David's Blog

DISCLAIMER: I am not a doctor and this is not medical advice. Do your own research and consult with a qualified professional before trying anything in here.

This year I hit a personal goal that I’d had for nearly fifteen years: running 5 kilometers in under 20 minutes.

Objectively speaking, this was no great feat – a time like this wouldn’t even win a recreational race. But on a personal level, it was one of the hardest things I have ever done. It took literally years of sustained effort, a lot of pain, and a lot of frustration.

I tried many things to improve. And I ran a lot of 5Ks. This left me with a decent sense of just how much each change incrementally improved my performance.

These are the things that ultimately moved the needle, along with my approximation of how much.

First Improvement: 4%

The first big improvement I saw was a 4% decrease in my time. This came from making the following changes:

• Running very slowly 80% of the time – slow enough that I could just breathe through my nose, or easily have a conversation – and very hard the other 20%.
• Increasing mileage to ~20mi per week
• Wearing carbon plated shoes while racing
• Taking 200mg caffeine before racing

Someday I’ll write a blog post about why running slowly is so important. This was honestly the first real breakthrough I had after ~10 years of stagnation. For now I’ll just say this: running slowly makes running enjoyable, reduces injury risk, and allows your body to slowly accumulate the adaptations necessary to actually get faster. This is a great book on the subject.

Wearing carbon-plated racing shoes also definitely helped. They objectively improve the performance of anyone who wears them. Note that not all carbon plated shoes are created equal, and some don’t provide any measurable performance improvement at all. (I’m looking at you, Hoka.) But the carbon-plated racers offered by Saucony, Nike, Asics, and Adidas all seem to be excellent. Wearing them during races instantly shaved ~20s off my time.

Taking caffeine also gave me a meaningful boost. I don’t normally drink coffee because it bothers my stomach, so I don’t usually have caffeine before running. But the research on caffeine and running performance is quite clear. It basically makes anyone who takes it faster. So I started taking a caffeine pill one hour before the start of races. This timing was based on a military research project which assessed optimal caffeine dosage/scheduling for peak performance. Fair warning: taking caffeine when you’re not used to it can be intense.

Second Improvement: 2.5%

The next leap forward came from a single change: running more miles. This is basically the advice you see posted all over Reddit.

I increased my mileage to 30-35mi per week, almost all at a slow/easy pace. After maintaining this mileage for a couple months, I ran another 5k and was ~30s faster.

Third Improvement: 5.5%

The biggest improvement by far came from three things:

• Running fewer days per week, but increasing run duration
• Running 1km intervals once a week
• Taking 1000mg Tylenol 1 hour before racing

I listened to a great Peter Attia interview with Iñigo San-Millan, one of the people who advised Tadej Pogačar, who won the Tour de France in both 2020 and 2021. One thing Iñigo said is that the body does not really start to develop cardiovascular adaptations unless it is getting low-intensity cardio in (a) 1+ hour doses and (b) 3-4 times per week.

To get my mileage up over 30 per week I had previously been running 6+ days a week, with many of the runs under an hour. I had the total duration right, but the dosing wrong. This changed.

I started running just 4 days a week, but increased the length of my runs so that my mileage stayed the same. All of my runs became 8+ mi and took over an hour to complete. Even when I did short, hard efforts, my intervals were sandwiched between warm-up and cool-down runs that brought the total distance to 8+ mi. Once a week I would also do a 10+ mi long run.

I mentioned intervals, and by this I mean running really hard for short, precise distances.

This was another thing I started doing consistently. But they were long intervals: 1+ km. I would run 5-6 of them with a slow jog between, usually for about 50-90% of the time it took me to run the interval. I’d try to sustain ~95% of my max HR during the intervals and otherwise not worry about pace. This is a workout that many serious runners swear by.

The training plans in this book were very useful and largely conformed to everything mentioned above (lots of slow running, all 8+ mi efforts, 1km interval workouts, a long run each week, etc.). I highly recommend it.

The final thing that I changed was I started taking 1g of Tylenol 1 hour before racing. This was also a recommendation from a Peter Attia interview:

[Tylenol] is a pretty remarkable performance enhancing drug. [. . .] We’re talking about a 1-2% performance boost. It’s not entirely clear if it comes from the [core] temperature reduction or the pain reduction. But it was absolutely part of my stack during time trials. [. . .] I felt like I was getting a 1, 2, maybe perhaps even a 3% gain from Tylenol. I would take 1000mg an hour before the race. [Source]

I’m not sure how much this helped. But it was one of the things that I changed.

Conclusion

There were other important changes that I made – e.g. I had to get much better at identifying whether or not I was injured, as opposed to just sore/stiff – but these were the things that actually seemed to make me faster. I hope they can be useful to someone else!