UGCTT_sakurabas ear -I like the sound of this.
No you don't, because you're Sakuraba's ear, and you can't hear it.
No you don't, because you're Sakuraba's ear, and you can't hear it.
Spent 4 years do stats at college.... NEVER AGAIN
Bearded Collie Herds Orca -MdGeist -
Statistics aren't going to predict who's going to be judging a fight.
Judges and their errors add to error variance, or unpredictability, but they do not make prediction impossible. They simply constrain the predictability of the system, and probably not as much as you'd think. Bad decisions stick out in your mind because you're a human, but most judging is pretty good.
I barely got by in statistics with a C for a reason.
But I can tell you one thing: picking winners is alot easier than going through the trouble of putting together statistical models.
MdGeist -Bearded Collie Herds Orca -MdGeist -
Statistics aren't going to predict who's going to be judging a fight.
Judges and their errors add to error variance, or unpredictability, but they do not make prediction impossible. They simply constrain the predictability of the system, and probably not as much as you'd think. Bad decisions stick out in your mind because you're a human, but most judging is pretty good.
I barely got by in statistics with a C for a reason.
But I can tell you one thing: picking winners is alot easier than going through the trouble of putting together statistical models.
OK, then "C" you later! ;)
MdGeist -Bearded Collie Herds Orca -MdGeist -Statistics aren't going to predict who's going to be judging a fight.
Judges and their errors add to error variance, or unpredictability, but they do not make prediction impossible. They simply constrain the predictability of the system, and probably not as much as you'd think. Bad decisions stick out in your mind because you're a human, but most judging is pretty good.I barely got by in statistics with a C for a reason.
But I can tell you one thing: picking winners is alot easier than going through the trouble of putting together statistical models.
Like how Rashad was weighting down my Maia-Silva parlay... Crazy shit happens sometimes, this idea sounds interesting though. 
Bearded Collie Herds Orca - In 2012, Nate Silver and his team were able to create a sophisticated statistical model able to predict the presidential election incredibly accurately.
My roommate and I would like to do something similar with MMA fighting. Like a presidential election, an MMA fight is a contest between two opponents. Unlike team sports, elections and MMA matches are highly suitable targets for Bayesian statistical modelling, which is the method Mr. Silver used. Bayesian modelling is discussed here: http://en.wikipedia.org/wiki/Bayesian_inference; I'm trying to avoid TL;DNR.
While admittedly not as skilled as Mr. Silver, my roommate and I are graduate students in psychology with advanced training in statistical analysis. His expertise is in neural network modelling in MATLAB, and his formal work is with machine learning. Mine is in frequentist and Bayesian inferential statistics and particularly logistic regression. We've both taken several graduate and undergraduate classes and I've been TA for graduate statistics. I'm fluent in SPSS and R statistical packages, and I've also coauthored several scientific papers in which my primary task was to run the statistics and create results sections. This isn't to brag, but rather to establish that I'm no schmuck with this stuff.
We're looking for several knowledgeable (about MMA) and reliable folks to help us build a database of fighter statistics that we could use as data for our model. This would include collecting statistics for your favorite fighters and importing them into a shared, premade spreadsheet (on Google Drive). You don't need to know how to do statistical analysis, you just need to have some free time and be comfortable entering data. We'd make the skeleton of the database, and you'd just add data into the cells. Then we'd use our mad skillz to make that data our bitch and tell us who's gonna win fights.
After testing several models against each other, we'd like to use the best model to predict fights and bet on them. Even if we were write only 70% of the time, we'd make out like bandits. I bet we'd do we better than that, and if you'd made a contribution to the database we'd share the models predictions for your own use.
I'd be happy to discuss details. Supercalo's not invited...he'd enter in fake data and watch the world burn.
so, you're like asking for helping in building a betting model for free?
BshMstr -Bearded Collie Herds Orca - In 2012, Nate Silver and his team were able to create a sophisticated statistical model able to predict the presidential election incredibly accurately.
My roommate and I would like to do something similar with MMA fighting. Like a presidential election, an MMA fight is a contest between two opponents. Unlike team sports, elections and MMA matches are highly suitable targets for Bayesian statistical modelling, which is the method Mr. Silver used. Bayesian modelling is discussed here: http://en.wikipedia.org/wiki/Bayesian_inference; I'm trying to avoid TL;DNR.
While admittedly not as skilled as Mr. Silver, my roommate and I are graduate students in psychology with advanced training in statistical analysis. His expertise is in neural network modelling in MATLAB, and his formal work is with machine learning. Mine is in frequentist and Bayesian inferential statistics and particularly logistic regression. We've both taken several graduate and undergraduate classes and I've been TA for graduate statistics. I'm fluent in SPSS and R statistical packages, and I've also coauthored several scientific papers in which my primary task was to run the statistics and create results sections. This isn't to brag, but rather to establish that I'm no schmuck with this stuff.
We're looking for several knowledgeable (about MMA) and reliable folks to help us build a database of fighter statistics that we could use as data for our model. This would include collecting statistics for your favorite fighters and importing them into a shared, premade spreadsheet (on Google Drive). You don't need to know how to do statistical analysis, you just need to have some free time and be comfortable entering data. We'd make the skeleton of the database, and you'd just add data into the cells. Then we'd use our mad skillz to make that data our bitch and tell us who's gonna win fights.
