Frank Horwill



  • These articles were first published many year's ago and whilst some are as relevant today as they were when new, many are now mostly of historical interest as modern research and coaching methods have superseded them.

Double or Quits

By Frank Horwill

Doubling or even trebling the training load can produce amazing results, but you have to follow strict rules

Cyclists do it, swimmers do it, and, strange to say, even great Finnish runners of the past did it. What is it? Crash training. What's it about? It is the doubling or even trebling of the training load in quantity and quality for a period of not less than two days and not more than seven days, followed by an equal number of recovery days. And does it work? Well, one very good example was the spate of shattering world records achieved by the Chinese women runners under the direction of the autocratic coach, Ma Junren.

The scientific explanation for its success revolves around one word: super-compensation. Most athletes train severely one day and have one or two days of easy training to recover. This does get results, but they are far less than putting together several days of intensive training, followed by a 75-per-cent reduction in normal training for the same period that has been crashed. By exercising strenuously for consecutive days, recovery never lasts for more than 22-23 hours, ie, the time between workouts. This results in a heightening of the training stress, and recovery, once it finally occurs, produces much-greater-than-usual training responses. These raised responses are often called super-compensation.

Apart from the obvious success of Chinese women athletes, what other evidence is there that it works? Well, the great Finnish runners of the 1970s practised a form of it on a monthly basis, eg: Week 1-severe-100m per week; Week 2-active rest-25mpw; Week 3-moderate-75mpw; Week 4 light-5Ompw; Week 5-severe-125mpw, etc. It will be noted that this routine is mathematically precise. Week 2, after the severe week, is 75-per-cent less. Week 3 is 25-per-cent less. Week 4 is 50-per-cent less, and Week 5 sees a 25-per-cent increase on Week 1, with the remainder of the month pro rata. Eventually, work done in Week 1 will be doubled. This will take five months to achieve.

The eight rules

The pioneering work on crash training was done by Dr Peter Snell (double Olympic gold medalist and world record holder for 880 / mile). This was followed by the University of Western Australia giving runners a 10-day crash cycle. Then Dave Costill and his colleagues at Ball State University in America joined the research using 12 collegiate swimmers. As a result of this pioneering work, certain precise procedures and findings were indicated:

  1. Crash training can boost oxygen uptake (V02 max) by as much as 7 per cent a time. To get the same boost from normal training may take as long as 6-12 weeks.
  2. Never crash train for more than seven days at a time.
  3. Before and during crash training the carbohydrate intake should be as high as 800g a day. This can be achieved by topping up the normal intake of daily carbohydrates with a liquid carbo-loader, about 200g a day. If this is not to one's liking, rice (22.2g / oz), flour-based foods, vegetables and fruit (especially raisins, dates and currants) should be added to the normal diet. It is also recommended that a carbohydrate snack with a small amount of protein be taken within 30 minutes of ceasing training and every 30 minutes thereafter for the next two hours.
  4. The anti-stress vitamins B and C should be doubled before and during crash training. Due to excessive sweat loss, potassium levels must be maintained by drinking pure orange juice with all meals.
  5. A clear pattern as to the regular use of crash training should be evolved:
    1. A seven-day crash session not more than once a month;
    2. A four-day crash session not more than once every three weeks;
    3. A two-day or three-day crash programme not more than once a fortnight.
  6. Athletes with a history of injury and/or illness should not be considered for crash training.
  7. Never give crash training to an athlete unless it is fully explained and full cooperation is assured.
  8. It is must be stressed that for every day of crash training an equal number of recovery days MUST follow.

A case history

Here is an example of crash training which I have used.

Gerald is a 21-year-old medical student with times of 4mins / 1,500m and 14:30 / 5km. He particularly wanted to do well in the British Universities Cross-Country Championships. His average weekly mileage was 45 miles, of which one track session a week was at 5km pace, lasting 5km in duration, eg, 5 x 1km in 2:55 with one minute recovery.

He agreed to crash training 14 days before the race for a period of seven days. His mileage was doubled to 90 miles a week. Nutritional requirements outlined earlier were emphasised. His one 5km session a week became two, well separated. To accommodate the increased work, he did a 35-minute morning run, and double that in the evenings.

On completion of the seven-day crash cycle, he had one day off, and for the remainder of the week did only 22 miles of running (half the normal routine). He had two complete days' rest before the race. In the BUSF cross-country championships he finished 10 places up on the previous year but, more to the point, he defeated five runners, three of them internationals, who had consistently beaten him all season. He was subsequently selected to run for the British Universities against an England team.

Minimising the injury risk

What about the possibility of injury? The evidence shows that injuries occur more often in those who train for long periods of consecutive days. My own research indicates that athletes given the same training task to achieve after seven days of consecutive training, which was also given on the first day of the cycle, produced an increase in the pulse rate. Thus the same task required greater effort and was indicative of stress. As a result of these findings, the traditional day off on Friday before a race on Saturday was altered to Wednesday, and if a race took place on Saturday, there was only light training on Friday.

A further cause of injury is a sudden increase in the training load, eg, 50 miles to 100 miles per week, which continues for several weeks. The body can withstand a seven-day boost but not a seven-week one!

If coach and athlete have misgivings about crash training, they should first experiment with a two-day cycle and progress to longer ones. For instance, if an athlete does track work on Sundays, Tuesdays and Thursdays, this can be altered to double-load track work on Sunday and Monday, rest on Tuesday and half the normal load on Wednesday and Thursday.

Harry Wilson, Steve Ovett's coach, was a firm believer in crash-training weekends at Merthyr Mawr for the GB team. Three training sessions a day were done on Saturday and Sunday, most of it on the murderously steep sand dunes. It was calculated that, including the Friday night run, the total mileage for the weekend was around 50. In that period, the mid-1970s and early-1980s, British middle-distance running was at its zenith. How sad that many present-day distance runners in Britain have expressed fear on going to these crash-training weekends !

Crash training can be applied to all sports where fitness is a major factor. A simple routine for a team sport where training is done three times a week is to double the number of sessions to six one week and then do two light ones the following week, alternating on this basis for six weeks.

With the exception of the 800m, 1,000m, 1,500m and mile, British records for the other distance events are pedestrian compared to current world records. Crash training could be the way to close the gap.