Odds are very good that if you look at the SILVA scores page, you’ll see some ratings that don’t seem to make very much sense. Paul Sass higher than B.J. Penn? Matt Mitrione higher than Junior dos Santos? Renan Barao higher than Urijah Faber??
Before you print out a list of SILVA scores just to throw them into a fire, let me do my best to explain why SILVA has plenty of ratings that don’t particularly make sense.
The way SILVA works is by evaluating a fighter based on his record, by assigning each fight a “Victory Score” based on the strength of the opponent. After doing this, SILVA is broken up into three tiers:
Fighters with a SILVA score of 43.00 or higher are considered by SILVA to be “elite” fighters. These fighters have demonstrated that they can beat quality competition on a consistent basis. For the most part, these fighters don’t have very many losses, and those that do have a ton of wins to balance out the losses. Case in point: Matt Hughes (45-8).
These are fighters who don’t quite win consistently enough, or haven’t faced enough quality opponents, to be placed into the elite tier. Instead, these fighters are sorted by the level of competition they have shown a capability of beating. This is why B.J. Penn is the top-rated fighter in the middle tier of the welterweight division. While Penn loses too often to be considered an elite fighter by SILVA, he’s definitely shown that he can beat great opponents. This is in contrast to a fighter like John Makdessi, who is undefeated but has yet to beat a good/great UFC opponent.
The middle tier consists of all fighters whose SILVA score is between 10.00 and 42.99.
These are fighters with a SILVA score of below 10.00. Simply put, these fighters lose too often to be put on the same level as their counterparts. Examples of low tier fighters include Matt Brown (12-10), Tom Blackledge (10-7), and James McSweeney (4-8).
I was specifically asked why SILVA doesn’t rate Melvin Guillard particularly highly despite being 8-1 in his last nine fights. Guillard rates in the middle tier, because he hasn’t won enough fights to offset his eight career losses. And in the middle tier, the most important thing is to beat the highest-level opponent you possibly can. Here are the SILVA scores of six of the fighters Guillard has recently beaten:
- Evan Dunham – 32.54
- Shane Roller – 32.51
- Dennis Siver – 30.51
- Jeremy Stephens – 30.33
- Waylon Lowe – 29.36
- Gleison Tibau – 28.55
Now, Guillard’s current SILVA score is 31.54, but the system still holds his UFC 79 loss to Rich Clementi against him. If that loss isn’t considered, then Guillard’s SILVA score would be 33.21. It’s just enough to place Guillard among the top 15 lightweights in the UFC.
Ultimately, unless a fighter’s winning percentage is very high, SILVA wants to see the fighter beat a great opponent before assigning him a great SILVA score.
To use another example, let’s look at Josh Koscheck. Koscheck has a SILVA score of 32.60. This ranks him 19th among current UFC welterweights. That might sound like a horrible rating, but take a look at who Koscheck has actually beaten in his last ten fights:
- Anthony Johnson
- Yoshiyuki Yoshida
- Frank Trigg
- Dustin Hazelett
- Paul Daley
- Chris Lytle
I don’t know about you, but I’m not overwhelmed by that list; not when Koscheck has also lost to Thiago Alves and Paulo Thiago in that period.
Now, you may be wondering… if this is the way SILVA works, how in the world did Matt Mitrione get a SILVA score of 50.40 (second-best in the UFC heavyweight division)? It’s simple – fighters in the top tier are evaluated by their consistency more than anything else. Because Mitrione started his career 5-0, with all five fights in the UFC against tough opponents, SILVA considers him to be ahead of the curve among heavyweights. If Mitrione, for example, lost his next two fights, he would be knocked down to the middle tier. And in the middle tier, his lack of wins against high-level opponents would prevent him from being highly-rated at all.
Hopefully this sheds a little bit of light on the way SILVA works. It may be a fancy system with fancy numbers, but in the end, just like in real life, what matters is who a fighter has actually beaten in the cage.