After testing several models against each other, we'd like to use the best model to predict fights and bet on them. Even if we were write only 70% of the time, we'd make out like bandits. I bet we'd do we better than that, and if you'd made a contribution to the database we'd share the models predictions for your own use.
I'd be happy to discuss details. Supercalo's not invited...he'd enter in fake data and watch the world burn.
so, you're like asking for helping in building a betting model for free?
Nope, I'm asking for folks to do the grunt work data entry so that I can develop an accurate prediction model and share it's secrets with them. Reciprocal Altruism.
Just looking at the most recent event: I don't see how it would be possible to predict with statistical models that Maia would grind out Fitch to a decision, much less win at all based upon Fitch's record of excellent sub. defense.
The data would more than likely suggest that Fitch would win, just based upon the fact that Fitch had 17 fights at WW in the UFC compared to Maia's 2 fights and has spent an overwhelming amount of time controlling guys in top position, probably more time spent in a dominant position during his entire career than any other fighter in the UFC.
How can you predict winners with statistical models when fights are heavily determined by the motivation and psychology of the individual fighter at a moment in time when there are so many other variables that are effecting their mindset? Which is something that played very heavily in the outcome of the Evans/Lil Nog fight and is something you can't measure.
If you can give me some direction I would be happy to lend a hand. Very interested to see how accurate your results turn out to be.
It's modeling FYI
Bearded Collie Herds Orca -BshMstr -Bearded Collie Herds Orca - In 2012, Nate Silver and his team were able to create a sophisticated statistical model able to predict the presidential election incredibly accurately.
My roommate and I would like to do something similar with MMA fighting. Like a presidential election, an MMA fight is a contest between two opponents. Unlike team sports, elections and MMA matches are highly suitable targets for Bayesian statistical modelling, which is the method Mr. Silver used. Bayesian modelling is discussed here: http://en.wikipedia.org/wiki/Bayesian_inference; I'm trying to avoid TL;DNR.
While admittedly not as skilled as Mr. Silver, my roommate and I are graduate students in psychology with advanced training in statistical analysis. His expertise is in neural network modelling in MATLAB, and his formal work is with machine learning. Mine is in frequentist and Bayesian inferential statistics and particularly logistic regression. We've both taken several graduate and undergraduate classes and I've been TA for graduate statistics. I'm fluent in SPSS and R statistical packages, and I've also coauthored several scientific papers in which my primary task was to run the statistics and create results sections. This isn't to brag, but rather to establish that I'm no schmuck with this stuff.
We're looking for several knowledgeable (about MMA) and reliable folks to help us build a database of fighter statistics that we could use as data for our model. This would include collecting statistics for your favorite fighters and importing them into a shared, premade spreadsheet (on Google Drive). You don't need to know how to do statistical analysis, you just need to have some free time and be comfortable entering data. We'd make the skeleton of the database, and you'd just add data into the cells. Then we'd use our mad skillz to make that data our bitch and tell us who's gonna win fights.
After testing several models against each other, we'd like to use the best model to predict fights and bet on them. Even if we were write only 70% of the time, we'd make out like bandits. I bet we'd do we better than that, and if you'd made a contribution to the database we'd share the models predictions for your own use.
I'd be happy to discuss details. Supercalo's not invited...he'd enter in fake data and watch the world burn.
so, you're like asking for helping in building a betting model for free?
Nope, I'm asking for folks to do the grunt work data entry so that I can develop an accurate prediction model and share it's secrets with them. Reciprocal Altruism.
seriously, the data is pretty easy to find, especially for someone who claims to have an advanced degree in the subject...
i'm pretty sure this is a bunch of BS.
MdGeist -
Just looking at the most recent event: I don't see how it would be possible to predict with statistical models that Maia would grind out Fitch to a decision, much less win at all based upon Fitch's record of excellent sub. defense.
The data would more than likely suggest that Fitch would win, just based upon the fact that Fitch had 17 fights at WW in the UFC compared to Maia's 2 fights and has spent an overwhelming amount of time controlling guys in top position, probably more time spent in a dominant position during his entire career than any other fighter in the UFC.
How can you predict winners with statistical models when fights are heavily determined by the motivation and psychology of the individual fighter at a moment in time when there are so many other variables that are effecting their mindset? Which is something that played very heavily in the outcome of the Evans/Lil Nog fight and is something you can't measure.
You're intuitive interpretation of data might lead you to believe these things, but statistical inference works and that's why science works. MMA prediction is not the hardest problem a science geek has ever set out to solve.
Goin to bed...if anyone wants to help or just hear more, PM me.
BshMstr -Bearded Collie Herds Orca -BshMstr -Bearded Collie Herds Orca - In 2012, Nate Silver and his team were able to create a sophisticated statistical model able to predict the presidential election incredibly accurately.
