clock menu more-arrow no yes

Filed under:

BE Analytics: How has 10-8 scoring evolved over the years?

New, comments

Here’s looking at you, Texas.

MMA: UFC 196-Holm vs Tate Mark J. Rebilas-USA TODAY Sports

Back in August 2010, then-rising star Junior dos Santos opened the UFC 117 main card against Roy “Big Country” Nelson, who was riding high off his Ultimate Fighter 10 win and subsequent 1st round knockout of Stefan Struve. The action started slow in the first 1:55 with Nelson pressing dos Santos against the cage for a good chunk of time.

Then dos Santos got free, unleashed, and unloaded.

Cigano battered Nelson in the remainder of the 1st round with multiple rockings, wobblings, and one official knockdown that almost looked like two after a Nelson stumble. Nelson shot a takedown to try to survive at one point and hugged dos Santos’ leg for dear life in a separate sequence later. In all, it was mostly a standup round that went 32-5 in power shots landed, 18-5 in jabs landed for a 10-9, 10-9, 10-8 dos Santos round.

A slow-starting beatdown round with two 10-9s. That’s where the sport was at in the not-too-distant past.

The decision to award a 10-8 or not can be critical in MMA, especially with fewer scoring opportunities in 3- and 5-round fights as compared to the 8, 10, and 12 rounds of higher-level boxing. MMA’s 10-Point Must System was originally borrowed from boxing, a sport where the vast majority of 10-8 scores come because something other than the soles of a boxer’s feet touches the canvas. If that phrase sounds familiar, MMA also borrowed its original definition of a grounded fighter from its boxing brethren.

It’s extremely rare to see 10-8 scores in boxing come solely from one fighter’s domination of the other without any knockdowns, which is exactly what MMA’s had to overcome since 2001. It couldn’t follow the general boxing principle that touching the canvas costs a fighter a point. Going to the canvas is not necessarily a bad thing in a sport whose essence is to be able to fight at distance, in the clinch, and on the ground. In fact some fighters (*cough* Fabricio Werdum *cough cough*) have been known to let a strike take them to the ground in an effort to move the action to the mat.

Back in the early years, the idea of losing a point by touching the canvas wasn’t part of MMA’s judging criteria, but the idea that a fighter had to utterly demolish the opponent to earn a 10-8 any other way seems to have remained. Which is the focus of today’s piece - how MMA judging views and implementation of the 10-8 scoring criteria have changed over the years from 2001 to today.

From the original Unified Rules of MMA through 2012, a 10-8 round was defined as when a fighter “overwhelmingly dominates” the opponent. Just like boxing without a knockdown, it was an extremely rare score, with some commissions going so far as to tell their judges not to use it.

The first modification to the 10-8 criteria came at the 2012 annual meeting of the Association of Boxing Commissions (ABC) when it was liberalized to winning by a “large margin.” The unwritten two D’s of 10-8 scores – Dominance and Damage – originated and started spreading in the 2013-2014 timeframe. A third “D,” Duration, was approved last summer and implemented this January along with a dramatically expanded written 10-8 definition that essentially boils down to always scoring a 10-8 if a fighter has all three D’s and considering a 10-8 if a fighter has two.

When boxing judges score a 10-8, it’s unanimous over 93% of the time, according to Matt Podgorski of the Pod Index. It’s essentially an objective exercise the vast majority of the time, a luxury MMA doesn’t have. 10-8 scoring in MMA is a subjective exercise originating with a somewhat strict criteria (“overwhelmingly dominates”) and an even stricter interpretation. That strict interpretation, frequently combined with only three rounds with which to work, was a recipe for seeing fighters lose the fight while actually winning “the fight.”

Quant Time

Have actual MMA judges responded to the written and unwritten relaxation of the 10-8 scoring criteria since 2012? If so, how much? It’s time to get a little nerdy up in here.

To get at that question, scorecards are used from two sources. I manually collected and entered round-by-round scores for every available UFC, WEC, and Strikeforce bout in Nevada and California from 2001 through part of 2015. FightMetric took over the process in January 2016 and has been collecting scores for every available new UFC event since. So the mixture isn’t perfect – California and Nevada in earlier years and a mixture of jurisdictions from 2016 on – but that’s what I have to work with, and we’ll also look at some judge-specific results towards the end that hopefully don’t depend on jurisdiction

Overall, there were over 1,200 fights with just shy of 3,000 rounds and almost 9,000 judge-rounds (one round is three judge-rounds since three judges score it). 96.2% of scores given out were 10-9 with 10-8s coming in at a puny 3.6%. For perspective, that means on average there was one 10-8 score from a single judge every three complete fights that went to 3-round decision.

