The marathon long run is overrated.
In my experience, too many beginner runners (those training to run slower than 3:45) focus on trying to get in multiple 20 or 22 milers in their training segment at the expense of improving more critical physiological systems. More importantly, scientific research has shown that runs of over 3 hours offer little aerobic benefit compared to runs of 2 hours while significantly increasing injury risk.
As such, rather than cramming your marathon training schedule with multiple 20-22 milers that increase your injury risk and recovery time without decisive aerobic advantages, you should focus on improving your aerobic threshold, teaching your body to use fat as a fuel source, and building your overall tolerance for running on tired legs through accumulated fatigue.
Since the long run is such an ingrained element of marathon training, and suggesting they are overrated sounds blasphemous to many veterans, I am going to provide you with scientific research, relevant examples, and suggestions on how to better structure your training to help you run your next marathon faster.
The science of the long run
Most beginner runners training for the marathon are averaging anywhere from 9 minutes to 12 minutes per mile on their long runs (3:45 to 5-hour finishing time). At a pace of 10 minutes per mile, a runner will take roughly 3-hours and 40-minutes to finish a 21-mile run. While there is no doubt that a 21-mile run (or longer) can be a great confidence booster, from a training and physiological standpoint, they don’t make too much sense. Here’s why:
Research has shown that your body doesn’t see a significant increase in aerobic development, specifically mitochondrial development, when running over 90 minutes. The majority of physiological stimulus of long runs occurs between the 60 and 90 minute mark. This means that after running for 3 hours, aerobic benefits (capillary building, mitochondrial development) aren’t markedly better than when you run for only 2 hours. Therefore, a long run of over 3 hours builds about as much aerobic fitness as one lasting 2 hours.
Furthermore, running for longer than 3 hours significantly increases your chance of injury. Your form begins to break down, your major muscles become weak and susceptible to injury, and overuse injuries begin to take their toll. This risk is more prevalent for beginner runners whose aerobic capabilities (because of cross training and other activities), exceed their musculoskeletal readiness. Basically, their bodies aren’t ready to handle what their lungs can.
Not only are aerobic benefits diminished while injury risk rises, recovery time is significantly lengthened. The total amount of time on your feet during a 3-hour plus run adds considerable fatigue to the legs, which leads to a significant delay in recovery time. In the long-term, this means you can’t complete more marathon specific workouts throughout the following week, which I believe, and research has shown, are a more important component to marathon success.
Why is the 20-mile long run so popular
Given the overwhelming scientific evidence against long runs of over 3 hours, why are they so prevalent in marathon training?
- First, many people have a mental hurdle when it comes to the 20 mile distance. The marathon is the only race that you can’t easily run in training before your goal race.Therefore, like the 4 minute mile and the 100 mile week, the 20 mile long run becomes a mental barrier that feels like an obtainable focus point. Once you can get that 2 in front of your total for the day, you should have no problem running the last 10k, or so your mind believes. Unfortunately, this just isn’t true from a physiological standpoint.
- Second, the foundation for marathon training still comes from the 1970’s and 1980’s at the beginning of the running boom. Marathoning hadn’t quite hit the numbers it has today (you could sign up for most marathons, including Boston, the day before the race) and the average finishing time at most races was closer to 3 hours (today that number is near 4 hours). As such, the basis for how to train for a marathon came from runners who averaged close to 6 minutes per mile for the entire race. Therefore, 20 and 22 milers were common for these athletes as a run of this distance would only take them about 2.5 hours to finish at an easy pace.
- Moreover, the 20-mile distance is synonymous with “hitting the wall” or “bonking”. Hitting the wall frequently occurred at 20 miles because your body can store, on average, two hours of glycogen when running at marathon pace. Two hours for a 6-minute per mile marathoner occurs almost exactly at 20 miles.
In short, the basis for a lot of our understanding of marathon training is passed down from generation to generation without regard for the current paces of today’s marathoners. Therefore, we also need to reassess where the long run fits into the training cycle and how we can get the most benefit from training week in and week out.
How to train smarter
I suggest that you downplay the role of the long run if you’re training to run 3:45 or slower and focus instead on improving your aerobic threshold (the fastest pace you can run aerobically and burn fat efficiently) and utilize the theory of accumulated fatigue to get your legs prepared to handle the full 26 miles, without needing to run the full distance.
