UFC 157 – Betting the Stats

I’m not a huge fan of making fight predictions. Because I hate being wrong more than I like being right. But I sure do analyze a lot of data on fighter performance, and it is fun to try to spot the ingredients of an upset all within the context of larger fight trends in MMA. So let’s approach this as economists.

The betting lines are essentially the “market price” for a certain outcome. They are extremely valuable in sports, just as they are in business. But competing against experienced odds makers is generally a foolhardy exercise, because betting lines have proven to be very accurate throughout history, including in a volatile sport like MMA.

When it comes to gambling, the trick is to try to be less wrong than the market. Hopefully, also with enough margin to cover “the vig” that prevents us from breaking even by playing both sides of a given bet. The bookmakers make a living on volume, not on making bets themselves, so it’s the market that has to pay to play. In thermodynamics, the three fundamental laws can be summarized as “1. You can’t win. 2. You can’t break even. 3. You can’t get out of the game.” Gambling comes scarily close to this system, in that if you choose to enter the game, just breaking even becomes extremely difficult. And the longer you play, the more you will lose.

But it’s not impossible to win. Last week I nailed all my picks. But the reality is that everyone will be wrong if they try their hand at prediction often enough. Please keep in mind, this exercise is as much about me committing to certain conclusions and identifying potential opportunities against the betting lines as it is about trying to make someone (or myself) some cash gambling. In order to truly test your knowledge of something, you’ll have to take firm stance: hypothesize, test, confirm or reject, and then reevaluate. Without a clear and testable hypothesis, hindsight bias will surely blur the picture of how accurate our interpretations were. We’ll see an upset and shout “I knew that was going to happen!” But really? If you knew, why not take the easy underdog money? What really happened was that an upset potential was identified, but we were unwilling to pay the price to play. Without making the prediction or the bet and standing by it, there’s no way to really test if you’re right. That’s what making a bet represents; it’s the hypothesis that you believe a betting line is improperly valued, and you’re willing to put your money where your mind is.

So let’s look at the betting lines for the main card at UFC 157 this Saturday. What do they tell us? What do performance numbers and the basic Tale of the Tape tell us? Most importantly, where do these stories diverge?

 

The main event has Rousey as a -1200 favorite. I haven’t yet analyzed data for the women’s division…busted. But at that line, really, there’s nothing left to do but sit back and watch history be made. Overall, when betting lines get that far skewed, history has favored the underdogs. A line like this implies a 93% chance of victory for the favored fighter. Yet historically, these fights end up being more competitive than that. However, the sample size is also extreme in nature, so we’re still left with the same conclusion: sit back and watch.

Which brings us to the rest of the card. When fights are handicapped with a clear favorite (anything more than -200 for the favorite in my opinion), upsets occur roughly 30% of the time. And we have here four fights with clear favorites. The macro view tells us that it’s fair to expect that one of these fights will result in an upset. So if we take that approach, we can look at the matchup stats to see where the best opportunity lies.

 

Machida-Henderson: Machida a -225 favorite

These are the most skewed matchup performance stats on the main card, and yet the line is the closest. Machida holds advantages in literally every striking metric, and whole lot of ground metrics as well. The big problem is Henderson is 42 years old with poor striking defense, facing a guy with fast, accurate, and powerful hands. Remember that Fedor launched a Machida-esque combination that dropped Henderson, before a Hail Mary around the back punch gave Hendo the miraculous comeback knockout.

There’s been 18 knockdowns scored between them, plus five more that they’ve received. This is the light heavyweight division after all. Between the two, it’s Machida who has 11 knockdowns to Henderson’s seven. Sure, Hendo has the H-Bomb and a Puncher’s Chance. But Machida is far more elusive than Henderson’s prior opponents, and has the best striking defense on the entire card. The numbers suggest there’s good reason to believe Machida can become the first ever to finish Henderson by (T)KO, especially if Henderson comes forward and walks into Machida’s sweetspot. Hendo may want to get this to the ground to avoid “point fighting” with Machida, but the reality is that he simply cannot afford to stand and trade. Unfortunately, if the stats play out, it may mean we never get to Henderson face off against Jon Jones, which could actually be a better matchup for Dan.

The betting line should remain tempered due to the fact that these are Light Heavyweights, and memories of Henderson’s epic H-bombing of Bisping remind us of his power. Machida has been knocked out once against Rua, so it’s not impossible that his chin is now vulnerable. But based on these numbers, Machida makes a good stats-based pick with a reasonable return.

 

Faber-Menjivar: Faber a -400 favorite

Pretty steep odds belie a reasonable statistical matchup. Faber has a significant reach advantage, and better power to date. But Menjivar has pushed a higher pace overall, while Faber has tended to lag opponents in volume. Judges notoriously score rounds on activity, not effectiveness, and that means higher paced fighters have a better chance in the lower weight classes where a one-punch KO is less likely. But we also have to remember that Menjivar’s stats have come from a mix of lower tier opponents, and he lags Faber in most skill metrics.

