Ask the Expert…What's Up With Your Rankings?

image desc

Q: I’m interested in how you compute your Ranking system? What is the program analyzing and how does it determine specific rankings within each weight class and pound-for-pound? I’m be eager to learn what analytics you are using to compute. A most interested fan: Sean Lavin

A: Much-maligned and almost universally misunderstood, FIGHT!’s Computerized Ranking System has been the source of controversy and consternation since the magazine launched in 2007. Although some of the most commonly raised concerns are answered each month in the magazine, there is still an obvious need to explain the logic and calculations that underpin our rankings system. Once we do that, we think our critics will have a better understanding of and appreciation for our rankings and what we hope to accomplish with them.

1) Data Nerds
The first step toward understanding FIGHT!’s rankings is to understand how our fighter database works. Our database was not created with the goal of offering every result from every fight card— and already do a good job of that. Our database exists to feed information into the rankings formula. That’s why our database is “missing” fights—we only enter fight cards promoted by what we call qualifying organizations. These are organizations that consistently book top-level talent, or, in some instances, have a track record of developing talent for top-level shows. Once results from a qualifying fight are entered into our system, a calculation is made based on the strength of the organization in which the fight took place, the quality of opponent, and the outcome of the fight.

2) Card Value
The strength of organization is the only element of our rankings where subjectivity factors in. When creating an event in our database, FIGHT! staff determine the value of the card. The lowest value is typically assigned to regional shows and developmental leagues. The middle value is assigned to national and international promotions that fill undercards with third-tier fighters and prospects. Like the top of a pyramid, the highest value is reserved for a handful of promotions that are historically home to the very best fighters in a given weight class.

The relative value of a win or loss is also dependent on the quality of a fighter’s opponent. If a fighter beats a more highly ranked opponent, his value will increase greatly. A fighter will not benefit as much from a win over a lower-ranked or unranked opponent. Conversely, a loss to a lower ranked or unranked opponent will damage a fighter’s value significantly. The only way a fighter can lose points after a win or gain points following a loss is if highly ranked and lowly ranked competitors fight to a split decision.

3) Solid Results
The third factor in our rankings calculation is outcome. A stoppage is worth more than a decision but there are artificial ceilings and floors built in to account for so-called fluke knockouts or submissions.

Anomalies Vs. Logic
Does our system have flaws? Sure. Since it is bound by results, Jon Jones’ value suffered following his DQ loss to Matt Hammill even though he was clearly the superior fighter. Does our system have anomalies? You bet. But there is always underlying logic.

FIGHT!’s Computerized Rankings System isn’t perfect, but we prefer its cold math over pundits subjectively shuffling names around on paper. Love ‘em or hate ‘em, at least our rankings get our loyal readers talking around the water cooler every Monday.

As far as the pound-for-pound rankings? That’s all computed based on reader votes.

Visit after every major card for updated rankings.

Comments are closed.