An Interesting Analysis of Some Elites’ Training History

Weekly Mileage, and Performance

Prof. Tim Noakes, in chapter 6 of the 4th edition of his book Lore of Running, spends 123 pages (13% of his book’s 931 pages) chronicling and discussing “the training methods of runners who, in the past century and a half, have joined a select band of world-class athletes.” So large is this chapter, by far the largest chapter in the book, it could easily be published as a separate book. Noakes spends so much time examining the training of elites because he believes it “…is invaluable, as it synthesizes the training wisdom acquired through the hard effort, substantial heartbreak, and occasional joy of the hundreds of thousands of individual runners.” Noakes presents these athletes not just because of the level of excellence they achieved, but also because a history of their training methods exists. This is significant because “the majority of great athletes record only the barest details of their training methods for posterior”. These athletes are middle and long distance runners, racing at distances from 1500 meters up to distances of 700 km. Prof. Noakes’ examination of the training history of these athletes reveals some fascinating information. Let’s take a look at some of the more interesting items about weekly mileages and performance.

Lessons from Ron Hill

Ron Hill was a British marathon runner in the late 1960s and early 1970s and was heavily favored to win the 1972 Olympic Marathon (though he did not do so). His two autobiographical books “describe his running career in unmatched detail”.

In 1969 and 1970 Ron ran the world’s second fastest marathon (at the time) in 2:09:28 and ran 8 of his fastest 15 marathons – 2:09:28, 2:10:30, 2:11:13, 2:12:39, 2:13:42, 2:14:35, 2:15:27, and 2:16:48. His serious marathon career continued until around 1978.

Ron was a dedicated trainer, often running 120-130 mpw, but his experiments with very high weekly mileages did not go well. About his poor showing in the 1972 Olympic Marathon he wrote, “I wondered what would happen if I went beyond my 120 to 130 miles per week. Would I reach another plane of fitness and capability? I had to find out…But I was never really happy. A lot of time, I felt slightly fatigued…”

The increased mileage resulted in a 2:16:31 in the 1972 Olympic Marathon, good enough for just 6th place. For three months following the marathon, Ron attempted to keep his weekly training volume high but was plagued by sickness and poor racing results. Finally, in April of that year he reduced his training to 65 mpw and recovered enough to run a 2:12:39 in July and a 2:14:35 in August.

This same pattern was repeated throughout the remainder of Ron’s racing career. Ron eventually settled on a training period of increasing mileage for 10 weeks followed by 4 weeks of an average of just 40 miles per week.

Prof. Noakes correlated Ron’s racing record with his training record and discovered the following.

“Hill’s racing record again confirms that heavier training is as likely to produce worse racing performance as it is to produce better performances…Significantly, if we were to apply statistical techniques to analyze the relationship between Hill’s training load and his marathon performances, we would be forced to conclude that there was no statistical relationship between the amount of training he did for his various races and his subsequent performances in those races. In fact…Hill clearly showed a training distance above which his marathon racing performance fell precipitously as he trained ever harder. According to this analysis, Hill’s optimum training volume (that is, the training volume that produced his fastest marathon times) was about 100 mpw. Higher training volumes produced progressively slower performances.”

On this same topic, Hill writes, “Winning the Enschede at the end of a (4 week) rest period, on an average of 56 miles a week, and in a 2:18:06, on a hot day, and with very little effort, made me think. Yes! It made me think that 120 to 130 miles per week perhaps weren’t absolutely necessary for good marathon performance.”

Lessons from Alberto Salazar

Alberto Salazar achieved international fame by winning the New York City Marathon three consecutive times, from 1980 through 1982, and the Boston Marathon in 1982. In 1980, he ran and won his first New York City Marathon in 2:09:41, the world’s fastest marathon debut. In 1981, Salazar set a world record of 2:08:13 in his second marathon and set the American marathon record at 2:08:51 at the Boston Marathon in 1982.

“Following his early marathon successes, which he achieved relatively soon after impressive performances (particularly at 10,000m) on the track, Salazar increased his weekly training distances from 176 to 208 km (from 109 to 129 miles). He reasoned that more distance would allow him to run faster marathons. The result was that his training intensity fell, he was running tired most of the time and his performances in both the 10,000m and the marathon fell off alarmingly, culminating in a disappointing run in the 1984 Olympic Marathon (15th place, 2:14:19).”

It’s possible that Salazar actually ran higher weekly mileages than Noakes writes above. The Wikipedia biographical article on Salazar states, “Salazar recounts falling into a “more is better” mindset which reasoned that: if 120 miles per week yielded a certain level of success, then 180 or even 200 miles must bring even better results.”

Unfortunately for Alberto, “he never again ran a world-class standard marathon, indicating that, like Edelen and Hill, his feverish period of racing in the early 1980s took something from his body that he never regained. Salazar described his illness thus: ‘My immune system was totally shot. I caught everything. I was always sick, always run down…I was sick constantly. I had 12 colds in 12 months.’”

