I started in Durban, but I finished my Comrades 2000 at Half Way. Partly because there were a lot more (very) old people in it this year, so that I thought it unlikely that I would retain the Founder's Trophy for being the oldest runner to finish, which I have "won" the last 2 years. Partly because I've finished an Up Run inside the normal 11-hour cut-off, so I had nothing to prove. Partly because I have read up in Sue's textbooks how many calories a guy of my weight needs to run a mile and, adding on enough for the loss of elasticity and efficiency which is inevitable with age over around 35 years and for the Comrades' five big hills, I know that its famed aid-stations only have half the amount of carbohydrate on offer which I need. And, not least, partly because I've been so busy on family duties AND writing all these long essays exposing the International Olympic Committee's absurd Drugs-in-Sport "policy" and riding other hobby horses that I've averaged only 10 miles a week in training for the Comrades' 55-mile run over the past 6 months.

I'm so unfit for racing that I'm effectively having to start from the beginning again, like I did before the first London Marathon almost 20 years ago (which was after an absence from running of 27 years, excluding the 4 seasons of weekly 90 minute interval training I did which is more generally known as soccer refereeing). I began by looking up in my diary what I did then. In the 21 weeks, I went running 51 times ; I averaged 19.3 miles/week ; it was 7 weeks before I ran over 10 miles in one outing ; another 4 weeks before I could do that again ; after that, out of 16 outings before the London, 12 were over 10 miles and 2 of those, one 6 weeks before and one 12 days before were 20.7 miles, in 3:08 and 3:06 respectively, i.e. just about 9 minute miles. The average rate over the whole pre-London period was 8:28 per mile. After hitting the Wall with a terrible clatter at 23 miles, I finished in 3:49:57, i.e. 8:46 per mile, for an Age-Graded score, on the 1994 WAVA tables, of 62.33%, only just into "Local Class". If I had understood what I should have done to carbo-load, though, having got to 23 miles at around 3:13, I should have finished nearer, say, 3:43/8:30 per mile, virtually the average pace of my so-called "training" runs.

I was then ready to die, happy in the knowledge of my great achievement of having run a marathon. Then, six weeks and no training of any kind later, I read in the Results Book that I had been beaten by 175 men in my age group - I had been 4th in the inaugural UK National Junior Cross-Country Championships in 1948 - a dozen over 60s and 110 women, that total including a 15-year old girl who had somehow beaten the Olivetti computer system. Worse, or so it seemed at the time, 2 walkers had finished ahead of me. Being a bit athletic, I had known that there is a 50km walk in the Olympic Games, but, being English, I hadn't realised that 50km is 5 miles further than 26 miles 385 yards, nor that the Olympic racewalkers go through the marathon distance in 3:05 !

This would not do. I don't remember making a conscious decision to forget all I had learnt from my friend Chris Brasher in his weekly articles in the Sunday paper he wrote for, but I see from my log that, apart from my first 2 weeks back on the roads - one outing, 5.4 miles, 45:08, 8:22 per mile, then 3 times 5.4, averaging 7:48 - there was only a single week in the next 17 when I was slower than 7:30. That was when I ran a total of over 50 miles in seven consecutive days which included 3 at 12 miles each. In the remaining 16 weeks before my next race - the inaugural City of Birmingham marathon, I had been born there 54 years previously - I averaged sub-7 minute miles in 4 of them and another 4 were at less than 7:05. On my own cobbled-together training ideas, I had lowered my PB by 36 minutes to 3:13:30, an average of 7:23 (74.4% AG), and that after a first half, which included the long hill up to the City Centre, in 88 minutes. I didn't have a watch with a memory feature then, but I remember the 88 minutes because it was the first time I had looked at the watch in the race and I did a double-take because I didn't believe what I saw first time ! I also remember a pack of runners coming past as we descended the hill on the way back and the second Half taking me 1:45, i.e. 8 minutes per mile. Fortunately I hadn't recognised that the first half was at 6:43 pace, otherwise I might have DNFed there and then. Three weeks after that I ran an actual Half in 85:37, followed by the first Leeds marathon two weeks after that in 3:08 and then yet another hilly marathon, Barnsley, in 3:10 three weeks later.

The other thing I remember about that race in Birmingham was struggling through nearby Solihull where the ex-Brumagem upper crust live. It was at about 23 miles, I was all on my own and so there was no doubt about the object of their curiosity when I passed two blue-rinse, probably right-wing voting, matrons. One turned to the other and said in her ex-Brummie accent "Some of them are VERY old, aren't they?"

The following year, 1982, I doubled my weekly miles up to an average of 46, with 10.2 and 13.6 miles per outing two or three times a week regularly, but all at 7 to 7:20 pace. Except for one week when I ran 13.6 miles on 6 consecutive days, then 8.5 miles on the Sunday morning and 10.2 in the afternoon to bring the week's total to 100.3 (and at an average of 7:34). In the Birmingham race that year, I ran 2:58:13 (81.1% AG) and the same half-marathon, this time 4 weeks later, in 82:54 (83.0% AG). Although that is and will remain my PB for the Half, 14 years later I ran 90:59 at the age of 69, which is 87.4% AG, equal to 78:36 for a 55-year old.

I felt a bit guilty all through the 80s not doing the proper training which those who aspire to competitive racing should, i.e. the Long Slow Distance run over hills and dales at the weekend. Worse, early in 1984, I abandoned my training routes of the previous 3 years, which had involved crossing main roads at least 8 times per outing - diagonally to avoid having to stop or slow down ! - and I settled instead for a 2-mile (road) loop, one mile down one mile up, round which I'd go typically 5 or 6 times, all at around 7:15 pace. I kept the total weekly distance at around 45 miles and that cocktail produced AG performances between 82% and 84% when I ran well.

