September 8, 2026
The Long Run: How Far, How Slow, and How Often
Most long runs fail because they are too fast or too long for the week around them. Here is how to set the distance, the pace, and the frequency.
The long run is the session people treat like a private time trial. I still see it on Strava around here: a "long run" at something close to half-marathon effort, then three quiet days, then a complaint that the plan is too hard.
It is not a shorter race. It is the one day you stay out long enough that the easy pace actually has to hold up. If you cannot talk, it stopped being a long run somewhere around kilometer 6.
What you are actually training
Easy days build the base. Tempo sits near threshold. Intervals are the sharp work. The long run is the one that asks you to keep a low intensity going after a normal easy run would have ended.
Stephen Seiler and Espen Tønnessen looked across elite endurance training and kept landing on the same split: about 80% of sessions sit clearly below lactate threshold, not in the grey zone people call "comfortably hard." That is the zone the long run belongs in. If it becomes a second quality day, you have not added a long run. You have deleted an easy day. (Seiler & Tønnessen, 2009, Sportscience)
Haugen, Sandbakk, Seiler and Tønnessen went through world-class marathon logs later and found the same shape. Weekly long runs are a key session, and most of the year's volume is still low intensity. Their typical long run is 30 to 40 km, a little slower than marathon pace. That is a description of people running 160 to 220 km a week, not a target for someone doing 35. Copying the distance without the week around it is how a Sunday wrecks Wednesday. (Haugen et al., 2022, Sports Medicine Open)
Jack Daniels is blunter about the amateur version of that problem. In Daniels' Running Formula he keeps easy runs between about 30 and 150 minutes, and he does not want a single run to eat more than about 25 to 30% of the week's mileage if you are around 40 miles a week or less. A 32 km long run on a 45 km week is not "committed." It is half the week, done tired. (Daniels, 2013)
How slow
Slow enough that the first 15 minutes feel slightly embarrassing.
Daniels puts easy running at roughly 59 to 74% of VO2max. In practice that is a pace you can hold a conversation through, not four-word sentences. Haugen and colleagues place easy and long runs in their lowest two zones, well under marathon pace for almost everyone who is not already racing at 3:10. If you only have one number, use the slower half of your VDOT easy range, especially after the halfway point. A recent race in the VDOT pace calculator beats "it felt fine at the turnaround."
Wind changes this. A hard westerly on the A2, or just the open stretch past the polder, can make true easy pace feel like work. Carl Foster's session-RPE work is useful here: the load of a run is not the pace on the watch, it is how hard the whole session felt, times how long you were out. A "slow" 18 km into a headwind can be a harder day than a calm 18 km at the same split. If the talk test fails, slow down. The split is not the workout. (Foster et al., 2001, Journal of Strength and Conditioning Research)
Do not run it at goal race pace. A short marathon-pace piece at the end can make sense late in a marathon block, once the easy long run is already a habit. It is a seasoning, not the meal. For a first long run, or a first half, skip it.
If "easy" itself is the thing you keep missing, the pace piece is here: how slow an easy run should actually be.
How far
Set the distance from the race, then cut it if either of these is true: it is more than about 30% of the week, or it is going to take you past roughly two and a half hours. Daniels' 150-minute ceiling is the one I trust more than a round number. Past that, you are mostly collecting soreness.
Rough ranges I actually use:
A 5K block does not need a heroic Sunday. Something like 8 to 12 km, 45 to 70 minutes, is plenty. The long run helps, but it should not steal the one faster day.
For 10K, 12 to 16 km is usually the right size. Long enough that you notice it, short enough that Tuesday's session still exists. If the rest of the week is the missing piece, the 10K training plan is the better place to start.
A first half can peak around 18 to 21 km. You do not need to cover 21.1 km in training to race it, and you do not need that peak every Sunday. On three runs a week, one of them is the long one and the other two stay short. See half marathon training on three runs.
