Strategic MMA betting analysis

[Sacrelicious]

So I'd like to open the debate here to anyone who wishes to discuss their MMA betting strategy. Hopefully we can devise some sort of a think tank here.

Anyway, first thing to consider, is this "edge", what is an edge, how to we determine it?

Well I have several ideas.

1) Hog Analysis

2) Execution of kicking strategy

3) Looking at a fighters record

Now if anyone is good at math, please calculate an equation for this.


Also, we have BET SIZING. Personally I have a rather large bankroll, so I like my small underdog fliers to be around the 5-7 unit range, a nice moderate play for is is around 20 units.

[NunyaBidness]

The problem is all 3 are such different criteria, but we should be able to come up with a workable power rating.

I suggest something like (H*Hg²) + (Kl - Kr) + (W -L³)

Hl = Hog Length, Hg = girth (obviously much more important)
KL = Kicks landed Kr = Kicks received
W= Wins L=Losses

Obviously, the least # of losses is the most important factor, as everyone knows undefeated fighters rarely lose (as evidenced by the fact that so far they have averaged losing 0% of their fights).

Unfortunately this model is not perfect, as division size affects our hog size without affecting the other features.

So, we need to remove this from our model, I'm open to suggestions.

[Crassus]

The problem is all 3 are such different criteria, but we should be able to come up with a workable power rating.

I suggest something like (H*Hg²) + (Kl - Kr) + (W -L³)

Hl = Hog Length, Hg = girth (obviously much more important)
KL = Kicks landed Kr = Kicks received
W= Wins L=Losses

Obviously, the least # of losses is the most important factor, as everyone knows undefeated fighters rarely lose (as evidenced by the fact that so far they have averaged losing 0% of their fights).

Unfortunately this model is not perfect, as division size affects our hog size without affecting the other features.

So, we need to remove this from our model, I'm open to suggestions.

What if we modified the Hl and Hg to proportional hog length and girth given the fighters height and/or weight?

[Sacrelicious]

Stool samples, is there any way we can get some of these?

[NunyaBidness]

What if we modified the Hl and Hg to proportional hog length and girth given the fighters height and/or weight?

Right, but how do we do this without getting a survey of hog size related to size in the general population. Obviously hog size varies in proportion to total size, but what is the formula? And what would the shape of the curve look like? Furthermore, what would the first derivative of this curve be describing exactly? I feel like we're close to something here.

[MD]

Right, but how do we do this without getting a survey of hog size related to size in the general population. Obviously hog size varies in proportion to total size, but what is the formula? And what would the shape of the curve look like? Furthermore, what would the first derivative of this curve be describing exactly? I feel like we're close to something here.

800 dudes multiplied by mean jerk time divided by four dicks at a time, or [T = (800 x mJT) / 4]...

Wait, what were we talking about?

[NunyaBidness]

[Crassus]

Right, but how do we do this without getting a survey of hog size related to size in the general population. Obviously hog size varies in proportion to total size, but what is the formula? And what would the shape of the curve look like? Furthermore, what would the first derivative of this curve be describing exactly? I feel like we're close to something here.

General population might be misleading, MMA fighters as a population might be better? You could make some basic assumptions at first and make it so the ideal fighter has a certain ideal ratio of length and girth per inch/lb of the fighter. We could find the four ideal hogs of fighters at HW, MW, LW and FW (flyweight) to come up with that ratio and then tweak as more information comes in.

[MD]

General population might be misleading, MMA fighters as a population might be better? You could make some basic assumptions at first and make it so the ideal fighter has a certain ideal ratio of length and girth per inch/lb of the fighter. We could find the four ideal hogs of fighters at HW, MW, LW and FW (flyweight) to come up with that ratio and then tweak as more information comes in.

Like the Ape index, but for dicks?

By God, it's just crazy enough to work.

[JIBBBY]

Obviously, the least # of losses is the most important factor, as everyone knows undefeated fighters rarely lose (as evidenced by the fact that so far they have averaged losing 0% of their fights).

I'm gonna try and expand on this..

Depends on how many fights that undefeated fighter has fought and what level of competition he has fought to date.. I don't put much weight into an undefeated fighter with less then 10 fights that has been fighting in sub par MMA organizations.. Once reaching a big MMA venue and then facing stiffer and more experienced competition those O's can and do go in a hurry..

For example just last night with WSOF, we had 2 undefeated fighters on the main card with around 10 fights on their record.. One lost and one won, both in different fights... 50% win ratio..

I see this happen alot with padded undefeated records with around 10 fights anyways.. With 10 wins and being undefeated fighters get noticed and then begin to get really tested.. Alot more to MMA betting then just numbers and percentages..

