Friday, November 15, 2013
Jiu jitsu math - why it doesn't always work
I'm not much of a numbers person, but I remember a few things from high school algebra. In the math world, if a>b and b>c, then we also know that a>c - every time.
So in the jiu jitsu world, if player a beats player b, and player b beats player c, then logically, player a should also beat player c. Right? Right?! This is jiu jitsu math 101.
Only it doesn't always work. Now, if I had to place a bet and I had no other information to go on, I would still side with jiu jitsu math. In the above scenario, I would pick player a to beat player c. But there is a decent chance I'd be wrong. Among competitors of the same skill level, I would expect this math to fail about 25% of the time. That makes this a pretty lousy jiu jistu theorem.
Indeed, I've lost several matches that I expected to win based on jiu jitsu math. And every time, I have found myself dazed and surprised. It makes no sense! Just ask my algebra teacher, Mr. Scata - a>c! Except every now and then, c beats a.
So why does jiu jitsu math fail? There are a couple of reasons.
- Randomness and chance play a role in jiu jitsu matches. According to some smart math people at MIT, there are more possible outcomes to a game of chess than there are known atoms in the universe. So as a form of kinesthetic chess, a given jiu jitsu match can go about a gazillion different ways - at least when the 2 individuals are close in size and ability. Now, if I were to compete against Gabi Garcia 100 times, I would most definitely lose to her 100 times. But there are plenty of other people with whom I would expect to go 50/50, 60/40, 25/75, or some other split. Just because you beat someone once, doesn't mean you will beat them every time. Conversely, just because someone beats you once, doesn't mean you won't get her next time.
- People match up with each other differently. Some people do ok handling big, strong folks but have trouble against faster scrambly people. Some people do great when they can get on top, but get smashed on bottom. Depending of your style, success in a tournament can be dependent on the attributes and style of your opponent.
- Familiarity plays a role. I tend to drill and train with the same people a lot - so much so that we get to know each other's moves. Some folks who know me well know exactly what I am going for and can shut me down before I even start. And sometimes I can do the same to them. How you perform against folks who you are super familiar with cannot accurately predict how you will perform relative to each other when rolling with new folks.
What about you? Have you ever underestimated an opponent due to jiu jitsu math? Or have you ever defied the laws of jiu jistu math and defeated someone you weren't "supposed to?"