FightMetric

November Insanity Tournament Pick 'em Bracket

noreply@blogger.com (FightMetric) — Tue, 03 Nov 2009 15:15:57 PST

November is a crazy month for MMA, with seven major league events within a three week period. To best appreciate this rare bounty, we've created a game similar to what you might play during March Madness. Of course, we've had to tweak the scoring system so it works in a situation where winners don't continue on in the tournament. It took a little creativity, but after testing nearly a dozen different scoring options, we've settled on one that should make for a great gameplay experience.

Head on over to Bloody Elbow to get all the details, download your bracket, calculate your score, and even enter to win a prize for the highest point total.

Check us out on MMA Live

noreply@blogger.com (FightMetric) — Fri, 30 Oct 2009 08:05:44 PDT

Just as we saw after Forrest Griffin vs. Rampage Jackson, the demand for stats is greatest in the wake of a controversial decision. The traffic to the report for Machida vs. Shogun broke our site records by several times. We're also happy to see the stats used in a number of places, including on MMA Live over at ESPN. Check out the video about 13 minutes in for a shout-out and some air time for the FightMetric numbers.

Machida vs. Shogun Stats

noreply@blogger.com (FightMetric) — Sat, 24 Oct 2009 23:59:51 PDT

The stats are now posted for the fight between Lyoto Machida and Mauricio "Shogun" Rua. Stay tuned for some analysis of the numbers and a list of a whole host of things that happened to Machida for the first time in his career.

Meet the 2009-10 FightMetric Fellows

noreply@blogger.com (FightMetric) — Mon, 19 Oct 2009 10:04:01 PDT

We were thrilled to receive such a positive response to the FightMetric Fellowship Program. There was an abundance of qualified applicants, but in the end, we could only choose a few. We are proud to introduce the people chosen for the 2009-10 Fellowship Cohort:

John Henry

John is an Assistant Professor of Mathematics whose research has focused on extreme value theory with applications in environmental/ecological and actuarial science. Dr. Henry received his PhD in statistics from Oregon State University in 2008. John is also an MMA fighter, winning the light heavyweight title in his last amateur fight with Brass Knuckles Cage Fighting Promotions. He has recently turned pro, and is currently training with the HIT Squad in Granite City, IL.

Reed Kuhn

Reed first got hooked on MMA as a teenager in the early 90's, but rediscovered the sport in grad school while working on statistical analysis of the NFL. Since 2005, he has not missed a single televised MMA event, has traveled as far as Germany to attend live UFC events, and credits The Ultimate Fighter show for making his wife a fan (and thus, partially accepting of his MMA habit). He is a strategy management consultant by trade, with a BS in Physics, a Master's in Systems Engineering, and an MBA from Duke University specializing in Operations and Decision Sciences.

Leland Roling

Leland is currently a Staff Writer for BloodyElbow.com, one of the most popular blogs on the Internet covering the sport of mixed martial arts. He is also a full-time professional web developer who builds data-driven websites that make the Internet work for you. In his free time, he is President of the Melvin Manhoef Fan Club, watches way too much MMA, and passionately awaits the Chicago Cubs World Series title, which, quite frankly, will probably kill him.

Brian Reich

Brian received a PhD in biostatistics from the University of Minnesota in 2005, and is currently an assistant professor of statistics at North Carolina State University. His research interests include modeling spatial and spatio-temporal data, environmental applications, Bayesian
methods, and nonparametric statistics. For more information, please view his website.

Henderson vs. Cerrone

noreply@blogger.com (FightMetric) — Sun, 11 Oct 2009 21:17:21 PDT

We've now posted the in-depth report for the fight between Ben Henderson and Donald Cerrone. Reading the reactions and live scoring online, it seems most everyone agrees that Henderson won Rounds 2 and 3, while Cerrone won Rounds 4 and 5. The Effectiveness Scores do reflect those conclusions. The question is about Round 1. What do you value more, Henderson's takedowns and ground strikes or Cerrone's tight submission attempts?

