Written by Jacob Tsypkin
The CrossFit Games leaderboard is an underappreciated, underutilized thing.
Though it is the bane of my existence during competitions – I feel like I spend half of my time slapping phones from athletes hands as they incessantly, obsessively check the standings – during the rest of the year, it is perhaps the most important tool for program development available to us, apart from the basic principles of program design.
As of this writing, we are privy to five years of data from what I consider the “Modern Era” of the CrossFit Games (2011 and on – “Open Era” or “Froning Era” would be equally appropriate.) Across the Open, Regionals, and CrossFit Games, this totals one hundred twenty-four scored events from which coaches and athletes may glean understanding.
These events lend us direct insight into performance, i.e. how well an athlete must perform in a given event if he or she expects to be competitive. But with a little bit of digging, we can also use this information to create a deeper understanding of the sport and how to train for it. This is done by drawing correlations. In the case of this article, the correlations drawn are between particular events at the 2015 CrossFit Games, and final placement at same.
To determine the correlations between individual events and the final placement, we took the athletes placing in those events and determined the correlation to their overall finish in the Games. In this method we are simply comparing rankings to rankings, with no regard for how close the athlete was in either real performance or points.
What the data actually tells us is how strongly related an athlete’s finish in a certain event is to their final placement in the 2015 CrossFit Games. With this information, we can draw conclusions about which tests and abilities are most important for an aspiring CrossFit athlete to excel at, and use those conclusions to impact program design.
Let’s take a look at three interesting, and potentially meaningful relationships from the 2015 CrossFit Games.
1) The most classic CrossFit event had the greatest predictive power
In a landslide victory, the Triangle Couplet was the event most strongly correlated with final placing for both men and women.
I don’t suppose I should have been surprised by this, but I was nonetheless. It makes sense, of course. Events such as this one do a good job of precisely what they claim: test overall fitness. At relatively low levels, one can take the cross section of a few different monomodal tests – for example, max back squat, 30 muscle-ups for time, and 5k run – and estimate an athlete’s capacity. But the more those numbers improve, the less bearing they seem to have on the athlete’s ability for multimodal events.
Training Implication: Practice your sport
There’s no way around it. Sport specific preparation is king in all athletic endeavours, and the CrossFit Games are no exception. Challenging multimodal workouts, mostly couplets and triplets with occasional pieces consistent of four, five or more parts, in a variety of time domains, executed at high intensities, are completely irreplaceable. Even if you’re of the opinion that they’re a poor way to develop general health and fitness, you must make them a cornerstone of your training if you wish to be competitive at any level of CrossFit Games competition.
2) Strength and Endurance are of equal importance
For both men and women, the most conventional strength event (the clean & jerk) and the most conventional endurance event (Pier Paddle) had effectively the same correlation to final placement.
Although these events were not quite pure strength or endurance events – the clean & jerk testing the athletes durability under fatigue nearly as much as their strength, due to it’s placement in the competition, and Pier Paddle testing the athletes adaptability to a modality which was new to most of them – it is still fair to say that they did an effective job of testing the athletes maximum strength and their aerobic capacity. If we are willing to take this claim as true, it suggests to us that the CrossFit Games does a good job of assessing athletes at both ends of the bioenergetic spectrum.
Training Implication: Equal Measure for Equal Measures
By now, I think most people interested in training seriously for the CrossFit Games are aware of the importance of developing strength and endurance in equal measure. But this measurement hammers the point home: you cannot rely on either your ability to move weight, or your ability to endure, in isolation. If there ever was a time when specialists could succeed in competitive CrossFit, it is long past. Train both your maximal strength and your aerobic endurance, year round, with slight emphasis in the areas you are weakest. Avoid biasing weaknesses so much that you create deficiencies elsewhere.
3) Men, Women, and Barbell Proficiency
Strength/weightlifting events were substantially more closely correlated to victory for women than for men. On the women’s side, the Snatch Speed Ladder was the third most predictive event and the clean & jerk the sixth most predictive. For men, those events ar tenth and ninth respectively.
For both men and women, however, “D.T.” was the fourth most predictive event.
The first correlation is a bit difficult to unpack in isolation. Does it suggest that strength and/or weightlifting ability is less critical for men than women? Viewed on it’s own, but in light of the second correlation, it may prove to be the case.
The second correlation is extremely straightforward: strength/power endurance are very, very important to success in competitive CrossFit, regardless of gender. Given that female athletes tend to be more aerobically dominant than male athletes, it intuits well that higher levels of strength for female athletes may have higher carryover to strength and power endurance, whereas male athletes, capable of greater force production, must dedicate more time and effort to specifically developing the ability to perform at high volumes with moderately heavy weights. In other words, strength on it’s own is not more important for women, but the effects of strength development for other aspects of performance for female athletes may mean that maximum force production has greater additional value for them than it does for their male counterparts.
Training Implications: More Reps for Men, More Weight for Women
If we accept as true that in most cases, female athletes are more biased towards aerobic capacity, men towards force production, we may reasonably conclude that men should spend more time developing their endurance, both generally through conventional aerobic training and specifically with moderately heavy high rep weightlifting, where women can dedicate more energy to developing top levels of strength and power, trusting their natural disposition towards endurance to do more of the proverbial heavy lifting than it does for men.
When applying this principle to training, there are three important points to consider. (1) This is not a universal principle. There will be female athletes with a much higher proclivity for strength and power, and male athletes who are naturally endurant, but do not easily develop strength. This will be dependent primarily on their athletic background and genetic predisposition. In other words, it’s pretty much out of your hands. (2) The differences in training should be relatively minor, except in very extreme cases. Male and female athletes training should generally be similar, with only a few changes here and there to account for general and individual differences. (3) The higher level the athlete, the more true this distinction is likely to be. Newer athletes without particularly glaring imbalances should generally utilize a well rounded program, designed to develop abilities in all domains of the sport.
Each year, the pool of data available to aspiring coaches and athletes in the competitive CrossFit community grows. With a little work, correlations such as these can be drawn, and the information we glean from them becomes more and more valuable. Effective programming is data driven. Don’t neglect the information at your fingertips.
Author’s Notes: I am not a mathematically inclined person. My good friend Matt Nolan is responsible for drawing the correlations cited in this article. I am responsible for the conclusions inferred from those correlations.