You be the Coach: How should John pace his marathon?

The marathon is a fickle beast: at 26.2 miles, there’s ample opportunity for things to go wrong. Fueling, pacing, health, and even your mindset can be ravaged by the marathon.


One of the most difficult aspects of running the marathon – especially if you’re trying for a specific goal time – is knowing how to pace yourself.

And not just the pacing strategy, but the actual pace you’ll attempt on race day.

I ran into this problem myself (my God, that pun is perfect) during my first marathon in 2008. I was confident I could run somewhere around 2:37 – 2:45, but an 8-minute spread at that level is enormous.

Instead of respecting the distance and being conservative (NYC is not an easy marathon course, after all), I had many miles under 6:00 which were simply too fast.

And the inevitable happened: I hit the wall hard, eventually slowing to a 7:09 24th mile.

I finished in 2:44:38 – not a bad debut marathon but not the time I was hoping for and certainly not how I wanted to feel during the race.

It was painful. Everything hurt. A senior citizen passed me with less than a mile to go. As you can imagine, it wasn’t the “strong” race I wanted.

But it did give me perspective. That first marathon helped my next marathon in a number of ways:

  • I took marathon fueling much more seirously
  • My pacing was far more strategic
  • Training before the race was more effective and consistent

And it worked: I ran a 5+ minute PR and achieved my stretch goal of running 2:39:32 at the Philadelphia Marathon.

Today I want to feature a question from a member of the Strength Running community. Someone who has run several marathons but always struggled with hitting the wall.

And I want your advice on how he can pace himself.

What marathon pacing strategy works best?

John has run three marathons. And each time, the first half is a lot faster than the second half. Sound familiar…?

He also hits the wall around the 19-mile mark in every previous race.

Here’s John’s question and more background information:

In 2014, my last marathon was 3:24:56. But that was with a 1:35 first half and a 1:49 second half!I recently ran 1:29:35 in the half marathon so a lot of the calculators say I should be able to run a 3:06 marathon. Is that for real?!

Even more recently I ran a half marathon in 1:26:41. If I nail down my race nutrition, what pace should I target so I don’t go out too fast and crash?

What may be most helpful for John is explaining two things:

  1. What his average pace should be for his next marathon based on the information we have.
  2. How he should execute that pace during the race itself.

He’s running the Brighton Marathon in the UK, which is fairly flat. So there’s no need to account for hills like at the Boston Marathon (which really throw a wrench into how consistently you can pace yourself).

Leave your answer below – and I’ll choose my favorite answer for a free copy of my newest book!

Was this post helpful?

Then you'll love the free email lessons I've never released here on the blog. Enter your email and you'll get:

  • The exact strength exercises that prevent injuries
  • Workouts that boost your speed (even for beginners)
  • Pacing strategies, coaching Q&A, and more


  1. I think Jon has to not get caught up in the hype of the race. remember all the training runs, and focus. Feel his pace. he should check himself at mile 3, 5, 7, 9.. is he running to hard or fast. He should know what he needs at these mile markers. if he can’t remember. he can write them on the back of his hand. That’s what I have been doing. makes it much easier to see if I’m where I should be . He has time to correct his pace and settle in early. BEFORE the big mileage happens. Keep the routine he has had for fueling during the race. When Mile 19 hits. just go with easy through it. Remember you Got this! Concentrate on the finish. Recheck how you are feeling, if its good, raise yourself a little, but don’t get cocky!! Most importantly, Finish Strong! Thanks

  2. My pacing recommendations are this. Always start out the first 2-4 miles about 15 seconds slower then your planned pace. So Looking at your half times, I would target a pace of around 7:00 min/mile. So after the first 2-4 miles of 7:15 min/miles start picking up the pace. The important thing is to use your effort to really zero in on your pace. For the first 20 miles the pace should feel easy. During the last 6 miles is when you can start racing and increase your effort to 90-100%.

    Keep in mind it is a 26 mile race, not 13, not 20 but 26. So you need to save some energy and do not go all out in the first 20. Relax and enjoy the first 20. As I am sure you have heard before, a marathon is a 20 mile easy run with a 10k race at the end!

