It's Marathon Week, and...

It's Marathon Week. Columbus is set for Sunday. As usual, I'm obsessing about things like the weather (it may rain) and my tenuous taper tactics. My greatest concern is a weighty one. Tipping the scale now at around 161, I've lost a couple of pounds since my creatine-heavy high (in August and September), but I'm still fatter than I've been for a marathon in quite a while. 

Time for some further analysis.

In 2002, I wrote an article for Marathon & Beyond entitled, YMMV (Your Mileage May Vary). The subtitle is, The Quest to Determine the Most Relevant Training Elements Begins with Examination of One Runner’s Data. Therein, I analyzed several different training elements to try to determine which ones correlate most closely with marathon performance. Spoiler alert: Overall training mileage and mile-repeat pace mattered the most. I republished the article in this here blog, so you can read the entire thing here. 

Other factors I studied were the number of previous races within the training period, age, and weight. There were a few surprises, but one of them was not the strong correlation between weight and marathon performance. The more I weighed, the slower I ran. 

I decided to recreate the chart with more current data. I studied the marathons I've run from 2015 through May, 2025. These are marathon numbers 101 through 131 on my all-time list. I added data for my weight and age, and performed the correlation analysis. 

Here are the results. First, the good news: Age.

Surprisingly, the percentage of variation is extremely low: 0.47%. This appears to show that there is no difference in my marathon performances over the ten years of data. The percentage is lower than it was for my previous analysis. It should go without saying that I am quite pleased about this.

Now, let's see about weight.

Here, the correlation is slightly higher: 5.72%. Just like before, the fatter I am, the slower I am. But wait a moment. That's not all that awful. Last time, it was 33%. I'm slightly encouraged.

I am not sure why both of these factors have decreased since last time. It may be that I'm working with a set of data that has less variation, or that the data here covers a shorter overall time span. I don't know, but I'll continue my examination, looking at some of the other factors.

There is (a little) hope.