# Modeling: optimal marathon performance on the basis of physiological factors

> An annotated edition of this paper on Fermat's Library, where scientists and readers explain its ideas, equations and history in the margins. Annotated edition: https://fermatslibrary.com/s/modeling-optimal-marathon-performance-on-the-basis-of-physiological-factors

- **Authors:** Michael J. Joyner
- **Topic:** Biology
- **Original:** https://s3-us-west-1.amazonaws.com/fermatslibrary/uploads%2F1502011581693-elc6sv3w8c-e5e1dcf1a7e5d1c7031f0059ae53b894%2F5d7becac8e72576c6e9a828a6315982c16b7.pdf
- **Pages:** 5
- **Annotations:** 11 comments in 10 threads
- **Featured:** Fermat's Library Journal Club
- **Annotated edition:** https://fermatslibrary.com/s/modeling-optimal-marathon-performance-on-the-basis-of-physiological-factors

## Annotations

### Annotation 1 (page 1)

Beside the passage: “JOYNER, MICHAEL J. Modeling: optimal marathon performance on the basis of physiological factors. J. Appl. Physiol. 70(Z): 683-687, 1991.-This paper examines current concepts concerning “limiting” factors in human endurance performance by modeling marathon running times on the basis of various combinations of …”

**Annotation** by Luís Batalha (2017-10-15):

> Marathon records have been dropping consistently over the past years and there's a lot of discussion on whether we are approaching some sort of biological limit (a lot of people believe a sub-2 hour marathon is impossible) or if it's still humanly possible to keep beating records.
>
> ![](http://www.sciencemag.org/sites/default/files/styles/inline__699w__no_aspect/public/marathon_graphic.jpg?itok=t0aaf8-5)
>
>
> In this paper Michael J. Joyner models the marathon running time according to several physiological factors and is able to predict a time of 1:57:58 under limit conditions.

### Annotation 2 (page 1)

Beside the passage: “tained in competitive events of 2-3 h. Running economy then appears to interact with Vo2mar and blood lactate threshold to determine the actual running speed at lactate threshold, which is generally a speed similar to (or slightly slower than) that sustained by individual runners in the marathon (14).”

**Annotation** by Luís Batalha (2017-10-17):

> On May 5th 2017, Nike organized a marathon to [attempt to beat the 2 hour mark](https://www.wired.com/story/nike-breaking2-marathon-eliud-kipchoge/) (Breaking2 project). The Nike Sports Research Laboratory worked for 3 years with 3 of the best marathon runners on the planet and planned a race under ideal conditions: special shoes to improve running efficiency, flat track without many sharp turns so that athletes wouldn't lose time on curves, ideal temperature and even finely tuned the configuration of the group of athletes to minimize wind drag.
>
> [Eliud Kipchoge](https://en.wikipedia.org/wiki/Eliud_Kipchoge) ended up running the marathon in 2:00:25, only 25 seconds away from breaking the 2 hour mark (due to the controlled conditions of the race the record is unofficial).
>
> If you want to learn more about this experiment watch this video
> [![](https://i.ytimg.com/vi/c6FS3D4a_kA/hqdefault.jpg)](https://www.youtube.com/watch?v=c6FS3D4a_kA)

### Annotation 3 (page 1)

Beside the passage: “JOYNER, MICHAEL J. Modeling: optimal marathon performance on the basis of physiological factors. J. Appl. Physiol. 70(Z): 683-687, 1991.-This paper examines current concepts concerning “limiting” factors in human endurance performance by modeling marathon running times on the basis of various combinations of …”

**Annotation** by Luís Batalha (2017-10-15):

> Currently the World Record is 2:02:57 and was set by [Dennis Kimetto](https://en.wikipedia.org/wiki/Dennis_Kipruto_Kimetto) in 2014.
>
> ![](http://running.competitor.com/files/2014/09/Kimetto202-600x338.jpg)

**Reply** by Tareq Yousef (2020-09-05):

> "The current world record time for men over the distance is 2 hours 1 minute and 39 seconds, set in the Berlin Marathon by Eliud Kipchoge of Kenya on 16 September 2018, an improvement of 1 minute 18 seconds over the previous record also set in the Berlin Marathon by Dennis Kipruto Kimetto" - wikipedia

### Annotation 4 (page 1)

Beside the passage: “where LT is lactate threshold, RE is running economy, and VO, is 0, uptake. This model is then used in conjunction with a range of well-accepted values for these variables in elite male distance runners to estimate the physiologically optimal marathon performance.”