10-8 Analysis - Round-by-Round Scores in MMA

For the remaining analyses, the only 10-7 in the data – Marcos Rosales’ 2nd round score for Forrest Petz’s beatdown of Sammy Morgan at UFC Fight Night 6 – was treated as a 10-8 in the sense that the round wasn’t a close margin (except for Adalaide Byrd who had it 10-9, but remember what her perspective might’ve been back in 2006).

10-8 Analysis - Number of 10-8s in a Round

Compared to boxing, MMA judges rarely saw eye-to-eye when 10-8 scores were in play. While only seven rounds out of 100 had some sort of 10-8 score, it was a single judge almost 60% of the time. Unanimity in 10-8 scoring happened only 14% of the time, roughly one time in seven; a far cry from boxing’s more than 93%. It almost looks like when one MMA judge decided to give a 10-8, each additional judge was roughly a 50/50 proposition, going from 59% to 27% to 14%.

Is this just subjectivity in action? Or maybe an understanding of what makes a 10-8 combined with a fear of using or possibly overusing it? It’s hard to say exactly, but one thing we can do is look to see how the written and unwritten culture of 10-8s has influenced it’s scoring from the early years of the sport to more recent times.

An easy first pass is to simply compare the rate of 10-8 scores from 2001-2012 with 2013 and later. Since it might’ve taken some time for the new view of 10-8 scoring to be taught and disseminated – and the view itself might’ve steadily changed over recent years – I also look at the 2015 and later time period.

In the chart below, the early years of MMA judging are on the left and two comparison groups for the later years (2013-2017 and 2015-2017) are on the right

10-8 Analysis - 10-8 Scores over Time

When 2013 and 2014 are included in the more recent years, there’s a slight increase in the frequency of 10-8 scoring relative to the early years. But the real improvement (colored orange to show statistical significance) seems to have come in the last three years with 38% growth in the rate of 10-8 scores from the early days.

The problem with simply comparing frequencies is we don’t know exactly what was happening in fights in the early years and more recent times. Were there more beatdowns? Fewer beatdowns? More time standing, on the ground, in the clinch? Faster or slower paced fights?

If recent fights were more competitive (probably the case), then the jump in 10-8s is that much more impressive in terms of the consideration judges are giving when they write their scores down. If recent fights were more lopsided (probably not the case), then the increase could be warranted and judges may not have actually changed their underlying 10-8 thought processes at all

I try to account for this in two ways. First, by looking only at rounds where at least one judge has scored a 10-8. This signals that at least someone thought the round wasn’t close. Then we can look at how the other judges behaved in those situations. Second, by looking at how judge behavior has changed relative to the likelihood of a 10-8 score from a statistical model calibrated only to the early years (2001-2012). This lets us see how judges behaved in the early and later years relative to the same underlying model that accounts for what happened in each round.

With the first method - using rounds where at least one judge had a 10-8 - a second judge also had a 10-8 score 38% of the time in the early days. In later years, it jumped to 45% (2013-2017) and 46% (2015-2017). This isn’t an enormous or significant difference and tells us that there still seem to be a good amount of instances where individual judges are doing their own 10-8 thing.

But the story certainly changes when looking at 10-8 scores across the board.

10-8 Analysis - Unanimous 10-8 Scores

It used to be that when at least one judge had a 10-8, there would be unanimous agreement only 1-in-11 ½ times (8.7%). In the last three years, that number’s skyrocketed 180% to roughly 1-in-4 (24.4%).

Junior dos Santos pounded on Roy Nelson in the 1st round back in 2010 and took home two 10-9s for his trouble. Fast forward to 2016, and Miesha Tate earned unanimous 10-8s for her dominance with just enough damage in the 2nd round against Holly Holm at UFC 196. I was a room with 15-20 people who laughed at me for suggesting it might be a 10-8 round. Judges Marcos Rosales, Chris Lee, and Glenn Trowbridge had adjusted to the new officiating reality (as you’ll soon see below) while casual fight fans were stuck in a past that had been repeatedly ingrained in their brains for years.

The work on 10-8 judging is by no means finished but the move from 1-in-11 ½ unanimous to 1-in-4 seems to suggest that change can happen, that judges can adapt to the new environment so long as athletic commissions are willing to get on board, modernize, and keep their judges trained in the industry’s latest and encouraged to be at the forefront.

In 2017, nobody wants to be the equivalent of dial-up Internet when fiber optic’s available. It’s embarrassing. But things are never as simple as they seem in athletic commission world.

