Data Scientist Previews 50/50 UFC San Diego Fights

BCDC Data Scientist, Brandon Bergstrom, is 22-9 with over 13.8U on these 50/50 fight picks. This weekend’s fights are great to look at from a data perspective since both matchups are toss up fights and have exciting, well-known fighters. These selections went 1-1 two weekends ago for the UFC fights.

Cynthia Calvillo vs. Nina Nunes

These are two female flyweight fighters who are on the struggle bus together. Their Dominance Metric for their last three fights is below. You can see how we calculate our dominance metric here:

Calculating Dominance Score
Example KO win = 5 Points
Example Split Decision Win = 1
Example U Decision Win = 2.78
Example U Decision Loss = -2.78
Not all wins are created equal. Historical data is how we create the point system for different outcomes.

The best piece of advice might be to take a look at their profitability. Nina Nunes is 135 dollars profitable in the UFC where Cynthia is 55 dollars profitable.

Let’s be honest, it’s not our favorite fight to look at for a 50/50 preview… but it’s the one we have to do. Nina is 2-2 as an underdog, and Cynthia is 4-3 as a favorite. It really is a toss-up fight. I am surprised that Cynthia is so heavily favored by the sportsbooks.

Our models project the fight to go to the ground and not go the distance. As a data scientist, my job is to use the models and data to give you the best insights I can.

Cynthia Calvillo is 5-1 against fighters with 108 or less for Total Time in Octagon. Nina Nunes is 1-2 against fighters with 4 or more for Sig. Str. Landed 1 min.
Cynthia Calvillo is 6-1 against fighters with 4 or less for Total Wins.

These three stats show me enough that Nina cannot handle a volume striker in her career. Cynthia can handle an inexperienced fighter comparatively speaking by stats. She 6-1 against bad or inexperienced fighters, essentially.

My prediction is Cynthia Calvillo (-155), but I will be curious to see who the models project as the winner.

Devin Clark vs Azamat Murzakanov

The next closest odds to a 50/50 fight is Devin Clark and Azamat Murzakanov. This is a light heavyweight showdown sure to bring fireworks to the arena. Azamat is a currently undefeated prospect who will take on UFC veteran Devin Clark, who recently won by knockout.

I want to check out their fighting at similar odds before and use our Bayesian simulation test to see who will win this fight. Right now, the stats don’t add up for Devin.

Devin Clark is 3-6 against fighters with 3 or more for Sig. Str. Landed 1 min.
Devin Clark is 2-4 against fighters with 7 or more for Sig. Str. Attempts 1 min.
Devin Clark is 2-5 against fighters with 2 or more for Sig. Str. Head Landed 1 min.
Devin Clark is 2-3 against fighters with 6 or more for Sig. Str. Head Attempts 1 min.
Devin Clark is 7-6 against fighters with 3 or more for Total Strikes Landed 1 min.
Devin Clark is 3-5 against fighters with 7 or more for Total Strikes Attempts 1 min.

This is a bad look for Devin Clark, because the sportsbook knows him very well. He has gone 1-4 as an underdog and 0-3 at the very specific odds he is at this weekend. Devin Clark is a gamer, though, and I do see his tree trunk legs helping him out this weekend.

This graph shows how each fighter does across multiple statistics, which might help you see who has an edge.

Looking at striking, Devin Clark is obviously much better in all categories. However, is this going to be enough? Will he be able to overcome the sportsbook underdog odds and face down an undefeated fighter? Light Heavyweights are 148 – 274 in over 422 fights in the UFC since 2011. This means that the underdog does historically win. However, I will be curious to see what BCDC analyst, Grant, and the models have to say this weekend. For now, unless someone can convince me otherwise, I am riding with Azamat Murzakanov to win this fight.