ultramarathon longest run

How Long Should Your Longest Run Be Before An Ultramarathon?

Share This Article

When I give training talks, I can inevitably count on being asked, “How long should my longest run be?” When I get this question, I usually retort with, “Why do you need a long run?” Silence ensues. The silence stems from the fact runners rarely think about why they need a long run. And even more specifically, why they need a super-long long run in preparation for an ultramarathon.

“How long should my longest long run be?” is an inherently flawed question and assumes the following:

  1. The longest long run is somehow a singular prerequisite for the event. This is similar to how the 16- or 18-mile long run has become ubiquitous in marathon training.
  2. The longest long run builds some magic fitness not otherwise attained in training.
  3. It is universal for everyone.

Here’s the truth: there is no magic long run distance that will automatically qualify you to be ready for an ultramarathon. I’ve coached successful ultramarathon runners whose long runs are as little as 20% and as much as 80% of the distance. Yes, long runs are important. They build aerobic capacity, reinforce your musculoskeletal framework and increase fat oxidation. On top of that, they improve a whole host of other aspects important to ultrarunning.

But you know what? All these areas improve through all the training you do, not just the long runs. You shouldn’t do a long run for a few narrow and very specific physiological adaptations. Physiology does not work that way. You accomplish those adaptations during the entirety of training. They do not result from one (or even a few) run(s), no matter the length. So how do you find out how far you need to go during a long run? Here’s how to get to the answer.

Take a quick look at the numbers

In my experience, ultrarunners put far too much pressure and emphasis on the one single longest long run. They feel they ‘must’ do a 50-miler in advance of a 100-miler or 20 miles in advance of a 50k. From a physical standpoint, the singular longest long run matters very little. This is because the amount of training you can do on one single day, even if it’s a really big day, is limited. What’s more, any single day is only a tiny fraction of  the weeks and months of training accumulated during the entire training process.

Hypothetically, let’s say you are training for a 100k event and you come up with some arbitrary (yes, that word is intentional) logic that tells you that you need a 7-hour long run in training. If you have been training for 4 months, at an average of 10 hours per week, that means you have 160 hours of training in the bank. In this case, your 7-hour long run constitutes 7/160ths of your total training load, or about 4.3% of your total training volume.

That might look like a lot on paper. But, consider that you would probably run about 3 hours on that day anyway. As a result, the difference between your arbitrarily-long 7 hour run and a normal long run is only 4 hours. That’s 2.5% of your total training volume. Now consider you will likely have to recover more during the week following the really long long run. This reduces the overall training volume for the entire period, further reducing the impact of the extra time and effort you put into the really long long run.

Why longest long run doesn’t matter

My point with the arithmetic exercise is this: from a physiological standpoint, it does not matter how long your longest long run is. If you are doing things right, you will have weeks and months and hundreds of hours of training to build up to your key event. Completing one single run that is a couple of hours longer matters very little.

Don’t pick your long run volume arbitrarily based on your race distance (as in the case of the 16-mile marathon long run). Similarly, you should not pick your long run volume based on some narrow physiological rationale. Instead, look at your current training and then examine what you’re trying to accomplish – beyond the physical – during your longest runs.

Build your bridge from both sides

Now that I’ve made the case for how not to think about your long run, let’s take a good look at good reasons to do long runs. Then we’ll whittle down to ways to determine what your longest long run should be. Training is like building a bridge. On one side, you have your current fitness, capacity for work and musculoskeletal resilience to handle long runs. On the other side is the event you are training for, with all of the physical and mental challenges it will throw at you.

Your job during training is to bridge the gap between where you are now and where you need to be on race day. You have to consider both sides of this equation to determine the length of your longest long run. The amount you can reasonably handle puts you on the near side of the bridge. Good rationales for doing a long run puts you on the other side, scheming for what you should prepare for as the event approaches. By looking at both, you can come up with the answer to this mystifying question.

Step one: The long run reasonability test

I dislike overgeneralized formulas that point to magic solutions to training problems. The overused “you should only increase your weekly volume/long run by 10%” rule is one of them. Early in the annual training cycle, any moderately experienced runner (which nearly all ultrarunners are) should be able to increase their volume more quickly. As training goes along, a 10% increase is much more risky.

When I have a new athlete who has a reasonable amount of training experience, I use the chart below as a starting point to determine how long of a long run they can reasonably handle for any given week of training. I call this, the long run reasonability test (creative, I know). I customize this chart for each athlete based on their experience. With more experienced athletes I can be more aggressive. I also look at injury history – I’m less aggressive with athletes who have a history of injuries – and a host of other factors. Irrespective of the athlete, the gist remains the same.