  3. PRAVEEN KUMAR says:

    1> John’s average pace should be around 7 min/mile ( with a tolerance limit of 15 seconds).
    2> Although his past records shows that he run first half faster than the second half, but for better result he should change this strategy and try to start a bit slower and should try to complete first half in about 1:38 and second half in 1:28 targeting his Marathon in a net time of 3:06

  4. Stephanie says:

    A marathon isn’t 2 halves. So get that out of your head. Starting out at a reasonable (7:50) pace and gauging how you feel is best. If it’s a warm day, you’ll go slower and need more water and electrolytes. If it’s cooler (yay) you might be able to pick up some speed in the last 6. Pay attention to 1. how you feel and 2. your heart rate. Recheck every few miles. Hopefully you’ll feel great and can pick up the pace after 4/5 miles 730/40, then miss the wall. If you reach 18 and feel good ratchet it up a bit more 720/30. Got a cool day? Still feeling good at 20? Reach for the pain!! Now it’s a 10k, go for it. Negative splits all the way in. Please spend the 72 hours before the race hydrating and resting well in order to have lots in the tank! A few sharpening drills each day will keep your legs fresh.

  5. Cristiaen Schaap says:

    After PBing my marathon yesterday,today I went to do some comparisons. Turned out that a friend and me both PBed on a Half 5 weeks before our Full. On avg. our races times were 2 x HM + 12.30. A different one is of a friend of mine and myself PBing on a Half 2 weeks before the Marathon. Then the calculation turned out to be 2 x HM + 15.18

    So I would advice John as the following:

    1. His average pace for the marathon should be 7.02 minutes per mile (based on a 42.5 course because as you know it’s difficult to run the ideal line in marathons)

    2. Build up the pace. 7.10 in the first 10 miles. 7.03 in the next 10 miles and then 7.00 in the last 10K.

  6. John should try to run an even pace throughout the race. He already knows that starting too fast leads to an awful second half of the race. If he starts too slow he would feel much better during the race but would not be able to fully realize his potential and get the time he’s capable of. Running calculators suggest John is capable of 6:55 for FM based on his recent HM time, however his marathon fitness might not be there. So I’d recommend sticking to an even pace of 7:00 for 21 miles. If at that point John feels good he might up it a notch.

  7. Jamie Abbey says:

    It feels like John should be targeting a 7 minute mile effort. His most recent half marathon (1.26.41) suggests he has the ability to hold sub 7 pace but his last marathon indicates he struggles to maintain even pace throughout.

    Achieving a 7 min mile marathon is a perfect stepping stone going sub-3. He should aim to go out in 1.31.45 (which should be comfortable enough), then maintain an even pace to 20 miles, which he should hit at the 2.20 mark. He then has just under 44 mins to run the final 6.2.

    The reward for John’s 7 min mile effort will be automatic entry into the London marathon next year, where he can go ahead and smash his sub 3 in front of thousands of spectators.

  8. Instead of breaking it up into halves, maybe it would be beneficial for John to break it up into fourths. Since it can be difficult to hit his goal pace in the beginning of the race, since everybody is usually really excited, if he focuses on trying to stay as close as possible to his goal, for the first 6 or so miles then it will be much easier for him to naturally adjust his pace back down to what his goal is. Since it’s only 6 miles of slightly fast miles it likely wont affect him too much and hopefully he can slide into a normal pace. In the latter parts of the race he should definitely practice fueling methods so mile 19 “aint nothin but a thang”. Basically I think that if John calms down the first 6 miles, then he will have a much better time settling into his goal pace and not overdoing it to fall flat later on!

  9. Howard Elakman says:

    First you have to see the runners FORM to be sure that he is not wasting any energy. Be sure that he has a strong body and can handle 26 miles. Training for about 4 to 6 months is required. His half is about a 6:30 pace. I would slowly pump him up to about a 6:50 pace for 3 hour marathon. I like to see negative splits and would sit down and work out where to start, about a 7 minute pace, and work him down to a 6:30 pace and average it out, so that he averages a 6:50 pace.
    I like to have them run one long run every week in training, and start at 14 miles and increase 2 miles each week to 20 miles. When this is accomplished I raise it up to 16, 18 ,20, 22. I continue raising it up by 2 miles each month until he is running 26 Miles. If he is feeling strong at 20 he can increase his pace, but he should be averaging a 6:50 pace or better. It really depends on how he feels, the weather, running conditions, etc. With a 4 to 6 month training period he should have no problem averaging 6:50 or better.

  10. Matthew Parham says:

    His marathon PR is about mine, but no wall, 5 minute split instead of 14, and his half PR is 10 minutes faster. If he fuels and doesn’t go out too fast he should be a lot faster than me. If I was him I’d target 3:08:30 overall, based on the 1:29:35 half. I wouldn’t project from the 1:26 because it’s very recent and he probably trained off of the 1:29 paces. My strategy would be to go out at 7:30 for 10 miles, run target pace of 7:12 for the next 10, then finish at 6:50 for 6. Hopefully I carb-loaded and drank enough gatorade and ate several gels, otherwise that 6 won’t happen. If he’s like me he’ll need to check himself a lot those first 10 miles to avoid going too fast. My first marathon I had an hour positive split, so it could be worse. I think it’s always good to have a B goal too. If he’s my age, the B goal would be a BQ at 3:15. That way if he can only maintain pace the last 6 miles he still feels like he accomplished something.

