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today as they were when new, many are now mostly of historical interest as modern research and coaching methods have superseded them.

Training for 5,000m

By Frank Horwill

Hold your breath! The computerised prediction for the 5km world record in the year 2028 is 12:09.39, and for 2040, 11:56.19. That’s just for men. Women’s records for those years will be 13:41.56 and 13:37.75. Will there be some revolutionary change in training methods to bring about these times?

The evolution of training regimes has been slow and they are not expected to change dramatically for the next fifty years. However, what is changing is the age for peak performance. The age when an athlete was expected to run a sub 4 minute mile for the first time in 1964 was 24 years. It’s now 21 years. Thirty years ago it was virtually impossible for a 20 year-old to be ranked in the first three of the all-time best 5km performances, this is not so today.

The Nobel Prize winning physiologist, A.V. Hill, analysed the 5km event in 1932 as being 80% aerobic and 20% anaerobic. This means that given 10 training sessions, eight of them would be aerobic and two would be anaerobic. Not only do we have to define those terms, we must also decide which types of running included in those descriptions are going to be the most beneficial. Aerobic running ranges from jogging (100% aerobic) to 3km speed (60% aerobic - 100% VO2 max).

The first may be running at 12:00 / mile, while the second may be at 4:16 / mile. A big difference - but both are covered by the description - aerobic.

Similarly, anaerobic running starts with 200m (95% anaerobic) and extends to 1,500m speed (50% anaerobic - 110% VO2 max). The first could be run at 25mph, while the second at 15mph, another big difference, but both considered anaerobic.

If we take the intermediate stage of aerobic running we get half marathon speed (94% aerobic - 80% VO2 max). And, if we do the same with anaerobic running we will find that it is 8OOm speed (67% anaerobic - 130% VO2 max). We are now in a position to draw up a "Priority Table" from which ALL training sessions must be selected.

Aerobic training

  • Half marathon speed (6:00 / mile - 94% aerobic - 80% VO2 max)
  • 10km speed (5:45 / mile - 90% aerobic - 90% VO2 max)
  • 5km speed (5:30 / mile - 80% aerobic - 95% VO2 max)
  • 3km speed (5:1 5 / mile - 60% aerobic - 100% VO2 max)

Anaerobic training

  • 200m speed (95% anaerobic)
  • 400m speed (83% anaerobic)
  • 800m speed (67% anaerobic)
  • 1,500m speed (50% anaerobic)

The times given per mile in the aerobic table are, of course, examples. An athlete who can comfortably run the half marathon distance at 5 mins/mile, will pro-rata be running 3km at 4:15 / mile (7:52.5 / 3k).

We can safely say that all the speeds indicated in the aerobic table should be used in a 14 day training cycle. The two most beneficial speeds in the anaerobic table are at 800m and 1,500 metres pace. The inclusion of 800m metres speed may surprise some, however, the father of work-physiology, the great Olaf Astrand, discovered that the 800m metres event brings about more lactic acid in the system than any other middle distance event and suggested that all distance runners train and race at that speed because they will be able to deal more efficiently with lesser amounts incurred in longer races. This is particularly the case in a race where an athlete injects a surge of speed for 400m metres or longer. In the 1984 Olympic 5kmm final. Aouita went from 4:16 miling to 4:02 for the final mile.

When an athlete moves up to the 5km distance from 1,500m what can be expected? There is, in fact, a foolproof formula for predicting potential times:

  • 1,500m time x 3 + 3 mins = poor endurance 5km time
  • 1,500m time x 3 + 2:30 mins = good endurance 5km time
  • 1,500m time x 3 + 2:15 mins = excellent endurance 5km time.

Let us take actual performances:

  • Zola Budd: best 1,500m: 3:59.96 - best 5km: 14:48.07 - that’s 3 x 1,500m time + 2:48.
  • Dave Moorcroft: best 1,5O0m: 3:33.79 - best 5km: 13:00.41 - that’s 3 x 1,500m time + 2:18.
  • Tim Hutchings: best 1,500m: 3:36 (converted from mile time) - best 5km: 13:11.50 - that’s 3 x 1,500m time + 2:23.

It will be seen that all things being equal, a runner with the time of 4:00 for 1,500m has the possibility of running from 15 mins for 5km to 14:15 for 5km. Such a runner will not improve further until the 1,500 metres time improves. On this basis, a runner like John Mayock, with a time of 3:31 for 1,5OOm, has the potential to run well under 13 mins for 5km and to create a new UK record!

Another major factor with 5km pace running is that it must not be forgotten that the race is now 12 consecutive laps at 61.5 secs per 400m, p1us 200 metres. An athlete who does 13 x 400m in 61.5 secs and jogs 400m recovery (2-3 mins), is not getting used to sustained running effort. There may be a hundred or more athletes in the UK who could do such a session - however, it is not specific to 5km racing. Nearer the mark would be 7 x 800m in 2:03 with 400m jog recovery, better still would be 5 x 1,200m in 3:04.5.