**Annotation** by Bruno Santos (2017-10-18):

> We make ATP in a three-step process: Glycolysis, Krebs Cycle, and Electron Transport Chain (ETC). The products of Glycolysis feed into Krebs, which subsequently feeds its products into ETC. When one glucose molecule is broken down completely, a small amount of ATP is made during both Glycolysis and Krebs, but most of the ATP is generated at the end of ETC. The drawback is that ETC is much slower than Glycolysis. ETC is very effective at making enough ATP to sustain low to moderate intensity exercise. When intensity is high we need more ATP that ETC can produce as its maximal output.  When the concentration of lactate in the blood starts to climb, our brain senses this and we start to feel nauseous. Within a few minutes we are forced to drop the intensity, ATP demand reduces, Glycolysis is slowed, lactate is cleared from the blood, and all is back to normal.
> A plot of lactate concentration vs. percentage of V02 max is produced and the lactate threshold is identified as the point of inflection.

### Annotation 5 (page 1)

Beside the passage: “The VO 2 max values from Pollock’s study (24) of 19 elite runners were used. These ranged from 71.3 to 84.4 ml l kg-l l min-l and averaged 76.9 ml l kg-’ l min-‘. On the basis of these data, the low, average, and high vo2 max values used for predictions of running speed at lactate threshold were 70,77, and 84 ml. …”

**Annotation** by Luís Batalha (2017-10-15):

> VO$_{2}$ max is the **maximum rate of oxygen consumption measured during incremental exercise**. VO$_{2}$ max is reached when oxygen consumption remains at a steady state despite an increase in workload. VO$_{2}$ max is usually expressed as a relative rate in millilitres of oxygen per kilogram of body mass per minute (e.g., mL/(kg·min)).
>
>
> ***Why is VO$_{2}$ important for running performance?***
>
> Oxygen is used in a cellular process called aerobic respiration to convert biochemical energy from nutrients into adenosine triphosphate (ATP). The chemical energy stored in ATP can then be used to drive processes requiring energy including locomotion. Thus VO$_{2}$ max basically sets maximum value for the rate of energy available for running.
>
> ***Measurement***
>
> Accurately measuring  VO$_{2}$ max involves a physical effort sufficient in duration and intensity to fully tax the aerobic energy system. In general clinical and athletic testing, this usually involves a graded exercise test (either on a treadmill or on a cycle ergometer) in which exercise intensity is progressively increased while measuring:
>
> 1) Ventilation
>
> 2) Oxygen and carbon dioxide concentration of the inhaled and exhaled air.
>
>
> ![](http://www.southdenvermedicine.com/portals/2229/Skins/IH-SDI/images/VO2_Max_Testing.png)
>
> Since pushing the body to VO$_{2}$ max can be dangerous there have been developed several ways to estimate the VO$_{2}$ without reaching the maximum of the respiratory and cardiovascular systems.
>
> The Cooper test is an example:
> $$
> \text{VO$_{2}$ max}=\frac{d_{12}-504.9}{44.73}
> $$
> where $d_{12}$ is the distance (in metres) covered in 12 minutes.
>
> There are also several consumer fitness devices (Garmin Forerunner 620, PulseOn) that use similar methods to estimate the VO$_{2}$ max.

### Annotation 6 (page 2)

Beside the passage: “Lactate threshold data from several sources were used (4, 7, 14, 20, 22, 25). Numerous criteria, techniques, and nomenclature systems for the lactate threshold and related physiological events have been used (12). However, it appears that, regardless of definition, most runners sustain a pace in the marathon that …”

**Annotation** by Luís Batalha (2017-10-15):

> **Blood Lactate Threshold is the exercise intensity at which the blood concentration of lactate and/or lactic acid begins to exponentially increase** (often expressed as a percentage of VO$_{2}$ max)
>
> During power exercises such as sprinting, when the rate of demand for energy is high, glucose is broken down and oxidized to pyruvate, and lactate is then produced from the pyruvate faster than the body can process it, causing lactate concentrations to rise. High lactate concentrations can be toxic to the body and results in an excessively low pH in the bloodstream.

### Annotation 7 (page 2)

Beside the passage: “Lactate threshold data from several sources were used (4, 7, 14, 20, 22, 25). Numerous criteria, techniques, and nomenclature systems for the lactate threshold and related physiological events have been used (12). However, it appears that, regardless of definition, most runners sustain a pace in the marathon that …”

**Annotation** by Luís Batalha (2017-10-15):

> A good analogy for VO$_{2}$ max and LT is to think about  VO$_{2}$ as the size of the car's engine and LT as the redline or the maximum engine speed.

