2016 CrossFit Games Correlation Analysis – Men

Individual Athletes during the Ocean Swim at the 2016 CrossFit Games
©2016 CrossFit Inc. Used with permission from CrossFit Inc.

It’s time to get a little geeky now that the CrossFit Games have concluded. In the first of a three part series, we will look at the correlations between events and the final ranking.

Correlation is defined as ” dependence or association is any statistical relationship, whether causal or not, between two random variables or two sets of data.” In simpler terms, it tells us how similar or dissimilar two events are. It can also show that there is no linear relationship between two events.

To do the analysis, we took the placements for each event as well as the final ranking and ran a correlation analysis. The results can then tell us which events are most similar or dissimilar from the final results. We can also see how the events compared to each other.

In this first analysis, we looked at the men’s results.

1RM Deadlift Ladder Doesn’t Matter

Out of all 15 events, the only event that had a negative correlation to the final rankings was the Deadlift Ladder. With a -0.13 correlation, the negative correlation is weak at best. Because it’s so close to 0, there really isn’t a linear relationship at all.

Essentially, the strength of an athlete’s 1RM Deadlift did not have any predictive relationship to how they finished the week. Only one athlete finished in the top 10 in the Deadlift Ladder and the final rankings (Alex Vigneault).

Climbing Snail Had the Highest Correlation

There were a few events that had strong correlations to the final standings. Events like 100% (0.72 correlation) and the Ranch Mini Chipper (0.77 correlation) had strong correlations. However, the strongest correlation was the Climbing Snail with a 0.78 correlation.

There are some interesting things about these three events.

  • None of the events contained a barbell. Only 5 of the 15 events used a barbell. The other 10 events did not. So the better a person was without the barbell proved to be beneficial at this year’s Games.
  • Two events required fast, high output for a short time domain. 100% lasted 3 minutes. The Ranch Mini Chipper lasted 4-5 minutes. Both required very little pacing and a high lactic acid threshold.
  • New or uncommon elements. The Snail was a brand new element at the Games this year. The D-Ball and Med-Ball GHD Sit-ups are not new, but they aren’t as common as other movements. The athletes that excelled at these events were those who could quickly adapt to things thrown at them with little to no practice.

Ocean Swim, Squat Clean Pyramid and Plow Pull Had Weak Correlations

Outside of the Deadlift Ladder with a negative correlation, these three events had the weakest positive correlations. The fact that the Squat Clean Pyramid along with the Deadlift Ladder, the two strength-based events, had some of the lowest correlations shows that the Games this year was more about endurance and high output training that tested lactic thresholds. The Games did not favor the strongest athletes.

This is likely what Dave Castro keeps referring to when he talks about how the Open, Regionals and Games tie into each other. The Open tested aerobic capacity with low weights (for the most part). Regionals tested strength and the Games tested endurance and the ability to adapt quickly to new challenges.

Events Weren’t Highly Correlated

While there were three events that were highly correlated (0.70 or greater) to the final standings, no two events had a correlation greater than 0.70. The highest correlation between two events was 0.67 and that was the Ranch Mini Chipper and Climbing Snail. Even the two strength-based events, Deadlift Ladder and Squat Clean Pyramid, only had a 0.41 correlation.

The lack of high correlations between events appears to indicate that the 15 events did not overlap that much on how they tested the athletes. There were not two events that were so similar that the same athletes did well (or poorly) on both. If nothing else, it shows that Dave Castro did a good job at mixing up the events and testing different metabolic pathways, skills, etc.

Swimmers Can’t Deadlift

Or maybe it should be titled, deadlifters can’t swim. Either way, the Deadlift Ladder and Ocean Swim had the most significant negative correlation at -0.38. It’s not a strong correlation by any means, but it does show that those with the biggest deadlifts didn’t do so hot in the water. Maybe it’s a reflection of training styles or the time devoted to working on strength versus swimming.

Leave a Reply