Archie Karas Turning $50 Into $40 Million: A Timeline
Archie Karas walked into Binion's Horseshoe in 1992 with $50. Three years later he'd won $40 million. Then he lost it. The timeline is the moral.
The deal
December 1992. Archie Karas had $50. He was a backgammon player from LA, experienced in gambling but broke. He walked into Binion's Horseshoe in downtown Las Vegas and sat in a poker game.
The stakes were $300-600 limit hold'em. High enough to sting, low enough that he could afford a few buyins before his $50 was gone. He played smart. He won a pot. Then another. By the end of the night, he had $100.
Most people would've pocketed the money and called it a win. Archie played another session. He increased it to $200. Then $500. Then $1,000. The specific dates don't matter. What matters is the trajectory.
Round 1: The Poker Run
For the first eighteen months, Karas played poker in Binion's. He was a good player in a room where the average skill was mediocre. He compounded his wins. Most gamblers lose because they take their winnings out. Karas never took a day off. He lived on nothing. He slept in his car. All his winnings stayed in action.
By June 1994, he had over $1 million. This is six months in. He's scaled from $50 to $7-digit bankroll through consistent poker play.
Here's the subtle dangerous part: he's been winning the entire time. He hasn't experienced a meaningful losing streak. He's made better decisions, encountered worse players, and the variance has been favorable. He's developed an unearned confidence in his ability to navigate risk.
Round 2: The Baccarat Pivot
Karas gets bored with poker. Too slow. He moves to baccarat, a game where the house edge is baked in. You're not playing against other players. You're playing against math.
In baccarat, a player or banker wins slightly more often than the 50-50 you'd expect because ties are removed. The house edge is about 1.06% on banker, 1.24% on player. Over thousands of hands, the house gets paid.
Karas plays baccarat with stakes of $100,000 per hand. He's betting more in one hand than he started with in total. He wins. Then wins again. The house edge exists but variance allows him to win in any specific session or specific week.
By mid-1994, Karas has $10-12 million in winnings.
Round 3: The High-Limit Room
As stakes increase, the player pool gets wealthier but not necessarily more skilled. High-limit play has better players relative to low-limit play, but it's still not a competitive ecosystem. Karas is playing against rich people, not against poker professionals.
He continues to win at baccarat. The runs extend into the fall and winter. He's been profitable for three years straight. He flies to Vegas, sits down, and wins another million. The pattern is established. He's unstoppable.
Round 4: The Inflection Point
Mid-1995. Karas has accumulated roughly $40 million in winnings. He's played every day for three years without a significant losing streak. The variance has been monumentally favorable.
Now the mathematics changes. He's been playing positive-variance games (poker against worse players) and negative-variance games (baccarat against the house). The positive variance can't continue forever. The skill edge in poker eventually reaches zero when you're playing against people with comparable skill. The negative variance in baccarat compounds over time.
Round 5: The Collapse
Karas starts losing. The specifics don't matter. What matters is that losing once creates psychological pressure to win it back. He increases stakes. He plays more sessions. The losing accelerates.
By 1996 and 1997, Karas has lost most of the $40 million. By 1998, he's lost it all. He's back to being broke. The run is over.
Round 6: The Moral
The Archie Karas timeline is the perfect illustration of how variance compounds over time. In the short term, a good player with good luck can win enormous amounts. In the long term, the math reasserts itself.
Karas won because (1) he had skill advantage in poker and (2) variance was favorable in baccarat. When the variance reversed and the skill advantage evaporated (playing against similarly skilled high-limit opponents), he lost.
The timeline also shows the psychology of compounding wins. Each win justified the next bet. Each big score made the next bigger score feel natural. The risk increased in ways that weren't conscious. He just kept doing what had been working until it wasn't.
Most gamblers will never win $40 million. But most gamblers will experience the same pattern in miniature: a winning streak that lasts long enough to feel permanent, followed by a loss streak that lasts until the bankroll is gone. The scale is different. The mechanics are identical.