### Annotation 8 (page 3)

Beside the passage: “(2:12:45) (2:05:46)* (1:57:48)* Values in parentheses (hours: minutes: seconds) represent time to complete a marathon at estimate of marathon running speed directly above. All values were obtained using data from Table 1 along with Eqs. 1-3. They were corrected (slowed) by - 10% to account for effects of wind …”

**Annotation** by Luís Batalha (2017-10-15):

> Note that now only one of these values is below the World Record (2:02:57)

### Annotation 9 (page 3)

Beside the passage: “On the basis of these estimates, two basic interpretations of the model presented in this paper seem reasonable: First, one could argue that substantial improvements in the marathon world record are “physiologically” possible at this time. Second, one could argue that the disparity between the present world record and …”

**Annotation** by Luís Batalha (2017-10-15):

> Note that although this might look like an achievable improvement, getting to 1:57:48 means decreasing 4.1% the current world record time! When Usain Bolt smashed the record for the men’s 100-meter, bringing it down from 9.74 sec to 9.58 sec it represented "just" a 1.7% improvement!
>
> In fact there has never been a single 4.1% drop or higher in the history of Marathon records.
>
> ![](https://i.imgur.com/2BImGm5.png)

### Annotation 10 (page 4)

Beside the passage: “In this context, it was assumed that two factors would distort the running economy curves during faster running for *periods >lO min. First, there is a slow upward drift in VO, during prolonged exercise (2, l7), while the running economy data used in this model were collected during 5- to lo-min exercise bouts. The …”

**Annotation** by Luís Batalha (2017-10-17):

> A lot of experts believe one of the biggest factors that will allow runners to improve their times in the future is a coordinated pacing effort (cooperative draft).
>
> This would be a result of teams unifying under the same country or sponsor and then running in a formation. Even if they do it just for a portion of the race this would  reduce substantially the overall wind drag and save a lot of time (similarly to what happens in cycling).
>
> ![](https://www.mobiefit.com/blog/wp-content/uploads/2017/05/kipchoge_eluidrabbits-sub2hr17-pacers.jpg)

## Full text

Extracted from the PDF. [Annotation N] marks where each annotation above sits.

#### Page 1

Modeling: optimal marathon performance

on the basis of physiological factors

MICHAEL J. JOYNER Department of Anesthesiology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota 55905

JOYNER, MICHAEL J. Modeling: optimal marathon performance on the basis [Annotation 1] of physiological factors. J. Appl. Physiol. 70(Z): 683-687, 1991.-This paper examines current concepts concerning “limiting” factors in human endurance performance by modeling marathon running times on the basis of various combinations of previously reported values of maximal 0, uptake (VO,,, ) lactate threshold, and running economy in , elite distance runners. The current concept is that vo2,, sets the upper limit for aerobic metabolism while the blood lactate threshold is related to the fraction of vo2 max that can be sustained in competitive events greater than -3,000 m. Running economy then appears to interact with \io, max and blood lactate threshold to determine the actual running speed at lactate threshold, which is generally a speed similar to (or slightly slower than) that sustained by individual runners in the marathon. A variety of combinations of these variables from elite runners results in estimated running times that are significantly faster than the current world record (2:06:50). The fastest time for the marathon predicted by this model is 15758 in a [Annotation 3] hypothetical subject with a VO, 1118X of 84 ml l kg-‘. min-‘, a lactate threshold of 85% of VO 2 max, and exceptional running economy. This analysis suggests that substantial improvements in marathon performance are “physiologically” possible or that current concepts regarding limiting factors in endurance running need additional refinement and empirical testing.

maximal oxygen uptake; lactate threshold; running economy; human performance

PHYSIOLOGISTS have long been interested in modeling optimal human performance in various running events on the basis of world records (23,26). Although a variety of approaches to this problem has been used (23, 26), recent mathematical models of running performance have been improved by the recognition that maximal 0, uptake (90, ,,) cannot be sustained in competition for S-10 min (23,26). This approach is supported by experimental data demonstrating that submaximal variables including the blood lactate threshold and running economy (0, cost to run a given speed) are powerful predictors of endurance running performance (3-5, 7, 11, 14, 16, 22, 25). Along these lines, this paper attempts to extend the current models of human distance running performance by considering how VO, m8X, blood lactate threshold, and running economy interact as determinants of performance in the marathon.

The emerging concepts concerning the limits of marathon performance are that vo2 max sets the upper limit for aerobic metabolism and that the blood lactate threshold is related to the fraction of Vo2max that can be sus-

tained in competitive events of 2-3 h. Running economy then appears to interact with [Annotation 2] Vo2mar and blood lactate threshold to determine the actual running speed at lactate threshold, which is generally a speed similar to (or slightly slower than) that sustained by individual runners in the marathon (14).