  11. I think he could go 3:00-3:05. He needs to keep in mind the first half should feel very comfortable and treat the distance like a progressive run. Start around 7:10 pace for the first 5k, spend a majority of the race right at 7:00, and work down to 6:50 by the last 10k. Fueling is paramount to making this happen as are workouts done the same way. Fast finish long runs are a huge strength builder-run the last 5-10k at your goal marathon pace. You teach your body what that pace feels like on tired legs.

  12. Jon Niehof says:

    I’m assuming the 3:24:56 is a PR. I’m also assuming he’s had a solid marathon training build-up including 16-20mi runs, with a bit of race pace, and they’ve gone fine.

    What does John want?

    If a “safe” race is the goal, finishing strong at the distance, then start at 7:40-7:50 pace–average pace for that PR…and take it to ten miles. Should be almost trivial. Then cut down to 7:30. At the 20, take stock and either hold on (finishing for 3:19, or 7:35 average) or cut to 7:05-7:10 (i.e., the pace the predictors say you should be able to hold for the whole thing) and finish in 3:15 (7:26 avg.) Big negative split and conservative start, but if John really wants best chance of a good result, it’s a safe way to go. Gives experience holding that 7:05 pace on really tired legs and would be a good setup for 3:05-3:10 next time.

    If the goal is “best chance of hitting 3:10 and accepting the chance of a blowup” (assuming John is 35-39…) going out at 7:20 or so to the 10mi and either holding for 3:12 or cutting down to 7:10 for 3:09:30 is a possibility.

    (I use the ten as a breakpoint because it’s where the work starts. If you ask me at the twenty if I can work a bit harder, answer is almost always NO. At the ten, the question is: do I feel almost as if I just started? If yes, can probably lay more out.)

  13. I think an important factor to take into account is actual race logistics. I feel this is sometimes underestimated. When you are running by yourself, or in a training group, it is easy to maintain pace but in an actual event there are some elements that come into play. Especially if you are not an elite runner starting at the front of the pack!

    I ran the Rotterdam Marathon yesterday. Prior to that my half-marathon record time was 1:47:00, and I was comfortable running a 5min21s / km pace, which would let me finish a marathon in 3:45:00. This seemed like not a bad target.

    I had also decided to enjoy the race, and to fit my race strategy to avoid running into the wall. On the day itself, it was quite warm so I really made sure that at each fuelling station I took time to drink properly and take my nutrition. This worked well, I ran the marathon relatively comfortably, but it added about 45 seconds every 5 kilometers (6 minutes in total).

    Next to that, you also tend to start in waves based on your predicted end-time. I found that many times I had to hold back a bit, due to not being able to pass slower runners. And, at the start you lose some time due to the number of people. My estimate is that this added another 6 minutes. Brining the total to 3:57:00.

    My actual end-time was 4:02:33. The other 5 minutes l feel I lost in the last 10k due to some fatigue.

    He is more experienced than me (this was my first marathon), so let’s say he only needs to add 9 minutes. I would take the time of his calculators (3:06:00) and add these 9 to get to a “No wall comfortable end time” of 3:15:00, requiring an average pace of 4min37s/km (7min26s/mile) including time for refuelling.

    Working back from that, if he can go off on a pace of 7min6s/mile and that feels comfortable, he can finish at 3:15:00 without hitting the wall. Still a big gain on his previous marathon time, but with the added bonus of having a fun run that will inspire him for the next step.

  14. Those race calculators are great to show deficiencies in aerobic power: Especially when the data from half marathon to full marathon don’t fulfill in real life.
    A half marathon you can run on sugar/carbs only, a marathon you can’t. That’s why many good 10k/21k people do not meet their expectations in the marathon and hit the wall typically around 30-35k, as this when the sugar is depleted.
    As we know nothing about John’s training, and whether he has improved his fat burning aerobic capacity, it’s hard to give him advice on pace. I’d tackle it very very conservatively. Since it’s been 2 years since his last marathon and his overall fitness and speed has improved, a 10-15min improvement in the marathon could be possible. If he goes for a 3:15 marathon time until 32k, and if he has anything left in the tank at that point, use it in the last 10k.