My roommate and I would like to do something similar with MMA fighting. Like a presidential election, an MMA fight is a contest between two opponents. Unlike team sports, elections and MMA matches are highly suitable targets for Bayesian statistical modelling, which is the method Mr. Silver used. Bayesian modelling is discussed here: http://en.wikipedia.org/wiki/Bayesian_inference; I'm trying to avoid TL;DNR.
While admittedly not as skilled as Mr. Silver, my roommate and I are graduate students in psychology with advanced training in statistical analysis. His expertise is in neural network modelling in MATLAB, and his formal work is with machine learning. Mine is in frequentist and Bayesian inferential statistics and particularly logistic regression. We've both taken several graduate and undergraduate classes and I've been TA for graduate statistics. I'm fluent in SPSS and R statistical packages, and I've also coauthored several scientific papers in which my primary task was to run the statistics and create results sections. This isn't to brag, but rather to establish that I'm no schmuck with this stuff.
We're looking for several knowledgeable (about MMA) and reliable folks to help us build a database of fighter statistics that we could use as data for our model. This would include collecting statistics for your favorite fighters and importing them into a shared, premade spreadsheet (on Google Drive). You don't need to know how to do statistical analysis, you just need to have some free time and be comfortable entering data. We'd make the skeleton of the database, and you'd just add data into the cells. Then we'd use our mad skillz to make that data our bitch and tell us who's gonna win fights.
After testing several models against each other, we'd like to use the best model to predict fights and bet on them. Even if we were write only 70% of the time, we'd make out like bandits. I bet we'd do we better than that, and if you'd made a contribution to the database we'd share the models predictions for your own use.
I'd be happy to discuss details. Supercalo's not invited...he'd enter in fake data and watch the world burn.
so, you're like asking for helping in building a betting model for free?
Nope, I'm asking for folks to do the grunt work data entry so that I can develop an accurate prediction model and share it's secrets with them. Reciprocal Altruism.
seriously, the data is pretty easy to find, especially for someone who claims to have an advanced degree in the subject...
i'm pretty sure this is a bunch of BS.
Lol, I could find the data myself, sure, but I'm busy with my graduate work. If you don't believe me that's fine, don't help.
MT11 -It's modeling FYI
correct! I'm on a spelling roll tonight, but I'm tired.
Shinsplint - If you can give me some direction I would be happy to lend a hand. Very interested to see how accurate your results turn out to be.
Cool, PM your email address.
Bearded Collie Herds Orca -MdGeist -
Just looking at the most recent event: I don't see how it would be possible to predict with statistical models that Maia would grind out Fitch to a decision, much less win at all based upon Fitch's record of excellent sub. defense.
The data would more than likely suggest that Fitch would win, just based upon the fact that Fitch had 17 fights at WW in the UFC compared to Maia's 2 fights and has spent an overwhelming amount of time controlling guys in top position, probably more time spent in a dominant position during his entire career than any other fighter in the UFC.
How can you predict winners with statistical models when fights are heavily determined by the motivation and psychology of the individual fighter at a moment in time when there are so many other variables that are effecting their mindset? Which is something that played very heavily in the outcome of the Evans/Lil Nog fight and is something you can't measure.
You're intuitive interpretation of data might lead you to believe these things, but statistical inference works and that's why science works. MMA prediction is not the hardest problem a science geek has ever set out to solve.
Goin to bed...if anyone wants to help or just hear more, PM me.
Statistical inference would have never predicted that big underdogs like Bigfoot, Lil Nog, or Maia would have won at UFC 156.
Based on the data that was available at the time, you would have been lucky to just get one of your picks rights if you were just relying on your system alone without using intuition.
Outside of team sports, statistics will not predict winners any better than a knewledgable fan.
So is the idea to assign a number to each of the variables such as "wrestling ability" and "striking ability," and so forth?
The problems I foresee with that would be inconsistency from one "grunt worker" to the next, incorrect assessments by the grunts, and your own inability to identify key factors which determine fight outcomes.
If you can't already beat the bookies by manually analyzing the fights individually, I don't think there is any chance this will help you.
If you get to the stage where you are actually prepared to use this system to place real world bets, I hope you will be bold enough to publish your results, whether good or bad.
You will never be able to do a statistic model for a sport like MMA trust me wayyy to many variables. If your really looking to do something like this with the purpose for betting do a sport like basketball or baseball where there are many games and numbers can be crunched to give an accurate representation of what will and wont work. There are plenty of models for team sports out there. Check out some betting forums to get ideas.
B_Goetz - So is the idea to assign a number to each of the variables such as "wrestling ability" and "striking ability," and so forth?
The problems I foresee with that would be inconsistency from one "grunt worker" to the next, incorrect assessments by the grunts, and your own inability to identify key factors which determine fight outcomes.
If you can't already beat the bookies by manually analyzing the fights individually, I don't think there is any chance this will help you.
If you get to the stage where you are actually prepared to use this system to place real world bets, I hope you will be bold enough to publish your results, whether good or bad.
No, I agree, subjective ratings wouldn't work. We'd use measurable statistics to build our models.
If it does get to the stage where the model is used for betting, I would certainly post successes/failures.
I hope that you have contacted Kirik.
He has that entire database on this website.
You do not need to crowd source this.