Anyone who's been coming here for a while should be able to guess what the Effectiveness Scores have to say about Round 1. But is the answer satisfying?

Fight Night 19 Stats...and a 10-8 Reminder

noreply@blogger.com (FightMetric) — Thu, 17 Sep 2009 09:04:25 PDT

We've now posted the stats from last night's Fight Night card:

Condit vs. Ellenberger

Quarry vs. Credeur

Maynard vs. Huerta

Also, just a quick reminder of how FightMetric calls 10-8 rounds using the Effectiveness Scores: A round is only called 10-8 when a fighter scores higher than 100 AND his score is more than six times that of his opponent. In the case of Round 1 of Condit vs. Ellenberger, only one of those conditions was met. Ellenberger scored 188 points, but his score was 2.5 times Condit's score. While Condit did get knocked down twice in the round, he did enough in his moments of lucidity to avoid the 10-8.

FightMetric Fun Facts: Volume III

noreply@blogger.com (FightMetric) — Wed, 02 Sep 2009 12:25:46 PDT

This installment of F³ has some fun numbers from our completed Reach Data Collection Project. Now that we have every recorded reach measurement in UFC history, we can share some interesting pieces of information.

Fact: The average reach measurement in the UFC is 73 inches.

Fact: The longest recorded reach in UFC history belongs to Tim Sylvia at 86 inches. Sylvia's reach was also reported to be 84 inches one time, but was measured five separate times at 80 inches.

Fact: The shortest recorded reach in UFC history is 64 inches, reported for both Andy Wang and Jason Reinhardt.

Fact: The fighter with the longest, consistently measured reach is Jon Jones. He has been measured three times, all at 84.5 inches.

Fact: The largest discrepancy between a fighter's shortest and longest measured reach is 11.5 inches. Jorge Gurgel was reported to have a 78.5 inch reach at UFC 63. At UFC 73, Gurgel's reach was recorded as 67 inches. His remaining five measurements were all recorded as 69 inches.

Fact: The most consistently measured fighter in UFC history is Jeremy Horn. Horn was measured six times and was reported at 74 inches every time.

Announcing the FightMetric Fellowship Program

noreply@blogger.com (FightMetric) — Wed, 19 Aug 2009 08:39:05 PDT

It has always been the stated goal of the FightMetric project to advance the understanding of Mixed Martial Arts and share findings with the community at-large. Unfortunately, we haven't been able to devote as much time to advanced analysis as we'd hoped, due to the inevitable obligations of running the business. Over time, we've gotten numerous offers of help from many of you who were interested in running their own analysis. Unfortunately, we didn't have a good way to include those that wanted to get involved. Now we're opening up the opportunity for advancement to you.

To further this effort, we are very proud to announce the launch of the FightMetric Fellowship Program. This is a chance for people who want to get involved doing the in-depth research on MMA data and publish their findings. Those accepted into the program will have unfettered access to our entire database of fight statistics. There's enough data in there to keep any analyst busy for quite a long time and it's a wide open field in which to make a contribution. No one has yet shown us what the data truly says about MMA. This is an open call to be one of the people who can make that happen.

We're very excited about this program and believe it has tremendous potential. There are plenty of very talented people out there who could do amazing things with our data. All we need to make it a success is you.

We plan to make this an annual program, accepting new cohorts of fellows for each academic year. For the inaugural cohort, the deadline to apply is September 10, 2009. All the details and application materials can be found on the site's FightMetric Fellowship page.

Thank you for your continued support of FightMetric. We're as interested to see what you can come up with as anyone.

Fun Facts from Silva-Griffin

noreply@blogger.com (FightMetric) — Sun, 09 Aug 2009 14:54:30 PDT

* Silva landed 13 of 25 total strikes, for an accuracy percentage of 52%. That is his lowest hit rate since coming to the UFC and the third lowest of his career. He landed 51% against Ryo Chonan and 50% in his first pro fight against Luiz Azeredo. Silva has still never had a fight in which he landed less than half his attempted strikes.