  15. Indira Avila says:

    He should go out slower. Around 7:40 Pace which will feel comfortable at the beginning and when he hits 13 start working. In my opinion he should aim for a 1:40 half. Then Aim for around 7:20-7:30 pace for the last half.

  16. It depends on his goals? Is he looking for a Boston time ? I think he should start at a 7:30 pace for the first 5 then drop down to a 7:15 pace if he feels good. Then be all he can in the last 5k. I think it’s important for him to have a nice consistent race and a challenging but very achievable time instead of going for his ultimate time. Next time he can go for it!

  17. Most experienced marathoners I know use a guideline of half x 2 plus 10 minutes if you’re a sub 4 hr runner, and target a slight negative split. Based on this, and using the 1;29;xx half, I’d recommend John target a finish time of 3:10, with splits of 1;36 1st half, 1;34 second half.
    The key will be holding back in the first few miles. Keep a pace at 7:20-7:25 for the first 3 miles, and settle in at that 7:20 ish mark up until the half split. From there, pick it up just a little to a 7:10 pace.
    Anything at or below 7:00 requires a self-warning to back off and reset the pace, especially in the first half.
    Mentally refocus between 20 and 23, take stock of his comfort and energy levels, and maintain that 7:10 pace.
    If he’s feeling very good at 23, with just over 5K to go, it’s ok to trend down toward 7:00 / mile – but only if very comfortable.
    With this plan he should roll in just under 3:10.

  18. James Valadez says:

    Personally I think that he should run a conservative first half of his marathon. Looking at his split, he was too aggressive in the beginning and paid for it later. Looking at running calculators, he’s capable of roughly a 3:03. Since the marathon is so long, obviously running a negative split is the way to go. I think he should run around a 7:30 minute pace for the first couple of miles to keep it easier and then gradually speed up to his goal pace, probably working down 5 seconds/mile until he hits goal pace.

  19. What about pacing his marathon mostly by feel?
    The first half should feel good, like a 50% race effort. He should relax, and enjoy the race and people cheering at him.
    Then, from 21 to 30k, he can start to feel that he’s working harder, at 70% effort. He will not run faster, but just slow down a lot less than during his previous marathons. Same for the 30 to 40k bloc, he should feel a 90% effort. And give everything he has left during the final 2.2k.
    No stress, not being a slave to his GPS watch. He just needs to check his stats from time to time, especially at the beginning to make sure he doesn’t start too fast.
    I was running similar HM times during my last marathon build (1h25), and then ran the full in 3h04h46 using this “pace by feel” strategy. I passed the half mark in 1h31m30s, so the second half was only 2m15s slower. And the most important, focusing on enjoying the race and feeling good made this marathon a great experience.
    Admittedly, I didn’t own a GPS watch at the time, so I didn’t really have a choice. I bought one 3 months later, in order to pace myself properly for a sub 3 marathon. Long story short, I became addicted to my Garmin stats instead of continuing to run by listening to my body, and got myself injured.
    Jason, I remember you wrote a nice piece a while ago, on ditching the GPS watch and running by feel, and I think this is best done during races.

  20. Its my first marathon in less than two weeks, and as my experience is not as advanced as others but I find I can keep a constant pace through out but only if I start
    at a much slower pace. However I’m not trying to beat a PB.
    I though that in your article if the runner starts at a slower start he should be able to finish with a faster second half. (Purely because I’d lap time is so much faster than mine anyway).

  21. With a 1:26 half marathon he should be able to run a 3hr marathon if he has done enough long slow runs and total mileage during his training.
    His average pace should be 6:54 per mile. Use the first 6 miles to warm up. Run these at 7:02-7:05 per mile. Next 3 miles at 6:54, then 6 miles at 6:45 if it feels comfortable enough. Last part at 6:54. Just keep it there and try to run under 3hrs. With correct nutrition it should not be a problem to run the second part fast

  22. Not capable of 3.06 unless fitness has changed dramatically.

    Think 3.12- 3.15 is more accurate target.

    Aim for 7.20-7.25 per mile from start to finish. Go out as close to 7.20 pace as possible and hope to average 7.25 at end. Test in training- do a long tempo at marathon pace- maybe 14-16 miles and see how it feels, is it sustainable.

