Edward Thorp’s Card Counting: Why Shuffling Changes the Method

← All case studies · Edward Thorp biography

Method study · Finite decks

The key is not remembering a winning streak. It is whether previously dealt cards change what can come next.

The setting

Thorp’s blackjack work starts with a finite deck. Once a card has been dealt, it is unavailable until the deck is restored or shuffled back into play. The composition of the remaining cards can therefore differ from its starting composition. Counting systems compress that information; they do not identify the next card.

This dependency is why the dealing process matters. A shoe that retains dealt cards creates a different information problem from a game that restores the full distribution before every round. A live video feed alone does not establish which process is in use. The rules, shuffle procedure and number of cards dealt must be checked.

What the case teaches

Imagine a toy deck containing two red and two black cards. Before any card is drawn, red has probability 2/4. After a red card is removed without replacement, the next red has probability 1/3. If that card is replaced and the deck mixed, the probability returns to 2/4. This is a small mathematical illustration, not a blackjack shoe or a reported Thorp hand.

Does a losing streak make the next hand more likely to win?

A streak alone does not establish a change in the distribution. Card composition and the dealing procedure are the relevant information.

Explore the next question

Sources

Source pages checked 5 October 2026. Historical results refer to the dates stated; examples marked hypothetical are Casino101 calculations.