* If you want to consider Silva's head movement, consider that Griffin landed 1 of 35 strikes thrown to Silva's head, or 3%. The average is 29%.

* With his three knockdowns against Griffin, Silva becomes the only fighter in UFC history with two fights containing three knockdowns. He also did it in the second fight against Rich Franklin.

* The three knockdowns also moves Silva into second place for most knockdowns in UFC history, behind only Chuck Liddell.

DREAM is Up-to-Date

noreply@blogger.com (FightMetric) — Wed, 05 Aug 2009 14:21:35 PDT

It was a much bigger deal when we completed the historical database of the entire modern era of the UFC, but we're still proud to announce that every fight in DREAM history is now complete and included in the FightMetric database. So why choose DREAM as the next initiative after UFC? A few good reasons:

1. After completing the UFC, it was important to complement this information with data from a Japanese organization. This will allow us to do some study on the differences between American and Japanese MMA.

2. DREAM has run enough events to get a significant enough data set.

3. They haven't run so many events that the data collection project would take more than a month.

4. They are the heir-apparent to Pride. While the fans would love to see full Pride stats, it's more important to keep up with current fights and fighters.

5. Who doesn't love Shinya Aoki?

So now that the data collection is done, we have to ask: For which DREAM fights do you want to see reports?

FightMetric Fun Facts: Volume II

noreply@blogger.com (FightMetric) — Thu, 30 Jul 2009 09:25:31 PDT

In the first installment of F³ we talked about fighters in The 100 Strike Club; those that have landed at least 100 strikes in a single fight. This installment will deal with takedown defense.

Fact: There is only one fighter in modern UFC history to have a perfect 100% takedown defense rate (meaning he was never taken down) with at least 10 attempts against him.

Fact: There is only one fighter in modern UFC history to have a not-so-perfect 0% takedown defense rate (meaning he was taken down every time) with at least 10 attempts against him.

The perfect score belongs to Corey Hill, whose opponents were 0-for-10 on takedown attempts.

The not-so-perfect score belongs to Elvis Sinosic, whose opponents were a perfect 15-for-15 on takedown attempts.

GSP's Five-Rounders

noreply@blogger.com (FightMetric) — Sun, 12 Jul 2009 14:28:17 PDT

We have a wealth of new material posted today, including the stats for the fight between Georges St. Pierre and Thiago Alves, the Yoshihiro Akiyama vs. Alan Belcher fight, and updated career reports for St. Pierre, Alves, Dan Henderson, Michael Bispin g, and Frank Mir. Make sure to check out the Mir report to see the utter dominations put in by Brock Lesnar.

After watching St. Pierre control Alves for five rounds it's interesting to take a look at how this performance stacks up against his other five-rounder, the fight against Jon Fitch.

There are a few things that stick out:

1. Alves did a much better job of getting back to his feet. GSP needed 10 takedowns against Alves to control the fight, compared to just five (plus a reversal) against Fitch.

2. Fitch did a much better job of keeping GSP in his guard. Against Alves, GSP landed 10 takedowns and passed to half-guard 10 times.

3. This was a much more reserved GSP, focusing more on control than ground striking. St. Pierre landed only 13 big head strikes on the ground against Alves. He landed 43 of those against Fitch.

4. GSP took a lot fewer shots, but he landed fewer too. In all, Thiago Alves landed 30 significant strikes, while Fitch landed 38. But at the same time, GSP landed only 60 significant strikes against Alves compared to an astounding 129 significant strikes against Fitch. So he exactly doubled Alves' total, while more than tripling Fitch's total.

5. While no doubt a terrific, dominating performance, the fight against Alves doesn't measure up to the performance GSP put in against Fitch. The total Effectiveness Score of 814 he tallied against Fitch is still the single-highest score in our database.

FightMetric Fun Facts: Volume I

noreply@blogger.com (FightMetric) — Thu, 09 Jul 2009 07:42:26 PDT

Now that we have completed our historical database of every fight in the modern era of the UFC, we'd like to share some of the more interesting and unique data points that we've found. We're going to run a regular blog feature called FightMetric Fun Facts, or F³ for short. We hope you enjoy this unique perspective on the sport of MMA.

