In bridge, the key difference between computer shuffling and hand-shuffling is that computer shuffling generally produces a more truly random distribution of cards, while hand-shuffling often results in a less random distribution, with a higher likelihood of certain card patterns due to human limitations in achieving a perfect shuffle, often requiring significantly more shuffles to reach true randomness; this can lead to complaints about "unfair" hands when using computer shuffling, as they may have more extreme card distributions than what players are accustomed to with hand-shuffling.

Key points about the difference

Randomness

Computer shuffling is designed to create a completely random order, while hand-shuffling can often unintentionally introduce biases due to imperfect technique.

Distribution of hands

With computer shuffling, you are more likely to see extreme hand patterns like singletons and voids, as the distribution is truly random, whereas hand-shuffling might produce a flatter distribution with fewer extreme hands.

Number of shuffles required

To achieve a truly random hand with manual shuffling, a player would need to shuffle the deck many times (around seven times according to mathematical calculations).

Current Trends

A majority of clubs now use computer/machine dealt hands.

All National and International competitions are administered using Computer dealt hands.

The main advantages of computer/machine dealt hands are

1. Hands are “Better Shuffled”

Computer generated hands are more random generally resulting in more distributional hands.

To create a random deal manually, players must shuffle the deck at least seven times to guarantee that the cards are properly mixed. Inadequate shuffling often leads to "flatter" distributions.

2. Board accuracy

If cards are put back incorrectly then reference can be made to a hard copy of the hands to reinstate.

3. Ability to review hands

Hands can be available at the end of a session to take away with a printed hand score.

Hands can be downloaded from the club’s website.

Makeable and optimum contracts are shown.

Hand Statistics, i.e. Average HCP, singletons, voids etc. are available for review.

Hands can replayed using the latest software, playing from any of the four positions, and experimenting with different leads.

Take a look at the image below. From a new pack of cards, you will see the results of human shuffling for every 15 seconds of shuffling. It takes around 2 minutes of shuffling to completely randomise the cards so that no 2 cards from the previous deal are touching.

On average the most people shuffle for is less than 15 seconds and some of us are better than others.

Sophisticated techniques are neither random nor any better. For example

• Pre dealing into piles before and or after the shuffle
• The “rifle” shuffle. rifle shuffling is no better because after 8 perfect rifle shuffles the deck is back to where it started!

Computer dealing truly randomises hands. This is why some people think the hands are “not fair”. It’s simply that we have been brought up in a bridge world of manually shuffled hands which tend to be flatter and not randomly distributed. This does mean more instances of long distributions, singletons and voids, but statistically you should have the same chance of picking up 13 spades as a 1 NT hand.

Reg Hull – March 2016
Many thanks to Ian Dalziel for producing the table above and providing the content in
Mr Bridge magazine No159 March 2016