Obviously this applies to Mens MMA and not Womens MMA..



A betting strategy thread.. I'm so lost... Forgot those Algebra and Calculus classes I took in College... Will try to comprehend this though...

[MD]

I'm gonna try and expand on this..

Depends on how many fights that undefeated fighter has fought and what level of competition he has fought to date.. I don't put much weight into an undefeated fighter with less then 10 fights that has been fighting in sub par MMA organizations.. Once reaching a big MMA venue and then facing stiffer and more experienced competition those O's can and do go in a hurry..

For example just last night with WSOF, we had 2 undefeated fighters on the main card with just under 10 fights.. One lost and one won, both in different fights... 50% win ratio..

I see this happen alot with padded undefeated records with under 10 fights anyways.. Alot more to MMA betting then just numbers and percentages..

Obviously this applies to Mens MMA and not Womens MMA..



A betting strategy thread.. I'm so lost... Forgot those Algebra and Calculus classes I took in College... Will try to comprehend this though...

Hm, makes sense.

Nunya, is there a way to normalize L (losses) when W (wins) is less than 10 and UFC (the organization they fight for) is a sub-par organization? If W = 9, L = 0, and UFC = Cage Warriors, we could run into some serious trouble. According to JIBBBY's data, fighters in this spot only have a 50% win ratio.

I feel like JIBBBY might have just unearthed a flaw that could have busted our entire algorithm.

Nice work, Jibster. It's going to take finer minds than mine and Nunya's to pull this all together.

[JIBBBY]

^^ Can always count on you MD to troll my mindful posts.. Right on Q as usual...

[MD]

^^ Can always count on you MD to troll my mindful posts.. Right on Q as usual...

I'm not trolling you, JIBBBY. Believe it or not, you have plenty to contribute to this thread.

[MD]

Can we include the variables H+ and H-, with H being hands, and +- being the willingness to throw them? To account for paradoxes in our system, we'll create a new value, the Newell, as a measure of how many hands a fighter has. We can safely assume that all fighters have a Newell value of 2 and just leave it out of the formula until we come across a fighter with a Newell score != 2.

[Grabaka]

We need to include what Sacko and Bigday are betting into the equation. The Sk and the Bd variables

[NunyaBidness]

Some good ideas presented so far. Keep em coming.

I came to this forum to see if anyone had built an MMA model, here 5 years later we're finally putting it together.

[Sacrelicious]

The problem is all 3 are such different criteria, but we should be able to come up with a workable power rating.

I suggest something like (H*Hg²) + (Kl - Kr) + (W -L³)

Hl = Hog Length, Hg = girth (obviously much more important)
KL = Kicks landed Kr = Kicks received
W= Wins L=Losses

Obviously, the least # of losses is the most important factor, as everyone knows undefeated fighters rarely lose (as evidenced by the fact that so far they have averaged losing 0% of their fights).

Unfortunately this model is not perfect, as division size affects our hog size without affecting the other features.

So, we need to remove this from our model, I'm open to suggestions.

We need something in this equation to represent a fighters proclivity to kick as opposed to punch. I am convinced that kicking is really a big advantage, legs are much longer than arms.

[MD]

Can positive and negative edges nullify each other?

Let's say that my ROI is 10%, and BIGDAY's is -5%. If BIGDAY bets on Weidman, how much should I hedge on Machida?

[NunyaBidness]

We need something in this equation to represent a fighters proclivity to kick as opposed to punch. I am convinced that kicking is really a big advantage, legs are much longer than arms.

That's already in the model as we are measuring kick differential, and ignoring punches altogether. As Cecil Peoples would say, punches don't win fights.

[NunyaBidness]

Can positive and negative edges nullify each other?

Let's say that my ROI is 10%, and BIGDAY's is -5%. If BIGDAY bets on Weidman, how much should I hedge on Machida?

It depends on who Sac is on.

[JIBBBY]

Picking who wins the MMA fights has so many variables to consider.. Compu box stats in previous fights just aren't gonna cut it, and either are fighter records and or other stats by themselves..

Trying to figure out a basic formula is gonna be a tough call guys as it really can vary from fighter to fighter I believe. This is mixed martial arts and simply said - there are so many ways, techniques and methods to win these fights.. This makes it hard to formulate... Who's strong at what and who is weak at what, is it punching or kicking, is it wrestling or Jits submssions, standing or take down defense, is it elbows and knees, Clinch and grinders, etc... etc...

In simple terms what is a fighter really good and bad at.. Who has that fighter been training with and at.. Apply that to the future style match up and arrive at a conclusion..

