Optimal pacing for an ultra marathon | Q&A w/Bob Shebest


A graph of my pace and heart rate during the 2015 TNF 50 Mile Championships. Notice the steady decline of my pace and heart rate throughout the race.

A graph of my pace and heart rate during the 2015 TNF 50 Mile Championships. Notice the steady decline of my pace and heart rate throughout the race.

What is the optimal pacing strategy for an ultra marathon, especially for a mountainous trail course? Since Run Rabbit Run 100 last month, this question has been of great interest to me: by being more conscious of it than other elites, I think I finished better than I should have given my talent, fitness, and ultra experience.


Bob Shebest at the Run Rabbit Run 100 check-in

An hour ahead of me, Bob Shebest (http://smoothflow.org) used the same approach and had similar success, with a second-place finish. In this Q&A, I wanted to dive much deeper into the topic. It does not have all the answers, but there is some valuable insight and advice.

Bob is a 41 year-old ultra runner, triathlete, and coach based in Santa Rose, CA. He has won the San Diego 100, Tahoe Rim Trail 100 (twice), and Pine to Palm 100, and is a four-time Hawaii Ironman finisher. His performances are more impressive given that he is a full-time sixth grade math and science teacher.

If you have additional questions for Bob or me, please leave a comment at the bottom of the page.

Q | Prior to Run Rabbit Run 100, you had competed in dozens of high-stakes ultras and triathlons. With this extensive racing experience, not to mention all your training, I would assume that you know your body very well. Why then did you wear a heart rate monitor for the first 42 miles?

A | I deliberated on the use of HR for RRR but decided it would be the smart thing to do, especially with the higher overall elevation and its unknown influence on my running. Because of my depth of racing experience, I believe in using HR to help hold me back early in longer A-priority events.

Having run Tahoe Rim Trail 100 on four occasions, I discovered how hard I could push the first half in order to minimize slowing in the second half. For me running at an average HR in the low 140s always seems to allow me to run well later.

I knew going in, that RRR would be the most competitive 100 that I’ve run. Therefore, using the HRM was even more important, because without it, it would be all too easy to go out too hard if I could not see that my HR was at unsustainable levels, specifically the high-140’s/low-150’s. It might feel sustainable, but I would very likely suffer a marked decrease in performance later. Using the HRM allows me to write a check that my body can actually cash.

Q | In retrospect, was having the HR data valuable to you?

A | Pacing off HR early was definitely worth it because it set me up to race well later. I generally only use the monitor for the first third to half of the race. By mile 42 I felt in a good place to start racing off of feel. The HRM had served its purpose.

My average HR at that point was 144bpm, my highest average HR at that point of any 100 I’ve done. So, I did push it a bit, but with low temps, and excellent fitness, I felt confident I was still running within myself. Had I not worn the monitor though, I can easily see how my average HR could have been 4-5 beats higher, and that would have resulted in significant slowing in the second half of the race, or worse.

Q | Based on HR data from prior 100-mile efforts, you had a goal HR for the early stages of RRR100. But what is your recommended pacing strategy for a runner who lacks these data points? Is there a method for determining a sustainable effort based on less relevant HR data (e.g. from shorter races or training runs), or based on “how you feel”.

A | A runner who doesn’t have reliable heart-rate data to use would be wise to still wear the monitor so they can capture the average HR from their race so they have that information moving forward. They wouldn’t even have to look at HR during the event.

While the athlete is in the process of creating optimal HR zones, s/he can continue to use rate of perceived exertion (RPE) as the pacing guide. Breathing must be kept in check on all the early climbs in order to stay in control and run within one’s self. Running quickly on the downs is a key to saving energy and banking a lot of “free speed.” At the start, you can key off other runners around you (and your ability) who you know always pace well.

Finally, I find it’s a good idea to not look at pace until much later in the event, after you’ve run a controlled pace for a long period of time. For example, in a 50k, I don’t use HR, and just pace off RPE, then flip over to race pace at about mile 20, to see how close to target pace I am then really push that last third to the finish.

Q | If you start too hard, why do you lose more time late in the race than you gained early in the race? And if everyone else around you is running more slowly, too, does it really matter?

A | Fatigue catches up to everyone, but to varying degrees. I learned in Ironman racing, it’s not about running the fastest marathon off the bike, it’s about who slows the least. When looking at data from 50-mi events, for example, I’ve compared by what percent I’ve slowed in the second half versus my competitors. Clearly, the folks who are doing the best are slowing the least.

In 100-mi races, I’ve simply concentrated on having a large body of aerobic work in the bank going into the race, so that I’m able to run quickly at an optimal aerobic heart-rate throughout the first 70 mi and still have enough in the tank to make an effective push to the finish.

Q | Earlier you mentioned rate of perceived exertion (RPE), but suggested it as a backup for heart rate data. Why is not as reliable in achieving peak performance? For example, can it be more influenced by race conditions (e.g. temperature, humidity), altitude, and emotional ebbs and flows?