With this information as a background, the purpose of this paper is to estimate an “optimal” human performance in the marathon on the basis of the following simple physiological model

marathon running speed

. = vo 2 max (ml l kg-l l min-‘) X %802 max

at LT X RE [km l h-l l VOW’ (ml l kg-l . min-‘)I

where LT is lactate threshold, RE is running economy, and VO, is 0, uptake. This model is then used in conjunction with a range of well-accepted values for these variables in elite male distance runners to estimate the physi- [Annotation 4] ologically optimal marathon performance.

METHODS

Previously reported values for VO, max, lactate threshold, and running economy in highly trained and elite endurance athletes were used to establish three estimates (low, average, and high) for each of the three variables. These values were used in different combinations to estimate the running speed at lactate threshold.

Maximal O2 Uptake

The [Annotation 5] VO 2 max values from Pollock’s study (24) of 19 elite runners were used. These ranged from 71.3 to 84.4 ml l kg-l l min-l and averaged 76.9 ml l kg-’ l min-‘. On the basis of these data, the low, average, and high vo2 max values used for predictions of running speed at lactate threshold were 70,77, and 84 ml. kg-l. min-‘. It is recognized that there have been occasional reports of Vo2max values >84 ml l kg-‘. min-’ in humans; however, this value is at or near the upper limit of values usually reported for elite runners in textbooks (1). The 70 ml l kg-‘. min-l value used for the lower limit was selected 1) so that symmetrical changes in VO, mru among the low, average, and high values would occur and 2) because there have been reports of a marathon world record holder with a VO 2 m8X value of ~70 ml l kg-’ l min-l 6, 24). 0161-7567191 $1.50 Copyright 0 1991 the American Physiological Society 683

#### Page 2

684 LIMITING FACTORS IN THE MARATHON

TABLE 1. Predicted vo2 at lactate threshold

vo2 -, ml l kg-’ l min-’

%VO,, at LT 70 77 84

75 52.5 57.8 63.0 80 56.0 61.6 67.2 85 59.5 65.5 71.4 Values are expressed in ml l kg-’ l min-‘. LT, lactate threshold.

Blood Lactate Threshold

Lactate threshold data from several sources were used [Annotation 6] (4, 7, 14, 20, 22, 25). Numerous criteria, techniques, and nomenclature systems for the lactate threshold and related physiological events have been used (12). However, it appears that, regardless of definition, most runners sustain a pace in the marathon that elicits blood lactate levels between 2 and 3 mmol/dl(4,14). This may explain why actual running speed for the marathon is generally above the onset of plasma lactate accumulation value (first increase in blood lactate above baseline) used by Farrell et al. (14) and generally below the onset of blood lactate accumulation (i.e., 4 mmol/dl) value used by Sjodin and Jacobs (25). Additionally, regardless of the nomenclature system used, it does appear that elite runners are able to run the marathon at speeds that require 435% of Vo2mm (7, 14, 23-25). On the basis of these considerations, it seemed reasonable to set the low, average, and high values for the lactate threshold at 75, 80, and 85% of Vo2 max, respectively. It should be noted that there are anecdotal reports of elite runners who appear to be able to sustain roughly 90% of their [Annotation 7] VO 2 max values during the marathon (5). Furthermore, slower runners may not be able to sustain running speeds associated with increases in blood lactate levels much above those observed at rest. This factor could act to lower the percent V02mar utilized to run the race and further prolong the time required to complete the distance by such individuals (4,14).

The 0, uptake (90~) ml l kg-’ l min-‘) at lactate threshold was then calculated by multiplying the percent vo 2 max at lactate threshold and VO, max (Table 1) (7, 14, 25). These values then served as estimates of the relative 60, values that could be sustained for a marathon.

