← Back to Paradoxes

The Hydra Game

The monster you cannot lose against

The Paradox

Imagine battling a hydra—a many-headed monster from Greek mythology. Every time you chop off a head, the hydra grows more heads in response. Worse: as time goes on, it grows them faster and faster.

Here's the counterintuitive truth: You cannot lose. No matter which heads you chop, no matter how the hydra grows, you are mathematically guaranteed to kill it in finite time. Every game ends in victory.

This isn't just a curiosity—it's connected to deep questions in mathematical logic. The proof that you always win is so powerful that it cannot be proven using standard arithmetic. It requires reasoning about infinity itself.

⚔️ Slay the Hydra

Click any head (green node) to chop it off

Heads
3
Turn
0
Total Chopped
0
📜 The Rules
  1. Click a head (green circle) to chop it off
  2. If the head is directly on the root, it simply disappears
  3. Otherwise, the hydra grows N new copies of the parent branch (where N = turn number)
  4. Your goal: reduce the hydra to just the root (brown circle)
🎉 HYDRA SLAIN!
You killed the hydra in X turns!
Click any green head to begin your battle!
Estimated Ordinal Complexity
ω²
Speed: 500ms
Head Count Over Time
Move History

🧠 Strategy Comparison

Compare different head-cutting strategies to see which wins fastest

Leftmost First
Always chop the leftmost available head at the highest level
-
Turns
-
Max Heads
Rightmost First
Always chop the rightmost available head at the highest level
-
Turns
-
Max Heads
Deepest First
Always chop the deepest head in the tree (closest to root = fewer regrowths)
-
Turns
-
Max Heads
Shallowest First
Always chop heads at the highest level first (most regrowth)
-
Turns
-
Max Heads

Why You Always Win

The proof uses ordinal numbers—a way to measure the "size" of infinity. Each hydra configuration has an ordinal that measures its complexity.

Ordinal Intuition

Think of ordinals as a super-powered counting system:

0, 1, 2, 3, ... ω, ω+1, ω+2, ... ω·2, ... ω², ... ω^ω, ...

Here, ω (omega) is the first infinite ordinal—the "size" of all natural numbers. But there are ordinals beyond ω!

The Key Insight

Every hydra tree can be assigned an ordinal. When you chop a head:

The counterintuitive part: the hydra can grow enormously during battle. Some configurations take more steps to kill than there are atoms in the observable universe. But finite is finite—you will win.

The Deep Connection

"Any proof technique that proves the Hydra always dies is strong enough to prove that Peano arithmetic is consistent."
— Kirby & Paris, 1982

Laurence Kirby and Jeff Paris proved something remarkable in 1982: the statement "every Hydra game terminates" is true but unprovable in Peano arithmetic (the standard axioms for natural numbers).

This connects to Gödel's incompleteness theorems. Some true statements about numbers require reasoning about infinity to prove. The Hydra game is a beautiful, playable example of this phenomenon.

To prove the Hydra always dies, you need ordinals up to ε₀ (epsilon-naught)—a number so large it satisfies ω^ε₀ = ε₀. Standard arithmetic cannot "see" this high.

Related: Goodstein's Theorem

The Hydra game is closely connected to Goodstein sequences. Take any positive integer, write it in "hereditary base-n" notation, bump up the base, subtract 1, repeat.

For example, starting with 4 in base 2:

4 = 2² → 3³ - 1 = 26 → 4⁴ - 1 = 255 → 5⁵ - 1 = 3124 → ...

The numbers explode astronomically... but Goodstein proved they always reach 0. Again, this requires ordinals to prove and is unprovable in standard arithmetic.

Timeline

1931
Kurt Gödel proves his incompleteness theorems, showing that some true statements are unprovable.
1944
Reuben Goodstein discovers his theorem about base-bumping sequences always reaching zero.
1982
Laurence Kirby and Jeff Paris prove Goodstein's theorem is unprovable in Peano arithmetic.
1982
Kirby and Paris introduce the Hydra game as an accessible illustration of their result.
Today
The Hydra game remains a favorite example in logic courses for teaching about limits of formal systems.

Sources & Further Reading