Numbers don’t lie… or do they?

Every day, I receive a motivational email from Runner’s World. Often their quote of the day really gets me thinking. One such recent quote is by the famous American runner Frank Shorter:

Numbers don’t lie. You always seem to remember your workouts as being a little better than they were. It’s good to go back and review what you do.

This blog is about critical thinking while running – at least it is if you read the tag line! It unites my passion for running with my critical eye for looking at numbers and patterns.

Let’s take a look back at my numbers from January. Below are a few points which I feel will develop into key themes as the months roll by.

Stated Targets
Will I hit my big target if I miss the incremental ones?

First and foremost is my slight disappointment in not achieving my stated target to run at least 150k during the month. My goal this year is to run 2015 kilometres. Based on simple division, this is approximately 168 kilometres per month, though I committed to 150 each month as I know that there will be some big 200k-plus months in the mix. Sadly, according to Runkeeper (which until now I have used as my de facto basis for measurement) I missed my mark by less than 8 kilometres – one solitary run – in January.

There’s a Japanese word: omoikomi (思い込み) that literally means to get deep into thinking something. It doesn’t carry the suggestion of deep thought or consideration that my clumsy translation suggests. This word usually has the nuance of prejudice or assumption. It can also mean a preconception or an unquestioned thought. For some unknown reason, I decided that Sunday was the last day in January. This was factored into my thinking about 2 or 3 weeks before the end of the month. On the last Saturday of the month (actually the 31st), I felt shattered from taking my kids to a theme park and let myself sleep early – thinking I could cover the last 8k and then some on the Sunday.

It wasn’t until the next day after my Sunday run when Runkeeper gave me a running total for January (and February!!) that I realized my mistake. Of course, as it was set in my mind that Sunday 1st February was the last day in January, it never occurred to me to open a calendar! I won’t make that mistake again! But to answer my own question, a few barely missed goals here and there is nothing to beat myself up about as long as it doesn’t become a habit.

Running total for January and February in Runkeeper
Running total for January and February in Runkeeper

Am I faster in the morning or at night?

I care about the answer to this question as I am aiming for a pace of 4:45 per kilometre. So I want to optimize my running times to increase my chances of hitting my targets.
First glance at the data, I was surprised not to be much faster in the morning than at night. But then I realized that my morning runs start in the dark.

As I would expect, my afternoon runs have been fastest (as well as closest to my target pace). But two runs in clear daylight this year do not a pattern make. Through the rest of the year I will be looking for this pattern to optimise my pace. Clearly with work and family commitments, running in the afternoons is a rare luxury for me. I will be looking forward to the end of winter when the sun will rise earlier so I can get daylight runs in before work.

Screenshot 2015-02-03 21.22.19

Numbers do lie (somewhat)
Which numbers do I believe? Or should I believe any of them?

I swear by measuring all my activity as I run. I like to compete against myself along similar courses and similar distances. In a later post, I will discuss the various tools I use. In this post, I will just merely point out a discrepancy between my two key running wearables (Runkeeper on my iPhone; and my Garmin Forerunner watch). A look at the data will show a number of differences between the distances, times, pace and calories.

Screenshot 2015-02-03 21.19.47

On the subject of target pace, my Garmin watch consistently reports a quicker pace than my Runkeeper app. To strip pace back to its constituent parts, there are two factors: total time spent and total distance covered.

Distance is supposed to be measured by GPS and so in theory, there shouldn’t be a discrepancy. However, on occasion, my Runkeeper app gets an anomaly reading as something interferes with the signal and causes huge bounces where I might suddenly “run” a whole kilometre in just a minute or even occasionally a few seconds. The craziest instances were in amongst skyscrapers in Hong Kong where the signal bounces off windows. I try to avoid running in areas that are too built up, but this is a hazard in Tokyo too! Anyway, I usually notice this when it is a big bump, and I go back later and correct them with the map editing feature in the desktop version of Runkeeper. But maybe it’s the smaller bumps that I’m not noticing. Not that I’m going to be so anal as to check my route online after each run. I just simply don’t have the time or inclination for that. So I will aim to paint a picture over time of any average discrepancies and factor these into my total measurements. I may also try the old-fashioned approach of getting a city map book and measuring my routes on paper with a piece of string and a ruler.

January according to Garmin Connect
January according to Garmin Connect
January according to Runkeeper
January according to Runkeeper

The other factor in pace is time. I know through habit that I tend to turn my Runkeeper on first as I start my run, and turn it off last afterwards. There is therefore a difference between these numbers too, but it is usually simply a matter of a minute or so. After my January 10th run, I remember forgetting (oxymoron?) to turn it off until I was already back inside the house! This was a good five or six minutes after the run, hence the large discrepancy between the Runkeeper “Duration” and Garmin “Time”.

The critic in me makes me question these numbers. Twelve days is not enough to develop a clear pattern. So as the year progresses, I will start to build a hypothesis and hopefully get closer to understanding which ones I can trust more, or whether the reality is somewhere in between. When I get round to it, I will also factor in past data from last year.

Being the closet geek that I am, there are other ways I measure my activity. I won’t go into them here for now, but I have noticed discrepancies with other data and am really starting to wonder what affects them.

Of course, I have to be careful not to disappear under my own pile of big data… 😉 But it would be great to hear your suggestions on what other influencers to measure: mood, impact of sleep, weather…9 As Douglas Hubbard says, you can measure anything!

3 thoughts on “Numbers don’t lie… or do they?

    1. That could be interesting. I’m not a big eater of gels, except for my really long runs. As I don’t do enough of those, it might take a really long time to get enough data.


      1. True. I have a friend who takes a 100mg caffeinated gel before his speed workouts. We always argue if that much really works and I also bring up the point that his body is just getting use to that much caffeine and it might not effect his performance.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s