Your Summer Slowdown Is the Weather, Not Lost Fitness: 16,727 Runs
Every summer runners panic that they have gotten slower. We checked 16,727 runs from 274 runners, each against their own normal. Hotter days were slower, and you stay just as slow at the same heart rate. That is the heat, not lost fitness.
The June Panic
It happens every year around now. The temperature climbs, your easy pace creeps up by fifteen, twenty seconds a kilometer at the same effort, and a quiet voice starts asking whether you have lost your fitness. The long winter base, the spring sharpening, all of it leaking away in a week of warm weather.
I do not believe that voice, and I had the data to argue with it. So I went and looked.
athletedata stores the weather for outdoor runs, so for a big chunk of our runners we know the temperature on the day of each run, sitting right next to the pace and the heart rate. That is enough to ask a simple question. When it gets hot, do you actually slow down, and if you do, is it your fitness going or just the thermometer?
What I looked at
I pulled every outdoor run we have with both a pace and a temperature on it: 16,727 runs from 274 runners, on Garmin, Strava, Apple Watch and the rest. For the main analysis I kept the 205 runners with at least 15 such runs each, which is 16,214 runs, so that every person had a stable picture of their own normal.
The temperature here is the daily average at the runner's home location, not a reading off the wrist at the exact minute they were out. Hold onto that, because it matters later.
Why I compared each runner to themselves
Here is the trap in almost any heat-and-pace stat you will read. If I throw every run from every person into one pile, the hot runs come disproportionately from people who live somewhere warm, run in the afternoon, or happen to be slower for reasons that have nothing to do with the weather. Compare strangers to each other and you measure who they are, not what the heat did.
So I did it the honest way. I compared every run to that same runner's own normal run. A "hot" run means hotter than they usually train in. A "slow" run means slower than their own average. Then I lined the two up, within each person, and pooled the result.
Hotter was slower, and it was not subtle
Within a runner, pace got slower as the temperature went up. Across the whole range it came to about 4 seconds per kilometer slower for every 10°C, which is around 1.2% of pace.
That average undersells the hot end, because the penalty is not a straight line. It barely moves while you are in your comfortable range and then bites hard once it gets properly warm. On the days that were about 10°C hotter than a runner's own normal, they were roughly 8 to 9 seconds per kilometer slower than their own average run. Over a 10 km that is more than a minute, gone, with nothing changed about the runner.
The fastest runs, by the way, were not the coldest. They sat a touch cooler than each runner's normal. Deep cold was slightly slow too, just much less so than heat. Cool, not cold, is where the legs work best, which is exactly what race directors have always known.
The part that proves it is the weather
Slower in the heat is one thing. The voice in your head says that is you falling apart. So I tested that directly.
First I asked whether the runs were slower at the same effort. I compared pace against temperature while holding heart rate constant, so I was looking at how fast you go when your heart is working exactly as hard. The slowdown was still there. At the same heart rate, runners were about 8 seconds per kilometer slower on those very hot days than on their average run. The heat tax barely shrank when I controlled for effort. So this is not you backing off because it feels hard. You held the same heart rate and the pace dropped anyway.
Then I asked whether it was fitness after all. I added each runner's training load on the day, our measure of how fit they were at that point in their build, as a control. The heat slowdown did not move. Same effort, same fitness, still slower when it was hot. There is not much room left for "you lost your base." The thing that changed was the air.
This is why I trust it. It is within one person, at a matched heart rate, after accounting for their fitness on the day. Three different ways of asking "is this really just the heat" and all three say yes.
The honest caveats
I am not going to oversell the size of it, because two things make our number a floor rather than the full bill.
The temperature I have is a daily average at your home, not the actual conditions on your skin. A hot day with a daily average of 23°C can be a brutal 30°C tarmac run at 5pm or a pleasant 17°C at dawn. That blurring drags the measured effect down, so the real penalty at the moment you are out there in the worst of it is bigger than the 4 seconds per 10°C I can see. I also only have temperature, not humidity or sun, and on a muggy day those do a lot of the damage.
These are everyday training runs, not races. In a race you are already at your ceiling with no slack to give, so heat takes a much larger bite, which is what the race data shows. And this is an average across 205 runners. The typical runner slowed in the heat, but not every single one did. Some barely budge, some fall apart. Your own number is the one that matters, and it is sitting in your own runs.
How this squares with the research
None of this argues with the science. It is a smaller, everyday-training version of a very well documented race effect.
The clearest work is Ely and colleagues, who looked at thousands of marathon finishers across a range of conditions and found pace slowing steadily as the wet-bulb globe temperature rose, with the back and middle of the pack hit harder than the elites. Wet-bulb globe temperature folds in humidity, sun and wind, which is why it bites more than the plain thermometer I am working with. The shape is the same as ours: hotter is slower, cool is fastest, and the effect is real and physiological rather than a trick of who runs when.
The mechanism is not mysterious either. When it is hot, your body sends more blood to the skin to shed heat, your heart rate drifts up at any given pace, and the same effort simply buys you fewer meters per minute. That is also why your watch shows your heart rate creeping up on warm easy runs, and why reading summer pace as a fitness verdict gets it backwards.
What I would actually do with this
Stop grading your summer runs by pace. The number on your watch in July is measuring the weather as much as your legs. Run by effort or heart rate instead, let the pace be whatever the heat allows, and judge the session by whether you held the effort you meant to hold. A coach that reads your runs in context can tell the difference between a hot slow day and an actually flat one, which is the whole point of training to heart rate rather than a pace chart.
Expect the slowdown and plan around it. The penalty accelerates in real heat, so the hottest part of the afternoon is where most of the damage lives. Early morning is where the pace actually comes back, faster as well as more pleasant, and it is where I would put anything that needs to hit a number.
And if you want to know whether you have actually lost fitness or just met summer, the answer is not in any single hot run. It is in your trend over weeks, in how your effort and your speed are tracking together once you take the weather out of it. That is a read you can only get from your own data over time, not from one demoralizing Tuesday in the heat.
The summer slowdown is real. It is just not the thing you are afraid it is.
athletedata connects your training data, reads the weather on every run, and coaches you on what your own numbers actually say instead of handing you a rule of thumb from other people. Get started here.
