Just recently, somebody questioned in one of our Krav Maga classes the number of fights which end up on the ground. I think it is still a relevant question to explore, even though the answer has been known for some time.

It immediately brings to mind the “ninety percent of all fights end up on the ground” myth.

Before we get to the answer, some background on the origin of this myth. First off, where does it come from?

The myth of 90 percent of fights end up on the ground

The origin of this claim appears to have originated from Rorion Gracie, who started claiming this back in the late 1980’s. A Google search quickly quotes Rorion Gracie stating quotes similar to this back from 1989. This was before UFC 1 and the Gracies weren’t as well known as they are now. Neither was Brazilian Jiu Jitsu.

Rorion seems to have been primarily looking to promote his family’s style when he claimed that ninety percent of all fights go to the ground. The angle he played seems to have been something along the lines of:

• 90 percent of all fights eventually end up on the ground according to an LAPD study.
• Gracie Jiu Jitsu specializes in ground fighting and we have a proven track record.
• Therefore, Gracie Jiu Jitsu is the best system for fighting.

A few years after that original article, UFC 1 came along, Royce Gracie won and that brought ground grappling and MMA in the limelight. The Gracies fared well in the twenty years since then, to put it mildly.

But where did Rorion Gracie get that number?

He apparently got it from a study done by Sergeant Greg Dossey.

Dossey’s study was specific to “Use of Force” issues surrounding the Los Angeles Police Department. Sgt. Dossey found that when LAPD Officers were arresting a suspect the arrest sequence ended up with one or both parties wrestling on the ground 62% of the time.

Let’s think about that for a moment. The number wasn’t 90%, it was 62%. And the context was a LEO arrest, not a “fight.”

A LEO arrest is all about capturing, controlling, and restraining a suspect who, very often, doesn’t want to “fight” per se but just wants to escape. Any “fighting” the suspect does is very often not aggression aimed at the LEO simply for the sake of fighting or injuring the LEO but rather to try to escape restraint.

The goal was restraint. This is seldom accomplished without engaging in some form of grappling.

A non-LEO is far less interested in or likely to be required to restrain anyone, particularly in the event of some sort of “street fight.”

To sum up, the study cited by Mr. Gracie is simply not applicable to “Joe Average” because the participants (on either side) are not representative of the participants of a “street fight,” “bar-room brawl,” or even a random criminal ambush. They’re the classic “apples to oranges” comparison. In other words, fictitious.

During those two decades, that 90% quote got tossed around all over the place until it eventually became accepted as a fact by the average martial artist, in particular by the BJJ or MMA practitioners. 

It even came to the point that if you questioned this “fact”, you were labeled a stupid traditionalist who didn’t know what real fighting was. Unfortunately, there are some problems with that quote.

Why is this a myth?

The first problem is that the logic doesn’t hold up.

One of the key problems we all have nowadays is that we are bombarded in the media with “studies” and “statistics” that supposedly prove all sorts of things. At least, that’s what’s implied when you read those kinds of articles. The issue is this:

Correlation does not imply causation

This is a classic mistake scientists, statisticians, and doctors are warned about and one they are strongly encouraged to avoid. Just because one factor is correlated to another, doesn’t mean it causes it.  Or to put it in relevant terms:

Just because 90% of the fights go to the ground and Gracie jiu jitsu focuses on fighting on the ground, doesn’t mean it’s the best system for fighting.

The logic just doesn’t hold up, even though it seems to at first glance.

The burden of proof lies with whoever poses a theory and Rorion used the LAPD study for this. Unfortunately, that study proves no such thing.

The Study THAT STARTED IT ALL

The often-misquoted statistic from a review of LAPD use of force incidents where:

“Nearly two thirds of the 1988 altercations (62%) ended with the officer and subject on the ground with the officer applying a joint lock and handcuffing the subject.”

Given the fact that the study cited is far too limited in its data source to be applicable to the vast majority of Joe Average “fights.” Frankly, there are statistics to be found to support either side of any argument if you look hard enough. But even beyond this – where is the mythic statistic that 90% of fights end up on the ground?

Again, unfortunately, nowhere.

There are simply no statistics kept in any Federal, State, or even Local collection which would track this metric. It’s not part of a check-list on any official report and most “fights” go unreported anyway. 

There have been any number of attempts by amateurs to create useful statistics about civilian “fights.”

The most interesting one is probably The Violence Project by James LaFond. He “interviewed” over 1,500 participants of violent encounters.

On the subject of grappling, he found 38% of the encounters involved or resulted in grappling. This figure includes clinches (both parties standing), throws (one party down), and floor fights (both parties down). This seems to be a solid number.

However, again, there are data collection and source issues which could bias or even invalidate the statistic entirely.

While not an entirely useless number it suffers from reliability issues which no researcher actually doing a scientific study would accept, most notably his sample set and normalization of it.

Taking into account statistical bias and margin of error, the actual percentage could easily be somewhat less or potentially greater.

There are many interested amateurs on reddit who sourced their “statistic” from random “fights” posted on YouTube or from their own experience, with a rather strong conclusion for such a rickety sample source.

No offence to these ‘researchers’, but the selection of the sample used is non-representative to anyone but them.

Another set of interesting sources which is often cited is in support of “most fights end up on the ground” is MMA, particularly UFC 1, and the ever popular Gracie Challenge videos. But their sample is still non-representative of the average “street fight.”

