UFC 83 and the elusive 10-8 round

“Dominance” is a word that gets bandied around quite a bit in the MMA world. It can mean different things if measured over the span of a fight, performance within a weight class or even across a whole career.

But the sport only has one clear, repeatable measure of dominance: the 10-8 round. In a landscape where nearly every round is scored 10-9, a fighter must do something extraordinary to deserve that 10-8, the scoring equivalent of winning two rounds at once.

Historically, 10-8 rounds are about as rare as a Chuck Liddell submission attempt. Examining publicly available records from the Nevada State Athletic Commission, we see that over 350 rounds have been judged in fights that went to decisions. Since each round was scored by three judges, that means more than 1,000 individually judged rounds in NSAC history. In all that time, less than 50 rounds have been scored anything other than 10-9, or about 4.5%.

Maybe that’s an appropriate number, maybe it isn’t. The problem really boils down to: Does anyone know what a 10-8 round should look like?

The fights at UFC 83 provided two good examples that put this problem into context.

Georges St. Pierre vs. Matt Serra

The first round of the Georges St. Pierre-Matt Serra fight was a blowout. St. Pierre landed 14 unanswered power strikes, scored two takedowns, and passed guard seven times. Serra landed no strikes of consequence and attempted no submissions from the bottom.

Officially, two judges scored the round 10-8 in favor of St. Pierre and the other 10-9.

FightMetric uses a proprietary algorithm that harnesses the power of thousands of fight stats and results. The output of this algorithm is a single statistic that acts as a raw measure of fighter effectiveness. The effectiveness scores for the first round were 116-12 in favor of St. Pierre.

To combat the vagaries surrounding the requirements for scoring a 10-8 round, FightMetric uses a simple rule, based on data around average round effectiveness differential. If a fighter puts up a score of more than 100 and his score is more than six times his opponent’s score, that round is called 10-8. In that case, Round 1 of St. Pierre-Serra was a clear 10-8, with St. Pierre’s score nearly 10 times that of Serra.

Taking a look at a dozen or so folks who live-blogged the event, only one of them scored the first round 10-8 for St. Pierre. That is not an accusation or indictment of the judgment of those who scored it 10-9; they scored it based on their interpretation of the rules, which is as valid as anyone else’s. And therein lies the problem. Since there are no official guidelines for what constitutes a 10-8 round, your guess is as good as mine, which is as good as the official judges’, as we see in…

Nate Quarry vs. Kalib Starnes

A bizarre fight with an even more bizarre score, none of the three judges at ringside could seem able to make sense of it. In one of the strangest judgments in MMA history, one judge scored it 30-27 (meaning three regulation 10-9 rounds), one judge scored it 30-26 (meaning one round was 10-8), but the last judge scored it 30-24 (meaning all three rounds were 10-8).

FightMetric didn’t have much more luck. I never thought it would come to this, but the fight between Nate Quarry and Kalib Starnes actually broke the FightMetric system. The stats for this fight were so unusual that the system simply did not have rules in place to handle the scores.

In the first round, Nate Quarry racked up an effectiveness score of 104. That’s certainly an above-average round, with the typical fighter scoring about 60 points per round. The curious thing was that Starnes scored just two points in the round. That means a ratio of 52-to-1 in favor of Quarry. That’s by far the widest margin in any complete round we’ve tracked.

As mentioned above, FightMetric calls a round 10-8 when a fighter scores over 100 and has a score more than six times his opponent’s score. What do you do with a score 52 times the opponent’s? Theoretically, the system’s concentration on diminishing returns at the upper end would mean a requisite score 36 times your opponent’s to get a 10-7 round. But because 10-7 rounds are all but unheard-of and there has never been an official 10-6 round in a major fight that has gone to a decision, there are no rules in place beyond that. This fight wasn’t just weird, it was epically, historically, confoundingly weird. The less said about it the better.

FightMetric is the world’s first comprehensive MMA statistics and analysis system. Visit Fightmetric to learn more about the system and for analysis of MMA’s closest bouts.

Follow Yahoo! Sports' MMA coverage on Twitter.
Updated Wednesday, Apr 23, 2008