So pace alone shouldn’t be enough to save him from a versatile fighter like Faber who can fight anywhere and advance through transitions. The line may be inflated a little due to Faber’s popularity, but he certainly has wrestling and superior grappling stats to control the fight on his own terms. And if Faber can use his reach and stiff jab to control the pace, he’ll negate Menjivar’s one true advantage.

The smaller fighters are much more likely to go the distance, and Menjivar is enough of a veteran to handle Faber for three rounds, even in defeat. Faber is a safe play for parlays, and if you’re trying to pick a Fight of the Night candidate, this matchup also has some good ingredients.

 

McGee-Neer: McGee a -230 favorite

Here’s one that more casual fans might be scratching their head on. Neer is old in “fight years” and is a 46-fight veteran doing his third serious tour in the UFC, but also coming off two brutal losses. McGee is a TUF winner and has a look befitting a man who trains at a place called “The Pit.” While his standing stats are mildly favorable, McGee has pushed the pace in his fights, staying active, and simply outworking opponents (despite getting screwed by judges in his last outing). McGee actually has the highest significant striking pace of any fighter on the main card. This fight could easily turn into a grinder, and I think that favors McGee.

The betting line isn’t steep, and that probably reflects the limit view we have of McGee’s fight game, and also the volatile nature of Josh Neer’s fight history. An upset wouldn’t surprise me, but I don’t see enough in the stat potential in Neer to warrant it. McGee has the tools to win this, but it might take all three rounds. As welterweights, the decision is a reasonable bet if you’re looking for a prop on going the distance, and McGee looks like a solid pick to win.

 

Koscheck-Lawler: Koscheck a -500 favorite

Another highly skewed betting line, and this time I may draw a different conclusion. Both guys have KO power in their hands, impressively powerful for welterweights. But I think Lawler’s lackluster record in Strikeforce is seriously misleading, and may be inflating Koscheck’s line. The fact is, Lawler’s skill stats in striking are slightly better than Koscheck’s in many ways, and Lawler is also a Southpaw who is five years younger than his opponent. This is Koscheck’s first fight on the wrong side of age 35, and he has been susceptible to flash knockdowns before. In fact, he’s been knocked down four different times, which is as many knockdowns as he’s scored. So if you fancy him a slugger, you have to also accept that he can be knocked out just as easily.

But this fight could (and should) hit the ground, and Lawler has lost rounds on his back before. Solid grapplers like Jacare Souza and Tim Kennedy beat him on the ground and negated his standup advantage. Koscheck has a solid wrestling pedigree and has a huge statistical advantage on the ground. So a smart Koscheck will avoid standing and trading and will get this down for ground and pound or even a submission.

That said, this extreme line opens up potential underdog plays on Lawler for the upset. If we’re looking for the ingredients of an upset, these are the best ones I see: dangerous hands and a striking advantage against an older fighter with knockout susceptibility.  We may see Koscheck play it safe and grind it out. But there are three rounds for Lawler to sprawl and brawl, and he only needs to land one clean shot for a huge upset. Avoid this one unless you’re comfortable the risk of a steep underdog. But from the stat sheet, if you’re looking for underdogs, Lawler makes a good bet on the main card.

 

So that’s a wrap. For the gamblers out there, remember that the favored fighters are favored for good reason. But this is cage fighting after all, so upsets can and do happen. Just ask King Mo, who was as a -1175 favorite until this happened. Just make sure you don’t pick your upsets based on hype, but rather real potential for victory. And never ever bet if you’re not willing to lose.

 

Note: the betting lines will change constantly, right up to fight time. I’ve taken straight up odds available at Bovada mid-week of fight week. In the future I may show 5Dimes odds. Fightnomics and the author bare no responsibility for bets placed on fights, financial or otherwise. The information presented here is for educational and entertainment purposes only. Never gamble if you can’t afford to lose.

Like Fightnomics on Facebook, or follow on Twitter @Fightnomics to hear when new research and blog posts are available. Raw data provided by FightMetric. 

 

Be Sociable, Share!
468 ad

2 Comments

  1. justin leong says:

    Great posting as usual. Its amazing the accuracy that can be attained from predicting fights from statistical analysis

  2. Hayk says:

    Very nice article and good job picking! You actually nailed all the picks :) I thought Koscheck – Lawler was going to be the upset, too.

Trackbacks/Pingbacks

  1. UFC on FuelTV Stann-Silva: Betting the Stats - Fightnomics | Fightnomics - [...] we’re going to play this game, please refer to my prior post on my philosophy of examining MMA betting …
  2. Homepage - ... [Trackback]... [...] Read More here: fightnomics.com/blog/ufc-157-betting-stats/ [...]...

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>