This summer I tried something new: running three 100-mile races in three consecutive weeks. The seed for this endeavor was planted some six months ago, when Mike Day made the observation on Ultra Adventures that the Massanutten Mountain Trails 100 (MMT 100), Old Dominion Memorial 100 (ODM 100), and the Old Dominion 100 (OD 100) were scheduled on consecutive weekends in 2007. A unique opportunity to run three 100-mile footraces in the Massanutten mountains of Virginia was too irresistible to pass up. This also posed an interesting challenge, as I have never raced the week after a 100 — never mind doing three 100-mile races back-to-back-to-back.
In this report, I examine the three Virginia 100s — coined the Virginia Triple — by comparing their respective altitude profiles, finishing times, and my physical performance (by analyzing recorded heart rate data).
Click to enlarge.
Altitude and heart rate profiles
The figures below display altitude and heart rate profiles, recorded with a Polar S625X heart rate monitor, for each of the 100-mile runs. The altitude data (shown in blue) and the heart rate data (in orange), plotted as a function of time, provide an illustrative snapshot of each of the races. Massanutten, for example, has significantly more climb than the two Old Dominion races, with a correspondingly slower finishing time (21:18). Old Dominion Memorial, the flattest course, is also the fastest (17:46), and that was on tired legs.
Click to enlarge.
Altitude and heart rate versus time
Altitude and heart rate profiles recorded during the three consecutive 2007 Virginia 100-mile runs. Altitude is in feet and heart rate is in beats per minute (bpm).
The following table summarizes each of the races.
|Data from the races of the Virginia Triple|
|Mean heart rate
Altitude profile comparison
Most people are interested in the altitude profiles and, in particular, how they stack up from race to race. The graph below compares altitude, plotted as a function of running time, for the three Virginia 100-milers. The relative profiles and finishing times vary significantly, and it is remarkable that these races are located in the same mountains with overlapping routes.
Click to enlarge.
Altitude profiles of the Virginia Triple
Massanutten has 19,500 feet of climb, Old Dominion Memorial has 10,680 feet of climb, and Old Dominion has 13,460 feet of climb.
Not shown in the altitude profiles (but reflected in the time to finish) is the course terrain. Massanutten is on very rocky trails, while Old Dominion Memorial is almost exclusively on country roads. Old Dominion has a mixture of trails and roads. These factors have as as much — if not more — influence on the finishing time as the total altitude climb discussed above. I ran each of these races fairly well, with no significant bonks or “down time.” The finishing times are therefore good indicators of the races’ relative difficulty, with Massanutten being the most difficult and Old Dominion Memorial being the easiest.
Heart rate comparison and trends
Let’s now look at the heart rate data. In the profiles above, it is apparent that I generally go out hard (near or slightly above anaerobic threshold), which is followed by a steady decline in heart rate beginning at approximately five hours and continuing over the next 10 or so hours. The last third of each race (with exception to the finishing kick) is run at a relatively lethargic 120 bpm, or roughly 62% of my maximum heart rate (MHR). I went out much harder at Massanutten than I did at the others (being the first in the series, this was run on fresh legs), while the start at the others (Old Dominion Memorial, in particular) was somewhat more conservative.
Heart rate histograms
Click to enlarge.
Heart rate histograms
Histograms of heart rate data from the 2007 Massanutten, Old Dominion Memorial, and Old Dominion 100-mile runs. Each race exhibits a bimodal distribution in heart rate. The vertical lines indicates the mean for each curve.
A more quantitative comparison of the exertion during each of the races is made by considering the distribution of the heart rate data, shown in the histograms to the right. During Massanutten, my heart rate reached 180 bpm (92% MHR) during the earliest sections, with a mean heart rate of 136 bpm (70% MHR) for the entire race. At Old Dominion Memorial, my heart rate never exceeded 160 bpm (83% MHR), with a mean of 132 bpm (68% MHR). The pace at Old Dominion was somewhere in between, with a maximum heart rate of 170 bpm (87% MHR) and a mean of 133 bpm (68% MHR). In all three races my heart rate seldom dipped below 110 bpm (56% MHR).
What is most striking from these data, however, is the unusual distribution in heart rate for each race. Rather than a Gaussian (a.k.a. “bell curve” or “normal”) distribution, the heart rate data are distributed bimodally, with a peak at a relatively high heart rate (whose actual value depends on the race) and a second peak centered near 120 bpm (62% MHR).
The upper peak reflects the intensity during the early protion of the race when things are fresh. For Massanutten, this peak is relatively broad and is centered at 158 bpm — a mere 10 bpm below my estimated anaerobic threshold. At Old Dominion Memorial the upper peak is situated much lower, near 140 bpm (not much higher than the overall mean), reflecting an overall lack of energy after the previous weekend at Massanutten. I was apparently more recovered for Old Dominion, as evidenced by the fact that a fair amount of the race (although not nearly as much as Massanutten) was run above 150 bpm.
The lower peak in the heart rate distribution, centered about 120 bpm, corresponds to the pace during the latter third of the race. The heart rate converges to this 120 bpm pace for all three races, regardless of how hard I went out initially, the race terrain, weather, or degree of rest (I was well-rested for Massanutten; I was not for the others). Similarly, the overall mean heart rate for each race is relatively constant (132–136 bpm).
The steady decline
Click to enlarge.