  23. Kelly McNair says:

    A 1:16 HM equates to a 3:01 marathon for an average pace of 6:54. However, I would not advise he go for this just yet. I would be more conservative and set him up for a marathon race based on his 1:29 HM. The estimated finishing time would be a 3:05 for an average pace of 7:05. This would still be a significant PR. I’m assuming he ran a 1:24 while in marathon training and not specifically in half marathon training. I’d like to know his total volume and the length of his long runs. Given that the upcoming marathon is flat, I’d advise that he go out at 7:15 for 3 miles or so, settling into 7:05’s come mile 4. From here, I recommend he hold pace, reevaluating at 20-22. If he has gas in his tank, he could start to drop the pace but only if he’s confident he’s not going to bonk. This was fun! Thanks for letting us give it a go!

  24. The correct answer, which I’m not seeing in any of the responses here, is to acquire more information. Just because one has run a 1:27 half doesn’t mean one is ready to run a calculated conversion, 3:05 marathon.

    When I was competing in college, I knew very little about training strategies. I just followed what I was told by the coach and this was in the low mileage era of 20-25 years ago and obviously pre-internet. After college, I decided I wanted to do a marathon and did what I was used to, 20 mpw. I ran a 1:30 half on about 20 mpw as a tune up and figured I was ready to do the marathon. My longest run in my life to that point was the half marathon. As one can imagine, it didn’t go so well. I “hit the wall” at about 17 miles (on 3:10 pace until that point) and crawled in to a 3:30 finish.

    Point being, the marathon is a totally different animal that even a half marathon. The mileage requirements to actually achieve what the calculators the poster referenced are close to 70+ mpw for most people.

    If the poster is training as I did in college on 20-30 mpw and hoping to run a marathon based on a conversion from a half marathon time on the calculators, he’s in for a world of pain.

    If the poster is training 100 mpw I would say go for that calculated time of sub 3:05.

    The goal, needs to account for the training in totality, not just previous race results at shorter distances.

  25. If he has done the distance — i.e. weekly runs around 20 – 25 miles, and his body is used to running for 3 hours or so, he should aim for a uniform 7 min/mile pace. That’s easy to remember and calculate, even when you are tired, and is consistent with his half-marathon performance. He may slow down slightly at the very end. Last year I turned 70 and adopted a similar (but slower) strategy in the Paris marathon, and managed 2:01 for the first half and 2:04 for the second, which was as close to a consistent pace as I was going to get.

  26. Ken Hurst says:

    Given his history of hitting the wall and not having run a marathon in two years, I recommend he simply try to fix the nutrition issue and not worry about time. Based on his half marathon times, he should be able to run about 7:20 per mile overall and I would recommend he do it by hitting 7:30s for the first ten miles, dropping to 7:20s for the next ten, and then if he feels able, dropping the last six to about 7:10s. If he successfully does that then the next marathon he can think more about a faster time.

  27. Mike Sampson says:

    In order to give a realistic pace to strive for I would need additional information about his training. Weekly mileage, amount of key workouts per week, long run length, and strength training all need to be accounted for. I think that strength training such as wall squats and lunges could help out alot for those hard miles towards the end of a marathon when legs start to feel fatigued. Also long runs ending in marathon pace miles will help prepare for the end of the race. With the times he has posted for the half marathon he should be able to run a sub 3:05 marathon but it sounds like his endurance is not to the level of his speed for shorter races. I would aim to break a sub 3:10 marathon with a pace of 7:15 per mile. If after 20 miles he still feels fresh I would start to speed it up towards the end of the race finishing strong.

  28. I worked with a fantastic coach who said GOAL marathon time, granted you have trained appropriately (long runs, runs for “specificity of training,” ect) can be determined by doubling your half time and adding 15-20 minutes. Based on his best half, that would be about 3:09 – 3:14. However, based on his previous “bonk,” I would definitely focus on pacing a negative split race and probably shoot for a more conservative goal, more based on what he is running in training than basing it on his half times. With a history of bonking, I’m not sure I would rely too heavily on a fast half to predict what might happen in those late marathon miles. More information would be needed to determine a pace goal – what is he running his long runs at and how does he feel during them? Does he end feeling like he can run more? Is he running long run with negative splits? Is he practicing hydration and nutrition on the long run?
    Without being able to specify a pace, I usually try to pace my marathons +10 second over goal pace needed to hit my goal time for at least the first 6-10 miles to settle in, then start whittling away closer to goal pace, and then eventually under goal pace if everything is in check. I also practice this on every single long run – not necessarily at that goal pace.

