Someone who has just learned the rules sits at the same table as someone who has played for ten years, and they play for an evening. When the chips are counted at the end, the beginner has more.
This happens all the time in Texas Hold’em, often enough that it is not worth being surprised by. It does not mean the beginner has a hidden talent, and it does not mean those ten years were wasted. What it means is this: one evening’s worth of hands cannot show the skill gap between two players.
1. Over a hundred hands, the stronger side is behind 43% of the time
This has been measured on a large amount of data. A study by Potter van Loon, van den Assem and van Dolder, published in PLOS ONE in 2015, took a full year of real-money no-limit Hold’em records from a large online poker site (October 2009 to September 2010, about 456 million player-hands) and ran a test:
First, using the first six months of results, they ranked every player who had played at least a thousand hands and picked out the best 1% and the worst 1%. Then, from the hands those two groups actually played over the following six months, they drew the same number of hands for each and compared who won more — 25,000 times over.
The result:
| Hands drawn for each group | Chance the best 1% is behind |
|---|---|
| 10 hands | 47% |
| 100 hands | 43% |
| 1,000 hands | 29% |
| 10,000 hands | 4% |
Over a hundred hands, the top-ranked group still loses to the bottom-ranked group 43% of the time — not far from a coin flip. By the estimate in online vs live, a live table deals around seventy hands an hour, so a whole evening comes to two or three hundred hands: roughly a 29–43% chance that the stronger player ends up behind.
The authors treat “the stronger side ahead three times out of four” as the point where skill starts to outweigh luck, and they put it at about 1,500 hands. And that is for the two groups furthest apart in the data; compare the best and worst 10% instead and it takes around 6,500 hands. Playing live once a week for three hours, 1,500 hands takes nearly two months, and ten thousand takes nearly a year.
To be clear, this table is not beginners against pros: everyone in the ranking had already played a thousand hands, so the worst 1% are people who have played plenty and kept losing. The gap between someone who has just learned the rules and a professional is probably wider, which means fewer hands to tell them apart; but the direction is the same — the fewer the hands, the closer the result is to a coin flip.
Lab studies point the same way. A study by Meyer and colleagues, published in the Journal of Gambling Studies in 2013, seated players six to a table (three average players, three experts) to play 60 hands for real money, with the deal controlled in advance. Who finished with more money depended mainly on the cards they were dealt, and the expert group did not win more; the one difference was that when dealt bad cards, they lost less. On that basis the authors argue that poker should, at least under certain conditions, be treated as a game of chance.
2. Skill is a little extra every hand; luck is a whole pot at once
Why does it take so long? Start with how big the gap between these two groups actually is.
The same study worked it out: on average, for every 100 hands played, the best 1% won 21 big blinds more than the worst 1%. Per hand, that is 0.2 big blinds. The skill gap between the strongest and weakest groups in the data comes down to that much on a single hand — not even one big blind.
Luck works on a completely different scale. Get all in with both players holding 100 big blinds and run it out, and that one hand is won or lost for 100 big blinds.
A single all-in is worth about four hundred and seventy hands of accumulated skill edge. And the deal decides who wins the all-in: you can work out your equity correctly and the cards can still land in the share where your opponent wins. Is Texas Hold’em gambling? works out how likely it is to lose several of those in a row.
So in the short run, who wins and who loses comes down mostly to which way the few big pots go. Skill is present in every hand but small in every hand; only once enough hands stack up does it rise above the swings of the big pots.
3. Results take thousands of hands to speak, so you cannot wait for them to improve
The same data holds one more number: of all the players, 58.9% played fewer than a hundred hands in the entire year. For most people on that site, their whole experience of the game sits in the second row of the table above: results close to a coin flip, strength impossible to see.
For someone just starting out, that means two things.
The first is how to read your own results. Winning your first few evenings does not show you play well; losing them does not show you are not cut out for it. A beginner can beat a pro — that part is true — but it is a result of too few hands, not evidence of talent. And the reverse holds too: a gap that takes thousands of hands to show will not come over to your side because you won tonight.
The second is how to improve. If you wait for results to tell you what you did wrong, you first have to play a thousand-odd hands, and even then the results will only tell you that “something overall seems off”, not which decision it was. So improvement has to run the other way: first learn how each decision should be made, then check whether you actually made it that way — no waiting for results.
Nor is that 0.2 big blinds won through some single inspired hand. It is many small decisions, each a little better, repeated thousands of times. So what you practise is not any one big hand but the few decisions that come up in every hand — and the first one you meet is whether to enter the pot preflop, and how.
Why might a beginner beat a pro? Because spread across each hand, the skill gap is a fraction of a big blind, and the luck of one big pot can outweigh hundreds of hands of it. That does not mean skill does not matter. It means skill is paid for in hands, and those hands have to start piling up before you can see anything.
For how to make the preflop decision on every hand, start with the complete preflop guide; or start practising right away — the preflop trainer deals you a spot and checks your answer immediately, no thousand hands of results required.
Read this next: if skill is paid for in hands, which habit should you fix first? Why you shouldn’t limp in covers the most common one — calling one big blind to see the flop when nobody has raised — the three things it gives up at once, and what to do instead.
If the rules still have gaps, or you want to know what to read next, go back to what Texas Hold’em is; it ends with a map that puts every guide on this site in order.