If I need to hype up this weekend’s fights with an artfully-crafted introduction, you’re on the wrong website. We’ve got three days of MMA nirvana coming up and technical experts have been throwing out kickass breakdowns left and right. So here’s my version of technical: Statistics with models behind the scenes (no, not those models).
The current model is overfit and version 2.0 will fix this issue and incorporate more flexibility, but version 1.0 still does pretty well so let’s get down to it. As always, the model is doing its thing behind the scenes and I’m trying to capture the simple statistics that appear to explain the predicted win probabilities.
Saturday: UFC 194
Jose Aldo (65.8%) over Conor McGregor (34.2%)
Distance is what we need to analyze most here as Aldo and McGregor spend 3:38 and 3:00 at distance respectively out of every five minutes in the cage (2:28 is the featherweight average). The key metrics are McGregor’s head and body volume and head accuracy and Aldo’s defense, leg volume, and leg accuracy.
McGregor throws 50.9 power strikes to the head per five minutes (P5M) at distance, more than double the featherweight average and 2 ½ times Aldo. He lands at a 41% clip to Aldo’s 29%. He also tenderizes the body 2 ½ times more than Aldo with a solid straight left and gut-busting shins.
Aldo’s powerful shins are gravitationally attracted to the legs – 4.8 attempts P5M at distance and landing 4.1 times (86%). Impressive as are his kicks, his defense is just as outstanding. Aldo only gets hit with 20% of his opponent’s power strikes and dictates the action so as to get them to throw 10.4 fewer attempts P5M than an average featherweight. When it comes to the all-important brain, opponents only land 13% of their distance power head strikes, or 2.2 P5M. Compare this to 8.8 for McGregor and the championship skills of Aldo’s distance defense become readily apparent.
McGregor’s obviously no slouch at distance and his fans should focus on this area as that’s where the fight will most likely take place if Aldo so chooses. The lineal champ defends 94% of all distance takedowns. In over 50 minutes of in-cage action, Chad Mendes and Frankie Edgar were only a combined 2 of 15 for god’s sake.
If the fight moves out of the distance position, Aldo spends 53 seconds of every five minutes on the ground and has top control 78% of the time. The 29 seconds he spends in the clinch P5M are mostly being pressed against the cage by opponents and defending takedowns.
If the fight ends up on the ground, it should be all Aldo, and neither guy’s much of a clincher. Basically, we’re probably in for a distance treat with possibly a little Aldo ground action.
In terms of judging tendencies, the two judges with enough data to analyze in the past are Junichiro Kamijo and Sal D’Amato. Kamijo is the only judge I’ve analyzed who doesn’t seem to be influence at all by missed strikes, which should be an advantage for Aldo’s defense. Sal D tends to give more weight than usual to touching jabs relative to power shots, which if anything, would seem to be an edge to Aldo’s 16.8 head jab attempts P5M at distance and 5.4 landed versus McGregor’s 14.0 and 2.4.
I’ve got Aldo as the likely winner, most likely by decision. Just for fun I’ll call it 49-46, but don’t sleep on a knockout from either fighter.
Chris Weidman (53.6%) over Luke Rockhold (46.4%)
Can you believe that in nine fully-documented bouts, 19 rounds, and 78:17 of total fight time, Luke Rockhold has never busted up an opponent’s face? Come on Patti, the man truly is sensitive. Weidman, on the other hand, is a bloodmonger who busts people up at 2.4 times the normal middleweight rate.
Weidman spends his time at distance and on the ground. If he’s clinched up with you, he’s generally throwing bombs and trying to take your ass down at 1 ½ to 3 times the normal rate (for lower and upper body takedowns) while landing 50 and 100%, respectively. Of the 27 seconds P5M that Weidman’s in the clinch, he’s had his back to the cage only 4% of the time. 4 percent! Compare this to Rockhold’s 51% and roughly average clinch takedown defense and we start to see how Weidman might get Rockhold to the mat where he has ground control 99% of the time and has only been reversed once.
If the fight stays standing and at distance, Rockhold has better power volume (3.3 more strikes landed P5M) and 3% better accuracy while Weidman has similar edges in the head jab department. Rockhold mixes things up with twice as many body strikes and lands over three times that of Weidman. He’s better with defensive accuracy to the head, body, and legs with the most important being that opponents land only 19% of power head shots on Rockhold compared to 24% on Weidman. If we combine defensive prowess with opponents’ volume, Rockhold absorbs two fewer power shots to the skull P5M at distance.
Rockhold’s never been taken down at distance but Weidman’s takedowns are on a championship level, landing 52% versus a 28% average with twice the volume. Statistically, it looks like Weidman should be able to press Rockhold against the cage and land some takedowns. The real takedown numbers battle is at distance where Rockhold’s defense has been impeccable and Weidman’s offense is outstanding.
