Deep vs Shallow Study: The Key to Rapid Poker Improvement
Uri Peleg is arguably the best poker coach in the world.
He just released a game-changing course called Upswing Lab 2.0 here (and today is your last chance to get the launch week 25% forever discount).
To help you understand the insightful way Uri approaches poker, he wrote this article about a critical topic for modern poker players: deep vs shallow study.
How Online Poker Changed Everything
Poker has undergone a huge transformation in the last 20 years.
Before online poker, it was a game played with a dealer and a deck of cards.
Moving the game online transformed it in three big ways:
- The pace of play: Online, you can play multiple tables at once, with each hand dealt far faster than in a live setting. That means you can gain experience 40 times faster, or more. In my first five years online, I played the equivalent of 200 years’ worth of live hands. Naturally, this led to players improving at a much faster rate.
- The player pool changed: Online poker attracted a different kind of player. People who might not excel at reading “poker faces” or picking up subtle body language could now thrive purely through strategic and analytical thinking. A game that once favored live social skills suddenly became a playground for math-minded problem solvers.
- Lower stakes: In a casino, you might need $1,000 just to take a reasonable shot at a poker table. Online, you could start with just a few dollars. That opened the game to college students, hobbyists, and anyone looking to start small and climb the ladder.
What followed was a rapid shift in strategy as players uncovered parts of the game that had barely been explored before. Those at the front of the curve made a fortune.
Poker itself evolved quickly — and poker theory evolved right along with it.
My Introduction to the Game
I first discovered poker around 2011. At that time, many strategic concepts were still poorly understood. Everyone made plenty of mistakes — a lot of mistakes — and if you had even a basic grasp of solid strategy, making money was surprisingly easy.
One example from PokerStars sticks with me. There was a player with a screen name something like “Aurora” who played 24 tables at once, which was the maximum allowed by the software. After raising preflop, he folded to 3-bets roughly 75% of the time.
That number was so obviously too high that he became an easy target. Any time he opened a hand, you could reraise him with almost anything — 7d 2c, no problem. And because he was on nearly every table, all day, every day, there was a steady stream of opportunities to exploit him.
The dynamic around him became so obvious that whenever someone 3-bet him, I would cold 4-bet bluff.
Then one day, by chance, I was invited into a Skype strategy group that he was part of. We became friendly, and out of genuine curiosity I asked him:
“Why are you folding so much to 3-bets? Don’t you see the garbage people are reraising you with?”
His answer floored me.
“What do you mean? I’m folding the correct amount.”
It turned out he simply had the math wrong. And honestly, why wouldn’t he? At the time there was no single, easy-to-access source for this kind of information. We all had to figure it out for ourselves.
Janda, MDF, and Mistakes in Theory
Another friend of mine bought a massive strategy book from one of the leading theorists of the time — Matthew Janda. In it, Janda built a deep and elegant strategy around the concept of minimum defense frequency (MDF).
My friend studied the book, worked on the concepts, and fully integrated them into his game.
But he couldn’t win.
As we eventually discovered, the core premise the entire book rested on was actually wrong. Nobody knew that back then — not even the author.
Fast forward a few years, and poker training sites became more polished and accessible. Then came the next haymaker: poker solvers (like Upswing’s own Lucid Poker solver) entered the market.
The Solver Era Begins
With solvers, you no longer had to wrestle with abstract theory. Now a computer could show you exactly how the game should be played. And just like that, the poker world shifted again.
No more “street poker.” No more guessing at the shape and logic of strategy.
A new race had begun, and the game suddenly became accessible to a different kind of player. You didn’t need to invent the answers anymore — you just had to study them, integrate them into your game, and execute.
But solvers were a double-edged sword.
And that brings us to the focus of this chapter: deep vs. shallow study.
Deep vs Shallow Study (Part 1)
Why Depth Matters
A solver doesn’t just think about the hand in front of it — it thinks deep. Every decision is made with the turn, the river, and multiple possible runouts already in mind. And poker offers an enormous range of possible decisions, across countless boards and situations.
With that much complexity, it was easy to get lost in the weeds — or, as I liked to say, disappear down a rabbit hole — when working with solvers.
The real skill was learning to spot the shape of a strategy: focusing on the most important and impactful elements, while setting aside the details that didn’t matter as much.
