Hi-Lo 13 predicts whether the player’s first two cards total above 13, below 13 or exactly 13. It is an optional side bet, not the Hi-Lo card-counting system. Odds apply only to the stated decks, card values and paytable.
Published two-deck model
One documented Microgaming version used two decks. Aces count as 1 for this side bet and ten-value cards count as 10. Hi and Lo pay 1:1; exactly 13 pays 10:1. The main blackjack hand is separate.
Exact two-deck probabilities
Two decks contain 104 cards and 5,356 unordered two-card combinations. Direct enumeration gives 2,500 combinations above 13, 2,408 below 13 and 448 exactly 13.
| Wager | Win probability | Net payout | House edge |
|---|---|---|---|
| Hi: above 13 | 46.676624% | 1:1 | 6.646751% |
| Lo: below 13 | 44.958925% | 1:1 | 10.082151% |
| Exactly 13 | 8.364451% | 10:1 | 7.991038% |
The values depend on drawing two cards without replacement from a full two-deck shoe. Another deck count, ace rule, threshold or paytable changes them.
Not a counting system
The ordinary Hi-Lo count assigns tags to ranks to estimate the richness of undealt cards. Hi-Lo 13 instead resolves a wager from a two-card total. Sharing the words “Hi-Lo” does not make the side bet dependent on a running count, true count, penetration or bet spread.
Before wagering
- Confirm the deck count, ace value and paytable.
- Check whether the virtual shoe resets after every round.
- Keep side-bet expected loss separate from the main blackjack return.
- Do not use these percentages for another version.
How to reproduce the exactly-13 calculation
In this model, two decks contain eight cards of each value from 1 through 9 and 32 ten-value cards. Exactly 13 can be made only by the unordered value pairs 3+10, 4+9, 5+8 and 6+7. Their combination counts are 8 × 32 = 256 for 3+10 and 8 × 8 = 64 for each of the other three pairs. The total is 256 + 64 + 64 + 64 = 448.
The number of possible unordered two-card draws is 104 × 103 ÷ 2 = 5,356. Therefore the win probability is 448 ÷ 5,356. Counting unordered pairs avoids counting 3 then 10 and 10 then 3 as two different combinations while treating other pairs only once.
Turn a win probability into expected cost
For a one-unit wager that wins b units of profit with probability p and otherwise loses the unit, expected profit is b × p − (1 − p). The house edge is the negative of that expectation. Exactly 13 therefore has an edge of 1 − 11 × (448 ÷ 5,356), approximately 7.991%. A 10:1 win returns 11 units including the original stake; it does not return only ten units.
On 100 one-dollar exactly-13 bets, the long-run expected loss is about $7.99 under these assumptions. That is not a forecast for a 100-round session, and it is not the probability of finishing ahead. The occasional ten-dollar profit can coexist with a negative overall expectation.
If you also make 100 initial ten-dollar main wagers at a hypothetical 0.5% main-game edge, their expected loss is $5. Adding the side bet gives about $12.99 of combined expected loss. The main-game percentage is illustrative; use the actual rules and strategy to estimate it. Keep initial-main-wager and separately staked side-bet calculations distinct.
Read the historical source cautiously
The linked historical paytable documents the two-deck setup and payouts, but its above-13 and below-13 probability labels are reversed relative to direct enumeration. The table on this page follows the counts shown here. The historical listing does not confirm current availability or a present-day product’s rules.
In a finite shoe, these full-shoe probabilities would need recalculation after exposed cards are removed. A familiar Hi-Lo main-game count is not automatically a correct trigger for this particular side bet. Where the game resets the full shoe each round, preceding results do not change these starting probabilities.
Further reading
Side bets increase total exposure. Set affordable limits before play.
