Pre-Race Thoughts, Part II - The Positives...I guess

This was supposed to be a more optimistic post than the last. Maybe it will be by the time I'm done, but my outlook is decidedly negative at the moment for a couple of reasons. First, my grandfather passed away over the weekend. It wasn't unexpected -- he was 88 and had been quite sick for some time -- but it was nonetheless jarring. He was the grandfather that every kid wishes they had. He was goofy, had an enormous belly, and did magic tricks. He was also full of wit until the day he died. (The doctor checking his mental acuity on that day asked him if he knew what day it was. He replied, "What, you don't know?") I don't normally run for anyone but myself, but this weekend I'll be running for Gramps.

The second reason for my negativity is that, yet again, I got sick. Now, given the short amount of time most of you have known me, you've probably concluded that I'm either perpetually unhealthy or, as Mike G suggested, this is all psychosomatic. Neither is true! And, for the record, I am also quite a cheery and optimistic person. (Now you're thinking I'm in denial.) I am as flabbergasted by this series of illnesses as anyone. The short of it is this: I have a cold and sinus infection, and I feel miserable. [Note: I wrote this on Wednesday. It is now Friday, and I'm feeling quite a bit better.] I'm not complaining or making excuses in advance of the race. Writing about it is a form of therapy and maybe the only thing I can do that makes me feel better. On the brighter side, all signs of my bronchitis are almost completely gone (still some coughing/wheezing at night). On that positive note, let's turn to the good stuff.

Number Crunching
I'm not ashamed to admit it...I like numbers. I also like running. So, naturally, I had to find a way to combine these two objects of affection. In baseball, you might have heard of Sabermetrics -- "the search for objective knowledge about baseball."  Sabermetrics uses novel statistics to improve our understanding of the game. These are tools designed to better assess players' and teams' performance and include both the simple (e.g., OPS - on-base plus slugging percentage) and the complex (e.g., WAR - wins above replacement).

In running, the need for such analytics is seemingly small. We have had the best performance measure available to us for centuries - our finishing time. While baseball needs elaborate equations to determine whether Player A is better or worse than Player B, in running all we need is a stopwatch. But there still seems to be a gap in our understanding of how we can run our fastest in any given race. We might, through trial and error, figure out an approach to training and racing that works pretty well. But is it the ideal approach? Is there something more we could have done had we had the right information?

When preparing for my 800 meter training this year, these were the types of questions dancing around in my mind. To begin to answer them, I wanted to do an analysis, but I also wanted to keep it simple. I thought about finishing time. Finishing time is a function of speed, and speed is a combination of how far we stride and how rapidly we turn over from one stride to the next. (Distance per stride x Strides per minute = Distance/Minute, or velocity.) There must be a trade-off here - in order to stride farther, we have to sacrifice turnover, and vice versa. Given this, I wanted to know two things:

1. Is there an optimal stride length/rate combination for me in the 800?
2. What was the length/rate combo in my 800 race earlier this year, and can it give me insight into what my training should focus on this year?

But where do I get the data for this, you ask? Well, um, I watched some video...a lot of video...video of over 70 of my past 800 meter races, to be exact. (By the way, this was done a few years ago, before I had a kid. So, I was only neglecting my wife, not my son, while poring over these videos.) I took lap splits and literally counted my steps. A lot of effort for something pretty meaningless, sure, but I mentioned I like numbers, right? After it was collected, the data just sat in a spreadsheet for a while. It was high time to put it to use.

I calculated the average stride rate for each race (total number of strides divided by finishing time) and an average stride length (total number of strides divided by 800 meters). I then plotted these two numbers against each other to see where things fell.

I discovered a few things from this analysis:

• There's quite a bit of variability in both stride length and rate. Average length has varied between 1.90 and 2.10 meters, and stride rate has gone between 92 and 108 strides per minute.
• There's some clustering by year (i.e., there's more variation between years than within years). This might have been due to differences in fitness/strength or race strategy or both. 
• To my original question, there doesn't seem to be an optimal combination. Check out the graph below. The dashed green line represents the combination of stride length and rate that equates to a 2:00 800. Anything above it is a sub-2:00 performance; anything below is over 2:00. So, the farther above the line you go, the faster the race. You can see that my fastest times came with all sorts of different combinations. This is interesting and not really what I expected.

As I mentioned, the other thing I want to know is what I needed to do to get under 2:00 this weekend. Looking at the data from my January race (circled in the chart), the shortest distance to the 2:00 line would be to increase both turnover and stride length slightly. But perhaps the easiest way to get to the line is to work on what seems to be the biggest weakness -- stride length. You can see in the graph below that the average stride length in my January race was among the shortest of my running career.

Fortunately, I did this analysis shortly after my last race and was able to base some of my training plan on the findings. I made the assumption that stride length is partly a function of explosive leg strength. I knew that this would likely improve on its own through any 800-meter training program, but I wanted to supplement the training with something more targeted. Conveniently, I have a book called "Explosive Running," that gives examples of various plyometric exercises designed to improve leg strength and power. (By the way, check out my endurance library, and literally check out a book if you'd like.) After doing my research, I added some simple one- and two-leg jumps to my twice-weekly lifting sessions. I haven't done anything like this in a long time, and the only other time I routinely did these types of exercises, I ran my PR.

[Interesting anecdote about that PR: I was injured after the '02 XC season and didn't run a step until January '03. I started up a conservative running plan but added box jumps after most of my workouts. By the end of February, I was running 2 seconds faster than I'd ever run the 800 before. It's a small sample, but my coach and I were confident that the jumps had something to do with my sudden and unexpected improvement.]

All this analysis must seem like overkill. I'm clearly over-thinking something quite simple. But if I have the data, why not try to learn something from it? Sure, I could go on forever repeating the same training plan, making little tweaks here and there as I learn, but I'd rather innovate. We'll see this weekend how much, if at all, this little addition to my training program helped. If nothing else, at least I'll have another data point to add to the spreadsheet.

Running doesn't need a bunch of nerds to come in and invent new metrics with confusing acronyms; there are enough of us already here to do it ourselves. We all obsess over numbers anyway (mileage, times, elevation, rankings, etc.), why not take it to the next level? Isn't there something you'd like to figure out about your own racing or training? Maybe it's time to start gathering data.