Running Economy

Three running economy curves relating running speed to V,Z were established. Raw data for each of the 12 subjects studied by Conley and Krahenbuhl (G. S. Krahenbuhl, personal communication; see Ref. 3, Fig. 1) were examined. Running economy values from the two most economical subjects (lowest 0, cost for a given running speed) were averaged, and a linear regression equation between running speed and VO, was calculated. Likewise, running economy values from the two least economical subjects (highest 0, cost for a given running speed) were averaged, and a regression equation was calculated. Finally, an average linear regression equation between running speed and 60, was established using the mean values of all 12 subjects (Fig. 1). This allowed estimates of running economv at faster sneeds. Use of a linear

model appears justified on the basis of the work of Hagen et al. (15). Because the highest speed used in the measurement of running economy by Conley was 17.7 km/h (which is below the current world record speed of -20 km/h for the marathon) and because reports of running economy data at higher speeds are generally anecdotal, individual examples of running economy in world class athletes were obtained to support the extrapolation of the Conley data to higher speeds. These individual values are plotted in Fig. 1 and represent previously unpublished observations for best running economy observations made in several individuals between -19 and -24 km/h (J. T. Daniels, personal communication). These values and those of Conley were obtained during brief periods (5-10 min) of treadmill running, so the effects of wind resistance and the upward drift in VO, that occurs during prolonged exercise are not considered in the regression equations used to construct Fig. 1 (10, 17).

Running speed at lactate threshold for each of the three values of running economy was then estimated using the i702 values at lactate threshold (Table 1) and three regression equations relating running speed and vo, (see RESULTS). These estimates of running speed and time were then slowed by - 10% to account for added effects of the 1) 7-8% reduction in running economy that would probably occur as a result of wind resistance during overground (compared with treadmill) running (10) and 2) the 2-3% increase in VO, that would occur from 10 min to 2 h of running (2, 6, 8, 17). This resulted in 27 estimates of running speed at lactate threshold (Table 2) ranging from a combination of the three “lowest” values to a combination of the three “highest” values for each variable. Marathon time was estimated by dividing the marathon distance (42.195 km) by the calculated values for running speed at lactate threshold and converting to

- 28 ';, 26 1 9 24

ti c3 z z Z 3 CY

12- -*' a . - . Low

10 I I 1 I I I I I I 1 35 40 45 50 55 60 65 70 75 80 85

i/o, (ml l kg-l . min- ‘1 FIG. 1. High, average, and low running economy curves plotted from raw data of Conley and Krahenbuhl (Ref. 3, Fig. 1). Actual data were collected on treadmill at speeds of 14.48,16.09, and 17.70 km/h. A, Unpublished observations of best individual running economy values for running speeds >17.7 km/h observed by Dr. J. T. Daniels in more than 20 years of testing elite runners. These data suggest that linear extrapolation of treadmill running economy curves to speeds >17.70 km/h is generally appropriate and that regression equation for high running economy values used in calculations is accurate for high-speed running. A discussion of how treadmill and overground running may 1.m ’ l 1 1 1. 1 1 tuner is inciuaea in text.

#### Page 3

LIMITING FACTORS IN THE MARATHON 685

TABLE 2. Estipated marathon running speeds and times on the basis of VO, lactate threshold and running economy

fro, at LT, ml l kg-’ l min-’ Low RE

Running Speed, km/h

Avg RE High RE

52.5 14.40 15.18 16.42

(2:55:49) (2:46:47) (2:34:11) 56.0 15.28 16.10 17.35 (2:45:41) (2:37:15) (2:25:55) 57.8 15.74 16.56 17.84 (2:40:51) (2:32:53) (2:21:55) 59.5

61.6

63.0

65.5

67.2

16.16 17.00 18.29 2:36:40) (2:28:55) 2:18:25) 16.70 17.56 18.85 2:31:36) (2:24:10) 2:14:18) 17.05 17.93 19.23 2:28:29) (2:21:12) 2:11:39) 17.68 18.58 19.89 2:23:12) (2:16:16) 2:07:17) 18.11 19.03 20.35 (2:19:48) (2:13:02) (2:04:24)* 71.4 19.17 20.13 21.46

(2:12:45) (2:05:46)* (1:57:48)* [Annotation 8] Values in parentheses (hours: minutes: seconds) represent time to complete a marathon at estimate of marathon running speed directly above. All values were obtained using data from Table 1 along with Eqs. 1-3. They were corrected (slowed) by - 10% to account for effects of wind resistance and upward drift in Vo2 that occurs with time. For details see text. * Performances that exceeded current world record.

hours, minutes, and seconds. It is assumed for discussion purposes that the weather and race course would also be “ideal.”