Fact: Only seven fighters have landed 100 significant strikes (or more) in a modern UFC fight. Two of them did it in five rounds:

Georges St. Pierre vs. Jon Fitch
Rich Franklin vs. David Loiseau

Four others did it in three rounds:

BJ Penn vs. Sean Sherk
Forrest Griffin vs. Hector Ramirez (more than 50 were leg kicks)
Chris Lytle vs. Kevin Burns
Martin Kampmann vs. Thales Leites

But the current record for most significant strikes landed in a single fight belongs to Tim Sylvia in his fight against Wesley "Cabbage" Correira. Sylvia landed more strikes on Cabbage in under seven minutes than any fighter has in any fight of any length in modern UFC history.

Modern UFC Complete, What Next?

noreply@blogger.com (FightMetric) — Wed, 08 Jul 2009 15:25:01 PDT

If things have seemed quiet around here the last few months, it's because we have been hard at work behind the scenes. One of the major projects underway was completing part of our data collection project, encompassing every fight in modern UFC history. We are proud to announce that every UFC fight from UFC 28 until today has now been scored and entered into our database. We are working with strategic partners to deploy all of this data to the MMA community. While we're not able to show all the numbers to you right away, over the course of the coming days and weeks, we'll be sharing some interesting facts with you that could only be ascertained using this rich, comprehensive data set.

Our data collection project is ongoing, with no end-point in mind. The modern era of the UFC is clearly the most useful and important data to have, but there is much more data to be collected. What we'd like to know from you is what data you think should be next. Please share your thoughts with us using the poll below.

Check Us Out in the UFC 100 Magazine

noreply@blogger.com (FightMetric) — Tue, 30 Jun 2009 12:36:42 PDT

It's a little like playing Where's Waldo, but if you look closely, you can find FightMetric in the premier issue of new UFC magazine.

Can't find it? Check out the feature on The 20 Greatest Fights Ever, starting on page 108. The yellow band at the bottom of the subsequent pages contains some interesting factoids from the UFC's history. And if you follow it all the way until the end on page 116...there we are. How could you miss it?

Why Won't Judges Call a Round 10-10?

noreply@blogger.com (FightMetric) — Thu, 25 Jun 2009 14:38:30 PDT

With the judging at the TUF 9 Finale raising so much ire this week, much of the focus has been on two fights that were not broadcast on Spike. We have now posted the stats for two controversial decisions on the undercard, the fight between Melvin Guillard and Gleison Tibau and the fight between Brad Blackburn and Edgar Garcia.

Both fights produced scores with the same issue: Two of the rounds were too close to call, with the third round producing a clear winner. In both cases, the system produced a score of 30-29. Of course, to do this, you have to be willing to call a 10-10 round. This is no problem for an objective stat system but seemingly impossible for a human judge. It wasn't always that way. Back at UFC 33, the first UFC event held in Las Vegas, there were three fights that had judges call at least one 10-10 round. Only one event since has had as many as two fights with a 10-10 round. The judges at UFC 33 are the same exact ones working shows today. What changed?

What Good is Top Position...

noreply@blogger.com (FightMetric) — Sun, 21 Jun 2009 13:18:28 PDT

...when you're getting beaten by the guy on the bottom? Round two of last night's fight between Diego Sanchez and Clay Guida was an object lesson in why top position is not always better. While Guida did maintain top control, he got outstruck 16-3 in power shots to the head, thanks to a bevy of elbows coming from Sanchez on the bottom. Joe Rogan made the comment that those might have been the most effective elbows from bottom that we'd ever seen, but we've seen those elbows used to pretty good effect before. Anderson Silva vs. Travis Lutter and Kenny Florian vs. Joe Lauzon are some other good examples of the devastating effect of this under-utilized weapon. They very well might be the most effective strike that most fighters will never use.

