"Is Ludo luck or skill?" usually gets answered by whoever just lost. The dice decide every move, so it feels like luck. But every roll also asks you to choose which token to move, and choices can be measured.
So we measured them. We took the rules of our Ludo, seated simple playing habits against each other, and played more than 420,000 games. Each habit played 20,000 games against three opponents, taking every seat in turn so nobody gained from rolling first. In a four-player game the fair share is 25%: anything above that is skill, anything below is a habit costing you games.
The rules we measured
These are the rules of our game, which differ from some house rules:
- A token leaves its base only on a six.
- A six earns another roll, and so does a capture. Three sixes in a row lose the turn.
- Landing on an opponent's token sends it back to its base, except on the eight safe squares (the coloured start squares and the stars). Land on two and one of them goes back. Tokens in their home column are out of reach.
- Two of your tokens on one square do not form a block in our version. Opponents can pass them and land on them.
- A token enters the centre only on an exact roll, and the first player with all four tokens home wins.
If your family plays with blocks, or needs a capture before entering the home column, the exact numbers will differ. Which habits help and which hurt should mostly hold.
Can a habit beat random play?
First, each habit against three opponents who pick a legal move at random. Every rate here is accurate to about ±0.3 percentage points (one standard error).
| Habit | Win rate | vs fair share |
|---|---|---|
| Random legal move | 25.1% | ×1.01 |
| Bring a token out on every six | 14.0% | ×0.56 |
| Keep two tokens in play | 23.4% | ×0.94 |
| Always move your lead token | 38.4% | ×1.54 |
| Capture whenever you can | 40.1% | ×1.60 |
| Land where you can't be hit | 47.6% | ×1.90 |
| Our computer, Normal | 62.1% | ×2.48 |
| Our computer, Hard | 73.2% | ×2.93 |
"Land where you can't be hit" means: before moving, look at where each token would land and prefer a square that no opponent sits one to six squares behind (a safe square or your home column counts as out of reach). If no move like that exists, any move will do. That one check nearly doubled the win rate. Capturing came second, and racing your front token third.
Two habits made things worse. Bringing a token out on every six dropped the win rate to 14.0%. A six is your biggest step, and our best guess is that spending it on a new token leaves several half-finished tokens on the board as targets. Getting a second token out as soon as you can, and then stopping was nearly neutral at 23.4%, slightly below random. Releasing tokens is not where games are won.
So is Ludo luck or skill? Both, measurably. A player who only checks whether they can be hit wins almost twice their share. Our Hard computer, which weighs captures, safety, progress and getting home, wins nearly three times its share. Ludo is luck over one game, but a good habit shows up over many.
Against a good player, the answer changes
Random opponents are easy. Next, each habit against three copies of our Hard computer, the strongest player we have:
| Habit | Win rate vs three Hard computers |
|---|---|
| Random legal move | 2.1% |
| Bring a token out on every six | 0.5% |
| Keep two tokens in play | 1.4% |
| Capture whenever you can | 3.9% |
| Land where you can't be hit | 7.2% |
| Always move your lead token | 10.2% |
| Our computer, Normal | 15.2% |
| Our computer, Hard | 24.5% |
Here the order flips. Racing your lead token is the best single habit against strong players, and capturing whenever you can falls to 3.9%. Good opponents hide their tokens on safe squares and keep their distance, so a player who waits for captures is mostly waiting. A token raced home is one they can no longer touch.
No single habit got close to a fair share against Hard. Winning needs several of them working together, which is what the Hard computer does.
Where luck still wins
- Even the best player loses a lot. Against three opponents moving at random, our Hard computer still lost 26.8% of its games. Against equals, it won 24.5%, its fair share and no more.
- Random play still wins sometimes. Moving at random against three Hard computers won 2.1% of games, about one in 48.
- Going first barely matters. With four random players, the seat that rolls first won 25.2% and the seat that rolls last won 24.6%.
- Games are long. A game took 410 to 485 rolls across the four players, and four Hard computers made about 25 captures a game between them. That is a lot of decisions, which is why small habits add up.
What to do differently
- Before every move, count six squares behind where you will land. If an opponent is in that range, look for another move. This was the most valuable single habit against casual players.
- Against strong players, race your front token home. Once it reaches the home column it can never be captured, and capture hunting stops paying.
- Take a capture when it is free, but do not wait for one. It helps more against careless players than against good ones.
- Do not spend every six on a new token. Use it to move a token on the board, unless you have nothing to move.
The Ludo rules and strategy page covers the basics. Then play Ludo free against the computer, on Hard if you want to face the strategy that won 73.2% here.
For more dice-and-decision questions, see what is a good Yatzy score and is Minesweeper luck?
