Here’s the key lesson: we see patterns in what’s already happened not what will happen. If the past doesn’t predict the future, then looking for patterns can’t help us.

A history of winners and losers

Let’s switch the back around the game played by the highest of high rollers. Baccarat resembles poker in some ways one person is designated the player and the house is called the banker. The player and banker each dealt cards according to predetermined rules. Everyone else bets on whether the player or the banker will win the hand.

It’s a game of great dignity played slowly and dramatically for very large sums of money. Because of this drama a variant of baccarat was a preferred game of James Bond. Despite all of its ritual the most common version of baccarat involves absolutely no strategy. The cards are dealt out according to fixed rules and neither the player nor the banker can do anything to affect what happens.

In its essence this game is the same as betting on flips of a coin, all you can do is decide about player or event banker and then you watch to see what happens. So what do high roller baccarat bettors do? They track the pattern of the game overtime. The casinos issue special cards on which the bettors can mark who wins each round or they have a computer screen attached to the table that shows the history of winners and losers.

Those screens show that history and as many as five or more different graphical formats so that people can look for patterns. If the banker and player alternate winning that generates an alternating pattern notice ping-pong. Other patterns are given evocative names that mirror their shape, a vertical line followed by horizontal line is called the Dragon for example.

We see patterns in what’s already happened

But, an alternating pattern or a repeating pattern or anything else tells you nothing about what’s going to happen next, in baccarat like in most gambling you can’t predict the future from the past. Here’s the key lesson: we see patterns in what’s already happened not what will happen, if the past doesn’t predict the future, then looking for patterns can’t help us.

Now suppose that you are attending a basketball game. On one of the teams is Lebron James one of the most talented athletes in professional sports. Lebron dribbles down the court and makes a jump shot an then a few minutes later makes his next shot then a minute later makes his third shot in a row. When it comes down the court the fourth time he rises up to attempt the jump shot and your heart rises with them or sinks, depending on whether you rooting for his team.

Let’s pause the game with Lebron in midair. What is the probability that this shop will go in? Lebron is an outstanding player, he makes about half of the shots. Given that is made three shots in a row does he now have a greater than 50% chance to make the forth shot. Many people answer yes! they say that Lebron has the hot hand and they predict that his chances of making the neck shot are considerably greater than normal.

But, does this nonrandom hot hand affect stand up to scientific scrutiny? More than two decades ago a team of researchers systematically tested a simple question: are basketball players more likely to make a basket if the previous shot was successful? They examined data from the previous year’s NBA season, tracking every shot taken by every player on one team, they also looked at data from the local college team.

The hot hand effect

They even brought both collegiate and recreational players into the gym and had them take hundreds of shots. The conclusion was clear and compelling: the probability of making a shot did not change regardless of whether the previous shot was made or missed. Since that time there’ve been hundreds of analyses of data from real-world basketball games as well as from other sports: the core conclusion of that first study has been corroborated repeatedly: the sequences of made and missed shots are essentially random.

Sports fans and players alike might find this conclusion unbelievable. Of course basketball players have a hot hand! we’ve seen it and we remember it, we’ve seen our favorite players make shot after shot where can’t miss. And what about all the other sports? A baseball player who hits home runs on three consecutive at-bats, a football quarterback who complete pass after pass or a golfer whose every putt heads straight for the whole.

How, then, can some scientists claim that there isn’t a hot hand effect? Aren’t our memories enough proof? Our memories are proof actually there is a problem. We think that there is a hot hand effect because we can remember examples of players making four or five or more shots in a row. When one of our favorite players does that we might leap out of her seat with excitement.

Those are vivid memories, but remember the representativeness heuristic in the previous lecture: we overestimate the probability of events that can be easily brought to mind. Let me illustrate this concept a different way: suppose Lebron James made 50% of the shots, not because of his skill but because of a random process, then we can imagine each game as a series of coin flips if he took 20 shots in a game there would be 20 flips of a coin.

Of course, every few games the coin would come up the same way over and over again in a long streak just by chance, but we would remember that streak and attribute it to a hot hand even though the hot hand of the basketball player is no more real than the hot hand of the coin flip. So does nothing matter is everything in a basketball game just random?

Not exactly! The players believe in the hot hand effect to, so players who make a few shots in a row will start shooting more and taking more difficult shots.