Penetration: How Deep the Shoe Goes, and Why It Decides Everything
Updated
Penetration is how much of the shoe gets dealt before the shuffle. It sounds like a detail. It is the single variable that decides whether counting is worth doing at all: moving from 50% to 75% multiplies the earn rate by nearly seven.
- Penetration = the share of the shoe dealt before shuffling.
- At 50%, a 1–8 spread earns about $1.51 an hour on $25 units.
- At 75%, the same play earns about $10.20 — nearly seven times more.
- It matters more than deck count, more than S17, more than anything else on the felt.
What penetration is worth
Six decks, a 1–8 bet spread, 60 rounds an hour, correct basic strategy with Hi-Lo. The only thing changing between rows is where the cut card sits.
| Penetration | Rounds at true count +1 or better | Units per hour | At $25 a unit |
|---|---|---|---|
| 50% | 33.8% | 0.060 | $1.51 an hour |
| 65% | 35.6% | 0.252 | $6.30 an hour |
| 75% | 37.0% | 0.408 | $10.20 an hour |
| 85% | 38.1% | 0.592 | $14.80 an hour |
The reason is straightforward once you see it: a count only becomes large when a lot of cards have gone. Shuffling at the halfway point removes exactly the part of the shoe where the money is.
Why the curve is so steep
The true count is the running count divided by the decks remaining. Early in a shoe you are dividing by a big number, so even a substantial running count produces a small true count. Late in a shoe you divide by a small one, and the same running count becomes a large edge.
Cutting off the last quarter of a shoe does not remove a quarter of the value. It removes most of it.
What to look for at a table
Watch where the cut card goes in. With six decks, a card placed with about a deck and a half left is roughly 75% — good. Two and a half decks left is around 60% and marginal. Three decks left is 50% and, on the numbers above, not worth counting.
Online live-dealer tables are usually the shallowest of all, and the main reason it rarely pays. A continuous shuffler is the extreme case: penetration is effectively zero.
One number here is provisional
The site currently uses 50% penetration as its standard assumption throughout, which is realistic for live online tables and pessimistic for a good physical one. It is a deliberately conservative default, and it is the reason the earn rates quoted around this site are low.
If you play somewhere with deeper penetration, read the table above rather than the headline figure. The difference is not a rounding error.
Keep reading
By Martin Kosek · every number on this site is computed by us, not copied.
← Back to guides