Almost every argument about a decision comes from the same place: the viewer is scoring the fight one way and the judges are scoring it another, using criteria most people have never actually read. The criteria are public, they are ordered, and once you know the order a lot of "robberies" stop looking like robberies.
The ten-point must system
Every round is scored on its own. The fighter who wins the round must receive 10 points, and the other fighter receives 9 or fewer. There is no running total carried between rounds and no credit for a round that was nearly won. A round you lose narrowly and a round you lose badly can both be 10-9, which is the single biggest source of confusion in the sport.
A 10-8 round is not reserved for a knockdown. It is available whenever one fighter wins the round by a clear margin — through volume, damage, dominance of position, or a combination. Judges have been repeatedly instructed that 10-8 rounds are under-used. A 10-7 round exists for a round of overwhelming, one-sided punishment, and is genuinely rare.
The criteria, in order
This is the part that decides close rounds, and the order matters more than the list:
- Effective striking and grappling. This is the primary criterion and in most rounds it is the only one that gets used.
- Effective aggressiveness. Only considered if the first criterion is genuinely even. Aggression means effective aggression — moving forward while missing is not scored.
- Fighting area control. Only considered if the first two are even. In practice this decides very few rounds.
Read that order again, because it is where most disagreements live. Cage control does not offset a striking deficit. Pressure does not offset being out-landed. Those factors only enter the conversation when what came before them is a genuine tie.
What "effective" means
Effective striking is judged primarily on impact, not count. A round where one fighter lands forty glancing shots and the other lands six that visibly hurt is not automatically a forty-to-six round. Judges are asked to weigh damage and the effect a strike had on the opponent, which is why significant-strike totals on a broadcast graphic often disagree with the scorecards.
Effective grappling is not measured in takedowns. A takedown that leads to nothing is worth very little; a takedown that leads to sustained control, damage, or a near-submission is worth a great deal. Threatening a finish counts even when the finish does not arrive. A fighter who spends three minutes defending a tight choke is losing that exchange badly, whatever the strike count says.
Why three judges disagree
Judges sit on three different sides of the cage. A round decided by body work looks very different from a seat where a shoulder is blocking the view, and a round decided by ground control looks different depending on whose back is to you. That is a structural feature of the sport rather than evidence of incompetence, and it is why split decisions in genuinely close fights are normal rather than scandalous.
The more useful question after a close fight is not "who won" but "which round was scored differently, and what was in it". Nine times out of ten the answer is a single round where one fighter had volume and the other had the harder moments.
What judges are not scoring
They are not scoring the fight as a whole; they are scoring five separate rounds. They are not scoring who looked more hurt at the end. They are not scoring who the crowd supported, who was the aggressor without landing, or who talked more before the fight. And they are not permitted to reward a fighter for a round they lost because they won the previous one more convincingly.