Is Hnefatafl Balanced? Depends on the Rules

Mostly, yes. Under Copenhagen rules the defenders win about 1.36 games for every attacker win across more than 13,000 recorded online games, and among the strongest players that falls to 1.07, which is about as even as chess. Other rule sets lean further, some toward the defense and some toward the attack, which is why the question has no single answer.

An asymmetric game has to earn its balance

Chess gets balance nearly free, since both players field the same army; its only asymmetry is the first move. Hnefatafl balances unlike things: numbers against mobility, a wall against a race. There’s no symmetry argument guaranteeing fairness, so every tafl ruleset is a set of dials, escape squares, king strength, turn order, board size, that either happens to balance or doesn’t.

History’s players likely tuned by feel and local habit, and we’ll never know how fairly their versions played, because their rules died with them. Modern reconstruction rediscovered the problem the hard way: the first popular twentieth-century rule sets, built from a flawed reading of Linnaeus, paired a big defending army with escape to any edge square, and the defense won constantly. A strong king with dozens of exits simply outruns a wall. Many cheap boxed sets still print rules of that family, and a new player’s “we tried hnefatafl and the king always wins” experience usually traces straight to them, not to the game itself.

What the modern data shows

The tafl community, centered on aagenielsen.dk, measures balance directly. The site counts only full pairs of online games, where each player takes both sides, and divides the stronger side’s wins by the weaker side’s. A score of 1.00 is perfect balance. The same page puts chess’s white-first edge at roughly 1.08 to 1.27. That evidence base drove the design of Copenhagen Hnefatafl in 2012: corner escape, the four-sided royal capture, the shieldwall, exit fort, and repetition penalties were chosen because tested combinations without them leaned one way or the other.

Here are the figures as of the site’s August 2026 update:

• Copenhagen Hnefatafl 11x11: defenders ahead at 1.36 across 13,640 games. The lean shrinks as players get stronger: 1.31 above a 1500 rating, 1.18 above 1700, and 1.07 above 1900 (1,418 games).
• Brandubh 7x7: defenders ahead at 1.62 across 2,318 games, and the lean doesn’t shrink with skill: 2.21 among players above 1900, though that’s only 364 games.
• Tablut 9x9: close to even. Two Sámi Tablut reconstructions measure 1.17 toward the defenders and 1.04 toward the attackers, over a few hundred games each.

So the traditional reconstructions on this site vary. Tablut plays near even, Brandubh favors the king’s side clearly, and Ard Rí has too few recorded games under this site’s rules to say. They’re played for history and variety rather than tournament fairness; each variant page says who its game suits. The is hnefatafl solved page covers why nobody can yet say what perfect play would do.

Two practical conclusions. For fair competition, play Copenhagen. For any variant, the real balancer is the format the community has always used: play pairs of games with sides swapped, so both players attack and defend, and the ruleset’s lean cancels out. The strategy hub covers making the most of whichever side you’re dealt, and the fastest way to form your own opinion is a swapped pair on the homepage.