RESULTS

The calculations of vo2 at the lactate threshold are presented in Table 1. Values ranged from 52.5 ml l kg-’ l min-l, when the lowest estimates of Vop max and lactate threshold were used, to 61.6 ml l kg-‘. min-‘, when the “average” estimates for Vo2max and lactate threshold were used, and 71.4 ml. kg-l l min-‘, when the highest estimates were used. Three running economy regression equations relating treadmill running speed (RS) to vo2 were generated on the basis of the data of Conley and Krahenbuhl (3) “high” running economy

RS (km/h) = VO, (ml l kg-‘. min-‘) X 0.2936 + 2.6481

(1

average running economy

RS (km/h) = vo2 (ml. kg-’ l min-‘) X 0.2878 + 1.5867

(2

“low” running economy

RS (km/h) = vo2 (ml. kg-’ l min-‘) X 0.2779 + 1.2499

(3)

Data from the treadmill running economy regression curves and predicted values above 17.70 km/h are plotted in Fig. 1. Also plotted are the previously unpublished individual data points collected at faster running speeds in elite runners. The correlation coefficient for Eqs. 1-3 was 0.99.

Running speed at lactate threshold during actual mar-

athon running was then estimated for each of the nine VO, values at lactate threshold in Table 1 by using the three equations above and correcting the estimates by 10% to account for the effects of wind resistance and VO, drift (2,6,10,17). The 27 combinations of running speed at lactate threshold (3 X 3 X 3) and estimated marathon time on the basis of these values are presented in Table 2. Values ranged from 14.40 km/h (estimated marathon time = 2:55:49) for the three low values to 17.56 km/h (2:24:10) for the three average values and 21.46 km/h (1:57:58) when the high values were used.

DISCUSSION

In this paper, a set of reasonable but upper-limit assumptions was used to predict the fastest possible marathon time on the basis of.currently available information about the interplay of VO, mar, lactate threshold, and running economy as “limiting” factors in endurance exercise performance (4, 5, 11, 14, 15, 20, 22-25). This approach yielded a predicted “best” marathon time nearly 9 min faster than the present world record, with 3 of the 27 estimates of marathon time being faster than the present world record (2:06:50).

On the basis of these estimates, two basic interpretations of the model presented in this paper seem reasonable: First, one could argue that substantial improvements in the marathon world record are “physiologically” possible at this time. Second, one could argue that [Annotation 9] the disparity between the present world record and the predictions in Table 2 suggests that either factors in addition to the three explored in this paper limit elite running performance or the linear extrapolation of data collected at slower running speeds in less-gifted athletes is open to question. In either case, the possible shortcomings of the proposed model must be addressed.

Potential Limits of the Model

Genetics. Little information is available concerning the genetic factors required to attain the extremely high vo 2 max? lactate threshold, and running economy values needed to run the faster times in Table 2. Although a large part of championship athletic performance has been attributed to genetic endowment, there is no information about the population frequency of the characteristics that, in combination with prolonged intense training, predispose individuals for success in endurance running. If, for example, the genetic likelihood of a very high . vo 2 max, lactate threshold, or running economy value is 1 X 10D3, then the probability of the same individual having all three values is 1 X lo-‘! (For discussion see Ref. 1, p. 291-292, and Ref. 5.)

Are exceptional values for more than one variable mutually exclusive? It may be that exceptional values for one variable are mutually exclusive of exceptional values for another of the factors. Two possibilities come to mind. First, the ideal marathoner may fail to achieve the extremely fast predicted times (<2 h) because of an inability to increase fat utilization enough to avoid glycogen depletion or intracellular acidosis during an effort requiring a VO, of >70 ml. kg-l l min-’ for several hours (18, 19). Second, it may be that high VO, mLII values are

#### Page 4

686 LIMITING FACTORS IN THE MARATHON

incompatible with excellent running economy or lactate threshold values (5, 22). In the study of Pollock (24), a group of elite runners successful at 1,500-lQ,OOO m had higher (-79 ml l kg-’ l min-l) values for Vozmax than elite marathon runners (~74 ml. kg-’ l min-l), with the marathon runners averaging ~2 ml l kg-l l min-’ of VO, less than the distance runners at 19.3 km/h. In summary, on the basis of currently available information, it is not possible to determine whether exceptional values for one variable exclude exceptional values for another variable. . Running economy. Although the assumptions about vo 2 max and lactate threshold are well documented in the literature, less is known about running economy. It must be emphasized that the VO, vs. running speed regression lines in Fig. 1 are based on treadmill running data collected during lo-min trials at speeds of 14.48, 16.09, and 17.70 km/h (241,268,295 m/min, and 9,10, and 11 mph, respectively) (3). Figure 1 and the equations used to calculate running speed at lactate threshold are based on the assumption that running economy continues to increase in a linear manner at speeds ~17.70 km/h (15). This assumption, although supported by limited data (J. T. Daniels, unpublished observations; Refs. 21,24), is tenuous because of the lack of systematically collected running economy data on large numbers of elite subjects at running speeds between 18.0 and 22.5 km/h (300-375 m/min, 12-14 mph).