Marathon is where people get romantic. Build toward the high 20s, and only toward 30 to 32 km if the week can carry it and the time stays inside that two-and-a-half to three hour window. If 26 km already takes you close to three hours, hold the duration and stop adding kilometers. The Netherlands-specific starting point is in the first marathon piece. How the rest of the week grows is in building weekly mileage without getting hurt.
Add a kilometer or two, not five. Every third or fourth week, make the long run shorter on purpose. A ramp with no landing is just a ramp.
How often
Once. That is the unfashionable answer, and it is the one that holds.
Seiler's 80% rule does not leave room for two long efforts in seven days. If you run three times, one of them is the long run. If you run six, it is still one. A "medium-long" on Wednesday because Sunday felt short is how the easy days stop being easy.
Miss it, and leave it missed. Foster's point about load is the practical version: stacking a long run onto a tired next day does not give you the session back, it gives you two expensive days. A short easy run, or nothing, is the honest fix. The week can move. The kilometers cannot be honestly made up.
Fuel, once you are out past about 75 minutes
This is the part people treat as optional until kilometer 16 goes quiet.
Asker Jeukendrup's review on carbohydrate during exercise scales intake with duration, not with how tough you feel. For work around 1 to 2 hours, about 30 grams an hour is often enough. Longer than that, the useful range moves toward 60 grams an hour, and only well-practised guts should chase higher. The amount is not the clever part. Practising the same drink or gel on the long run, before race morning, is. (Jeukendrup, 2014, Sports Medicine)
Water too, especially if the long run is the first warm one of the year. I have watched more Rotterdam build-ups fall apart from a dry April Sunday than from a pace that was 10 seconds off.
A few ways this goes wrong
The long run that needs two days off afterwards was too fast, too long, or both. Tired is the point. Wrecked means the rest of the plan just got worse.
If Sunday is half the week, shorten Sunday. Add a little to an easy day instead. The Haugen review is about elites, but the logic travels down: the week is the stimulus, not the single outing.
Starting at the pace you hope to finish is the other classic. The first 20 minutes decide the run. Easy, then still easy. Whatever you can still say a sentence at, at kilometer 16, was the pace the session was for.
What I would do this Sunday
Pick the distance that fits the race and the week you actually have. Run the first 15 minutes slower than your pride wants. Talk. Eat something if you will be out past 75 minutes. Stop when the clock or the percentage says stop, even if the loop has another kilometer in it.
That is harder to judge once a session moves, or last week's long run went badly. Runapt sets the long run from the race, the days you can train, and a VDOT easy pace, then rewrites the rest of the week if you miss it or the day comes back heavier than it looked on paper. The route is already there, which matters more than it sounds when you are trying not to turn the long run into a navigation project.
If you want that Sunday decided for you, build the first week free.
Sources
Daniels, J. (2013). Daniels' Running Formula (3rd ed.). Human Kinetics.
Foster, C., Florhaug, J. A., Franklin, J., Gottschall, L., Hrovatin, L. A., Parker, S., Doleshal, P., & Dodge, C. (2001). A new approach to monitoring exercise training. Journal of Strength and Conditioning Research, 15(1), 109–115. https://doi.org/10.1519/00124278-200102000-00019
Haugen, T., Sandbakk, Ø., Seiler, S., & Tønnessen, E. (2022). The training characteristics of world-class distance runners: An integration of scientific literature and results-proven practice. Sports Medicine - Open, 8, 46. https://doi.org/10.1186/s40798-022-00438-7
Jeukendrup, A. (2014). A step towards personalized sports nutrition: carbohydrate intake during exercise. Sports Medicine, 44(Suppl 1), 25–33. https://doi.org/10.1007/s40279-014-0148-z
Seiler, S., & Tønnessen, E. (2009). Intervals, thresholds, and long slow distance: the role of intensity and duration in endurance training. Sportscience, 13, 32–53. https://www.sportsci.org/2009/ss.htm