highest SCORE.
Read the Hi-Lo guide →Min-maxer · best in class
Wong Halves for shoes (betting ceiling). Hi-Opt II + ace side count for pitch games (playing ceiling). Zen for level-2 strength with no side counts.
See who wins each table →Read the numbers honestly
The public EV strips use one shared spread, public index sets, and no ace side-count model. EV/hr is the simmed mean units per round multiplied by the table minimum and rounds per hour. That setup can reward simple unbalanced systems because they are easy to execute in this simulator and are not penalized for missed true-count conversion. Harder systems still have real reasons to exist: Wong Halves can pull ahead in shoe games when the ramp is tuned to its betting correlation, ace-neutral counts can shine in pitch games with an ace side count, and team players may value Hi-Lo because everyone speaks it.
Effort vs Reward
X = subjective effort (1 easiest … 5 hardest). Y = simmed $/hr on a 6-deck H17 DAS LS shoe with a shared 1–15 spread (5M rounds each). Green = level 1, amber = level 2, rose = level 3.
Reading the Ratings
Betting correlation
How well the count knows when to bet big. King stat for 6–8 deck shoes, where betting is ~90% of your profit.
Playing efficiency
How well the count guides hit/stand/double deviations. King stat for single and double deck, where you can't spread as much.
Insurance correlation
How well the count spots profitable insurance — the one side bet that becomes +EV at high counts.
Rule of thumb: playing shoes? Prioritize betting correlation (Hi-Lo, Zen, Wong Halves). Playing pitch games? Prioritize playing efficiency (Hi-Opt II, Omega II). Just starting? Take REKO or KO and skip the division entirely.
Norm Wattenberger's simplification of KO (QFIT, the CVCX author). REKO stands for 'Ridiculously Easy KO'.
Shines at
2–8 decks
Who it's for: Beginners who want real results without true-count math.
How it works
Count 2–7 as +1 and tens/aces as −1, starting from a negative number that depends on the decks. Because the count is unbalanced, you never divide by decks remaining — the running count is the whole story. Every playing deviation triggers at the same number: +2.
Strengths
The easiest legitimate system in print: one set of tags, one index number, no division at the table. Gives up only a few hundredths of a percent to full KO.
Trade-offs
Unbalanced counts are least accurate early in the shoe, and the single +2 index leaves some playing value on the table versus true-count systems.
Simmed EV / hr by decks (H17 DAS LS · shared 1–15 spread)
1D
$290
2D
$164
6D
$51
8D
$27
Card tags
Green = count up when dealt · red = count down · T covers 10/J/Q/K
Harvey Dubner (1963), refined by Julian Braun and standardized by Stanford Wong. The most studied count in existence.
Shines at
Any decks
Who it's for: The default choice — every book, sim, and index table speaks Hi-Lo.
How it works
Count 2–6 as +1 and tens/aces as −1. Divide the running count by decks remaining to get the true count, which drives both bets and the Illustrious 18 / Fab 4 playing deviations.
Strengths
Best-documented system ever: published indices, ramps, and risk numbers exist for every game. Excellent betting correlation (0.97) for its simplicity.
Trade-offs
True-count division is real work at a live table, and playing efficiency (0.51) trails the level-2 counts in pitch games.
Simmed EV / hr by decks (H17 DAS LS · shared 1–15 spread)
1D
$166
2D
$76
6D
$19
8D
$8
Card tags
Green = count up when dealt · red = count down · T covers 10/J/Q/K
Olaf Vancura & Ken Fuchs, 'Knock-Out Blackjack' (1998).
Shines at
2–8 decks
Who it's for: Shoe players who want ~Hi-Lo results with zero division.
How it works
Hi-Lo tags plus the 7 counted as +1, making the count unbalanced. Two landmarks replace the true count: the key count (where your edge starts — bet up) and the pivot (+4, where big plays trigger).
Strengths
No division, and the built-in imbalance automatically compensates for depth in the shoe. Nearly matches Hi-Lo's win rate in 6–8 deck games.
Trade-offs
Key counts change with the number of decks, so you memorize a small table. Slightly weaker than Hi-Lo in single deck.
Simmed EV / hr by decks (H17 DAS LS · shared 1–15 spread)
1D
$290
2D
$161
6D
$59
8D
$32
Card tags
Green = count up when dealt · red = count down · T covers 10/J/Q/K
Stanford Wong, 'Professional Blackjack' (1975).
