In Defense of the Five Round Non-Title Fight

Fightlinker posted an e-mail conversation with Rami Genaur of FightMetric.  He said: I saw your post that pointed to the Five Ounces article about adding more…

By: Mike Fagan | 14 years ago
In Defense of the Five Round Non-Title Fight
Bloody Elbow 2.0 | Anton Tabuena

Fightlinker posted an e-mail conversation with Rami Genaur of FightMetric.  He said:

I saw your post that pointed to the Five Ounces article about adding more rounds. I hope I can provide some numbers that will prove that this is a very bad idea. In fact, it’s so bad that it will almost certainly have the opposite effect: By adding more rounds, you are guaranteeing more decisions, and therefore, more controversial decisions. Check this out:

Before any fight starts, there’s about a 32% chance that it will end up going to a decision.

Once Round 1 ends without a finish, the fight now has a 52% chance of eventually going to decision.

If Round 2 ends without a finish, the fight now has a 79% chance of going to decision.

In championship fights, if it goes to Round 4, there’s now an 83% of a decision.

If the fight makes it all the way to Round 5, there’s a 94% chance of it going all the way. Only one fight in UFC history has ended in the fifth round (Ricco vs. Couture).

Unless every MMA fighter drastically improves his cardio, this can’t work. Guys lose their finishing ability as the fight goes on. Giving them more time just means more chances to throw weak punches and half-hearted submission attempts.

Rami’s a smart a guy and I think he’s doing some phenomal work with FightMetric.  However, I’m not sure I agree with his logic here.

For starters, he fails to mention a key fact in his e-mail.  Three-round non-title UFC fights are finished at a clip of 67% (720 instances, all since the introduction of the Unified Rules).  Five round, championship UFC fights, however, are finished 77% of the time (74 instances).  With 68% of title fights ending before the fourth round, extra rounds DO add more opportunity for the fight to finish.  Looking at it round-by-round like Rami doesn’t really prove anything because title fights as a whole are being finished in greater numbers.

Adding more rounds also softens the impact of judging mistakes.  The value of one round drops with each additional round you tack on.  Combined with the idea that title fights are finished more often on average, this directly contradicts the notion that five round fights will introduce more controversial decisions.

I’m not arguing all non-title fights should be five rounds.  There are safety concerns as you start to include additional rounds.  I do think that all non-title main events and all number one contenders fights should be required to go five, and the option should be available for two sides to agree to a five round fight.

Update: Rami sent me an e-mail.  He was kind enough to allow it be reposted here.

You’re right to question the logic of the numbers I gave FightLinker. After thinking about it some more, I realized that those don’t prove anything and already sent Ryan a retraction. If anyone asks, I blame it on eating a bunch of chicken wings.

Now that I’ve given it some more thought, I’m left with the conclusion that adding more rounds won’t have any effect on the percentage of fights that go to decision. You can’t really prove anything from the number of championship bouts in UFC history because the sample size is too small and I have a feeling there’s a bias in the sample toward mismatches. A better example is to look at the history of Pride. Pride fights went 20 minutes (albeit within three rounds) and those fights finished with exactly the same proportion of decisions as UFC fights. In fact, the proportion of decision vs. finishes is remarkably stable across a huge sample. For some reason, it seems that about one third of fights go to a decision, no matter how long they are. This leads me to believe that adding rounds wouldn’t matter at all as far as preventing decisions, it would just make that 33% of fights take longer to watch.

Rami’s correct about the Pride numbers, but I took it an extra step.  I looked at the Bushido and non-Bushido fights separately and came up with this:

Finishing %
Non-Bushido Events: 72.5%
Bushido Events: 58.2%

I’m not sure why the Bushido fights have such a low finishing percentage (a few theories: Pride’s matchmaking, Pride’s rule set/elbows outlawed, and lighter weights finish fights less.  It might also be an aberration.), but the non-Bushido fights would suggest that there is an increase in finished fights as you add time somewhere between 3-5%/5 minutes.

Share this story

About the author
Recent Stories