The second method – examining judge scores relative to a model calibrated to judge behavior in the early years (2001-2012) – has interesting results as well. The chart below shows how judges in different time periods scored 10-8 rounds relative to a model calibrated to the view of 10-8s in the early 2001-2012 years. For example, did they score 80% more 10-8s than would be expected? Or 25% less? That’s what the chart is showing and it’s no surprise to see judges in the 2001-2012 time frame scoring 10-8s roughly as the model expected (only 8.1% less).

If we apply the early view of 10-8 scoring to what happened in more recent fights, judges certainly appear to have been learning and adapting, scoring 85% more 10-8s in the last three years than would be expected with the early 10-8 mindset.

10-8 Analysis - 10-8s Relative to Expected

Part of the reason for the 85% increase relative to the early mindset is the underlying statistical finding that recent fights have been getting more competitive from a 10-8 perspective. Based on the way judges scored in the early years, the rate of 10-8 scores should’ve been 33% lower in 2015-2017 solely due to the fact that fights were that much less lopsided.

Fighters are more well-rounded and knowledgeable these days. Camps, training, and nutrition have improved. Wannabe tough guys off the street aren’t getting put in the UFC Octagon anym… err… well, you get my drift.

We’ll finish up with a look at specific judges. There are many in the data, but only 19 have at least 100 rounds scored. Sorted by their 10-8 percentage, they can be seen in the table below.

10-8 Analysis - Table of Judge 10-8 Frequencies

The table doesn’t encompass the judges’ entire MMA career, just a sample of the rounds I and FightMetric have collected.

The 10-8 percentages run the gamut. It can be hard to interpret them without knowing the underlying fights, except to notice that Doug Crosby’s percentage appears abnormally high, and with 405 documented rounds Cecil Peoples probably should’ve seen more than eight 10-8s.

Better conclusions can be drawn by controlling for what actually happened during the rounds in the same way as above, modeling out how judges behaved relative to what was expected with a 10-8 mindset from the early years, 2001-2012.

10-8 Analysis - 10-8s Relative to Expected (Individual Judges)

A handful of judges were excluded from the table if they had zero or just a few rounds scored in the 2015-2017 time frame. Of the 13 that remained, all gave 10-8 scores more often in recent years than would be expected with the early mindset, and only two regressed over time (Kamijo ever so slightly that it’s probably nothing and Crosby in a major way).

Eight of the 13 judges (Griffin, Lee, Cleary, D’Amato, Morse-Jarman, Hagen, Rosales, and Trowbridge) seem to have adapted enough to score 10-8 rounds over 100% more often than would be expected with the early mindset, with four over 200% more often.

Chris Lee and Lester Griffin appear to have made huge strides from the early days to today, although Griffin could use more data as his recent sample size isn’t terribly large. MMA Decisions doesn’t have him scoring an event since UFC 200, so that might be tough. The big blue line towards the bottom shows Doug Crosby judging to the beat of his own 10-8 drum in the early years. Or perhaps he was just ahead of his time? He seems to have regressed a lot when it comes to 10-8s, but he too could use extra samples of recent data.

MMA: UFC 204-Bisping vs Henderson Per Haljestam-USA TODAY Sports

It’s fun to complain about MMA judging. Decidedly less fun is witnessing bad decisions or having a fighter’s career or title shot hang in the balance of the losing end of one. It’s also easy to complain about MMA judging. Decidedly less easy is going through the steps to learn exactly what it’s like and what they’re looking for.

And that’s where we’re at. For those who have been trained up and are on the front lines trying to employ the 10-8 judging criteria as they know it, and the regulators and committee members trying to move the criteria away from a boxing mindset towards one more conducive to MMA, this is about the fighters. It’s about trying to give credit where credit is due within the confines of reality and the system they have to deal with.

10-8 judging in MMA has certainly statistically improved in recent years. Hopefully select judges who may be set in their ways and athletic commissions who may have lagged behind will get on board and say, “Dial-up Internet sucks!” Our sport and its fighters deserve fiber.

For those wondering, Texas had the fewest 10-8 scores for a state with at least two events in the dataset with one 10-8 in 117 documented judge-rounds scored (by Sal D’Amato in the 1st round of Curtis Blaydes vs. Adam Milstead). New York, with as much criticism as people love to hurl at its athletic commission, is off to a roaring 10-8 start with 19 such scores in 213 judge-rounds (not including UFC 210).

Paul is Bloody Elbow’s analytics and business writer and is a licensed MMA judge for the California Amateur Mixed Martial Arts Organization (CAMO). Follow him @MMAanalytics. Fight data provided by FightMetric.