📈 economics

Limit Continuous Poker: A Variant of Continuous Poker with Limited Bet Sizes

This paper introduces and analytically solves the Nash equilibrium for "Limit Continuous Poker," a simplified variant of Von Neumann's game with capped bet sizes, demonstrating how these limits influence bluffing and betting strategies while bridging theoretical models to real-world poker dynamics.

Original authors: Andrew Spears

Published 2026-06-02

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Andrew Spears

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a simplified version of poker played between two friends, Alice (the Bettor) and Bob (the Caller). They each put a small fee into a pot. Alice gets a secret number between 0 and 1 representing her "hand strength." Bob gets his own secret number.

In this game, Alice goes first. She can either:

Check: Pass the action to Bob. If they compare numbers, the higher one wins the pot.
Bet: Put more money into the pot. But here's the twist: Alice can't just bet any amount she wants. She is stuck between a minimum bet (let's say $L$ ) and a maximum bet (let's say $U$ ).

Bob then sees Alice's bet amount and his own number. He must decide to Call (match the bet and show hands) or Fold (give up the pot).

This paper, written by Andrew Spears, introduces a new way to study this game called Limit Continuous Poker (LCP). It sits right in the middle of two famous extremes:

Fixed-Bet Poker: Where Alice is forced to bet exactly the same amount every time (like a strict rule).
No-Limit Poker: Where Alice can bet any amount she wants, from a penny to her entire life savings.

LCP asks: What happens when Alice has a range of choices, but not infinite ones?

The Big Discovery: Finding the Perfect Balance

The authors solved the game mathematically to find the Nash Equilibrium. In plain English, this is the "perfect strategy" where neither player can improve their results by changing their mind, assuming the other player is also playing perfectly.

Here is how the perfect strategy works, using some simple metaphors:

1. Alice's Strategy (The Bettor)
Alice divides her secret numbers into three zones:

The "Bluff" Zone (Low Numbers): If Alice has a terrible hand, she still bets! But she doesn't bet the same amount every time.
- The Analogy: Think of a magician. If she has a very weak hand (a bad trick), she bets the maximum amount to scare Bob into folding. If she has a slightly less bad hand, she bets a smaller amount. She uses her biggest bluffs with her worst hands.
The "Check" Zone (Middle Numbers): If Alice has a mediocre hand, she just checks. She doesn't want to bet because she's not strong enough to win a big pot, but not weak enough to bluff effectively.
The "Value" Zone (High Numbers): If Alice has a great hand, she bets to win money.
- The Analogy: Think of a salesperson. If she has a "good" product, she bets a small amount. If she has a "perfect" product, she bets the maximum amount to extract the most profit.

2. Bob's Strategy (The Caller)
Bob has to decide whether to call Alice's bet. His decision depends entirely on how much Alice bet.

If Alice bets a tiny amount, Bob needs a decent hand to call.
If Alice bets a huge amount, Bob needs a very strong hand to call.
The paper shows that Bob's "cutoff point" (the minimum hand he needs to call) changes smoothly based on the bet size.

The Surprising Twist: More Options Can Be a Trap

One of the most interesting findings in the paper is about what happens when you increase the Maximum Bet ( $U$ ).

You might think, "If Alice can bet more, she should win more money, right?"
Not necessarily.

The paper shows that if you give Alice the ability to make massive bets, Bob gets scared. He becomes much more cautious. He starts folding even with decent hands because he knows Alice might be bluffing with a huge bet.

Because Bob folds so often, Alice's "middle-strength" hands (the ones that aren't great but aren't terrible) actually lose value. They can't win big pots because Bob won't call, and they can't bluff effectively because Bob is too scared to fold.

The Metaphor: Imagine a bully (Alice) who is allowed to punch harder. If the bully is allowed to punch too hard, the victim (Bob) might just run away immediately. The bully ends up with fewer opportunities to actually land a punch on a standing opponent. The bully's "average" punches become less effective because the victim is too scared to stay in the ring.

The Hidden Symmetry

The authors found a beautiful mathematical symmetry. They discovered that the game's value depends on a specific relationship between the minimum bet and the maximum bet.

They showed that increasing the maximum bet limit ( $U$ ) has the exact same effect on the game's fairness as decreasing the minimum bet limit ( $L$ ) in a specific way. It's like a seesaw: if you push one side down (making the max bet bigger), it balances out exactly like pushing the other side up (making the min bet smaller).

Why Does This Matter?

While this is a simplified game with no real cards or chips, the paper claims it helps us understand real poker:

Bet Sizing Matters: It proves that having a wider range of bet sizes (like in No-Limit poker) changes how players bluff and call, compared to games with fixed bets.
Stack Depth: In real poker, the amount of money you have (your "stack") limits your maximum bet. This paper suggests that deeper stacks make the game more complex and can actually hurt players with "okay" hands because opponents become more cautious.
Bluffing Logic: It confirms the poker intuition that you should bluff with your worst hands using your biggest bets, and bet smaller with your "okay" bluffs.

Summary

This paper takes a complex game, strips it down to its mathematical bones, and solves it. It reveals that in poker, having more freedom to bet doesn't always mean you win more. Sometimes, giving a player the ability to bet huge amounts makes their opponents play so defensively that the player's average hands become less profitable. It's a lesson in how rules and limits shape the psychology and math of competition.

Technical Summary: Limit Continuous Poker (LCP)

Problem Statement

This paper introduces and analyzes Limit Continuous Poker (LCP), a simplified game-theoretic model designed to bridge the gap between two classical continuous poker variants: Fixed-Bet Continuous Poker (FBCP) and No-Limit Continuous Poker (NLCP).

