Does Striking Accuracy Accurately Measure Striking?

Now imagine another fighter, this one who only throws strikes to the head while at kickboxing range. A strike thrown to the head while the fighters are standing at distance is much more difficult to land than the body shot on the ground. While standing, their opponent is in constant motion, keeping their head moving, and using their hands and arms to block any strikes that come their way. On average, these strikes only land about 33% of the time. If this fighter landed 50% of these strikes, we would say they were doing a great job. 

Real fighters throw all kinds of strikes, but they don’t all follow the same pattern. Some are kickboxers, who choose to throw most of their strikes at long range. Others are grapplers, who aim to throw their strikes in the clinch and on the ground. When we look at a fighter’s Striking Accuracy, we are seeing how often their land strikes, but we are mostly seeing the kinds of strikes they choose to throw. Those that throw more of those short range strikes will likely have a higher Striking Accuracy than those that throw mostly long range strikes. Their accuracy will be higher, but it doesn’t tell us if they are good or bad at striking precision. 

Last Saturday, at UFC Fight Night: Adesanya vs. Imavov, UFC introduced a new statistic designed to help explain if a fighter’s striking accuracy in a fight should be considered good or bad. It first looks at the mix of strikes a fighter throws and calculates an Expected Striking Accuracy. We know what to expect their accuracy to be because we know from our vast trove of historical data what the accuracy is of each kind of strike they throw. 

Once we have their Expected Striking Accuracy, we compare that to their actual Striking Accuracy to see if they are doing better or worse than expected. If they are doing better, the number will be positive. If it’s worse, the number will be negative. 

Let’s look at a real example from the fight between Tagir Ulanbekov and Clayton Carpenter at UFC 311:

In this example, both fighters had exactly the same Striking Accuracy: 49%. However, Ulanbekov threw almost exclusively head strikes, while Carpenter mixed his targets, throwing higher-probability body shots and leg kicks. Based on each fighter’s Expected Accuracy, we can see that Ulanbekov overperformed his expectation by 4 points, while Carpenter underperformed, also by 4 points. We call those values, +4 and -4, their Weighted Striking Accuracy because it takes into account the kind of strikes they threw and the weighted probabilities of landing each one. 

This statistic is also useful for evaluating how precise fighters are across their entire careers. If you look at the leaderboard for all-time Striking Accuracy, you will see Anthony Hamilton ranked #2, with a 65.6% career average, while Sean O’Malley is ranked #10 with a 61.5% career average. Both are excellent numbers, but looking only at Striking Accuracy hides the bigger picture:

Hamilton has almost exactly matched his expected accuracy. He has done as expected, but his Striking Accuracy appears high because the strikes he chose to throw – mostly ground strikes – were likely to land. Sean O’Malley is a different kind of striker. His game relies on pinpoint accuracy in his boxing, which is far harder to achieve. O’Malley outperforms his Expected Accuracy by 18.5 points, the highest number for any fighter in UFC history with at least 5 fights. As good as you thought O’Malley’s hands were, they’re even better when measured fairly and placed in the proper historical context. 

These Expected Accuracy and Differential numbers will appear in UFC broadcasts when the Striking Accuracy numbers don’t tell the whole story or when a fighter is performing far differently than expected. The additional information will help make us smarter about the sport and highlight the incredible precision of UFC strikers.