  29. Laurens Cloete says:

    John has at least two problems: endurance and pacing. Since the Brighton Marathon is this weekend he can’t do anything about his endurence now. The only thing he can do is ensure that he rests this week to get rid of some fatigue from that recent fast half marathon and to then run to his current potential by pacing correctly. For correct pacing he needs to pick a realistic goal. Although the calculators indicate that he could aim for 3:06 I don’t see him improving from a 7:49 pace in 2014 to 7:05 in one go. The calculators predict his time based on his achievement in shorter distances which doesn’t tell us if he has the endurence to run a marathon at his full potential. We don’t know his half marathon time in 2014 but assuming he could do a half marathon in 1:30 in 2014 his pacing for the marathon which saw him doing the first half in 1:35 was actually not that bad if he aimed for a 3:10 marathon finish. The problem was that although he had the speed to reach 3:10 he didn’t have the endurence. For next year’s race he should build a proper aerobic foundation by literally running hundreds of miles at a ridiculously low heart rate: 180 bpm minus his age… And then he needs to adapt his 16 week marathon training to average 26.2/3*7=61.1 miles per week during the peak about a month before the race.

    But, for this weekend I suggest he targets something slower than 3:06 and then running the first 5/6 miles at 10s a mile slower than goal pace, running the next 10 or 12 at 5s faster and then finishing at goal pace. He should also take the first mile or so very slowly. A minute lost in the first mile is not the end of the world, there are 25 miles to make that up and he has the speed to do it. The 3:06 goal may be tempting but I suspect 3:10 will be a stretch. Rather go out aiming for 3:15. If he achieves this it will be a huge confidence boost and he will have built some experience in marathon territory, which is very different from running half marathons or shorter.

  30. Laurent Kurtzemann says:


    I usually use the +1km/h rule. That is to say that you run your marathon 1km/h on average than your half marathon. So for your athlete that would translate on an average speed of 13.5km/h for the marathon. That said, his marathon will last more than 3 hours, and lots of things can happen. Moreover, given his past results and history, i would use a more consevative appraoch until teh 32 km, and run 1.5km/h slower tahn half marathon pace (i.e. 13km/h)
    That would give the following:
    until 32km, run at 13km/h
    after that, see what happens: if things are ok, speed up to 13.5km/h, and if it’s too hard , keep going at 13km/h.
    in any case, he will beat his previous time and finixh under 3h15.


  31. Alexandre Charest says:

    I face the same problem. My marathon times are always longer than what calculators predict. The shorter the distance, the better my performance. My conclusion is that I’m a speedster and I need to focus on endurance in my training to help me maintain my speed for longer. For now my marathon time is at twice my HM + 30 minutes. I would suggest that aiming for 1:40 for the first half should be ok. Continue on that pace up to mile 20 and then, picking up the pace if there is still some juice left. Or increase the effort to keep the pace and finish with a nice PR.

  32. Carl Jess says:

    The pacing strategy that has worked best for me — best being to run the fastest time I thought possible, was to run as close to even splits as I could. Making this work means a lot has to be done right. The training leading up to the race is right, for me this means I’ve done the long runs and the tempo runs. So going in I BELIEVE a fast time is possible. Then hydration and nutrition have to be nailed during the race. Given all that then yes the half marathon times you post should support a 3:06 flat marathon.

    I monitor pace based on perceived effort and break the race into three pieces.

    For miles 0 – 10 it goes this way. The first 3 – 4 miles are about settling into a pace that feels right. John has already seen for himself what it is like to fade at the end of a marathon, so I’m sure he already understands that this part of the race is important. Don’t press the pace at all here, warm up and get into rhythm. Once you get to the 3 or 4 mile point then check the pace and see where you are and make an evaluation. The effort here should feel very easy. As in “I have no doubt I can hold this to the end”. Keeping that pace should gradually get more demanding so that at 10 miles it has moved to “I can hold this pace, but need to monitor it”. Meaning that holding pace is now not completely automatic.

    For miles 10 to 20 the effort to hold that pace will increase to the point of requiring constant focus. If your mind wanders so will your pace.

    Over this last 10K the effort to hold pace will be increasing exponentially. Holding pace for this leg will take a deliberate focused discipline. Somewhere in this stage I get to the point of “scanning the systems”. Constantly checking posture (am I keeping my head up), breathing (deep breathes don’t start breathing shallowly), cadence. Don’t kid yourself about how much effort will be expended during this stage. I always feel doubt that I can hold the pace. I get through that telling myself that I did the training I can do this. I often focus on very short goals during this segment. Hold pace to the next mile marker, the next aid station, or even the next intersection, then repeat that to the finish line.

