clock menu more-arrow no yes

Filed under:

UFC 162: A Look at Fighter Longevity in the Aftermath of Anderson Silva's Knockout

New, comments

Bloody Elbow's own Fighting Historian Patrick Wyman takes a broader look at the idea of fighter longevity and asks the question, when do fighters start to decline?


A guest post by BE reader and community member Patrick Wyman

Fighters decline as they age. This should be an uncontroversial statement, and it's been repeatedly backed up by in-depth analysis, as this piece from the incomparable Fightnomics firmly demonstrates. This statement, however obvious, leaves out the truly intriguing questions. At what point in their careers do fighters tend to decline? What are the factors that correlate to exceptional longevity or, alternatively, a rapid falloff from one's peak? Finally, what are the causal factors that contribute to this decline?

Several years ago, it was argued that fighters tend to decline markedly after the ninth year of their careers, measured from the date at which they started fighting. The piece wasn't perfect, and criticisms flew fast and furious; some were valid - flawed data sets and methodologies, for example - while others were not. More than anything, using winning percentage as the sole metric by which to measure peak and decline is exceptionally problematic. What about being on the wrong end of a bad decision - Sanchez-Kampmann comes to mind - or fighting up to one's talent level but being beaten by a fighter who's clearly better, a la Condit-GSP?

I'm unsatisfied with the current state of the issue. MMA and the large-scale analysis thereof is still in its infancy, especially compared to baseball, where these questions have been investigated in exceptional depth over the last several years. This piece is not intended to provide definitive answers to the question of career decline, but I hope to restart the conversation and provide a starting point for more in-depth investigations.

With that in mind, I've assembled a large database. Using FightMatrix's ranking system, I took the top 20 fighters in every division and examined the following variables:

1) length of career to date, rounded to the nearest quarter year

2) the age at which they had their first professional fight

3) their base style (BJJ, wrestling, Muay Thai, etc.), and the level of accomplishment in that base style

4) significant strikes absorbed per minute, as a way of getting a general idea of how much damage they've absorbed over the course of their career

5) total number of fights and fights per year

6) knockout losses

Examining the current top 20 at each weight allows us to see the broader context for fighters at an advanced stage in their careers. I then narrowed the sample down to all fighters who were at or near the nine-year point, and examined them individually. For many of them, there's no reason to believe that they're beginning to decline. Fabricio Werdum has been fighting for eleven years, for example, and doesn't seem to have dropped off noticeably; if anything, he's still getting better. This is true of a number of fighters, twenty to be precise: notables in this category include Lyoto Machida (10 years), Mike Pyle (13.75 years), and Gegard Mousasi (10.75 years).

This leaves us with a sample of twenty-five fighters who are currently ranked in the top 20 of their weight classes and who have noticeably dropped off from their peak form. To round out the sample, I added an additional ten fighters who were at one point ranked near the top of their weight class, but who have either retired or dropped out of the top 20, giving us a total of thirty-five.

The list is as follows: Frank Mir, Antonio Rodrigo Nogueira, Andrei Arlovski, Fedor Emelianenko, Tim Sylvia, Mirko Filipovic, Dan Henderson*, Antonio Rogerio Nogueira, Mauricio Rua, Rampage Jackson, Chael Sonnen, Wanderlei Silva, Randy Couture, Chuck Liddell, Tito Ortiz, Forrest Griffin, Anderson Silva, Vitor Belfort*, Yushin Okami, Rich Franklin, Martin Kampmann, Nick Diaz, Jake Shields, Robbie Lawler, Nate Marquardt, BJ Penn, Hayato Sakurai, Matt Hughes, Gilbert Melendez, Josh Thomson, Takanori Gomi, Clay Guida, Manny Gamburyan, Mike Brown, Kid Yamamoto, and Miguel Torres.

*An argument can be made that Henderson and Belfort are currently in their primes. I'll address the TRT issue at some length later in the piece.