While FBCP restricts the bettor to a single, predetermined bet size $B$ , and NLCP allows the bettor to choose any positive bet size, both models represent extreme cases of betting constraints. Real-world poker variants typically impose both lower bounds (minimum bet/raise sizes) and upper bounds (stack sizes or table limits) on bet sizes. LCP formalizes this intermediate scenario by constraining the bettor's action space to a continuous interval $[L, U]$ , where $L$ is the minimum bet and $U$ is the maximum bet. The primary objective is to derive the Nash equilibrium strategy profile for this parametric family of games and analyze how the constraints $L$ and $U$ influence optimal play, game value, and strategic convergence.

Methodology

The authors approach the problem using constructive game theory within a two-player zero-sum framework. The game involves two players, a bettor and a caller, each dealt an independent hand strength $x, y \sim \text{Uniform}[0, 1]$ . The bettor acts first, choosing to check (bet 0) or bet an amount $s \in [L, U]$ . If a bet is made, the caller decides to call or fold.

The methodology proceeds in three stages:

Equilibrium Selection and Structure: Recognizing that LCP admits a class of Nash equilibria, the authors select a specific equilibrium profile characterized by monotone strategies. They restrict the analysis to:
- A monotone calling strategy: The caller calls with all hands above a threshold $c(s)$ and folds below it for any bet size $s$ .
- A structured betting strategy: The bettor partitions hand strengths into bluffing ( $x \in [0, x_2]$ ), checking ( $x \in [x_2, x_3]$ ), and value-betting ( $x \in [x_3, 1]$ ) regions. Within the bluffing and value regions, bet sizes are assigned via continuous, monotonic functions ( $b(s)$ for bluffs and $v(s)$ for value bets) that map hand strengths to bet sizes.
Derivation of Indifference Equations: The authors derive a system of algebraic and differential equations based on equilibrium conditions:
- Caller Indifference: The caller must be indifferent between calling and folding at the threshold $c(s)$ .
- Bettor Indifference and Optimality: The bettor must be indifferent between checking and betting at the transition thresholds ( $x_2, x_3$ ) and must maximize expected value for value bets.
- Continuity Constraints: The betting functions must be continuous at the boundaries of the bet size interval $[L, U]$ .
Symbolic Solution: The resulting system of equations is solved symbolically (using SymPy) to obtain closed-form expressions for all threshold values ( $x_0, \dots, x_5$ ) and strategy functions ( $b(s), v(s), c(s)$ ) in terms of the parameters $L$ and $U$ .

Key Contributions and Results

1. Closed-Form Nash Equilibrium

The paper provides the first explicit closed-form solution for the Nash equilibrium of a continuous poker variant with bounded bet sizes. The solution is expressed using a change of variables:
$r = \frac{L}{1+L}, \quad t = \frac{1}{1+U}$
These parameters represent the pot odds faced by the caller at the minimum bet and the pot fraction remaining after a maximum bet, respectively. The equilibrium strategy profile is fully defined by rational functions of $r$ and $t$ , detailing exactly how the bettor partitions hands and how the caller sets thresholds.

2. Game Value and Symmetry

The authors derive a compact rational formula for the value of the game $V_{LCP}(L, U)$ , representing the expected payoff to the bettor:
$V(r, t) = \frac{1 - (r^3 + t^3)}{14 - 2(r^3 + t^3)}$
A significant finding is the symmetry property:
$V_{LCP}(L, U) = V_{LCP}\left(\frac{1}{U}, \frac{1}{L}\right)$
This implies that the strategic advantage gained by increasing the upper limit $U$ is mathematically equivalent to the advantage gained by decreasing the lower limit $L$ in a reciprocal manner. The value is monotonically increasing in $U$ and decreasing in $L$ , confirming that greater betting flexibility favors the bettor.

3. Convergence to Limiting Cases

The paper proves that LCP smoothly interpolates between FBCP and NLCP:

Fixed-Bet Limit: As $L \to B$ and $U \to B$ for any fixed $B$ , the LCP strategies and value converge to the known solutions for FBCP.
No-Limit Limit: As $L \to 0$ and $U \to \infty$ , the strategies converge to the known solutions for NLCP (Newman Poker).
This establishes LCP as a genuine generalization of both classical models.

4. Strategic Dynamics and Counterintuitive Effects

Through parameter sensitivity analysis, the paper reveals a non-monotonic relationship between bet size limits and individual hand values. While increasing $U$ generally increases the total game value, it can decrease the expected value for intermediate-strength hands. As $U$ increases, the caller becomes more conservative (raising their calling threshold) to protect against the bettor's ability to make larger bets with strong hands. This defensive adjustment harms the profitability of intermediate hands, even though the bettor has more options. Only the very strongest hands benefit uniformly from increased $U$ .

Significance and Claims

The paper positions LCP as a tractable model for understanding the impact of bet-sizing constraints on information asymmetry, bluffing, and balancing.

Theoretical Bridge: By solving the intermediate case between fixed and no-limit poker, the authors provide a unified framework to understand how bet size limits shape equilibrium strategies.
Real-World Insight: The authors claim that the model offers rigorous justification for intuitive poker concepts, such as the relationship between bet size and bluffing frequency (larger bluffs require fewer bluffing combinations to remain balanced) and the effect of stack depth (effective $U$ ) on strategic complexity.
Analytical Foundation: The closed-form solutions serve as a benchmark for validating computational solvers for more complex, multi-street poker variants.

The authors remain modest regarding the model's limitations, acknowledging that LCP abstracts away multiple betting rounds, non-uniform hand distributions, and imperfect information structures found in real poker. However, they argue that the analytical tractability of LCP allows for rigorous proofs of strategic principles that are often inaccessible in more complex settings.