Lessons from Bruce Fordyce

Scotsman Bruce Fordyce achieved his fame by multiple wins in the most competitive short ultradistance marathon race in the world – the Comrades Marathon held annually in South Africa. Bruce won this race eight consecutive times. Additionally he won the London-to-Brighton Marathon three times and he ran a world best of 4:50:21 for 50 miles during the 1983 London-to-Brighton race.

In 1978, during training for his second Comrades Marathon Bruce became injured and was only able to run a total of 285 km in January and February (compared to 472 km the previous year).

“This was the single most important event that shaped Fordyce’s thinking. When the injury resolved, Fordyce increased his training and subsequently finished fourteenth in that year’s Comrades Marathon. He is convinced that with a little more racing experience he might have done better.

Fordyce subsequently told me that this performance made him suspect that runners such as Newton and Medler had possible trained too much for the race. He was particularly impressed by Dave Levick’s Comrades performance in 1971 when, on a grand total of only 130 km of training in January and February, he ran one of the great Comrades, finishing second in 5:48:53 and going on to win the 1971 London-to-Brighton race in record time. Thus, Fordyce concluded that high training mileage was probably not necessary for a peak performance.”

Fordyce stuck by this belief for the remainder of his career. Prof. Noakes consolidates Fordyce’s training ideas into a nine-point plan for competing in the Comrades Marathon. Points specific to our study are:

Start gently in January and February regardless of fitness. Train hard over distances of 6 to 10 km, and do not run more than 110 km (68 miles) per week, with long weekend runs of up to 25 km.

Specific ultramarathon training only in mid-March. Fordyce began to move toward his peak by increasing his training distances to 130 km (80 miles) per week, and in April, which he considers to be the most important month, this rose to 176 to 192 km (109 to 119 miles) per week. He sustained this heavy training (beyond 130 km / 80 miles per week) for a maximum of eight weeks.”

In his concluding section Prof. Noakes summed up what he has learned from these elite athletes:

Remarkable Performances Achieved on Little Training

“Jim Peters in his marathon debut came within 4 minutes of the world marathon record despite training only 77 km (47 miles) per week…

South Africa’s best marathoner, Gert Thys, reports that he trains 140 km (86 miles) per week. Yet, his best five marathon times of 2:06:33, 2:07:45, 2:07:52, 2:08:30, and 2:09:31 made him the world’s fastest marathon runner of the twentieth century with an average time of 2:08:02.2…

On the other hand, Herb Elliot, Kip Keino, and Matthews Temane all trained considerably less (up to 110 km / 68 miles per week), yet all set world records at distances from 1500 m to 21 km.

Interestingly, the best modern short and long ultramarathon runners, including Bruce Fordyce, Yiannis Kouros, Eleanor Adams, and Ann Transom, seldom ran more than 160 to 180 km (100 to 112 miles) per week.”

Prof. Noakes concludes the discussion of weekly mileage and performance with these comments,

“Thus, it would seem that elite runners perform best in the marathon and ultarmarathon races when they train between 120 and 200 km (74 – 124 miles) per week, with an increasing likelihood that they will perform indifferently when they train more than 200 km (124 miles) per week, as vividly shown by the experiences of Ron Hill.”

What makes the above information so interesting? Well, several things.

First, it is commonly believed that all elites run high mileages. While this generally seems to be true and elites, as a rule, do seem to run higher mileages there have been some notable exceptions. I found it very interesting that the fastest marathoner of the twentieth century ran fairly modest (for an elite) weekly mileages of around 85 miles per week and that world records at up to the half marathon distance have been set in the modern area by runners training at less than 70 miles per week. 85 miles per week is more than 30% less than the typical 125 mpw that is suggested that elite marathoners all train at and 70 mpw is 30% less than the 100 mpw often suggested as the training volume of elites racing at distances less than the marathon.

The second thing I found interesting was the finding that the top elites in the world have performed best when running between 74 and 124 miles per week. This is a broader range of elite weekly training mileage than seems to be generally recognized, especially in light of the claim that all elites run high mileages. The low end of elite optimal training load (74 miles per week) is not only much lower than I’ve generally heard claimed for elites but is between 25 – 65% lower than the standard belief that elites perform best when training 100 – 125 mpw. Recall that the 74-124 mpw numbers come from a review of the training habits of the elites – it’s what they ran – and is not simply a theoretical number.

Another thing I found interesting was the results of comparing Ron Hill’s racing performance with his training load. In the 2 part series “Mileage Based Training: a logical analysis” I made the point that runners need to determine their own individual optimal training load and that running at a lower or higher-than-optimal training load would produce sub-optimal performance. This idea is supported by the analysis of Ron Hill’s training and racing history. The availability of Ron’s training and racing history allowed Prof. Noakes to not only determine the training load that resulted in Ron’s best performances but to show that when Hill ran above this level his performances were sub-optimal.

Of great interest is the finding that a too high training load may result in not just sub-optimal performance, but may permanently impair performance. The results of both Hill and Salazar indicate that too much training and racing during a brief period of time may permanently alter the body in some way so that the athlete is never again capable of reaching similar performances.