For example, you should focus on stringing out your workouts and mileage over the course of the week, rather than having 40 to 50 percent of your weekly mileage come from the long run, which increases the total amount of quality running you can do and decreases the potential for injury.
The question still remains, however, about how do you get your legs prepared to run for 26 miles?
The answer lies in the theory of accumulated fatigue.
- By shortening your long run to the 16 to 18-mile range and buttressing it against a shorter, but steady paced run the day before, you’re able to simulate the fatigue you’ll experience at the end of the race.
- In addition, when you have shorter long runs, you’re able to increase the total quality and quantity of tempo and aerobic threshold workouts throughout your training week. Instead of needing four to five days to fully recover from a 3-hour plus run, with a shorter long run, you can recover in one or two days and get in more total work at marathon pace or faster. Developing your aerobic threshold is the most important training adaptation to get faster at the marathon distance because it lowers the effort level required to run goal pace and teaches your body how to conserve fuel while running at marathon pace. The more work you can do to improve aerobic threshold and your ability to burn fat as a fuel source, the faster you can run the marathon.
- Finally, with a focus on shorter, more frequent long runs, you can implement faster training elements, such as fast finish long runs or surges, which allow you to increase the overall quality of your long runs. Running your long runs more intensely teaches your body how to run marathon pace while tired, and also increase your body’s ability to store energy for the end of the race and use fat as a fuel source more efficiently.
When you balance out the gains you can get from finishing a long run fast and upbeat with the potential drawbacks from an extended, 3-hour plus long run, you can see why a shorter, faster long run is the better training option for almost all marathoners aiming to finish over 3:45.
RunnersConnect Bonus
Download your FREE Long Run and Aerobic Pacing Calculator now.
Need help converting your race times to your optimal easy and long run pace? Download our FREE calculator and we’ll do the math for you.
35 Responses
Sounds a lot like the Hanson method. I’m training for my first marathon in October and in the early stages of their training program. The reason I haven’t tried a marathon before is my hesitation to do 20+ miles in training. That would destroy a runner at my level. I’d rather spread my miles out over the week, so the Hanson method is appealing. I’m no physiology expert, but it makes good practical sense to me.
Yup, it’s the same idea. I did an interview with the Luke Humphrey you may enjoy: https://runnersconnect.net/luke-interview
Jeff – I’ve become fascinated with aerobic threshold training lately and have been looking for information just like this for a while now. Sounds strange, but I’m glad you had the guts to write it. Knowing a few BQ guys locally, I suspect you’ll get some heat for this
Thanks, Russell. There is always a lot of feather rustling when you write things like this, but it’s important to open the eyes of all runners to the new literature and research.
Now, this isn’t the case for all runners. If your friends are 3:05 or 3:25 BQ guys running solid mileage, a 20-22 miler is fine. This principle applies more to the 40 mpw marathoner where a 20-22 miler becomes more than 50% of their weekly mileage.
If the majority of physiological stimulus occurs between 60 and 90 minutes, what benefit would a 3:15 marathoner get from doing a 20-miler? It would take nearly 2:40 to do it–far past that 60-90 window.
I’d argue very little extra benefit than an 18-miler. However, long runs also become a factor of your weekly mileage. Typically, a 3:15 marathoner is going to be doing decent mileage. Thus, the long run really isn’t a huge percentage of their training. This is compared to a slower runner (assuming slower = less mileage, which I know is not always the case) from whom a 20 miler could be 50% of their weekly mileage.
I just ran my 1st marathon using the SmartCoach from Runner’s World. I edited it a bit and only did 1 20 miler in training. I only ran max 45 miles a week but ran a 3:16. Thinking of training for another one and was going to try more 20 milers in the next plan….any suggestions? Do you have a sample training plan with fewer longer runs?
I don’t have a sample schedule since I am not a big fan of writing template schedule. But, we can write you a custom schedule: https://runnersconnect.net/training-plans/
I think you can crack 3 hours with 40 mpw. I consider that a reasonable number if you x train properly. Totally agree on the shorter long runs even for slightly faster guys maybe not the 2 40 and below runners. A lot of Europeans cap long runs at 30k which is 18.6 miles just because it’s a nice round number
Jeff, Great article. I used the Hanson Method and set a PR. Accumulative fatigue works. IMO it lead to better, quicker recovery therefore increased number, frequency and quality of my training runs. Thanks for the insight on the 20 mile idea. Helps to illuminate the reason behind the why of what we’ve been doing all of these years.