An impressive showing by Sanchez is made more impressive by the advances he has made in putting offense into his guard. What a difference this was from the guard he displayed against Jon Fitch, where Sanchez landed basically no strikes of consequence on the ground. In the Fitch fight, Sanchez concentrated almost exclusively on submissions. In the fight against Guida, Sanchez threw strikes where available and still managed to latch-on a few good submission attempts.

The scores from the fight show a victory for Sanchez 30-26, though that could very easily be 29-27, as Glenn Trowbridge scored it. Round 1 was a 10-8 round, using our objective criteria (score over 100 and more than 6x opponents' score) and Round 2 goes to Sanchez thanks to his offensive guard. Round 3 is basically a toss-up, which the system gives to Sanchez on damage, but could easily go Guida's way instead.

Also, check out the stats from Joe Stevenson vs. Nate Diaz and from Chris Lytle vs. Kevin Burns.

Rich Franklin = Nailbiter?

noreply@blogger.com (FightMetric) — Mon, 15 Jun 2009 19:39:36 PDT

Historically, Rich Franklin is not a fighter you associated with razor-thin decisions. For the better part of his career, Franklin either won big or lost big (thanks to Anderson Silva and Lyoto Machida). The exception to this rule was Franklin's fight against Yushin Okami two years ago. This was a very important fight for him, because a win would get him another crack at the middleweight belt, while a loss would set him back severely. Franklin fought a cautious fight and capitalized on Okami's slow start, just squeaking by for a decision.

Franklin's win over Wanderlei Silva on Saturday night marked his second consecutive close decision. The stats for the fight show a clear victory in Round 1 for Franklin, but then two close rounds that could have gone either way. This, following a coin-flip kind of decision against Dan Henderson, that could have easily been called a draw. Is this Franklin being cautious with his career, facing stiffer competition, or is this kind of fight one that we can expect consistenly from the new Rich Franklin?

Also, check out the stats from Dan Hardy vs. Marcus Davis to see if Davis has a case when he says he didn't lose, and look at the numbers from Cain Velasquez vs. Cheick Kongo to see just how many of those little punches to the head Velasquez landed.

UPDATE: By popular demand, we've added the stats for the fight between Spencer Fisher and Caol Uno.

Is Lyoto Machida the Best Fighter in the UFC?

noreply@blogger.com (FightMetric) — Sun, 24 May 2009 19:30:26 PDT

The answer to the above question is highly subjective, however, there are a few things that we can look at statistically to make some personal conclusions. We've now posted Lyoto's performance stats from last night (along with Hughes vs. Serra), but here were the facts going into the fight last night:

Machida has the lowest SApM (strikes absorbed per minute) of any fighter in UFC history by a very large margin. Coming into the fight last night, Machida's SApM was 0.58. After the fight, that dropped to 0.55. The next lowest SApM belongs to Anderson Silva, but it's not even close. Silva's SApM is 0.73, even after 25 minutes of not getting hit by Thales Leites. Just as a reminder, Fedor's SApM is 0.53.

Machida is the second-most accurate striker in the UFC (for fighters with at least 300 strike attempts), with a hit rate of 65%. The only fighter better is Silva.

Coming into the fight, Machida had the second-highest takedown success rate (for fighters with at least 10 attempts) at 82%. The only fighter better is Gray Maynard.

Machida's takedown defense rate is 84%, second best (among fighters with 25 takedowns attempted against them), trailing only Georges St. Pierre.

Machida has scored a knockdown on all but one of his UFC opponents (can you name the one he didn't?). After last night's thrashing, he's scored nine knockdowns in seven fights, a rate that matches Chuck Liddell at his peak.

He is now only the third undefeated fighter to hold a belt in the modern UFC era (can you name the other two?).

Again, there's no objectively correct answer to the question in the title. But these are some fun numbers to play with while making your own decision.

Seeking Volunteers for Beta-Test

noreply@blogger.com (FightMetric) — Sun, 24 May 2009 16:46:30 PDT

UPDATE: Thank you to those who volunteered. We will hopefully have more opportunities soon for others who wish to help.