Hourly mean heart rate versus time for the first 17 hours of each race. The horizontal lines indicates the mean for each curve.
The figure to the right displays hourly mean heart rate versus time for the first 17 hours of each race, which is about how long it takes for the heart rate to settle to the ultimate 120 bpm pace. Surprisingly, the rate of decline is relatively constant for each race. Going out hard or easy, on fresh legs or tired, has apparently little influence on the decline in pace.
This graph illustrates another point. Comparing the overall mean heart rate, as I have done in the table above, is misleading since Massanutten took much longer than the others. Nearly all of Old Dominion Memorial (17:46) is represented in the graph, whereas there are more than two hours omitted for Old Dominion (19:21), and over four hours missing for Massanutten (21:18). Most of this missing data represents time spent at 120 bpm, thereby reducing the overall mean heart rate reported for the longer races. The mean heart rate for just the first 17 hours, a fairer comparison, is also shown in the graph; it is clear that Massanutten was a significantly harder effort than the others.
Thoughts and discussion
Going out too hard?
It is apparent in the heart rate profiles above that I go out rather hard. Consider the heart rate profile for Massanutten. During the first four hours, most of the pace was above 160 bpm (82% MHR), with the pace steadily declining thereafter. At high heart rates (above, say, 150 bpm), the law of diminishing returns takes over, where a lot of energy is expended for little gain in speed. Rather than working to maintain a marathon-like pace for what ultimately amounts to a very small fraction of the race, a more conservation start might delay the decline in heart rate that inevitably occurs later on in the race. This logic is challenged, however, considering the rate of heart rate decline, which indicates that a slower start has little effect on the pace observed later in the race. While I may have gone out a little too hard at Masanutten, I “blew up” no faster there than I did at the other races. Indeed, that my heart rate at Massanutten was significantly higher than that of the others indicates a much stronger race overall, in spite of the harder start. The fact that Masanutten took so long (21:18) testifies to the difficulty of the course.
Can’t keep it up
As discussed above, the bimodal heart rate distribution arises from the two paces at which each race is run — the fast initial pace (which varies, but is typically 150–160 bpm, 77–82% MHR), and the slow (120 bpm, 62% MHR) pace maintained over the latter third of each race. A more normally distributed heart rate distribution, where the data are centered about the mean, would seem more efficient physiologocally. This can be achieved by not going out so fast in the beginning, or not slowing down so much at the end. Since the initial pace has little effect on the rate of heart rate decline or the bimodal nature of the distribution, the key is not slowing down.
These data point to a fundamental difficulty in training for 100-mile races. Longer training runs would surely prevent the decline in pace, but training runs longer than 12 hours or 50 miles are difficult to pull off. A standard “long run” in a typical marathon training plan is 21 miles, or 80% of the race distance. Ultrarunners seldom go beyond 50 miles, except during 100-km or 100-mile races. A 100-miler is therefore a journey of sorts into “no man’s land.” Little surprise, then, that performance fades after half a day of racing.
Let’s also not lose sight of the fact that I was passed by few people (only one, actually — Todd Walker) during the last half of these races, so I couldn’t have slowed down that much relative to the field. I suspect this behavior in heart rate is rather typical, and a comparison of other peoples’ heart rate during a 100 would be interesting indeed.
Is speed work necessary?
Since so much of a 100-miler is run at 120 bpm (62% MHR), well below anaerobic threshold, it is easy to discount the importance of speed work in training. Speed work, which in this context refers to tempo runs near anaerobic threshold, improves aerobic capacity and running efficiency. The benefits in preparing for a 10-km or marathon are obvious, since these distances are run near anaerobic threshold.
In a 100-miler, this type of training has both physiological and physochological benefits, and is essential for running a fast 100. Training at threshold ensures that plodding along at 120 bpm is still a repsectable clip. Moreover, if the body is made accustomed to running fast in training, 120 bpm in a 100-miler will seem easy. As David Horton says , “the more you can trash your body in training, the more you can trash your body in the race.”
Altitude and heart rate profiles from three consecutive 100-mile races, the 2007 Virginia Triple, are presented and compared. The main points:
- While sharing the same geography, the Massanutten Mountain Trails 100, Old Dominion Memorial 100, and the Old Dominion 100 are very different races, as reflected in their altitude profiles and disparate times to finish.
- Despite their differences, the heart rate behavior in the three races is remarkably similar:
- A fast start (near anaerobic threshold, 85–90% MHR) is maintained for less than five hours. A steady decline in heart rate ensues over the next 12 hours, ultimately converging to 120 bpm (62% MHR).
- There is little variation in the rate of decline, despite variable environmental (particular course, terrain, weather) or physical (prior exertion during the race, degree of rest before the race) factors.
- Going out hard, and then easing off, results in a bimodal distribution of heart rates.
- In a 100-miler, course difficulty (e.g., amount of climb, quality of footing) is the overiding factor in dictating finishing time. Interestingly, the race with the hardest effort, Massanutten, was the slowest. The race with the easiest effort, Old Dominion Memorial, was the fastest.
- Much of a 100-mile race is spent running at ca. 120 bpm, well below anaeorobic threshold. The goal, then, is to be well trained so that 120 bpm is still a fast running pace.
- “Q&A with David Horton.” Trail Runner (July/August, 2000), page 16.
Link to pdf (1.5 MB)