I had run my fastest marathon, 2:57:06, in 1986, when I had just turned 59, but that was only 84.1% AG, and, when I was 1st Senior Over 60 in Boston in 1989 in 3:03:46, that was only 83.9% AG. As my family commitments grew - Sue, Alan and Co arrived back in Leeds in 1990 - the mileage slipped into the 30s, but the better races were still in the 82% to 84% AG range.

Then in 1995 I bought a heart rate monitor. I hadn't given it much thought really, but I got it because George kept on nagging me that I needed one. First time I used it in a race was just after I'd turned 68 and, in a (track) 5k, I averaged 171 beats per minute and finished with it recording 176 bpm ! So much for the doctors' surgery and Polar booklet "220 minus age". The year before, all wired up and running on a treadmill, I had given in before I drowned in my own sweat when my heart rate was 164. My VO2 max was 63.4, "an outstanding figure for a man of our age" the man said. I found out later that he had tested Seb Coe back in the 1980s and his VO2 max was over 90 and a sustained heart rate of 250 bpm. Which explains a lot ! So now I set my training pace so that I aim always to run for at least 10 minutes over the 220 minus age "maximum", to strengthen my heart. Since that first time, I've never raced without my heart monitor on and I've found what heart rates I can sustain to race how far and at what speed. Like 168 for a 22-minute 5k and 145 for the marathon in Gateshead last year. This has enabled me to construct the following table showing race speed and heart rate:

Pulse bpm 171 163 156 153 151 147 146 136 130

All I need to do is to set the thing to bleep when I'm running too fast or too slowly and I don't have to do any mental arithmetic in a race any more based on mile markers in the wrong place.

It wasn't until I started getting heavily into riding hobby horses that I found I didn't have time to run all those 8, 10 and 12 miles in training, even though at around 7:30 miling, which I had covered in previous years. I just had to fit in what I could when I wasn't required to fetch and carry kids, walk the dog, etc. I have to say that it was with some astonishment that year, 1998, that I twice ran 10 miles in under my age in minutes off only 30 miles/week and virtually all the training outings 2, 4 or 6 miles only, but "flat out" according to the heart rate monitor. The first such run was at the start of the month, that was followed by a British M70 100km record a week later - only 75.3% AG, aaargh ! - first O/60 in the Leeds marathon the week after that - 3:41, aaargh ! - and then the 69:33 for 10 two weeks after that. I've been into the 87s Age Graded only on three occasions, as follows (the "Open Equiv" column being what that AG % calculates to for a 20 to 35-year old)

Year/Age Race Time AG % Open Equiv

1993/66 Abbey Dash 10k 40:10 87.1 30:58

1997/69 Brass Monkey Half 90:59 87.4 68:15

1998/71 Thirsk 10 69:33 87.3 51:10

In trying to see some pattern, I found that there was a division at 84% Age-Graded. In the 12 years since I started racing again in 1981, I always trained at a relatively fast pace compared with the Slow Distance recommended by all the experts in the Running magazines, but I had only run quicker than 84% Age-Graded on four occasions, never exceeding 85%. In the 7 years '93 to '99, I had run above 84% on 19 occasions, above 85% 4 times and over 86% another four times. The only difference I could spot was that the slower times were when I was running the greater training mileages, which averaged 45 miles/week and occasionally went over 60 mpw when I put in several 90+ weeks. When I could only get in 27 to 37 mpw in the 90s, my age-graded percentages were higher.

I shall never know, of course, because it's not possible to do any valid scientific research on the effect of any marathon training programme, let alone on just one individual, but this reduced mileage may have had two beneficial effects. Firstly, the less one runs, the less chance there is of an injury resulting from a slip on a wet surface or a stumble on an uneven one. Or, putting it the other way, the long, slow distance offers little or nothing, except extra hazards, to a runner of more than one year's experience. Secondly, running less often allows more time for minor injuries to heal before (already torn) soft tissues are subjected to the same stresses again which tore them in the first place. Or, putting that the other way round, I believe that running through an injury is risking committing athletic suicide and "recovery run" is a contradiction in terms. Far better to wait for another day until the little niggle has been healed for certain by your own natural processes, rather than go for a 10 mile run because that's what Bruce Tulloh's schedule in Runner's World says you have to do.

A suggestion. Write out your best times in your races over the various distances in the last 5 years and, comparing like with like, work out how much slower they are getting. If you're more than 5% slower overall - say 2 minutes for 10k, 4 minutes for the half-marathon or 9 minutes for the full marathon - then it's time to train differently from what you've been doing.

Whatever you change - e.g. going longer/shorter, slower/faster - try to get in at least 25 good hard miles a week. Since I've fallen below that, my AG figures have slumped around 5 percentage points and I cannot afford to carry a handicap like that now that I'm in the wrong, older half of the M70 age division. If changing for you means training faster but shorter, take heart from the fact that that's how two VS world record holders trained. John Keston averaged 31 miles/week in the three months before he set the M70 marathon WR, a lot of which was fast work, 200s and 400s, on the track. His 3:00:58, set eight weeks before he became 72, was 95.08% age-graded and worth 2:13:23 Open Class. Lou Gilchrist's W65 74:30 10 miles, off 25 miles/week, is the age- and gender-graded equivalent to a man running 2:23:49 for the marathon. There aren't many in the Club who have beaten that in the past or any who can now. Forget Runner's World. The best training manual by far is the book of Age-Graded Tables published by WAVA. When ordering, just let me know whether you want copies of the track events as well as the distance ones. O.K.?