Style match ups then and only then should then be the largest variable to consider in my opinion for picking....

This simply isn't like other sports.. I don't think you can blue print it like others... My take anyways..

[NunyaBidness]

Wait? So you're saying that the measurements of someones dick is suboptimal for handicapping fights? Go on. . .

[MD]

Picking who wins the MMA fights has so many variables to consider.. Compu box stats in previous fights just aren't gonna cut it, and either are fighter records and or other stats by themselves..

Trying to figure out a basic formula is gonna be a tough call guys as it really can vary from fighter to fighter I believe. This is mixed martial arts and simply said - there are so many ways, techniques and methods to win these fights.. This makes it hard to formulate... Who's strong at what and who is weak at what, is it punching or kicking, is it wrestling or Jits submssions, standing or take down defense, is it elbows and knees, Clinch and grinders, etc... etc...

In simple terms what is a fighter really good and bad at.. Who has that fighter been training with and at.. Apply that to the future style match up and arrive at a conclusion..

Style match ups then and only then should then be the largest variable to consider in my opinion for picking....

This simply isn't like other sports.. I don't think you can blue print it like others... My take anyways..

It doesn't need to be right 100% of the time, we just need to pick the girthiest fighter often enough to beat the market.

[MD]

Wait? So you're saying that the measurements of someones dick is suboptimal for handicapping fights? Go on. . .

This is the guy who said "they both have hands and can throw them"; if you're willing to ignore years of research in the field of penile analysis because of what some nutjob on a forum says about "clinch" and "take down defense", how are you any different from those crazy assholes who think that Obama's a Muslim?

There are a lot of people in this world with no idea what they're talking about. Don't let them get in the way of your vision.

[JIBBBY]

It doesn't need to be right 100% of the time, we just need to pick the girthiest fighter often enough to beat the market.

I know... I'd say look into and find the formula for kick boxing if there is one, tweak that formula to fit the MMA scheme and maybe ya might be onto something.....

Hey, but let me know when you experts and genius's arrive at the MMA formula that can beat the market.. I'd love to see that...

I find capping MLB to be much easier then MMA just for the record and in MLB all you have is stats and formulas... MMA is probably the hardest sport to bet on...

[mirinquads]

We're going to need a clause for Estevan Payan - As we all know, any Estevan Payan fight is an automatic all in play - thus 100% bet. Oh yes, we need a Ron Paul clause as well - Any -1000 favorite is an automatic all in play so 100% play. Doesn't matter if they have hands or not.
Oh yeah, and the 'Parlays equals Magic clause'. For instance [((100/110) +1) * ((100/100) +1) * ((125/100) +1) * ((100/210) +1)^Motherfucking-Magic] for 4 matchups with odds -110, +100, +125 +210.
This is going to get complicated, Jibby, care to weigh in on this?

[MD]

I find capping MLB to be much easier then MMA just for the record and in MLB all you have is stats and formulas... MMA is probably the hardest sport to bet on...

Exactly.

[MD]

We're going to need a clause for Estevan Payan - As we all know, any Estevan Payan fight is an automatic all in play - thus 100% bet. Oh yes, we need a Ron Paul clause as well - Any -1000 favorite is an automatic all in play so 100% play. Doesn't matter if they have hands or not.
Oh yeah, and the 'Parlays equals Magic clause'. For instance [((100/110) +1) * ((100/100) +1) * ((125/100) +1) * ((100/210) +1)^Motherfucking-Magic] for 4 matchups with odds -110, +100, +125 +210.
This is going to get complicated, Jibby, care to weigh in on this?

What you're looking for is the rule that states that when chest hair is greater than or equal to head hair in length, disregard F, where F is the entire formula, and then bet on the Armenian. It's called Killin'em's law.

[Sacrelicious]

Bobbing with long hair must be included in this model, as well as ineffective aggression, shocked that nobody has mentioned this yet.

Also, good call Nunya on ignoring the punches, seriously, they are an ineffective tool, but we do need a multiplier for tactical kicking in the (Kl - Kr) portion of this model, based on some hypothetical edge that we ought to, as a team, be able to calculate.

[Bumdeal]

I've been away most the weekend and have been missing out on this masterpiece thread. This is why I hate leaving the house.

[Sacrelicious]

(Kl - Kr)*Ks

Where Ks is kicking strategy, derived from spreadsheets and years of statistical analysis.

[Bad Tattoo]

[JIBBBY]

I've just popped a brain cell trying to figure out what is legit info and which is just silly humor on this thread.. Tricky thread to decipher I say...

So I leave ya all with this little gem and formula.......