In this context, it was assumed that two factors would distort the running economy curves during faster running for *periods >lO min. First, there is a slow upward drift in VO, during prolonged exercise (2, l7), while the running economy data used in this model were collected during 5- to lo-min exercise bouts. The experimental observations suggest that Vo2 during cycling or running at 70% of maximum increases ~0.1 l/min (6,8) from 10 min to 2 h of exercise. Such an increase might slow the time required to complete the marathon by 2-3%. Second, the best available evidence suggests that the 0, cost of sustained (5-7 min) high-speed overground running in elite runners is 7-8% higher than for treadmill running, probably because of the addition of wind resistance (10,ll). It would therefore seem likely that these two factors might operate together to slow the ideal runner on the order of [Annotation 10] 10% during overground running for 2 h. Additionally, when the estimates of performance are not corrected for the effects of VO, drift and wind resistance, the fastest predicted time falls to an unrealistically low l:47:13 and 9 of the 27 estimates of performance surpass the current world record. Finally, this discussion demonstrates the need for the systematic collection of running economy data at speeds >300 m/min in elite performers. The effects of wind resistance and the duration of the exercise bout on running economy at these speeds should also be considered.

AdditionaL sites of fatigue. The traditional concept has been that metabolic processes are the key determinants of fatigue during marathon running (18, 19). Although the large training-induced increases in muscle fiber oxidative enzymes (18, 19) seen across fiber types in the trained muscle of endurance athletes suggest a high degree of fatigue resistance, there are other potentially important sites of fatigue causing failure upstream from the

muscle fiber in the neuromuscular apparatus (13). Along these lines, 1) Are the “highest threshold” motor units recruited during fast running? and 2) If recruited, can they be trained to fire and contract repeatedly for several hours without fatigue?

Summary

A variety of approaches has been used in previous attempts to model human performance. These approaches have recently been evaluated in detail, and the newer models have been improved by the recognition that . vo 2 max cannot be sustained indefinitely (23, 26). However, in these models, further improvement in world record marathon running is predicated on increases in . vo . 2 max, because the values for the fraction of sustainable vo 2 max and running economy presented are held fairly constant (23, 26). In the model presented in this paper, the effects of altering various submaximal variables (running economy and lactate threshold) known to affect performance have been evaluated along with changes in . vo 2max (3-5, 7, 9, 12, 15, 17, 21, 23). The principal new concept advanced using this approach is that, on the basis of a set of well-documented and reasonable assumptions, an ideal marathon runner may be able to run substantially faster than the present world record. The fact that the current world record (2:06:50) is nearly 9 min slower than the predicted best time indicates that either the genetic probabilities against such a performance are immense or our level of knowledge about the determinants of human performance is inadequate. In either case, studies of how VO, maX, lactate threshold, and (particularly) running economy interact as possible determinants of performance in elite athletes are needed to provide new insight into the physiological determinants and limitations of human performance.

The author thanks Dr. Gary Krahenbuhl, Arizona State University, for generous use of the raw running economy data (Ref. 3, Fig. 1) and Dr. Jack Daniels, SUNY Cortland, for the use of previously unpublished and unique running economy data (Fig. 1). Actual reference times were obtained from Truck and Field News. The author also thanks Janet Beckman for secretarial assistance.

The author was supported by The Upjohn Medical Education Fund, Medical Student Research Training Grant HL-07479 and Postdoctoral Training Grant GM-08288 from the National Institutes of Health, and the Mayo Foundation. Received 20 February 1990; accepted in final form 11 September 1990.

REFERENCES

1.

2.

3.

ASTRAND, P.-O., AND K. RODAHL. Textbook of Work Physiology (2nd ed.). New York: McGraw-Hill, 1977, p. 291-292. CASABURI, R.,T.W. STORER, I. BEN-D• V, K. WASSERMAN. Effect of endurance training on possible determinants of vo2 during heavy exercise. J. Appl. Physiol. 62: 199-207, 1987. CONLEY, D. L., AND G. S. KRAHENBUHL. Running economy and distance running performance of highly trained athletes. Med. Sci. Sports Exercise 12: 357-360, 1980. COSTILL, D. L. Metabolic responses during distance running. J. Appl. Physiol. 28: 251-255, 1970. COSTILL, D. L. Inside Running: Basics of Sports Physiology. Indianapolis, IN: Benchmark, 1986. COSTILL, D. L., W. F. KAMMER, AND A. FISHER. Fluid ingestion during distance running. Arch. Enuiron. Health 21: 520-525, 1970. COSTILL, D. L., H. THOMASON, AND E. ROBERTS. Fractionalutilization of the aerobic capacity during distance running. Med. Sci. Sports 5: 248-252, 1973.