In summary, Prof. Noakes review of the training and racing performance of some of the top elites of the past two centuries revealed some fascinating information. The range of weekly mileage producing optimal performance by elites was 74 – 124 miles per week and some of the fastest performances ever recorded have been accomplished by elites running quite a bit less than 100 mpw. The data also support the belief that training at a too high training load results in a sub-optimal performance and may even permanently impair physical capacity.


An Interesting Analysis of Some Elites’ Training History — 8 Comments

  1. Pingback: An Interesting Analysis of Some Elite’s Training History: Weekly Mileage and Performance | Training Science

  2. I have always wandered if that could be explained that before running these fast marathons elite athletes had already accumulated relatively large endurance base in their early athletic years, and therefore indeed there is no more need for extremely high mileage in peak years, and intensity is the key factor. However, in the same time young athletes just turning to marathon maybe still should run higher mileage in their first marthoning years for endurance development, which would result in better performances and maybe lower mileage training in later years. What’s your opinion about this?

    • Hi, Renars.

      The idea that elites spent their development years building a large endurance base has been brought up before, usually by those espousing running high weekly mileages. There is no direct research evidence that I’m aware of that supports this idea. However, there is some indirect research evidence that opposes the idea that a “large aerobic base” explains the training of elites or is necessary for fast marathon running.

  3. I ran marathons during the latter part of the 55-59 and the early part of the 60-64 age categories, with a maximum training volume between 110 and 130 Km per week, and was unable to break three hours in five attempts. During my sixth and seventh attempts in the latter part of the 60-64 age category the training volume was reduced to a maximum of 75 to 85 Km per week, with key workouts spaced a minimum of four full days apart for adequate recovery. Of the total calendar days during the training period, a full fifty percent were complete rest days, and the other fifty percent were days I actively trained. The change resulted in two sub-three hour marathons: 2:59:15 in the spring, followed by 2:56:31 in the fall. I credit these results to more specialized training for marathons, and, equally importantly, more rest and recovery time between key workouts.

    • I’m really interested in your approach. I’m trying to figure out from your post what your typical “week” looks like. You said you have at least 4 days rest between runs. Does that mean you run only about twice in a week? Or have a got this wrong?

  4. The key workouts I do are specifically for preparation for marathons. These come into play in the last twelve to fourteen weeks prior to race day. Typically I train on a ten to twelve day cycle, alternating key workouts and affording myself adequate rest.
    THE FIRST KEY WORKOUT IS SPEED WORK, consisting of 1600m repeats, increasing over the weeks from three for the first few sessions, and working up to ten or eleven near the end of the fourteen weeks. These build resistance to fatigue at speed. These are done hard, but holding back a little. The point is not to kill yourself doing these, but to go fast enough and get in the required number to trigger the desired adaptive response. Ideally you should quit when you feel you still have one good repeat left in you. Recovery jogs between repeats can last for a time period up to but not longer than it took to run the 1600m. This workout becomes rather rigorous as the number of repeats increases, so speed day is followed by two days full rest, one day easy recovery run, and one more rest day as a MINIMUM. You might think in the interim you would lose fitness, but that has not been my experience. Keep in mind that rest is every bit as important as the workout. Your body needs time to recover and repair itself.
    THE SECOND KEY WORKOUT IS THE”LONG”RUN. Most suggest working up to 35Kms at a brisk but comfortable pace, to build adaptation to marathon distance. I followed this for a number of years, but I always found on race day that my system knew when I was at 35Km, and that was pretty much the point my body would hit the wall. Hitting the wall is basically when your system runs out of glycogen as a fuel supply, and has to convert to using fat, which requires energy to convert to glycogen, which forces you to have to slow down. My experience with this strategy has been that it makes no difference how many calories I ingest during the race, extra calories have not forestalled when I hit the wall. Trust me, I have done all manner of experimentation, and energy budget calculations, all to no avail.
    I NOW USE A DIFFERENT APPROACH. Knowing I have about two hours’ supply of glycogen, I make a point of doing my long run just long enough to deliberately burn off almost all of my glycogen supply (about two hours), and then I repeat the same workout the very next day, after making a point to eat lightly after the first day’s workout. The second day’s workout ensures that my glycogen supply is completely depleted, which TRIGGERS AN ADAPTIVE RESPONSE FOR MY BODY TO STORE MORE GLYCOGEN. I have found this to be a much more successful strategy. In my last two marathons I did not hit the wall, and I ingested minimal calories during the race, and I turned in personal best times. I call these two workout days my back-to-back runs. Once done, I allow two full rest days, one optional recovery run day, followed by another rest day as a MINIMUM before my next key workout day.
    I concentrate on key workouts, and alternate between speed days and back-to-back runs. Recovery runs are optional, and volume is for the most part irrelevant. As I said, 75 to 85Km per week, but some are lower. What matters is getting the key runs in, and getting the appropriate amount of rest thereafter. About half of all my time getting ready for a marathon is rest days. I have gone out to do a workout and felt in the first 10 minutes I was not fully recovered to do it, terminated it for that day, and allowed myself one more sleep. Sleep is magic!

  5. Les – you approached me at the end of the Comox Valley Half on Sunday and I wanted to ask you some questions! Is there any way for me to get a hold of you, perhaps by email?

Leave a Reply

Your email address will not be published. Required fields are marked *