    I deliberately did not specify the target pace. The pace calculators tell you what is “possible” based on their models. Those models are pretty good, but they are statistical models. Breakthroughs do happen and so do “off” days. I know John will have memorized the pace he needs to hit that 3:06 prediction. That predicted pace may be the perfect pace for that day or he may be capable of something faster. How he feels that morning along with that days condition should control the pace he shoots for.

  33. Right there with ya’ John…I’m actually in a very similar situation! I’ve run 10 half marathons and 6 fulls. My half PR is right at 1:30, and given that I should be closing in on a 3:10 full (which would BQ me). However, my PR in the full is 3:23 and I haven’t improved in my last three tries.

    I’m pretty confident that this isn’t a matter of nutrition. My thoughts are it’s partly a pacing issue and partly a training issue. In my last three marathons I’ve went out too fast – running at paces *I think* I should be capable of, only to hit the wall HARD between miles 18-20.

    What I’m thinking for my next marathon, and my recommendation to John, is to try to run a negative split. Meaning, you run the first half of the race slightly slower than the second half. You should be thinking you’re almost running the first half of the race TOO slow. It might take a try or two (or three), but if you (and I) can’t negative split, it likely means we went out too fast. This first race of trying to run a negative split will be an investment…one that will pay off as you look to hit your sub 3:10 goal in your NEXT marathon, not this one. Remember, nobody ever said marathon improvement involves instant success 🙂

    In John’s case I’m thinking he should run the first half at 7:32 (which will seem almost too easy) and the second half at 7:22 (which will be tougher but doable if he’s gone out conservatively enough to avoid a bonk). This is a substantial, 2 minute negative split, and gets John home around 3:14. Not the sub-3:10 that he was looking for but again it’s an investment…one that will instill the confidence he needs to chip another 5 minutes off during his NEXT marathon when he hits his sub 3:10 goal. If he can pull off a 2 minute negative split in this race, he can be more aggressive and shoot for a smaller 1 minute or even 30 second negative split during that next marathon.

  34. Neither John nor his coach have enough information to make a good recommendation. No matter how trained he is for a half marathon, it’s not clear he is trained for a full. However, assuming he has followed at least some sort of full marathon training ramp-up…

    John should schedule TWO marathons. The goal of the first should be to achieve a negative split–and run the first half slow enough to guarantee that. That’s probably something like a 5:00/km pace for John. Not only will that give him confidence that he doesn’t slow down at the end, it will give him information to pace his second, really-the-target marathon.

    John can then gauge how he feels when he finishes. Assuming he’s pretty tired in achieving a negative split (i.e. couldn’t keep that pace for another km), he can then use his second half pace as the target pace for his next marathon!

    If he doesn’t have anything other than the target marathon, he can do this in a training run, but it would need to be at or close to a marathon distance to be sufficiently meaningful.

    Best wishes, John!

    (We should all know what pace we need to run the first half of a marathon allows us to run the second half at the same speed or better.)

  35. John, it sounds like you need a pacing buddy to keep you honest. Find someone who is willing to run the race with you for the sole purpose of helping you stick to your plan. And your plan needs to include a goal pace for the first few miles (slower than goal pace), for miles 5-20 (your goal pace), and the last 6.2 (just fast enough to make up for the lost time in the first few miles). Don’t forget about fueling in your race plan! How many energy gels will you consume and when? About how much water do you need? If your pacing buddy knows your plan, then he/she can prompt you, keep you from going out too fast, and feed you an energy gel if you’re about to bonk. Good luck!

  36. Marathons are difficult things to run – as I’m sure John already knows – and there are so many factors to consider (and some you probably won’t/can’t) that race time predictors in my experience are not very helpful; they are likely to give you unrealistic goals.

    I’ve often had good 20 mile races a few weeks before a marathon but not been able to replicate either that pace or strategy over 26.2 miles.

    My advice: go out steady at a pace that should feel be comfortable. I would suggest aiming for 7 minute miles which should feel pretty easy. Personally I wouldn’t use a pace band, on 7s the maths of the miles should be easy to do: 7, 14, 21 etc… That’ll take you through halfway in around 1:30; if you’ve run a 1:26 (6:35s) half you should have some good leg speed to make up some time at the end if you are feeling good in the final five or six miles.

    But importantly listen to your body – you’ve spent lots of time training so you should hopefully be well in tune! A good, well practised nutrition and hydration strategy will help you for the final miles but a marathon is a long time for things to go awry. If all feels well focus on your breathing and enjoy it. Best of luck!