In terms of judging tendencies, we know a little about Marcos Rosales and Adalaide Byrd. Rosales tends to put less than normal weight on head jabs and power leg strikes, which gives a slight edge to Rockhold. He puts normal weight on ground control but values submission attempts around twice as much as a typical judge, which could possibly help to make any time Rockhold spends in guard not seem as bad.
As for my personal favorite judge, Adalaide Byrd, she gives virtually no weight to submission attempts and values clinch and ground control highly – two advantages for Weidman. She’s a pretty typical striking judge but tends to see small differences as bigger advantages or disadvantages. In other words, Byrd is a higher variance judge and that could make Saturday’s co-main ending pretty damn interesting.
So where will the fight take place? Can Weidman dirty things up a little and be the controlling fighter? Can Rockhold work small edges at distance? I’m salivating just thinking about it. This fight is basically a coin flip. I’ve got Weidman with a slight overall edge and both fighters most likely to win by TKO.
Jacare Souza (63.2%) over Yoel Romero (36.8%)
If Jacare wants to get this fight to the ground, he’s a lower body takedown beast from the clinch, and that barely includes the Chris Camozzi cake walks. Romero’s Olympic wrestling pedigree doesn’t show in his lower body takedown defense as he’s about perfectly average (although more recently Brad Tavares and Tim Kennedy came up with goose eggs).
Gunnar Nelson (56.6%) over Demian Maia (43.4%)
So far, Nelson’s been good at keeping his back off the cage (controlled in the clinch only 16% of the time) and has never been taken down from the clinch in a whopping one attempt. We all know what Maia probably wants to do. The odds give Nelson the edge and, either way, this fight should most likely go to a decision.
Max Holloway (62.7%) over Jeremy Stephens (37.3%)
* Channeling his inner Bart Scott * Can’t wait!
Urijah Faber (70.1%) over Frankie Saenz (29.9%)
We can do all things through the Fight Gods! With 100% certainty they will cause a Faber victory 70.1% of the time so we can eventually see Dilly/Faber 1.
Friday: The Ultimate Fighter 22 Finale
Frankie Edgar (53.8%) over Chad Mendes (46.2%)
Remember how Edgar vs. Faber had almost 21 minutes at distance, three minutes in the clinch, and only one minute on the ground? Might we be shaping up for that type of fight here? It seems so statistically – if Mendes doesn’t get an early knockout.
If their wrestling cancels out and most of the fight takes place at distance, they both land power head shots at a similar clip (10.0 P5M for Edgar to Mendes’ 9.3) but Edgar lands 2.3 more than his opponents while Mendes on net absorbs 0.4 more. Their defensive accuracy is pretty similar, but Edgar does a better job controlling his opponent’s volume.
Offensive knockdown power isn’t even close as Mendes’ knockdown rate and percent (see definitions below) are almost nine times higher than Edgar’s. As for defense, we all think of Edgar as susceptible to knockdowns early, but his overall knockdown defense rate is pretty damn good. He gets dropped at roughly half the rate and percent of Mendes and is below the featherweight average for both.
Tony Ferguson (54.9%) over Edson Barboza (45.1%)
Evan Dunham (62.4%) over Joe Lauzon (37.6%)
Ryan LaFlare (63.3%) over Mike Pierce (36.7%)
Thursday: UFC Fight Night: Namajunas vs. VanZant
Rose Namajunas and Sage Northcutt don’t have the minimum number of documented fights needed for a prediction, so the Paige and Sage bouts are quantitatively scratched from Thursday’s card. But hey, we’ve got the Lock of the Weekend with Aljamain Sterling to get us started. Don’t let me down, Funk Master.
Michael Chiesa (63.1%) over Jim Miller (36.9%)
Elias Theodorou (66.0%) over Thiago Santos (34.0%)
Tim Means (67.7%) over John Howard (32.3%)
Aljamain Sterling (92.8%) over Johnny Eduardo (7.2%)
Enjoy the glorious weekend!
Notes: Strike attempts are for an entire five minute round in each position (P5M) and are categorized as jab or power. A jab is just a non-power strike. Strikes are documented based on where they land or are targeted (head, body, legs), not the type that is thrown (punch, elbow, kick, knee). Visible damage rate is per five minutes the fighter is not on his back. It’s hard to bust up someone’s face while lying on your back. Damage percentage is per power head strike and distance head jab landed. Knockdown rate is per five minutes at distance or in the clinch off the cage. Knockdown percentage is per power head strike landed while standing. It’s really hard to knock someone down if they’re already on the ground. Clinch control is having the opponent pressed against the cage. Ground control is having top position. Submission attempts are per five minutes of ground control minus time spent in the opponent’s guard plus time spent with the opponent in guard. A bout closeness measure towards zero means a fighter is in blowouts (win or lose) and towards 100 means he is in very close fights.
Paul is Bloody Elbow’s analytics writer. All mistakes are his own and they’ve been known to happen sometimes. Follow him @MMAanalytics. Fight data provided by FightMetric.