Follow me on Twitter!

Saturday, March 30, 2013

Quick Hits: Open Week 4 Initial Thoughts

Despite our best efforts, neither I nor any of the commenters nailed 13.4 exactly with our predictions, but we were awfully close. As I (and probably many of you) expected, things got heavier and shorter this week. Personally, I expected them to go even heavier, but then again, it's the Open and surely HQ is trying to keep things inclusive.

So without further ado, let's move on to my thoughts on 13.4 (and a little analysis):

  • I think you'll see a lot of folks try this one twice. I know I plan to retry it. With only 7 minutes to work with and an unusual rep scheme, I think strategy can play a big role. Anyone who watched the Chris Spealler-Graham Holmberg demo on Wednesday saw that Graham went way too hard out of the gates and paid for it. However, I felt like I let myself go too far the other way: pacing too much early on simply doesn't leave you enough time to make up ground late. I got a score of 85 but I'm going to give it another go tomorrow.
  • Strategy aside, I think for the elite athletes, this workout comes down to who has the biggest engine. The clean & jerks aren't heavy enough that there's really a risk of failing a rep, but they're heavy enough that going unbroken isn't feasible after the early rounds. The key is who can pick the bar up a split-second faster each rep in those rounds of 15 and 18.
  • I'm surprised HQ decided to stick the clean and jerk first on this workout, if for no other reason than it does eliminate from the competition any athletes who can't clean and jerk 135/95 pounds. It's not an excessive load, but at the same time, sticking the toes-to-bar first would allow almost everyone to continue on. But I'd also like to expand on the design of this work out more:
Although we saw Graham struggle on the toes-to-bar, I believe that the design of the workout puts more emphasis on the clean and jerk. Similarly to what I did with 13.2, I watched videos of three athletes completing the workout with similar scores* and timed each of their rounds. Using this data, I got some rough averages for the speed of both the clean and jerk and toes-to-bar. These averages allowed me to do the leveraging analysis I discussed in my post "Why It (Usually) Pays To Be Well-Rounded."

Below is a chart showing the expected score of an athlete with the average rates of speed, then also two other scenarios: one with the clean and jerk speed decreased by 20% and the toes-to-bar speed increased by 20%, and another the toes-to-bar speed decreased by 20% and the clean and jerk speed increased by 20%.

So what this is saying is that if you're slower on the clean and jerks, it's very difficult to make up for it on the toes-to-bar. However, the opposite is not true. Going 20% slower on the toes-to-bar and 20% faster on the clean and jerks actually produces a higher score than being average on both of them. Part of this is due to the fact that a 135-lb. clean and jerk is simply harder than a toes-to-bar, yet both are valued the same. But another factor is that the clean and jerk comes up first, which means that most athletes will end up doing more clean and jerks than toes-to-bar when it's all said and done (all three of these athletes did). Let's look at the same leveraging analysis, but assuming the reps scheme was reversed (toes-to-bar, then clean and jerk).

You can see that the clean and jerks are still emphasized more so than the toes-to-bar, but they are much more even than with the original rep scheme. This isn't to say that the original rep scheme is bad, it's just saying that clean and jerks are clearly valued more than toes-to-bar. But perhaps that was HQ's intention.

One last set of graphs, just because I thought it was interesting. Here are the speeds for the three athletes on each movement as they progressed through the workout. Notice how much the clean and jerk speed degrades after the first 20 reps, yet the toes-to-bar stay much flatter. Even Graham didn't drop off by as much on his toes-to-bar as he did on the clean and jerks.

That's it for now. Good luck to all for the rest of week 4 and I'll see you Monday night for our last SWAG of the season.

*Due to time constraints, I'm just focusing on this workout from the perspective of high-level athletes (this is not me). The three athletes I used were Holmberg, Spealler and Travis Stoetzel (he had the highest score with a video submission at the time I wrote this). They averaged a score of 106. For athletes shooting for a score of 75, we should use different average speeds to break this down, which would produce a different result. However, I believe you'd find that the emphasis is even more on the clean and jerk for an athlete shooting for a lower score.


  1. A thought on your notion to reverse the order (in terms of likelihood of being eliminated for not being able to get one rep). Switching the order would not be helpful for newer female crossfitters, I don't think. MANY women at our gym were able to PR their C&J, but could not get their first toes to bar. Just a thought.

    PS I love the reps per minute graph, these are really helpful for strategy.

    1. That's interesting. I'd be curious to know the percentage of CrossFitters who can complete a legitimate toes-to-bar vs. the percentage who can complete a clean and jerk at 135/95. I wonder if it's different between men and women. There may be some ways we could get a proxy for that based on how many people got stuck at 27 reps on last year's 12.3 (15 box jumps + 12 push press + 0 toes-to-bar). However, that does limit the field to those who could get through 12 push press at a decent weight. Then again, anyone still remaining in the field by 13.4 must have been able to do at least 1 push press at 115 (in 13.2).

      I plan to try and grab all the scores this year once 13.5 is closed. At that point I could probably come up with some rough estimates for this comparison. Thoughts?

  2. I was going to make a similar comment to Jen's. I'd question assumption that T2B is easy for some folk who can't do a 135/95 CJ.

    The analysis I did of the 13.4 stats showed that there were ~180 men that registered a single rep of CJ (which doesn't tell us much - some that couldn't CJ likely didn't enter a score), but about 350 men who entered a score of 3, which were all likely ones that couldn't get past the first T2B. I know we had a couple guys at our box in that camp. Strong guys, but large guys that are beginner crossfitters.

    BTW, you might like the analysis I did which you can find here:

    1. Kim,

      I'll try to touch on this topic in my Open wrap-up post (which probably won't actually be until Monday due to work commitments this week). However, I think it's really tough to draw any precise conclusions with the data we have. We can see how many people did not post a score for 13.4 after completing 13.3 (it's about 10% for me and 19% for women), which gives us a rough estimate of those that can't do a 135/95 clean and jerk. Now, then if you look at the percentage that finished exactly 3 reps, you're limiting the pool of people only to those that CAN do a 135/95 C&J. There's no way to know how many additional people couldn't have done the T2B or the C&J at all. And of course, there are other reasons people drop out of the competition, so we can't say for sure that 10%/19% of the population can't do a 135/95 C&J.

      I'll be showing some numbers for attrition week-to-week in my next post. For women, 13.4 was definitely the one that hit the hardest, but for men, the drop-off was fairly typical.

      When I get time in the coming days, I'll check your blog as well. Always nice to see others are thinking about things from this perspective as well.

      Thanks for reading!