Hi Jeff, thanks for the articale. I think I will try to incorporate it into my training schedule. I do have one question (it may be silly), but when trying to improve your aerobic threshold and your ability to burn fat as a fuel source, how do you know when you are using your glycogen reserve or using fat as fuel? Is it typically after x amount of time? It would be nice to know mentally that after say 1.5-2 hours I know I am running on fat as fuel.
Thanks
It’s a good question. It’s important for me to point out that you never use only one mechanism as a fuel source (fat, carbs, etc). You’re always using a lit bit of all of them. It’s more the percentage of each that is used and what is being relied on primarily. As for telling when you’re using each, again, you’re always using each energy source, but the determination of what percentage of each is based on speed. The faster you go, the more glycogen you burn. Hope that makes sense.
Coach Jeff – I’m glad you brought this article out of the locker and blew the dust off…..as a senior runner (going on 64) I have to think hard about my volume. I ran 3 marathons in 2012/13 and improved my time on each one from 3:32:xx; 3:24:xx and finally 3:21:xx in Boston 2013.
My highest weekly mileage topped out at about 55m/wk and in fact I ran fewer 20m + runs in the 2nd & 3rd races instead concentrating on a hard 5k or cross country (5m) on Saturday followed by a 16-18 mile run on Sundays.
Twice I ran PB races Sunday morning 10k and 10 miles (around 41 mins and 69-70 mins respectively) and 12 -15 miles in the afternoon after a shower, some food etc, so I for one believe for the most part that it works.
I have one reservation and that is that I faded a bit in the last few km of Boston but still ran fairly even splits of 1:40:xx and 1:41:xx so a question: how do I avoid this drop off if I haven’t run the “extra long” mileage?
P.S I am planning on running Boston 2014
hi jeff,
thanks for all the good tips and excuse my english.
i signed in for your training plan sub 3:30 marathon on runkeeper.
my actual goal is a little faster i would say 3:10. i thought to adjust
your plan in speeding up the exercises to fit my assumed mararthon speed.
(i started running marathons in 2003 and finished around 15 til today)
reading your evaluation on the long exercises i stumbled over the
words ” for runners over 3:45″. sounds to me that not only i have to
adjust the training speeds for the exercises, but the milage too.
can you please give a little comment on that?
thank you
Tough to say. Typically, runners in the 3:10 ability range will be able to run more mileage and long runs, but this isn’t always the case. Here’s a more in-depth answer to why: http://www.youtube.com/watch?v=f0sqq0oWLxY
Jeff,
I’ve been wondering about this ever since I read this article.
A runner in my category (long run once per week of 14 miles at 12:40-13:20 pace), puts me over that three hour time zone each week. Since I am considering myself somewhat of a beginner, is this okay?
Great question. I think it’s ok. At some point there becomes a balance of needing to get enough miles on the legs when marathon training and that’s where I feel 16 is a little better than 14, even though 14 is already over that 3 hour mark. In addition, it relates back to percentage of overall mileage. 14 and 16 will likely be a moderate amount of your weekly mileage, but not anything crazy like 60% (which can be the case for some 20-22 milers).
Hi Jeff,
Thanks for the article, I agree that running slow will make you fast however your scientific basis you put forward in this article doesn’t fully back up what you are saying. The paper you reference talks only about mitochondria development and not about the other physiological developments that build towards aerobic development. The number of mitochondria present plateaued as the exercise period reached 90 min however even the author says that “a multitude of complex hormonal, metabolic, and physiological changes, often specific to the exercise task, must be influential in exercise training.” In other words for e.g. “capillary building”, as you mentioned, is not measured in this study and as we know is very influential in aerobic capacity. They only looked at one part of a complex system and so running greater than 90 min may develop these other factors. If you know of any papers that look at these other factors against exercise duration I’d love to see them as there are some many conflicting articles flying around I want to learn what is really happening.
Love to hear your comments on this, keep up the good work. Stephen.