#### Page 5

LIMITING FACTORS IN THE MARATHON 687

8. COYLE, E. F., A. R. COGGAN, M. K. HEMMERT, AND J. L. Ivy. MUScle glycogen utilization during prolonged strenuous exercise when fed carbohydrate. J. Appl. Physiol. 61: 165-172, 1986. 9. COYLE, E. F., W. H. MARTIN III, D. R. SINACORE, M. J. JOYNER, J. M. HAGBERG, AND J. 0. HOLLOSZY. Time course of loss of adaptations after stopping prolonged intense endurance training. J. Appl. Physiol. 57: 1857-1864, 1984. 10. DANIELS, J., P. BRADLEY, N. SCARDINA, P. VAN HANDEL, AND J. TROUP. Aerobic responses to submax and max treadmill and track running at sea level and altitude (Abstract). Med. Sci. Sports Exercise 17: 187,1985. 11. DANIELS, J. T. A physiologist’s view of running economy. Med. Sci. Sports Exercise 17: 332-338, 1985. 12. DAVIS, J. A. Anaerobic threshold: review of the concept and directions for future research. Med. Sci. Sports Exercise 17: 6-18, 1985. 13. ENOKA, R. M. Neuromechanical Basis of Kinesiology. Champaign- Urbana, IL: Human Kinetics, 1988. 14. FARRELL, P. A., J. H. WILMORE, E. F. COYLE, J. H. BILLING, AND D. L. COSTILL. Plasma lactate accumulation and distance running performance. Med. Sci. Sports 11: 338-344, 1979. 15. HAGAN, R. D., T. STRATHMAN, L. ,STRATHMAN, AND L. R. GETT- MAN. Oxygen uptake and energy expenditure during horizontal treadmill running. J. Appl. Physiol. 49: 571-575, 1980. 16. HAGBERG, J. M., AND E. F. COYLE. Physiological determinants of endurance performance as studied in competitive racewalkers. Med. Sci. Sports Exercise 15: 287-289, 1983. 17. HAGBERG, J. M., J. P. MULLIN, AND F. J. NAGLE. Oxygen consumption during constant-load exercise. J. Appl. Physiol. 45: 381-384, 1978.

18. HOLLOSZY, J. O., AND E. F. COYLE. Adaptations of skeletal muscle to endurance exercise and their metabolic consequences. J. Appl. Physiol. 56: 831-838, 1984. 19. HOLLOSZY, J. O., M. J. RENNIE, R. C. HICKSON, R. K. CONLEE, AND J. M. HAGBERG. Physiological consequences of the biochemical adaptations to endurance exercise. Ann. NY Acad. Sci. 301: 441-450, 1977. 20. HURLEY, B. F., J. M. HAGBERG, W. K. ALLEN, D. R. SEALS, J. C. YOUNG, R. W. CUDDIHEE, AND J. 0. HOLLOSZY. Effect of training on blood lactate levels during submaximal exercise. J. Appl. Physiol. 56: 1260-1264, 1984. 21. ITO, A., P. V. KOMI, B. SJODIN, C. Bosco, AND J. KARLSSON. Mechanical efficiency of positive work in running at different speeds. Med. Sci. Sports Exercise 15: 299-308, 1983. 22. LA FONTAINE, T. P., B. R. LONDEREE, AND W. K. SPATH. The maximal steady state versus selected running events. Med. Sci. Sports Exercise 13: 190-192, 1981. 23. P~RONNET, F., AND G. THIBAULT. Mathematical analysis of running performance and world running records. J. Appl. Physiol. 67: 453-465,1989. 24. POLLOCK, M. L. Submaximal and maximal working capacity of elite distance runners. Part 1. Cardiorespiratory aspects. Ann. NY Acad. Sci. 301: 310-322, 1977. 25. SJODIN, B., AND I. JACOBS. Onset of blood lactate accumulation and marathon running performance. Int. J. Sports Med. 2: 23-26, 1981. 26. WARD-SMITH, A. J. A mathematical theory of running, based on the first law of thermodynamics, and its application to the performance of world-class athletes. J. Bionech. 18: 337-349, 1985.

---

Source: [Fermat's Library](https://fermatslibrary.com/) — https://fermatslibrary.com/s/modeling-optimal-marathon-performance-on-the-basis-of-physiological-factors. More annotated papers: https://fermatslibrary.com/llms.txt