These are arranged challenge matches between two trained martial artists. Those two facts alone, “arranged challenge match” and “trained martial artist,” immediately invalidates the statistical findings from everyday “street fights” because, going by published statistics, people with any sort of martial arts training represent a tiny fraction of the general population.

That is not to imply that these challenge matches do not offer important insight in “style versus style” or “fight strategy versus strategy” debates, but only that the people fighting these fights are in no way representative of Joe Average walking down the street and getting jumped, mugged, or in a bar fight.

And that’s the problem with trying to use any of the currently available sample sources: they are non-representative. In other words, the claim of “most,” let alone 90%, is unproven.

WHAT Rorion Gracie GOT WRONG

The second problem is that Rorion misinterpreted the LAPD study he is quoting. Maybe he didn’t read it completely. Maybe he misunderstood it. Maybe he did it on purpose. Who knows? Either way, it doesn’t matter as he kept on using this study as proof of his theory about the superiority of his family style.

Unfortunately, the study doesn’t support his claim at all.

If you have invested a lot of time and energy in BJJ or MMA, you might have a knee-jerk reaction of “Bullshit!” while reading and I totally understand why you would feel that way.  

What about now?

I think there are way more people nowadays who think going to the ground in the street is a good idea than ten years ago, when MMA really shot to fame.

We now have an entire generation that grew up watching MMA as the dominant combat sport.

Sixty, seventy years ago, the dominant combat sport was boxing. What young men thought of as “fighting” was often primarily inspired by boxing.

Throughout the years, Asian martial arts got added to the mix and changed that view. In the last two decades, MMA (and BJJ) have been front and center in all the media when it comes to fighting. As a result, a lot of kids grew up thinking that this is how you should fight.

They will not learn that there are kinds of violence and situations where BJJ is not the answer but the one thing you should absolutely avoid.

To them, their kind of fighting is the real thing (which it is, all their fights prove this) but they fail to see that their experiences don’t encompass the whole spectrum of violence. But as most people don’t come into contact with those other kinds of violence, they don’t know any better and think their experience applies across the board.

Where does my theory come from?

From close to a decade of teaching those kids, teaching them both in my classes, seeing ad nauseam what’s written on the internet and what’s “taught” on Youtube and in other videos. From teaching security guards, bouncers and bodyguards. From teaching trained fighters who got into a dust up with multiple attackers. From teaching correctional services and law enforcement. But also from looking at as much CCTV or other footage of actual fights as possible: in the last decade, I’ve seen a huge rise in MMA techniques in street fights.

Before the UFC got big, I hardly ever saw that. Now I see it more and more.

Another factor I think is important is that there have never been more MMA gyms and schools. More and more (mostly) young guys take up the sport as their first introduction to martial arts and fighting. Just like several decades ago, they used to take up boxing if they wanted to learn to fight. So more and more men grow up with MMA as their primary source of instruction for all their fighting.

This is in no way a scientific theory. This is me sharing my experiences, giving you my opinion based on what I have seen, read and heard in the last twelve years. Which doesn’t mean I’m right, I could be way off. But I believe there may be some validity to the theory, which is why I still train in sports grappling and ground techniques and why I also recommend to our students to do the same.

This is why we offer Sambo classes at SGS Krav Maga.

I firmly believe you should at least have a solid base in ground grappling as taught in modern MMA. Not to fight like them, but to understand how they fight and know what to do about it. Just like I believe you have to spend some time boxing and doing Muay Thai:

• You haven’t been punched until you’ve been hit by a boxer.
• You haven’t been kicked until you’ve been kicked by a Muay Thai practitioner.

Those arts excel at these techniques, just like BJJ excels at ground grappling. However, that doesn’t mean their techniques apply across the board.

Try training on a slippery surface with techniques from boxing, muay thai and BJJ.

Conclusion

Know what your goal is and use the appropriate tool to achieve it. Meaning:

• If your goal is competition, then you train differently than for self-defence.
• If your goal is to choke people or tap them out on the mat, then you train differently than if you want to learn how to get up right away after you end up on the ground in the street.

Determine your goal first and only then pick your tools.

Regardless of how good other tools are for other goals, if they don’t help you for that particular job, they’re not a good match for you then and there. I believe the same is true for all fighting arts: determine first and foremost why you train and then pick the system(s) that helps you achieve that goal.

Unfortunately for those who like to quote the (ahem) “statistic” that 90% of all fights go to the ground, it is a complete myth. It is unproven at best simply because no one has been able to collect a representative sample to draw data from. Its original source is fictitious in that it was not “all fights” but rather LAPD arrests. Finally, the number is possibly false simply because no one knows what the “real” number is to say nothing if it is close to the oft-quoted 90%.

But what does all this mean for the modern martial artist or individual interested in Self-Defence? Absolutely nothing. No, you read that right, I wrote, “nothing.”

For years now it has been accepted wisdom that a well rounded martial artist must include some amount of grappling and ground-fighting in his training. This has not changed one little bit. Regardless of whether or not the actual number of fights that go to the ground is 20%, 38%, 62%, 90% or some other figure, it remains an important and significant number.

Let’s assume that Mr. LaFond is right in his 38% figure. Nearly 40% is a dramatic and significant percentage and only a fool would ignore a nearly 1 in 2 odds of a fight going to grappling.

For myself, I will continue to practice wrestling, clinching, Sambo, and other forms of grappling and, in the strongest terms possible, I recommend others do so too.

While you go through the selection process for that system, beware the marketing hype and unproven theories (including all of mine). Think things through and do your own research so you can at the very least be comfortable with your choice.

After all, it’s your butt on the line.