That skill came naturally to me. I’ve always disliked obsessing over tiny details, so I was forced to look for broad patterns instead. It turned out to be a strength — my friends and students who studied with me thrived using the same approach.
Enter: The Data Revolution
Then came another haymaker: data.
Tools for analyzing massive datasets became publicly available. Combined with solvers — which showed you how you should play — you could now see how people actually did play. It was a huge leap forward in technology, and one I never fully embraced myself.
I had a pretty good intuition for population tendencies, and like with any new advancement, I wasn’t motivated to dive in as long as I was winning without it.
But now there were players who would bluff not because a solver recommended it, or because an opponent’s heads-up display (HUD) stats showed they folded often — but because the data said it was profitable. It was a different level of thinking and execution, giving anyone who used it access to insights the best players in the world had always felt intuitively, now delivered in a scientific, measurable way.
Poker Today: Complex, Unsolved, and Full of Opportunity
That’s where poker stands today. It’s come a long way — and it still has a long way to go. The game is so complex that very few players get even the fundamentals right.
For anyone with talent and motivation, there’s still plenty of room to succeed.
At least… until the next big change.
Deep vs. Shallow Study (Part 2)
So how do you study deep? And what separates deep study from shallow study?
The best way to explain it is through an example — and it doesn’t even have to come from poker.
Let’s start with something completely different: the card game Magic: The Gathering.
Jon Finkel and Magic: The Gathering
Jon Finkel, one of the all-time greats in Magic: The Gathering, was playing a match.
On the board, he had a creature in play. In his hand, he held a card that could protect it from removal. At that moment, those were his only relevant cards.
He passed the turn to his opponent, who immediately played a card that would remove Jon’s creature.
What Jon did next was subtle — but it’s the kind of move that sets a player apart from almost everyone else in the world.
In one smooth motion, without hesitation, he let the creature die.
99.9% of players would have stopped their opponent, paused to think, and then decided whether or not to protect the creature. Even if they chose to let it die, the hesitation would have given away that they had a decision to make.
- Jon didn’t need to think.
- He had already considered every possible play his opponent could make — and decided in advance how he’d respond.
- By the time the removal spell hit the table, his mind was already past the decision point.
The result? His opponent believed, completely naturally, that Jon had no way to save the creature.
That’s what deep study looks like in action.
Back to Poker: Shallow Study
Shallow study looks something like this:
You’re dealt As Ks and reraise your opponent preflop. The board comes Qs Jd 3c.
- You bet, and he calls.
- The turn is a 4d, bringing in a flush draw.
- You bet again, and he calls.
- The river is an Ah, and now you’re not sure what to do.
After the session, you fire up a solver and skip straight to the river. You see that the solver checks your hand, and then mixes between calling and folding versus a bet.
“Cool — now I know for next time.”
That’s shallow study.
What’s deep study?
Back to Poker: Deep Study
Deep study means looking at this spot and thinking beyond just A-K.
- What if we had K-Q?
- What if we had A-T?
- What if we had A-Q?
You need to know how to play all of your hands on the river — not just one. And when you check, you also need to know: what is my opponent supposed to bluff with?
That’s not a trivial question.
What if he has J-T or K-J — are those ever winning for him?
- Should they bluff?
- Can they bluff?
- How does that affect our decision?
- And what if we’re playing a bit deeper?
- What if the positions are different?
The more you start asking these kinds of deeper questions, the more you begin to understand the DNA and structure of the game.
“What should I do on the river with my hand?” is really just a subset of the bigger question: “How do I play rivers in general?”
If you only know the answer to the first, you’ll be lost when the situation changes. But if you understand the second, you’ll almost always know the answer to the first — aside from a few nuances.
Most players study poker in a shallow way, which leaves them stuck at the level of imitation. To reach mastery, you have to study deep.
To lean more about how Uri Peleg recommends playing top pair, weak kicker, read: Top Pair, Bad Kicker? Here’s the Right Way to Play It.
Most players will keep studying the way they always have — shallow, narrow, and stuck at the level of imitation.
You don’t have to be one of them.
The Lab 2.0 is your blueprint for deep, structured study — the kind that builds real understanding, not just spot knowledge. It’s where solver precision meets real-world application, and where modern population data shows you how to actually win in today’s games.
If you’re serious about mastery, start here.