Let's begin with the question of peak and decline. I've used relatively generous criteria to assess the beginning of a fighter's peak, starting the clock at the time they had their first win over a notable opponent. The question of decline is more difficult, and my approach has been organic and flexible rather than rigid. I've used the following criteria to deduce decline: first, a streak of losses, especially by knockout; second, taking an uncharacteristic amount of damage in multiple fights; and third, stylistic stagnation or a reversion to one-dimensionality.

To use Fedor as an example of the methodology, his losing streak didn't start until the tenth year of his career, but I placed the end of his peak in the ninth year. While he knocked out both Arlovski and Rogers before being submitted by Werdum, he took far more damage than was usual for him, and he abandoned some of the celebrated nuances of his game in favor of swinging for the knockout. I've followed precisely this approach for every individual in the database, and while parts of the method are obviously subjective, this study's estimates for peak and decline are internally consistent from fighter to fighter.

Using these criteria, then, what do we learn about the average time of a fighter's decline? The average for the sample of thirty-five fighters is 9.5 years, with a standard deviation of 1.5. 9 is the most frequently occurring time of decline, followed by 8 and 10. While the "Nine-Year Rule" referenced above is a useful shorthand, it sells short the potential peak longevity of a large number of fighters.

In the cases where fighters declined sooner than that nine-year mark, the reasons tend to be clear. Frank Mir and Forrest Griffin are both outliers in terms of taking damage, Kid Yamamoto suffered a devastating series of injuries to his knees and elbow, Miguel Torres is an extreme outlier in the number of fights he had (even discounting the multiple unsanctioned and unlisted fights) and the age at which he started fighting, and Mirko Filipovic had already had a long career in K-1 and as a successful amateur boxer before transitioning to MMA.

What happens when we examine only fighters with exceptional longevity, in the sense that they declined very late relative to their peers? When we look at them as a group, they display distinct commonalities. First, they take very little damage, less than two strikes per minute (1.79). This is much lower than the number for our thirty-five On a fight-to-fight basis, this number doesn't tell us very much; sometimes guys get hit quite a bit in a bad stylistic matchup, and then come out and dominate their next fight while taking very little damage (Faber-Aldo, Faber-Mizugaki). In the aggregate, however, the damage adds up. Moreover, a guy who gets hit quite a bit in his fights is also likely to take more damage in innumerable sparring sessions over the course of his career.

Second, their backgrounds are almost exclusively in wrestling and BJJ. This point is related to the first: one tends to get hit less from top position or from an active, offensive guard than striking in the free-movement phase. While the fighters in this group might be good-to-excellent strikers (e.g. Rampage, Marquardt, and GSP), they can always impose a top-control gameplan against dangerous opponents. Again, on a fight-to-fight basis this is not especially informative. Over the course of a career, though, this capacity allows fighters to take dangerous opponents out of their comfort zones, maintain long winning streaks, and thus take less damage.

Third, they tend to have started fighting at a relatively young age, an average of 21. This aspect deserves in-depth investigation, but as a preliminary hypothesis I'd say it has something to do with the age at which men reach their athletic prime. While to my knowledge nobody has investigated this in MMA specifically, in other sports one's physical peak tends to come between the ages of 27 and 33. If we take a hypothetical fighter who had his first fight at 21, then he'd reach the nine-year point at age 30, still well within his athletic prime. Alistair Overeem (possible decline at 32, though I'm not prepared to say that definitively), Nate Marquardt (decline at age 32), Rampage (decline at age 32), and GSP (still going at 32) all fit this pattern nicely. When looking at fighters who fall off their peak at a more average point, however, the ages at decline are seemingly random: Andrei Arlovski clocks in at age 29, Rick Franklin at 34, Tito Ortiz at 31, and Chuck Liddell at 37.

Fourth, they tend to have better chins than their peers. This might be a product of taking less damage, but there doesn't seem to be a direct correlation between chin and damage taken. Guys with historically iron chins - Dan Henderson, Nick Diaz, Eddie Wineland, Carlos Condit, Josh Burkman, and Clay Guida, to name a few - don't necessarily take less damage, and in some cases (Diaz and Wineland) are actually extreme outliers in the other direction relative to their peers. It's likely that "chin", in the sense of avoiding a knockout, has its basis in environmentally- and/or genetically-determined brain and bone structure. Regardless, fighters with extreme longevity tend to have suffered fewer knockout losses and knockdowns than their peers.