A | I believe RPE is the gold standard by which we should all race. After using HR to learn how to get the most out of myself in terms of pacing, I’ve now been trusting myself more with just RPE for a large majority of my training and racing, including a win and course-record at San Diego 100 back in June (first time without an HRM in a 100).

Because Run Rabbit Run presented two new variables for me (higher overall elevation and the most competitive field I’ve raced against at this distance) I knew it would be especially easy to go out too hard, especially starting out on a 3mi black-diamond ski slope! The HRM helps me more effectively listen to my body early.

Where RPE has more subjectivity involved, HR is cut and dry—it quantifies the effort. I need that feedback sometimes. Once I got to mile 42 and my average HR was well established at 144bpm, I knew I’d done a good job pacing. And there was still over 10hrs remaining to race off RPE.

Q | We’ve identified three ways to measure pace: time splits, HR, and RPE. In no case have you advocated keeping them even for the length of the race. For instance, if you want to want to run a 16:40 100-miler, you haven’t advocated starting the race at 10 min/mile pace and trying to hold it for the next 100 miles. Similarly, if you think you can average 130 bpm for a 100-miler, you haven’t advocated locking it at at 130 bpm from the start. And of course with RPE, a steady effort is not conducive to a peak performance, e.g. finish as casually as you started. Do you have reason to believe that going out slightly unsustainably will achieve better results than a steady effort throughout, or are you just trying to be a practical racer and acknowledge that it’s too hard to race evenly?

A | Anecdotally, I’ve found a race process that’s consistently produced effective results: holding a relatively even effort for the first 50-70% of the race—on the ups, downs, and flats—that correlates with a HR range of approximately 140-147bpm. Doing so I seem to be able to dramatically reduce my percent slowing in the second half just by maintaining high cadence, staying positive, and pushing hard.

Do I sometimes wonder if I’d enjoy even faster times if I kept my HR in the 130s the whole time? I have. I also know that doing well in ultras is as much about racing tactics as it is about “racing within yourself.” Alex Varner, coached by Jason Koop, wrote in his record-breaking Lake Sonoma 50 race-report back in April, “I figured I’d try a new tactic and go out a bit more aggressively, try to keep myself near the front, and hang on as long as I could because those last 10-15 miles were going to suck no matter what and I might as well be near the front if I could.

Varner, a highly successful, experienced runner had the capacity to put out good power for the six+ hours he needed to get the job done. Rob Krar does the same thing — he remains conservative through about the half, and then gives it everything he’s got to the finish.

I feel it’s in our best interest to keep experimenting over time so we can more deeply internalize just how to get the most out of ourselves on race-day. “The will to win means nothing without the will to prepare.”

Q | Why do the DNF rates among elites seem so high? At RRR100 and UTMB, it was around 50 percent for the men. That’s higher than the overall field, despite the elites being the most capable of finishing the race. It seems like pacing must play a role — it seems like they all want to race at the front and hang onto the leaders, even if they shouldn’t be.

A | DNF rates are high for a variety of reasons. We’re often still dealing with issues from the last race (3-5 weeks ago in many cases) or we just lack the training specificity required to be successful, i.e. muscular endurance for 100-milers. The personality types that find themselves competing as elites are obviously a hyper competitive lot, and very hungry for results. We like dicing it up on the front. We feel we need to be there. That is, after all, where the action is.

Sponsors also influence our behavior, as does prize money. We want to put on a good show and be perceived as strong and deserving of sponsorship. We’re hopeful things will pan out in our favor and we’ll have that extra gear at the end. And if we’ve done the work, and we’re not still cooked from the last race, we’ll likely have that gear. It’s my experience that nine times out of ten we get the race we deserve.

Posted in on October 7, 2015
Tags: ,


  1. Ultramarathon Daily News, Wed, Oct 7 on October 7, 2015 at 7:39 am

    […] Skurka and Bob Shebest talk about optimal pacing strategies. These two guys know what they’re talking […]

  2. Paul on October 7, 2015 at 9:37 am

    Great Q&A, thanks man!

  3. ken michal on October 7, 2015 at 10:33 am

    Thanks for the interesting interview!

    From the interview, it seems that you’ve found your HR zones by feel and that 140’s are more an arbitrary number instead of based on your max HR… Is this true? If not, what is your max HR? Do you stay in the same zones early on in every ultra distance or do you slow things down even more for 100’s?

    All Day!

    • Bob Shebest on October 7, 2015 at 11:46 am

      Ken – Hey, good questions. As you know, so much of trying to improve in ultras is about good ol’ trial-n-error. I’ve learned that running ultras in the low 140’s is, currently, sustainable for me. It seemed at Run Rabbit Run that excellent preparation and cool temps, allowed me to push a little bit harder early. Perceived exertion is still very much a part of my race process.