Shines at
6–8 decks (pros)
Who it's for: Professionals maximizing bet-sizing precision on shoes.
How it works
Fractional tags (5 is +1.5, 2 and 7 are +0.5, 9 is −0.5) track the deck's value with near-perfect fidelity. Convert to a true count for bets and indices; many players double every tag to avoid fractions.
Strengths
The highest betting correlation of any classic count (~0.99) — in shoe games, where betting is nearly everything, this is the ceiling.
Trade-offs
The hardest count in common use. Fractions at table speed invite errors, and the extra edge over Hi-Lo is small unless your spread is large and your play is flawless.
Simmed EV / hr by decks (H17 DAS LS · shared 1–15 spread)
1D
$165
2D
$80
6D
$18
8D
$6
Card tags
Green = count up when dealt · red = count down · T covers 10/J/Q/K
Charles Einstein's count, published by Lance Humble & Carl Cooper in 'The World's Greatest Blackjack Book'.
Shines at
1–2 decks + ace side count
Who it's for: Pitch-game players willing to side-count aces.
How it works
Only 3–6 are +1 and tens are −1; the ace counts zero. Aces act like small cards for playing decisions but like big cards for betting — keeping them out of the main count sharpens playing accuracy, and a separate ace side count restores betting power.
Strengths
Cleaner playing decisions than Hi-Lo and a stronger insurance count (aces don't pollute the ten-density signal).
Trade-offs
Without the ace side count its betting correlation (~0.88) lags badly in shoes. It's really a two-count system in practice.
Simmed EV / hr by decks (H17 DAS LS · shared 1–15 spread)
1D
$198
2D
$109
6D
$28
8D
$13
Card tags
Green = count up when dealt · red = count down · T covers 10/J/Q/K
Humble & Cooper's stronger sequel to Hi-Opt I.
Shines at
1–2 decks + ace side count
Who it's for: Single/double-deck specialists chasing maximum playing accuracy.
How it works
Level-2 tags: 4s and 5s count +2, tens −2, aces zero. True counts run about double Hi-Lo's, so indices and ramps are scaled accordingly. Pair with an ace side count for bets.
Strengths
The classic playing-efficiency champion (~0.67, near the theoretical single-parameter ceiling). In hand-held games where deviations matter most, it converts more decisions than anything simpler.
Trade-offs
Double-value tags at speed are demanding, and shoe-game results barely beat Hi-Lo — the extra work only pays in pitch games.
Simmed EV / hr by decks (H17 DAS LS · shared 1–15 spread)
1D
$199
2D
$109
6D
$27
8D
$12
Card tags
Green = count up when dealt · red = count down · T covers 10/J/Q/K
Bryce Carlson, 'Blackjack for Blood' (1992).
Shines at
1–6 decks
Who it's for: Serious players who want one strong count for every game type.
How it works
Level-2 balanced count: 4/5/6 are +2, 2/3/7 are +1, 9 is −1, tens −2, ace zero. True-count converted; Carlson's book ships a full index matrix.
Strengths
Rare combination of strong betting correlation (~0.92) and top-tier playing efficiency (~0.67) without fractions — a genuine all-rounder.
Trade-offs
Six distinct tag values to keep straight, and like all ace-neutral counts it wants an ace side count to reach full betting strength in shoes.
Simmed EV / hr by decks (H17 DAS LS · shared 1–15 spread)
1D
$202
2D
$112
6D
$28
8D
$13
Card tags
Green = count up when dealt · red = count down · T covers 10/J/Q/K
Arnold Snyder, 'Blackbelt in Blackjack' (1983).
Shines at
2–6 decks
Who it's for: Players who want level-2 strength with no side counts.
How it works
Level-2 tags with the ace kept in the count at −1: 4/5/6 are +2, 2/3/7 are +1, tens −2. One count does everything — bets, plays, insurance — via true-count conversion.
Strengths
The best self-contained compromise: betting correlation ~0.96 (near Hi-Lo's ceiling) plus level-2 playing strength, with zero side counts to juggle.
Trade-offs
Half-ace treatment means it never quite matches Hi-Opt II's playing accuracy or Halves' betting precision — strong everywhere, champion nowhere.
Simmed EV / hr by decks (H17 DAS LS · shared 1–15 spread)
1D
$162
2D
$82
6D
$15
8D
$8
Card tags
Green = count up when dealt · red = count down · T covers 10/J/Q/K