Fifth, and there's no way of beating around the bush on this one, illegal anabolic steroids and TRT help fighters extend their careers. When we look at guys with extreme longevity, a disproportionate number of them have either failed drug tests or are currently on TRT. This list includes Hendo, Sonnen, Belfort, Marquardt, Josh Barnett, Overeem, and Rampage. Whether this has to do with aiding recovery from injury, allowing for extra training time, or simply a long-term advantage in strength and musculature - studies have suggested that the benefits of steroid use remain long after the user's finished with them - the pattern is clear.

Having examined longevity, what can we say about the causal factors that lie behind decline? I've already mentioned several variables, but let's run through them in more detail.

1) Damage. While there are obvious outliers, guys who get hit a great deal don't tend to have exceptional longevity. Forrest, Mir, and Chris Leben all come to mind.

2) Physical decline. Injuries pile up over time and recovery becomes more difficult. This has a direct effect on athleticism and speed, and it has an indirect effect in that it limits training time. Guys spend less time doing strength and conditioning and more rehabbing, while skills might atrophy or decline from lack of time in the gym. As an anecdote, I'd add that I was recently talking to my coach about one of his training partners, a top-10 guy in the UFC who just passed the nine-year mark in his career. He mentioned that his training for his last fight had been limited, that he'd almost pulled out, and that he wasn't able to impose parts of his gameplan, all because of a relatively minor injury that would've barely slowed him down several years ago.

3) The game starts to pass you by a little bit. Combinations of skills that were dominant at one time eventually get countered and surpassed. Limited training time due to injuries or more difficult recovery might accelerate this process, since they inhibit the acquisition of new skills and promote the atrophy of older skills.

4) While this is harder to quantify or define, and thus must remain a hypothesis, it's possible that the quality of competition starts to rise around the 8-10 year mark. At this point, a fighter's had enough time to develop a name and profile after several peak years. Recent examples of this might include Nick Diaz or Chris Leben, the latter of whom went from fighting Jake Rosholt on an undercard to getting battered by Mark Munoz in a main event in a two-year span. It's conceivable that the greater damage taken and the demands of training for higher-quality opponents accelerate the process of decline, and the fact that older fighters fight less might be seen as an attempt to minimize the negative effects of fighting better competition.

I make no attempt to argue that all fighters decline at x, y, or z points in their careers. It's only possible to speak in terms of general patterns in a sample; while the weight of evidence points to the majority of fighters declining between the 8 and 10 year marks, there are distinct outliers in both directions.

Given last Saturday's events, it seems fitting to end this piece with a brief discussion of Anderson Silva, the greatest outlier we'll ever see. He's a striker first, and has no desire to spend much time in top position. In my sample of fifteen fighters with exceptional longevity, he's the only one who didn't come to MMA from wrestling or some form of submission grappling. Nevertheless, he's taken very little damage, absorbing only 1.43 strikes per minute, more than two standard deviations lower than the average of his peers; this is one way of demonstrating the sheer brilliance of Anderson's striking over the course of his career. He started fighting in MMA relatively late in life (25, though he had two fights at age 22), and was dominant well beyond what we'd consider the likely end of his physical peak.

Part of Anderson's mystique lay in the fact that he seemed like an exception to a great many rules, perhaps especially the deleterious effects of aging. But explosiveness diminishes, reflexes slow down, and chins don't improve like fine wine. An inch more of distance between Anderson and Weidman, and maybe that hook looks as ridiculous as so many others over the course of Anderson's career. An inch more of backwards sway, and maybe that hook passes harmlessly by Anderson's face. Five years and hundreds or thousands of blows to the head ago, and maybe that punch doesn't put Anderson to sleep. But time waits for no man, not even Anderson Silva.