In another life, I could see myself as a sabermetrics junkie.  Some of you may be asking what in the heck that is. It’s essentially the use of statistics in sports, or in simpler terms think Moneyball.  When I was younger and baseball was my sport of choice, I’d spend hours pouring through the stats trying to decipher patterns. Baseball is a sport that lends itself to numbers crunching.

As my sport of choice shifted to running, which is a sport that is defined by numbers, I carried some of that passion with me.  Instead of crunching batting averages, or even personal bests, I used my geeky numbers passion to crunch miles run per week, percentages of training done at different paces, resting HR and how it correlated with training volume, and all sorts of numbers.  It wasn’t because I used it really, but itwas more of a tool to see if it was useful.  Turns out, I don’t really track any of that stuff now.  It wasn’t that useful… (Far simpler= more usable…Track mileage, and how athletes feel (good,bad,average) and it does the job).

The funny thing though, is that while running is defined by its numbers, there is a large human component that makes track a much trickier sport to quantify than something like baseball.  I’ve always reasoned that it’s because distance runners are living on the edge.  It’s a fine line between pushing your body to complete exhaustion and 99% of exhaustion in a race.  The same is true in finding the balance between aerobic and anaerobic development, or mileage versus speed, or going to the well versus training within your limits.

The funny thing though, is that while running is defined by its numbers, there is a large human component that makes track a much trickier sport to quantify than something like baseball.  I’ve always reasoned that it’s because distance runners are living on the edge.  It’s a fine line between pushing your body to complete exhaustion and 99% of exhaustion in a race.  The same is true in finding the balance between aerobic and anaerobic development, or mileage versus speed, or going to the well versus training within your limits.

Track is essentially about balance.

So it makes it hard to simply quantify a few variables and say if you run X then you will run Y.  Paradigms like this are why we were left with people like Lauren Fleshman being told she would never run that fast because her VO2max was too low (Of note, is I got told the same thing when my VO2max was right around 65…I think one of the lowest for running decently fast…)(http://www.runnersworld.com/runners-stories/why-science-and-running-dont-always-mix)

Having now coached HS, college, and pro runners, one of the topics that I thought could use some number crunching is progression.  How do world class athletes progress to world class athletes.  We all hear this idea that you “peak” as a runner maybe around late 20’s early 30’s.  Furthermore, now many programs and governing bodies are defined by progressing from the junior to senior level and hitting certain bench marks along the way.  When I dealt with the Irish federation for example, they had certain marks at each age and tried to establish a “pathway” that the athlete should progress upon.

The questions I wanted answered were simple.  Was there a “normal path” to reaching the highest level, which I highly doubted,and if not what did the variation in performance look like.  So what I did was look at the yearly progression of the best non-african runners in the women’s 5k.  I chose non-african because of the simple fact that the known “age” issues would skew the very variable I was trying to look at.  Additionally, I took out all  known dopers and even the suspected dopers.  I was pretty selective as I didn’t want any crazy drug induced spikes.  Of course there are probably still some druggies in there, but I did my best to eliminate that variable.  I further separated the data out into all runners and sub 15:00 runners to see if there was a difference.

This is what we are left with:

 

 

We’ll start with the highlights.

Is there a standard progression?

To answer my question on whether there is a standard progression, I would give a resounding no.  While I didn’t post the progression graphs, the progression was kind of all over the place.  You can see this reflected in the data above.  While the average age of their lifetime PR was around 28, there was a wide variance with people setting their PR anywhere from 20 to 35.  And though it’s not included on this data, athletes generally were able to keep within 10sec of that PR for 3 more years on average.  It’s possible that they could have kept within that range longer, but most when they hit mid 30s shifted the focus to the marathon which skews the data.

Breaking through to the next level?

We like to think that PR’s occur after yearly consistent progression with a nice smooth drop each year.  The reality though is that most of the time the year they set their PR or broke through to world class was a huge breakthrough.  On average, though with a large variation, athletes improved by around 16seconds in the year they ran their PB.  The first year athletes went sub 15:15, they improved by on average 24seconds to get to sub 15:15.  That’s a huge jump.  For our sub 15, runners, the first time they went sub 15, they improved their PB by a huge margin (21sec average).  Again, it points to the fact that breakthroughs are they way in which athletes either set PR’s or broke through barriers.

It’s quite interesting that the average Seasons best for the sub 15 runners for the 2 years prior to it was 15:09.  Once again, showing that breakthroughs were the norm.

Regression to the mean

Another interesting phenomenon was that if we looked at the average season best for the 2 years after the athlete PR’d, there was a big regression to the mean.  Not for everyone obviously, but the athletes essentially went to the levels they were the years prior to their PR.

Age Trends:

While I looked at a couple of age related trends one that stood out was the difference between the age in which they ran their first sub 15:15 clocking and the year they hit their PR. Not surprisingly, the sub 15 runners took longer to reach their PR (on avg. 3.13yrs) and ran their first sub 15:15 at an earlier age (24.8yrs) than our non sub 15 runners.

The questions to ask:

So what does this data tell us?  That progression is highly individual and that there is still hope for you to PR at the age of 35 if you are an outlier.  But really, it tells us that breakthroughs are possible and not to limit yourself.  Even world class athletes tend to rely on breakthroughs instead of chopping of a few seconds here and there.

But more importantly are the questions the data raises.  Why do breakthroughs occur?  A complimentary analysis of training would be a nice study for something like this. Was it years of hard training finally absorbing? Some change in training, recovery, diet, or mental attitude?  No one knows and the answers are probably highly variable.  Perhaps even more interesting would be what happened the year after the big PR.  Did athletes regress because of a shift in focus?  Or was it because they had a huge breakthrough the previous year and decided that since they went from a 15:10 athlete to a 14:50 one, they had all of the sudden start training like a 14:50 athlete.  Perhaps it was something else, but I’ve noticed in coaching that there’s a large tendency to want to start training harder once you have a breakthrough.  There’s this idea that now that I’m fast I have to train up to my race ability.

I’d be curious to see what others thought about the data above.

 

Get My New Guide on: The Science of Creating Workouts

    4 Comments

    1. Jared on September 4, 2013 at 6:25 pm

      How many of those PBs were set in Olympic years? Coaches and athletes naturally (consciously or subconsciously) peak in Olympic years (or should anyways!). How many had a bit of a 'breakthrough' the year of the Olympics and then regressed the year after? I realize the quadrennial plan isn't an excuse to run slow in non-Olympic years, but a shift in focus (more volume, not running your prime event at major meets, etc) in non-Olympic years may say something about an athlete's ability to peak at the right time and progress to Olympic participant –> finalist –> medalist –> champion.

    2. Anonymous on September 6, 2013 at 4:20 pm

      What did you end up with for sample size?

    3. Steve Magness on September 6, 2013 at 4:22 pm

      Sample size was right around 50.

      I'll have to go back and look through the data to see what years things occurred. That is a good point. Perhaps post olympics shifts play a role.

    4. Nathan Austin on October 12, 2013 at 7:28 pm

      This gives me additional hope for post-collegiate improvement in time.

      It also made me wonder if there are tendencies in the number of breakthroughs an athlete may experience throughout an entire career and as individuals get older if they become more or less frequent (comparing those who started training around the same age and continued similar progressions in training volume/intensity over time)?

    Leave a Reply