Colonel Blotto, explained.
Colonel Blotto is a family of allocation games in which players distribute limited resources across several simultaneous contests. Winning a field depends on both allocations, not on effort alone.
Why it happens
Ten tokens spent on one field cannot be spent on another. Since each field is worth one point here, widening an already secure victory earns no extra points. Ties split the point. The allocation that succeeds against one opponent allocation can fail against a different one.
Colonel Blotto games distribute limited resources across simultaneous contests. The value of an allocation depends on the opponent's allocation and the rule for scoring each contest.
Read the result
Use minus and plus buttons to redistribute the ten tokens. Submit only after the full budget is allocated, then inspect the revealed bot resources on every field. New round changes the hidden allocation. The bot uses a stated random allocation rule, not an optimal mixed strategy.
A worked example
Concentrating versus spreading a budget
Against an illustrative rival allocation of 4/3/3, putting 10/0/0 wins only the first field for a score of one.
Changing to 0/5/5 loses the first field but wins both other fields, producing a score of two.
The second allocation is better against that specific rival. A rival choosing 0/5/5 instead would tie it, showing why an example is not a universally winning strategy.
OPTIONAL DEEPER DETAILGo deeper: inside the model
Inside this model
Three nonnegative integer allocations must sum to ten. The bot generates a valid allocation by placing each of its ten tokens on a uniformly sampled field. Higher allocation earns one point, a tie earns each player half a point. The opponent is hidden until submission. New round generates a new allocation.
Where this idea is useful
A practical use
A campaign, sales team, or defender may need to distribute effort across multiple objectives rather than maximise effort on a single one.
A common misconception
“Winning one field by a huge margin offsets losing the others.”
Field value is fixed in this board, so surplus tokens in a won field do not create extra points. Other Blotto variants use unequal values or different contest rules.
What this explanation leaves out
- This discrete three-field teaching game is not Roberson's continuous equilibrium model. The disclosed random bot is not an optimal mixed strategy and winning a few rounds does not establish optimal play.
Why hide the opponent's allocation until submission?
Simultaneous allocation would lose its strategic uncertainty if the rival's current choice were visible in advance. Revealing it afterward lets you learn without turning the contest into a known-target puzzle.
Where are you spending extra resources to widen a win while leaving another important objective unsupported?
Associated thinkers
Associations marked provisional are awaiting source review.
Further reading
Explore the original research or the teaching reference behind this experiment.