Media MMA Math: Does it Work? Examples and Lessons

[Martialarts123]

Hey guys, check out this video I made about MMA Math and its overarching lessons. Let me know what you think!

 

[KingBeef]

Can't neglect bro-science!!!

 

[NoBiasJustMMA]

No it doesn't work. The best way to analyse a fight is to compare similar fighters or how people did against common opponents around the same time or similar styles to the upcoming fight. MMA math doesn't work cause it's just A>B and B>C so A>C.

 

[EVIL WAR GOD]

 

[Megatronlee]

It works. But you gotta do algebra and not just addition n subtraction. If that makes any sense

 

[SpedDaddyV]

It doesn’t work, as someone else said it’s based on styles

Example:
Fighter A is an absolute stud striker
Fighter B is a shit striker but great wrestler
Fighter C has solid striking and good takedown defense

Fighter A can destroy fighter C, but that’s because fighter C will likely stand and trade

Fighter B can destroy fighter A by relentlessly taking them down, but can’t take down fighter C and gets outstruck by them

Trivial example, but still. And there are other variables as well obviously... cardio, ring rust, age, sea level, etc.

 

[TCE]

Styles make fights.

 

[ignorant MMA fan]

No it doesn't work. The best way to analyse a fight is to compare similar fighters or how people did against common opponents around the same time or similar styles to the upcoming fight. MMA math doesn't work cause it's just A>B and B>C so A>C.

I wish I knew the flaws of this theory back when my parents got my report cards.

 

[BEATDOWNS]

lol at people saying kos and struve

Fulton by far...

 

[Kvothe_the_kingkiller]

Didn't watch the video, but I hope you came across these statistics in your research:

http://www.fightmatrix.com/2009/01/01/if-a-beats-b-and-b-beats-c-will-a-beat-c/
If A beats B and B beats C, will A beat C?
I ran the analysis with all mixed martial arts bouts that we have on record (over 68,000).
So how often does A beat C?
In the end, A prevailed over C at a 71% success rate (2988 of 4210).
​

 

[Kvothe_the_kingkiller]

The most interesting thing about that 71% is how utterly uninteresting it is lol.

By that I mean, if it were 51% it would be interesting (e.g MMAth doesn't exist), and if it were 95% if would be interesting (MMAth is highly predictable).

Instead, 71% is probably about what most of us would guess; statistically true (on a macro level) but far from inevitable.

 

[britelights1124]

This is Sherdog on Valentine's Day.

 

[smart.feller]

I think the only example of MMA math being correct and undisputable is that Fighter A > Fighter B > Joseph "The Ho Bag" Bochenek, therefore Fighter A > Joseph Bochenek, because "The Ho Bag" sucked at fighting.

 

[Deleted member 220895]

Math works with absolutes, MMAth works not off absolutes but it works based off of results not variables.

Scientific method is very accurate because of variables. They change, so does the result.

When one fighter fights a different fighter, variables change, and so does the result.

With mmath it only relies on its hypothesis.

MMath is used to prove hypotheses that contain many different variables not brought into rational thought, not facts.

 

[Klaavical]

Cornering hundreds of fights I have to say;

You just never know

 

[sadoquera]

i was watching joey beltran in that bare knucles tournament....and i got the same conclusion as u... that dude is definlty the goat

 

[93confirmed]

Potatoes.

 

[WhuffieCollector]

Didn't watch the video, but I hope you came across these statistics in your research:

http://www.fightmatrix.com/2009/01/01/if-a-beats-b-and-b-beats-c-will-a-beat-c/
If A beats B and B beats C, will A beat C?
I ran the analysis with all mixed martial arts bouts that we have on record (over 68,000).
So how often does A beat C?
In the end, A prevailed over C at a 71% success rate (2988 of 4210).
​

Yes, but it is a deeply flawed question in the first place because all we know is that that A is 1-0 and C is 0-1. Without more information A is likely a better fighter than C.

What's missing is whether A and C have comparable backgrounds, experience, styles, etc.

If you repeat that experiment with only fighters who have comparable experience and strength of opponents, then use the common opponents to do the test, I doubt you'd get anything close to 71%.

 

[BisexualMMA]

All I know is Matt Hughes lost to Dennis Hallman, and that nobody dude lost to Robin Phoenix. So Matt Hughes vs that nobody dude is probably a tie.

 

[dissectingaorticaneurysm]

Cody Pfister > Crazy Horse > KJ Noons > Nick Diaz > BJ Penn > Matt Hughes > GSP > Jake Shields > Dan Henderson > Fedor