Very interesting article. I am at the beginning of my marathon training for Las Vegas in November. I am still adjusting my training schedule. How would you rate doing 2 types of exercises per day in regards to the accumulation of fatigue? I am doing a 1 hour cross fit style workout every morning and then I started running in the afternoons. I am undecided yet about the lengths of those runs. From reading your article I would say it makes sense to bring those afternoon runs up to 10/12 miles per run and then to runs up to 18 miles on the weekend, but to skip the 20+ long run?!
I totally agree about the long run being over rated. I run 4-6 marathons a year and my last three this year have been 2:43, 2:41 and 2:41 last week. ( not bad for a young 46 year old ) … My preparation is all about keeping the average pace of all runs as close to sub 4 min/k average and my long run is generally 20-25k… But I generally run at least three runs over 20k a week. (As part of an average 100-110k a week total). I believe it is more important to run more at marathon pace rather than wasting time running long runs over 3 hours at a pace slower than your ideal marathon pace.
This year I have more of an effort to just run faster all the time and added about 15% extra mileage. It has seem me PB in every distance I’ve raced from 10k to Marathon. My Marathon last weekend was my 30th and faster yet… All this without long runs… 🙂
Hi. Scared but breathing a sigh of relief. My last marathon was 2000. 3:12 with no rhyme or resson or proper nutrition. Just running at age 32 after s baby. Since then busy with life I have never stopped running, running now age 44 and runnibg back to back 1/2 marathons at 1:31 . For my up coming marathon in december I have not run longer than 18 miles. I do however hammer out 10 to twelve miles the day after the half. I did read your other articles but how can I get it in my head I sm really capable of the 3:15 once again, now that I am nutritionally sound too. How do I get past all the chatter of those running theur twenty milers? Type A worried. 🙂
Dear Jeff,
I like your idea, I was following your training plans on RK preparing my marathon of 20 october for 3:30. I am 53, started running in 2007, first Marathon 2008. My PB was 3.40 in 2009, then always 3.43/46. Training was fine, I could follow it. Mid september also I had my best in half Marathon (1.40). I arrived on 20 October but after half race I was feeling so tired…..1:47 at half then 3:44 at the end. Either I was overtrained, either I did not discharge enough, anyway I was feeling so tired….after the race in the following weeks very tired. I am still trying to understand why I arrived so tired. I enjoyed the training so much (first time I was following a training for marathon…) but still cannot understand what happened… 🙂 I hope I can understand since I want to try next year again!!
Thanks for the article. Combining the information from this posting with previous ones seems to point to the optimal long run for the 3:45 marathoner being less than 12 miles, with total weekly mileage at about 40-60 miles. That’s actually not too hard to reconcile with many training approaches. Perhaps not as frightening or controversial as Jeff suggests.
How I arrived at those numbers:
1) 3:45 marathoner has an expected 5K pace of 7:30/mi (McMillan)
2) Optimal long run pace is 65% of 5K pace, 10:08/mi (Jeff, previous article)
3) Aerobic benefit capped at 2 hours, 11.85 mile long run
4) Long run at 20-30% of weekly mileage.
5-hour marathoner?
9 mile long run, 13:24/mi, 30-45 mi/week
So my question: There must be a point where the biomechanical inefficiency of running so slowly defeats the aerobic development efficiency of such a pace. I would imagine there is such a threshold regardless of how fast your race pace is. Has any research been done on this topic?
Great questions. First, to point out in point 3 – aerobic benefit doesn’t “cap”. It’s simply that the rate of return isn’t linear – i.e. it doesn’t grow exponentially.
To answer your other question. Yes, there is a point where you can run too slow, but I find this to be VERY slow and far outside what anyone would run. As you hint it, it actually occurs when it’s too “painful” to run that slow because of biomechanical inefficiencies. I haven’t seen any studies, but in my experience this is around 4 minutes slower than MP (but this could be off because of my talent level – it’s just a guess).
Hi Coach Jeff. Thank you for this great article. I followed your sub 4 hour marathon training on runkeeper. I was very consistent, but got injured with an abductor strain so I skipped the entire week of the 20 mile run but I was able to do every other training after that week. My marathon is this coming weekend. Will I able to complete with a sub 4 hr even though I missed the 20 miler week? This will be my first marathon.
Thank you.
