REKO (“Ridiculously Easy KO”) is Norm Wattenberger’s level-one, unbalanced simplification of Knock-Out. Its documented initial counts and ramps vary by deck count; generic positive-versus-negative advice is not the REKO method.
What REKO simplifies
REKO uses the KO card tags but reduces playing-strategy memorization to one index value, +2, across the supported deck counts. Wattenberger documents the method in Modern Blackjack and compares it with KO Preferred under stated simulation conditions.
Read the card-counting overview before using a running-count system.
Card-tag values
| Card | A | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10, J, Q, K |
|---|---|---|---|---|---|---|---|---|---|---|
| Tag | −1 | +1 | +1 | +1 | +1 | +1 | +1 | 0 | 0 | −1 |
Cards 2 through 7 are +1, eights and nines are 0, and aces and ten-value cards are -1. That is six +1 ranks, not five. These numbers are card tags or point values; an index is a count threshold for changing a playing decision.
Initial running counts and betting
Wattenberger’s published initial running counts are -1 for one deck, -5 for two decks, -20 for six decks and -27 for eight decks. They are not calculated as -2 multiplied by the number of decks.
Published REKO ramps can increase wagers while the running count is still negative. For that reason, crossing zero is not a universal trigger. Use the complete ramp for the correct deck count, rules, penetration and spread.
A practice packet with a nonzero starting count
For six-deck REKO, start a counting exercise at −20. Reveal 2, 7, K, A, 8, 3, 9, 6. Your running checkpoints should be −19, −18, −19, −20, −20, −19, −19 and −18. The packet contributes +2; the final running count is −18 because the initial −20 remains part of the calculation.
Do not restart at zero after a hand or ignore a seven because Hi-Lo treats it as neutral. REKO counts every seven as +1, regardless of suit. Eights and nines leave the number unchanged, but still remove cards from the undealt pack. Keep the same running count across successive rounds until the shuffle resets the configuration.
Check a whole deck without mistaking the answer for a bet
In a complete 52-card deck, the six positive ranks contribute +24. Four aces and 16 ten-valued cards contribute −20, while eight cards are neutral. The net contribution is +4 per deck. The following are arithmetic practice answers when every card is exposed; actual play usually stops before all cards are dealt.
| Decks | Published initial count | Contribution of all cards | Practice endpoint |
|---|---|---|---|
| 1 | −1 | +4 | +3 |
| 2 | −5 | +8 | +3 |
| 6 | −20 | +24 | +4 |
| 8 | −27 | +32 | +5 |
These different endpoints are expected; do not force them all to zero. A correct endpoint can still hide two opposite counting errors, so check intermediate packets as well. Changing the initial count without shifting every corresponding decision threshold and betting instruction changes the method.
One threshold does not mean one universal decision
REKO’s +2 is a running-count threshold for its specified departures from basic strategy, not a card tag, true count or instruction to double every wager. The published single-deck version adds five decisions, while the eight-deck version omits 12 against 4. The common numerical threshold therefore does not mean identical decision lists for every deck count.
First identify the hand, dealer upcard and applicable rule. Then check whether that situation belongs to the selected REKO chart and whether the running count has reached +2. If it does not belong to the chart, continue using the matching basic strategy. A surrender departure cannot be used where surrender is unavailable, and a double-down departure cannot override a rule that prohibits doubling that hand.
Classification and performance
QFIT classifies REKO as unbalanced, level one, ace-reckoned, not suit-aware and based on compromise indexes. Its summary reports betting correlation 0.98, playing efficiency 0.55 and insurance correlation 0.78. These correlations do not guarantee a player edge.
QFIT’s published comparisons show REKO performing close to KO Preferred under the stated rules, penetrations, spreads and five-billion-round simulations. That evidence does not make REKO the MIT team’s count: the MIT teams are associated with Hi-Lo, whereas REKO is a KO simplification.
Practice and limitations
REKO supports single-, double-, six- and eight-deck configurations. Its performance depends on the complete game configuration, not just whether a game uses fewer decks. Practise the exact configuration until tags, the initial running count, the +2 playing index and the documented bet ramp remain accurate under distraction.
Practise three tasks separately: count a known packet from the correct starting value; choose the applicable playing action using the correct chart; then identify the next wager from the chosen deck-specific ramp. Combine them only when each task is accurate. Our linked trainer uses Hi-Lo, whose seven is neutral: it can develop card-recognition habits, but its score does not verify REKO tags or REKO index decisions.
Further reading
Blackjack involves financial risk. Set affordable limits before playing and do not treat a count as a promise of profit.
