KO Card Counting: The Knock-Out System Explained

By CardCount Academy · Reviewed October 5, 2026 · 9 min read

KO, short for Knock-Out, is the best-known unbalanced counting system. It was published by Olaf Vancura and Ken Fuchs in 1998 with one goal: remove the hardest part of counting — converting the running count into a true count — while giving up as little accuracy as possible. You keep one number, and that number alone tells you when to raise your bet.

Before you start: select KO in the menu at the top of the site so every drill, quiz and the simulator use these values and starting counts.

The card values

The only difference from Hi-Lo is the 7, which counts +1 instead of 0. That one change is the whole trick. A deck now has 24 cards worth +1 and 20 worth −1, so counting down a full deck ends at +4, not zero. That is what "unbalanced" means.

Why the imbalance removes the division

In a balanced count, a running count of +6 is strong with one deck left and nearly meaningless with five left, so you have to divide. In KO, the count drifts upward by four points for every deck dealt, even when the shoe is perfectly average. If you start from the right negative number, that built-in drift does the division for you: the same running count means roughly the same thing early and late in the shoe at the moments that matter most.

Where to start: the initial running count

Because a deck adds up to +4, you do not start at zero. The standard starting point is:

Initial running count = 4 − (4 × number of decks)

With this starting point, every game — whatever the number of decks — ends at +4 if all the cards are dealt. That is a useful self-check when you practice with a single deck: start at 0, count it down, and you should finish on +4.

The key count: when you have the edge

The key count is the running count at which the player, on average, first has an advantage over the house. Below it, bet your minimum. At or above it, bet more. In a six-deck game the key count is −4.

The simplest version of the system — the authors call it KO Rookie — is exactly that: two bet sizes, small below the key count and large at or above it. It is a real, working strategy, and a good first target.

The pivot point: where KO is most accurate

The pivot is the running count at which KO agrees exactly with a balanced true count, no matter how many decks are left. With the starting counts above it is always +4, and it corresponds to a true count of +4.

Here is why. In a six-deck game you start at −20. In an average shoe the count rises four points per deck dealt, so with r decks remaining a neutral shoe shows 4 − 4r. If your running count is +4 instead, the shoe holds 4r extra high cards — four per remaining deck — whether r is one or five. At the pivot, the division has already been done.

Away from the pivot the match is approximate. Near the key count early in a shoe, KO slightly misjudges the edge compared with a true count. In practice the difference in results between KO and Hi-Lo is small, because the largest bets are placed near the pivot, where KO is exact.

Betting with KO

A full bet ramp rises from your minimum bet just below the key count to your maximum at the pivot. Our tools use a simple teaching ramp for six decks: one unit at −4 or below, rising in even steps to eight units at +4 or above. As with every ramp on this site, it is a learning aid, not advice — see the bet spread guide for what real bet sizing depends on.

Playing decisions

You play basic strategy almost all the time. The most valuable exception is insurance: take it when the running count is +3 or higher, in any number of decks. The complete system (KO Preferred) adds a short list of other plays that change at the key count or the pivot; the book lists them in a single small table. Learn insurance first — it is worth more than the rest combined.

KO compared with Hi-Lo

• Effort. KO has no deck estimation and no division. The cost is remembering a starting count and key count for each game you play.
• Betting accuracy. About the same. KO's tags correlate with the effect of each card on your edge slightly better than Hi-Lo's.
• Playing accuracy. Hi-Lo with a true count supports the full set of index plays. KO uses fewer and applies them at fixed running counts.
• Negative numbers. A six-deck shoe spends most of its time between −20 and −5. Some people find that tiring; others stop noticing after a few hours of practice.

Red 7: the other unbalanced count

Arnold Snyder's Red Seven count predates KO and uses the same idea with half the imbalance. Only the red sevens count +1; black sevens are 0. A deck adds up to +2, so you start at −2 for every deck in play (−12 for six decks). The pivot is 0, which corresponds to a true count of about +2: when the count reaches zero or higher, you have the advantage and raise your bet.

Red 7 keeps the count closer to zero and its pivot sits right where the advantage begins, at the cost of checking the color of every 7. Both systems are available in the system menu, and the systems comparison puts their numbers side by side.

Common KO mistakes

• Starting at zero in a shoe game. In six decks that overstates every count by twenty points.
• Counting the 7 as zero out of Hi-Lo habit.
• Dividing by the decks remaining. KO is used straight from the running count.
• Carrying a key count from one game to another with a different number of decks.

Practice plan

1. Select KO, then run the values drill until the 7 is automatic.
2. Count down a single deck from 0 and check that you end on +4.
3. Use the running count drill with six decks, starting from −20.
4. Play the simulator and call out "minimum" or "raise" before each bet, using the key count.

Sources and further reading

• Olaf Vancura and Ken Fuchs, Knock-Out Blackjack (Huntington Press, 1998).
• Arnold Snyder, Blackbelt in Blackjack (Cardoza Publishing) — the Red Seven count.
• Wizard of Odds — Card Counting: Introduction
• Peter A. Griffin, The Theory of Blackjack (Huntington Press).