I understand sentiment behind not doing multiple long runs of 20 miles because you need to do the other training sessions and you also do not want injuries however I did my first marathon with 15 miles as my longest run and I cursed not doing the longs runs of 20 miles, it hurt really badly, I managed to push through pain at 15 miles onwards but how can you motivate yourself to push from say 18 to 26.2 miles, I did start the marathon slower than I would normally but I do feel that it was lack of long runs that killed me early on. I do agree thought longer than 3 hours is no point but if you are going to run a marathon for more than 3 hours you should have spent at least that amount of time on your feet no?
I doubt it was the lack of long runs that had you feeling this way. If you only reached 15 miles in training, I have a feeling your overall mileage was very low. Had you been running 50-60mpw then I have no doubts the marathon would have been a breeze, even with 15 miles being your longest run. My suggestion is to look at the training plan as a whole and not just the long run.
Hi Jeff,
I am actually following your sub-4 hour plan for my first marathon. I seem to be straddling that 3:45 line you mention. In the plan there is only one 20 mile run. I live in Michigan where it is super cold and snowy. i do my long runs outside as much as I can. However, I know the weather is impacting my pace.
In your opinion, should I push to complete these long runs? Or stop at three hours even if I am not finished (assuming it never warms up here)
Thanks!
In addition, I am hoping to be just under four hours. I have been cross training to improve my strength. My 1/2 time is right at 2 hours.
I have been struggling to get my long runs in because of the weather. I managed to get a 19 mile run in but it was not pretty. I have been running 7-10 miles on regular training runs at an 8:30 min mile pace. The problem is I have not been giving myself enough rest days. My mileage per week is 25-40 miles per week. I feel like I never run on fresh legs anymore. I swim for cross training once to twice a week. Any advice?
Ok so I am somewhat of a believer with your thoughts of the 20 mile training runs. My best marathon was one that I felt somewhat undertrained for. However I am doing things a bit different with my next marathon. My question is should I run my long run day after my mp run? Should I only run 18 for my longest runs? I am hoping for a B.Q. My best marathon was a 3:57 2 years ago. I only need a 3:55 but I am hoping for an extra 5 mins so I am hoping for a 3:50. My best half is 1:47:51 I ran that last fall while training for The NYC marathon. I tend to get injuries a lot! I am sure it’s form all the training for marathons. I run on avg 30-35 miles a week as I train for marathons I tend to stay at about 45 during peak weeks.
I trained according to the Hanson’s method back in Jan 2014 in preparation for running Boston. I was as strong as ever, or so I thought.
I ran a 6 min PR for a half marathon just 4 weeks before Boston, but fell apart in Boston at (you guessed it) the Newton hills. I could tell at the starting line that things didn’t feel right – yet I pushed on, trusting my training and my recent racing. Maybe it was just “one of those bad days” or I didn’t taper properly (I didn’t taper per Hanson’s plan), but whatever the case, it got me doubting that this method will get me through the marathon distance.
So I’m a little late on this post, but I think you are right. I’ve ran the last 2 marathons with multiple 20-22 mile runs and followed recommended training paces to get a BQ and both times fell apart because I feel I wasn’t trained to run at race pace for a sustained amount of time. I arrived at the marathon injured both times. Perhaps I’ll try this method next time.
Found this article really encouraging as well as refreshing. Thank you for taking the time to write this and share your expertise. I ran my first marathon on regular short and medium runs because I was confident that I had built up enough endurance in my body after 10+ half marathons which my muscles are comfortable with. If I was to run a 20+ as part of my training, it would take a considerable amount of time to recover my energy. Instead I focus on long slow distances (LSD), strengthening and massage, with make sure I get extra carbs and protein in my diet. This works for me because I find that my energy and muscle power gets zapped quickly at a higher pace or very long distance, and am not a quick runner, a marathon mile is more 9+ than my 10k mile but I feel that I control the fatigue in my body much better when I consistently run medium distances. Come the marathon day I just take the first half real steady and then after 2 hours take a food supplement, focus on the music and really listen to my body. Last bit of noise: shoelaces, cannot underestimate the difference to how your bottom half feels by just the way the laces are tied and the tightness of shoe laces – no more uncomfortable pins and needles.
Hi Coach Jeff,
I’m planning on running my first full in about a month. I have run many halfs in the past.. I’m really anxious about running my first full. I know I have the mental capacity but I’m not sure if I have the physical capacity. I’m not sure what to do from now until then. Any advice?