  37. Assumptions: He has improved his endurance including at least 5 20 mile long runs in the last 20 weeks of preparations including a final long run three weeks prior to the race of 21-22 miles (I am swedish so I will discuss the rest in kilometers). The final long run(s) ideally included marathon pace running in the end. These long runs should not leave him sore or depleted from energy for days

    He has done regular marathon paced tempo runs up to half marathon distance as final “exam” to increase running economy at target pace.

    He has done regular speed work faster than marathon pace at least once a week (not faster than 21-10k speed) during the final 8-10 weeks leading up to the marathon.

    He has a solid fluid intake strategy where he knows what works for him and sticks to that schedule and those products.

    My experience from 8 marathons (including two with negative splits) and a PR or 2:50 (four times sub3hrs). I think he should aim for running the first 5 km’s about 20 seconds slower than his overall target pace, aim to hit the halfway point in 1:33-1:34. A sub 3hr marathon requires repeated stability of 1:25 half marathons without maximum effort so forget 3hrs for now. Aim for steady pace and I think there is a good likelyhood he can do sub 3:10.

    In 2010 I ran a 3:09:46 marathon hitting 1:35:51 halfway by the way, The goal was 3:15 but I felt so strong at about 23-25k so I managed to increase my pace on the second half. Best of luck to John!

  38. Going out in a 1:35 first half is pretty conservative if you’re capable of running 6-8 minutes faster for a standalone 1/2 marathon. If John is consistently running big positive splits and crashing before mile 19, maybe it’s not bonking. Maybe it’s the fact that he hasn’t trained to run with sustainable speed.

    Needs to work on pace during training runs of 20+ and do some fast finish 20-22+ milers with the last 3-4 miles on the track. Or run a ten mile race following a 10-12 mile training run so he can get pace down while he dials in nutrition in a race-specific environment.

    As far as pacing – he is very capable of running an easy 1:35 first half (staying comfortable under 7:20’s), and then coming home in 1:33-1:34 (7:10-7:15’s until the last 2 miles then dropping the hammer if it’s there.) That way he’s safely under 3:10 and feeling good. And if weather and his legs permit, he’s got a nice sub 3:05 Boston Qualifier.

    Crush it, John!

  39. Michael Green says:

    Congrats on half PR! Good job! Although race predicting is fun to look at, strongly recommend considering current marathon PR.(3:24). I would recommend setting a more conservative goal time/PR. 3:15 looks possible! . as far as pacing, the obvious starting out slower. Generally since we talk averages, some miles slower some faster (10_15 sec either way). Question 2. Set Garmin up as intervals for each mile split. This way you know each predicated mile pace and can run it 1 mile at a time. Good luck .

  40. Thank you Jason for putting my question out there and for everyone who took the time to share their wisdom. My marathon was yesterday so I thought I would report back on how it went…
    A couple of replies were asking about training – My plan started in January at 40 miles a week and peaked at just over 60 miles/week – with my 2 longest runs at 22 miles – I didn’t do much cross training or any clever training runs – more just tried to get in a mix of distances and speeds.
    I’d decide to aim for a 3:15 marathon, sub 3:20 would give me a qualifying time for London for my age group, and thought I would ensure my pace for the first half didn’t go any faster than 7m15s per mile (=3:10 marathon). I wanted a pace that I was fairly certain to be able to maintain to the end rather than blow up as in previous attempts at mile 19. So looking for an even pace through the race, with the option to speed things up if it went really well.
    My other strategy was to consume loads of fuel through the race – in my previous attempt I estimate I consumed around 300kcal, this time I got through 773kcal (a combination of energy gels and jelly babies).
    So on to the race – A frosty start to the day, but the sun came out and temperatures reached 10C/50F with very little wind and no rain – perfect. About 11000 runners and huge crowds lining the route.
    I tracked my pace on my Garmin – I hit target pace from the off (despite a slight climb at the start) my fastest mile was 7:05, slowest 7:23 – I was only 14 seconds slower on my second half of the race – coming in at 3h 10m 11s. The pace was reasonably comfortable throughout, I was consciously holding back over the first quarter when I saw the pace get too fast – so I think that helped keep something available for later.
    So a huge confidence boost to be able to run the whole race without hitting the wall or slowing down – I now know it can be done – assuming I do another (London 2017?) I may be tempted to push a little harder maybe aim for 3:05.
    If anyone fancies a marathon in the UK, I can thoroughly recommend Brighton Marathon, its well organised and a great route through the city and along the coast – entries now open for 2017.
    Thanks for all your tips – I hope my experience helps some others to achieve their goals.

    • Congratulations, John! Great race and very consistent pacing. Definitely 3:05 is achievable, good luck!