Spunk Odds Reference

Hi-lo guide

Cobra Hi-lo: the 7 in the middle

Cobra is the label on this page and has no effect on the rules. The page is about the 7, the only rank with the same number of cards on both sides, and about the beliefs people bring to it.

Six, one, six

Ranks run from ace = 1 to king = 13 here, and each next rank has chance 1/13. A 7 has six ranks below (A to 6), six above (8 to K) and the one tie. Under the strict rule each side wins 46.15% and a 1% edge sets both payouts at 2.1450x. The fair price would be 13/6 = 2.1667x.

Myth 1: "a 7 is a coin flip"

It is symmetric, but it is not even money. The tie takes 7.69% away from both sides, so each call wins less than half the time. That is why the payout is above 2x. Compared with a coin flip that pays 1.98x, the 7 call wins less often and pays more; at the same edge both lose 1% per unit on average.

Myth 2: "after three high cards, low is due"

In this model each next card is a fresh draw from 13 equal ranks. Whatever came before, the chance that the card after a 7 is lower is still 46.15%. The idea that a run of one result makes the other result "due" is the gambler's fallacy. Three higher cards in a row happen with chance 9.83% from a run of 7s; it is an ordinary event, not a debt the deck has to repay.

Myth 3: "the tie is a hidden extra cut"

The tie is already in the price. Expected return = 6/13 x 2.1450 = 0.9900, which is exactly 1 minus the 1% edge. If the tie were not priced in, the payout would be the coin-flip style 1.98x and the edge would be much larger; that is a pay table to avoid, and the tie rules guide shows how to spot it.

How long the losing runs get from a 7

Either call from a 7 loses with chance 7/13 = 53.85%, six ranks on the wrong side plus the tie. Over 50 calls the expected longest losing run is about 5.5 calls. The chance of at least one run of eight losses in those 50 calls is 13.74%, and for ten it is 3.94%.

From a 7, chance of the first winning call at round k (win chance 6/13)
Round kP(first win exactly at k)P(first win by round k)
146.154%46.15%
224.852%71.01%
313.382%84.39%
53.880%95.47%
80.606%99.29%

The 7 under the tie-wins rule

If ties count as wins, each side wins 7/13 = 53.85% and pays 1.8386x. The symmetry stays; only the shift from 6 to 7 winning ranks changes.

Is higher or lower better from a 7?

Neither. They have the same chance, the same payout and the same average loss.

Can tracking past cards help?

Not in this model. Each next rank is independent, so there is nothing to track.