We're looking for a few people to help test out a new feature. This will only take 10 minutes of your time and should be a lot of fun. If you can help us out, just send an email to info@fightmetric.com and we'll get you the particulars.

Worst Title Fight Ever?

noreply@blogger.com (FightMetric) — Sun, 19 Apr 2009 15:36:27 PDT

We've got the full stats up for the title fight between Anderson Silva and Thales Leites, but check out a post we did at Bloody Elbow for an answer to the question: Was this the worst title fight in UFC history? There is a suprising amount of competition for the (dis)honor.

Win, Lose, but Never Draw

noreply@blogger.com (FightMetric) — Thu, 02 Apr 2009 07:27:44 PDT

It's hard to advocate for draws. They're usually the worst resolution to a contest because no one feels like anything was settled. It's not fair for someone to work so hard and have "nothing" to show for it. But what's more unfair is to lose when you don't deserve to. Last night's main event between Martin Kampmann and Carlos Condit was a fight so evenly-matched that neither fighter deserved to lose. Looking at the in-depth report on the fight, we see a first for a match posted here: Both the ten-point must system and the overall score call this fight a draw. Each round was exceedingly close, with both Round 2 and the overall score falling within our four-point margin for error.

A draw is not a No Contest. In a No Contest, it's functionally as if the fight never happened. A draw means a hard-fought effort that was equal to the other fighter's. In a sport filled with disappointing matches where guys just fight not to lose, it's sad that one of two people really fighting to win has to lose because judges won't consider the third way.

Site Update

noreply@blogger.com (FightMetric) — Wed, 25 Mar 2009 12:07:18 PDT

There are a lot of very exciting things happening behind-the-scenes here, both short-term and long-term. We will hopefully be able to share some of this news with you pretty soon, but posting will be light for the immediate future. Thanks to everyone who sent requests for fight reports. Rest assured your requests are not denied, they're merely delayed.

Rampage-Jardine and Requests

noreply@blogger.com (FightMetric) — Mon, 09 Mar 2009 10:37:19 PDT

The stats are up for the main event fight between Quinton Jackson and Keith Jardine, but that's not all there is to this post. The previous post asked what kinds of reports people were interested in, if not the most timely information. The comments received were much-appreciated, and the most frequently requested thing was to focus more on the most controversial bouts, even if the fight or fighters were no long relevant. This is interesting because it's exactly how we started the FightMetric project, putting out reports on only the most hotly-contested decisions in MMA history.

One commenter rightly pointed out that our focus seems to be entirely on UFC matches, at the expense of other organizations. That would be correct. We are currently engaged in wrapping-up Phase II of the FightMetric project, which involves compiling data on every fight in modern UFC history. Until that's finished (and it'll be sooner than you might expect), we can't divert our attention away from that goal. I don't think it would surprise anyone that the commercial appeal of UFC data is far greater than the appeal for data from all other organizations combined.

So our more specific question is: What UFC fights would you like to see stats for? Take a look at the current offerings and then leave your requests in a comment. Anything besides Rutten-Randleman?

Again, apologies for this, we will get to all of the others in due time, but for now, consider this request to be UFC only.

What Do You Want to See?

noreply@blogger.com (FightMetric) — Mon, 02 Mar 2009 14:16:57 PST

After a much-needed vacation, we're back and have posted some new stuff leading up to UFC 96. Check out the report on the fight between Keith Jardine and Brandon Vera and a career* report for Keith Jardine.

This brings up an important question for us. For the past few months, we've focused coverage on the fighters in the top few matches of the upcoming UFC event. That's worked out great, but it means that we're running out of brand new material to feature. In general, the same few fighters make appearences at the top of UFC cards and we've already shown you what there is to see on them. Taking UFC 97 as an example, we've pretty much covered Anderson Silva, Chuck Liddell, and Shogun Rua.

We'd like to ask what you'd rather see: info on fighters deeper into the upcoming card or info on events long past with fighters that may no longer be timely? Leave us a comment and we'll try to accomodate.

*The report contains info on all fights for which there is video readily available