There are two ways of doing specific 5km sessions. The first is to start with slow repetitions with very short recovery and get faster. The second is to start with very fast short repetitions with adequate recovery and then to extend the distance of the rep. with the original recovery retained. The latter method has the advantage of running near to world-class speed, the disadvantage being that the process may take months or even years to come to fruition. Here are examples of the two methods:

Slow Pace - Short Recovery7 x 800m in 2:08 (13:20 / 5km) with 30 secs recovery

Fast Pace Short Rep. - Adequate Recovery25 x 200m in 30 secs with 60 secs rest. Graduating to 17 x 300m in 45 secs with 60 secs rest, thence to 13 x 400m in 61.5 secs with 60 secs rest, moving on to 10 x 500m in 77 secs with 60 secs rest. Ultimately to 3 x 1 mile in 4:06 with 60 secs rest!

With the first method (slow pace - short recovery), the aim is to speed up the rep. in time. The writer recalls Hutchings starting with 4 x 1 mile in 4:40 (14:35 / 5km) with 60 secs rest, and 12 weeks later he was doing 4:15 / miles (13:19 / 5km).

We can now select training sessions on the basis of eight aerobic work-outs to two anaerobic ones:

Day  Training Session
Day 1 Aerobic – run half marathon distance fast
Day 2 Aerobic – 3 x 2 miles at 10km speed with 90 secs recovery
Day 3 Anaerobic – 8 x 400m at 1,500m speed with 60 secs rest
Day 4 Aerobic – 4 x 1 mile at 5km speed (target speed) with 60 secs rest
Day 5 Aerobic – 3 x 1,500m at 3km speed with 3 mins rest
Day 6 REST
Day 7 If no race: anaerobic – 8 x 200m at 800m speed with 60 secs rest
Day 8 Aerobic – run 10km very fast
Day 9 Aerobic – run half marathon fast
Day 10 Aerobic – 7 x 800m at 5km speed with 30 secs rest
Day 11 Aerobic – 4 x 1k at 3km speed with 2:30 rest
Day 12 Start the cycle again

Now, that looks pretty tough and it is! But a study of Kenyan full-time 5km runners shows that this is precisely the type of work they do. Haile Gebrselassie (Ethiopia) told one of the writers’ athletes that his favourite session was 50 x 400m at 63 secs with thirty secs rest!

But, we have to study our bodies. For example, a Dublin 5km runner aged over 30 had the ambition of running sub 14-minutes. He was put on the above type schedule. He reported that he was not recovering enough from the previous day’s work when starting a fresh day.

He was put on jogging 30 mins the day after each of the above workouts. This meant that the 10 day training cycle was extended to 22 days. He regained his freshness and realised his ambition (13:58 / 5km).

A female 5km runner (15:49), took a different line. She calculated that after the third consecutive day of training she needed a day off. She was later able to extend this to four consecutive days and a day off. With perseverance and adjustment the whole cycle can be accommodated in its entirety within 12 weeks.

But supposing you are a runner who doesn’t believe in all this aerobic and anaerobic stuff, you train by instinct. Fair enough.

However, there are a few common sense matters to consider. If you want to run 15 mins for 5km, you have to know that’s 72 secs / 400m and get used to it for spells lasting 3-5 minutes’ duration with short rest.

Also, you will realise that a good 3km time aids the realisation of a good 5km. For example. Liz McColgan ran 8:34.80 for 3km in 1989, that’s 69 secs / 400m. She also ran 15:01.8 (72 secs / 400m), only 3 secs per lap difference. Training at 3km pace and improving 3km speed is paramount.

A further aid to 3km improvement is work at 1,500m speed. McColgan recorded 4:01.38 (64.3 secs / 400m), 5 secs per 400m faster than her 3km time.And what about over-distance work? A 5km runner should be able to convert the time for 5km with this formula - 2 x 5km time + 60 secs = 10km time.

Work at that pace needs to be done lasting for spells of at least 8 mins, eg 4 x 8 mins at 74 secs / 400m. And, of course, there is the old steady run lasting four times the duration of the 5km (4 x 15 mins = 60 mins).

What about racing? Usually, it’s the fifth to seventh 5km race which will realise a peak performance in a season. Aids to this will be the same number of under-distance races (1,500m / 3k), and perhaps a couple of 1Okms as a psychological endurance boost.

Sweat loss in 5km runners during the summer is extensive, sweat removes minerals, particularly potassium which has a direct affect on heart function.

Pure orange juice with all meals will restore the status quo. 5km running is tough, it needs fuel, and carbohydrates in liquid form should be consumed within 30 mins of finishing any training session as well as being abundant in all meals.