Increase in long run % What your long run could be
Current week 4:00
1 20.00% 4:48
2 20.00% 5:45
3 10.00% 6:20
4 0.00% 6:20
5 5.00% 6:39
6 5.00% 6:59
7 5.00% 7:20
8 0.00% 7:20
9 2.50% 7:31
10 2.50% 7:42
11 2.50% 7:53
12 0.00% 7:53
13 1.25% 7:59
14 1.25% 8:05
15 1.25% 8:11
16 1.25% 8:18


Before you dive in to the minutiae, focus on the ‘Increase in long run %’ column. It goes from 20% for a couple weeks, to 10% to 5% to 2.5% to 1.25%. For the math geniuses out there, yes, I’m cutting the increase down by half every few weeks. Quite perceptive of you! You also might notice that every few weeks the percentage increase is zero. This is to accommodate for recovery weeks, race weeks, or any other reason to back off of the progression. What have I determined with this test? Only what is reasonable for any moderately experienced runner to handle for a long run on any given (but not every given) week.

FAQs on Reasonability Test

Here are some questions you likely have at this point:

Question- Is this the perfect answer?

Answer: Nope.

Question: Is this the exact same formula for every athlete?

Answer: Of course not.

Question: Should I really do a 6 hour and 59 min long run on week 6?

Answer: No again. That’s a silly amount of precision. Be practical, round up or down as necessary.

Question: Should I follow this long run progression every single week?

Answer: Absolutely not.

The only thing you should use this type of chart for is as a guidepost to estimate what is reasonable for a long run during any particular week of training. In my example, if I had a runner that had a current long run of 4 hours per week, in 16 weeks they could reasonably handle about an 8-hour long run. In Week 8-12, the runner could reasonably handle anywhere between a 7- or 8-hour run. Is that what their long run ends up being? Sometimes yes, sometimes no, depending on the actual progression and the ‘good rationales’ exercise in the next section.

For this example (after going through the exercise in the next section), I would have the athlete target 1-2 long runs per month to be near these ‘reasonable’ targets. This means in Week 4, they could run a long run that is a little over 6 hours, and in Week 6 they could run a long run for 7 hours, etc. The other long runs are ~50-75% of these ‘reasonable maximums’.

To build from the other side of the bridge, and take the entirety of training into context, I ask the athlete a series of questions rooted in what goals they would like to accomplish during the long run.

Good rationales for your longest long run

Hopefully, my earlier point that there little physiological reason you need a lengthy long run resonated with you. If not, please reread the first section! Yes, long runs improve you physically, but so does the rest of training! The entirety of training is more important to focus on than one single long run. Your long run(s) represent a small fraction of your overall training; it is important to treat them as such.

So why even do long runs if the physical side does not matter too much? Good question. Long runs need to serve a purpose outside the physical. In my opinion, there are three good rationales for doing any one particular long run:

  • Work on your nutrition plan
  • Build confidence
  • Work through troublesome logistical scenarios, such as running through the night or in big temperature swings.

Step 2: Long Run Questionnaire

Therefore, to further hone in on what that longest of runs should actually be, ask yourself these questions:

  1. How long does my run need to be to practice my nutrition plan?
    1. The minimum for this is 4 hours for just about any ultramarathon runner.
  2. How long does my long run need to be to give me confidence?
  3. How long does my long run need to be to work on any logistical consideration (running through the night, big temperature swings, etc.)?

Give yourself a maximum window of 1 hour for each of these answers. After you have finished the mental exercise, you can line up all 3 reasons, complete with their timeframes and see what makes sense as compared to the ‘long run reasonability test’. Using my previous example, the athlete might answer the three questions as follows:

  1. How long does my run need to be to practice my nutrition plan?
    1. I feel I need 4-5 hours for this because after 3 hours, I don’t like sweet foods any longer.
  2. How long does my long run need to be to give me confidence?
    1. I feel I need 6-7 hours for this because the longest I have ever run is 6 hours.
  3. How long does my long run need to be to work on any logistical consideration (running through the night, big temperature swings, etc.)?
    1. I feel I need 5-6 hours for this because the race I am training for will start at 8 AM in the 40s and warm up to the 80s by 2 PM.

In this case, all of these timeframes meet the ‘long run reasonability test’, but at different points.

Reasonability supersedes any particular reason

From time to time when I go through this exercise, the timeframes to practice nutrition, build confidence or work on the logistical issues are mismatched. There simply isn’t enough time to meet the reasonability test and get a long enough long run in to accomplish one of the good rationales. In this case, bend on the time needed to work on nutrition, confidence or logistics. If you have been diligent on the reasonability test, that should supersede all. Find a way to increase your confidence by doing a shorter run, or may be back-to-back runs. Cobble together the logistics by starting a run in the middle of the night. The point is, don’t compromise reasonable progression for one or a few of the of the other good rationales you have for a long run.

The take home

The duration of your long runs and longest long run should have intent. That intent is not driven from some arbitrary and predetermined number that acts as a prerequisite for success. Similarly, the intent should not be driven from the physical rationale that you need a long run to improve your aerobic capacity, mitochondrial density, musculoskeletal integrity, fax oxidizing enzymes or any other physiological phenomenon.

Your long run intent should come from your current training status, how long you have, and what reasonable progression looks like. Then, you should take into account specifically what you want to accomplish during the long run, aside from the physical. Ultrarunners should be giving their long runs purpose that extends far beyond the physical gains they could get. I hope these exercises can put you on a path to realize not only how long your long runs and longest long run should be, but why you should be doing them in the first place!

