Just Noticing Patterns

When I was younger, I was part of a card-counting team. We traveled around the country playing blackjack professionally. Card counting isn’t illegal, but it’s unwelcome. If casino security thinks you’re doing it, you’re usually shown the door quickly. In some places I lasted thirty minutes. In others, a few hours.

But there was one casino where I logged an absurd amount of time: Little River Casino in Manistee, Michigan. I played there well over 100 hours without getting backed off. That kind of longevity isn’t an accident, and it wasn’t because the casino was naïve. It had more to do with how patterns were being interpreted on the floor.

One of the clearest tells of a skilled card counter is bet variation. In blackjack, you want more money on the table when the deck favors the player and less when it favors the house. As the count rises, your bets go up. As it falls, they go down, sometimes all the way to the minimum, or off the table entirely. To trained eyes, that kind of disciplined fluctuation can be suspicious, especially if it tracks the count too cleanly.

At Little River, I was often playing next to a group of regulars who were doing something very similar. They moved their bets up and down constantly. From a distance, it looked like classic card-counter behavior.

But they weren’t counting cards.

They believed in what they called “streaks.”

The idea is common among gamblers. If a few face cards come out in a row, the thinking goes, more face cards are likely to follow. If the table has been cold, it’s “due” to turn hot. So when they saw a run of tens or face cards, they’d press their bets. When they didn’t, they’d pull back.

The problem is that none of this corresponds to how cards actually work.

In a six-deck shoe, every card that comes out reduces the number of that card remaining. A jack is no more likely to follow another jack than a two is. Cards don’t clump together. They don’t form veins or layers. They don’t remember what just happened. Seeing several face cards in a row doesn’t increase the odds of seeing more; it slightly decreases them.

Card counting works on a different principle entirely. You’re not chasing what just happened. You’re tracking what’s left. When lots of small cards come out, you know there are proportionally more large cards remaining in the shoe. That changes the math. Blackjacks become more likely. Dealer busts become more likely. The player’s edge increases. That’s when you raise your bet.

Here’s the important part: from the outside, both behaviors looked the same.

We all varied our bets.
We all responded to what we were seeing.
We all believed we were noticing patterns.

But only one of those patterns had a real relationship to the underlying system.

The casino tolerated the streak players because, over time, they lost money. Their pattern recognition felt meaningful, but it wasn’t grounded in reality. They were attributing meaning to randomness. Meanwhile, the actual card counters blended in, hidden in plain sight, because the noise created by false pattern recognition masked the real thing.

That experience stuck with me because it reveals something uncomfortable about the human mind.

We don’t just recognize patterns; we impose meaning on them. And subjectively, true patterns and false patterns can feel identical. Confidence, repetition, emotional reinforcement, even short-term success don’t tell you whether a pattern is real. All they tell you is that your brain is satisfied.

This is why human pattern recognition is both indispensable and dangerous. It’s how we learn, plan, predict, and survive. But it’s also how we convince ourselves that random events are messages, that coincidences are signs, and that correlation is causation.

The lesson isn’t that pattern recognition is bad. It’s that patterns don’t justify themselves. They have to be tested against reality: against structure, causality, and long-term results.

At Little River, one group of players saw streaks and lost money.
Another group saw ratios and made money.
Both groups were “seeing patterns.”

Only one group was seeing the truth.