      “MAF” heart-rate is generally considered to be about 15-20 beats below your lactate threshold (Zone 2). I have no idea what my maxHR actually is (I did see 168bpm one time this year at the end of a hard hill session though) but I do know I generally run my tempo runs in the 140s and my hill sessions in the 150s to low 160s. For my age, this is about right. Running ultras in your optimized Zone 2 burns an good mix of fat and carbs, thereby sparing quite limited muscle glycogen, while permitting the stomach to function well enough to continue processing calories (while running!).

      It’s funny how my avgHR is very similar across 50mi, 100k, and 100mi, but in general, I feel you should be able to hold slightly higher HRs as the distance gets shorter and speed increases. The trick really is to keep the HR down on the climbs and higher on the descents. Gotta bank all that FREE speed Ken!!

      • Bob Shebest on October 7, 2015 at 11:51 am

        Addendum to above: Max Aerobic Function (MAF) HR is generally considered Zone 2. It’s an “all day” fat-burning zone. Whereas running at lactate threshold is often considered Zone 4, where you’re burning a lot more muscle glycogen, which we just don’t have near as much of, in terms of energy, as we do in our almost limitless fat stores.

        • Michael Heimes on July 17, 2018 at 5:37 am

          Find this oldie but goodie! Bob, did you mistype 140 when referring to your tempo? Maybe your definition of tempo is different than mine. My tempos typically climb from high 150s to even close to 170. Tempo effort has always been described to me as an effort that could be held for an hour if there was an hour-long race. So faster than my marathon pace by about 15-30 seconds per mile. With that said, for you to run Ultras at the same effort as tempos didn’t make sense to me.

  4. Daniel on October 7, 2015 at 12:53 pm

    Thank you for the great interview!

    Could you explain a little further about the average HR? When you aim to achieve an average HR of ~140bpm how does that play out in different terrain and how do you keep to that goal based on the information your heart rate monitor is showing?

    In a race I also try to stick in an aerobic zone based on heart rate. I show the current HR on my watch and have an upper limit of 150bpm. On climbs and flats I keep an eye on that and try to stay under the limit and on descents I run as fast as I can. I’d appreciate hearing your approach so I can make some refinements to my own.

    Thanks again!

  5. Bob Shebest on October 7, 2015 at 6:55 pm


    Hey, thanks for the question. I noticed at two successful 50milers with 10,000′ of cumulative gain that my average HR for both was 142. At my next 100 in summer 2013, I decided to use 142 as a ceiling HR. The only data I had on my watch for the first 50 was current HR and avgHR. You want to see that the HRM is giving you good data (current HR) and you want to keep an eye on how your race is progressing in terms of effort (avgHR). In that event, I came through the half at an avgHR of 139, which, as you’ll see is damn close to 142. Keep in mind that fitness also plays a role in how high you can sustainably push the HR. For 100s, I tend to be the most fit I am all year, so it stands to reason I was able to be close to 139 and still only slow by about 10% in the second half. At the same race, a year later, with greater volume of training, I was able to run through the half at an avgHR of 141 (that was about a 20min difference at mi50 over the year before. Again, only slowing by about 10% in the backside of the race.

    Being a strong downhill runner must be a priority. Anybody can push high HRs on the climbs, and most do. Then you see a lot of folks recovering on the descents and HR drops into the 130s or even the 120s. If you’re on a runnable section of down you want to try to stay within that assigned 7-10 heart-beat zone you’ve established for yourself. Steady and even on the ups. Same on the downs.

    Peak performance in racing directly correlates with the body of work you did for the event. Superior muscular endurance is the name of the game. 7-8 weeks out I like to be coming off a strong race, have about 2 weeks of rest in me before starting to 7-8prep to an A-Race. This way, your both fresh, and fit for the training. Then, I like to do 2-weeks ON, and 1-week BIG recovery. If I can do 2-3 two-week training cycles with something like 80-100mi with a lot of climb for mountainous ultras, then a very chill two-week taper, I always seem to arrive to the starting line excited and confident. Then, it’s just a matter of managing emotion and racing smart early. The training’s in the bank. Metabolic efficiency has been optimized through a volume of specific work (and a lower carb, higher fat diet). In the training weeks, having a solid hill session with longer, zone 4 intervals, a longer zone 3 tempo run, and, if you can manage it, two back-to-back long runs on Sat/Sun, really does the trick on race-day, making you bullet-proof in all aspects of the race.

    Good luck out there!!

    • Daniel on October 8, 2015 at 12:46 am

      Thank you, Bob, for taking the time to provide such a great answer. Your explanation of how to use HR really helps a lot and gives me a good idea of where I can improve on what I’m currently doing (downhill pace needs to be a focus).

      Also, thanks for the advice on training. Very useful and more than I expected based on my questions!! I appreciate it.

Leave a Comment