By Jason Koop,
Head Coach of CTS Ultrarunning

Share This Article

Comments 25

  1. Great article, Jason, thanks so much for your wisdom! I am currently building up for my first 100K, and I am torn between your method from the chart you posted and something that I read on Science of Ultra; which talks about there being very little physiological benefit (and much more risk) when pushing long runs past 3.5 hours. Here is the article: https://www.scienceofultra.com/blog//the-long-run

    Punching my current long run (3-3.5 hours) into your table, I would peak at a long run of 6 hours; which goes against the recommendations from the other article I referenced. What do you think about that?

    As an additional data point, my current plan for the 100K is to do Sat-Sun back to backs which would peak at a total of approximately 8 hours split between the 2 days.

  2. An excellent article. I am an old ultra cyclist. Ever ultra athlete should read and rethink what they are trying to accomplish with their training plan. Thanks.

  3. Pingback: Heat Acclimation: When to start and how long it takes - CTS

  4. Thank you. Plenty of helpful advice for me to consider and utilise. Made a lot of sense in line with experiences from the marathons/ultras one so far.

  5. Pingback: Double Day Training for Ultrarunners (Updated) - Jason Koop

  6. Pingback: Ultramarathon Daily News | Tue, Apr 30 - Yeovil Marathon

  7. Hey Jason,

    Thanks again from out the Netherlands. The 10% needs to be piramide down.great view..
    Than again always good yo read your thought.

  8. This is really great advice and direction, and it’s very consistent with my own experience. I ran a 50k comfortably last weekend with a just a single “long run” of 10 miles about 3 weeks out. BUT … this came after a solid base of consistent daily 4-5 mile runs for 4 months leading up to it.
    Thanks for a great article!

  9. Uncle Koop! Thank you for a brilliant post. This sits well with KoopCast episode 13.

    Long Runs should “break you” they should build you up, and give you confidence to make the next one a bit longer, or more successful!

    Love your postcast and articles!!

  10. Pingback: 3 Sports Psychology Strategies to Prevent Ultramarathon DNFs - Jason Koop

  11. Pingback: What Runners Should Do If You Missed a Training Day - Jason Koop

  12. Pingback: 3 Little Things That Go a Long Way for Ultrarunners - Jason Koop

  13. Pingback: Not Quite, Captain – Randy's Runs

  14. Great info. It’s like my answer to almost any training question – “It depends.”

    It’s not just the total training in the current cycle, but the accumulation of time/miles over years that really get you ready for a long race. That includes the ability to withstand the physical rigors of longer races as well as efficiency. As such, the numbers/importance of any single long run become even smaller.

    I don’t think Matt or Camille ever did/do run >4hours. I think they’ve both said their longers runs are/were 22-23 miles. As you allude to, the damage (recovery required) increases the longer you go and the benefits diminish. I understand the psychological and nutrition/gear testing benefits, but the physiological cost/benefit needs to be emphasized strongly.

    How do you address learning to run efficiently and fast (relatively) when tired?

  15. Thank you.
    This is excellent – and it makes perfect sense.

    Are going to – or have you already – addressed
    Question #2 How do I prevent my stomach from going south?

  16. Great content and suggestions 👊🏻

    It is so easy to hop onboard what is traditionally considered “right” about the Long Run, when a conversation that pulls individuality into the equation is necessary.

    LOVE the bridge analogy as well 👍🏻

  17. Wow. I’ve got 60+ running books in my library. A third of those have training components. Many for marathons, half dozen for Ultras. It’s the first time I hear advice such as this. And though Koop clearly expresses the importance of tailoring/customizing – and thus not necessarily look for a miracle plan – I find much logic in this approach. It’s quite novel too, I’d say. Nice that these ideas are not just for elites – but shared with us mere mortals. Many thanks.

  18. This is gold. I appreciate the long form answer to a popular/simple/complex question. Another factor not discussed (Jason could write a whole book on long runs…) is what that long run looks like. Is it like 10 hours at Disneyland walking, standing, eating and taking selfies? Is it a constant 70-80% effort day on a fueling plan? Does it include some tempo effort, hill strides or strong downhills? It is the most complex day to plan.

    1. Dean,
      Thanks for the comment. Most of your long run volume should be at the same intensity as your normal Endurance Runs. I don’t like to overcomplicate the long run with strides, hard downhills and the like unless there is a very specific reason.


  19. Agree. For self-coached or coached runners, the runner should know the purpose of every run.
    For those of us who run trail ultras but cannot get to trails for every run it is probably beneficial to train on a similar surface and profile and the long run seems to be a great fit for those adaptations. Running on roads/trails can be a lot different than moving up and down hills, running through/over mud, rocks, roots. Different muscles are engaged. I feel a lot different after running a road marathon vs a 50K.
    Coming from HM/M training plans where long runs were not just getting in the miles at easy pace what about pace? Just keep it easy/z1 particularly for the shorter 50K/50 mile races where race pace might be a little bit more than an easy pace but more towards moderate efforts.

  20. Pingback: Ultramarathon Daily News | Tue, Apr 30 | Ultrarunnerpodcast.com

Leave a Reply

Your email address will not be published.