Melber Flinn

Running sub 3hour marathons – stats, performance and analysis.



That is not an allegorical title and this blog contains no information or comment on NHS interim management. It is an entirely self indulgent and lengthy piece on running marathons. So if you’d prefer not to read on I’d quite understand.

I am a really keen runner. I had some talent at school, forgot all about it in my 20s and then took it up again around 30 to lose weight and get fit. I ran my first marathon in 2013 and like many first timers I was under trained and over confident, I went out too fast and died in the last 6 miles, finishing in 3:19. Six months later in the inaugural York marathon I ran a 2:58, and 18 months later I was a stone lighter with a structured training programme and I ran the York marathon in 2:48 to finish 27th out of a field of 7000. I was hooked, and began to think big about getting my times down further and starting to challenge for those higher top 10 places.

But running took a back seat in 2015, we bought a new house, completely refurbished it, and I left my old employer to set up Melber Flinn – clearly it was a year where I wanted to pack in the stress. Perversely one of the reasons I run is to have some alone, thinking time, and to de stress, but during the Melber Flinn set up period I was wracked with guilt, feeling that I couldn’t really afford an hour out each day to be training when I need to build a business and support my family. On January 1st 2016 I remember stepping in the scales and weighing 82.1KG, whilst not overweight it was the heaviest I had been in 4 years and it galvanised me into action. In January Melber Flinn was starting to take off and I felt I could reward myself with time to run again, I cut out carbs and went for a minimum 5 miles every day for 28 of the 31 days in the month, reinvesting in a fitness base and losing 4 KGs in the process. Over the course of this year the focus has been once again the 9th October York marathon, which is 20 miles from where I live, superbly organised and just the right size of event.

As I have got fitter over the course of this year I started to realise I might be able to challenge my 2014 PB of 2:48. For those that know me well, I love my statistics and analysis, and reviewing my GPS running data on Strava I could directly compare by 2016 fitness to my 2014 fitness because I was doing many of the same workouts and races. I could run 13Km at just under marathon pace in 2014, and I could do the same in September 2016, doing an interval session of 5 * 1KM with a 105 second rest in between each one, I finished 26 seconds faster in 2016 than 2014. I ran a half marathon 2 weeks before York in 2013 in 1:20:09 and I ran a half marathon 4 weeks before York in 2016 in 1:20:30. Training volume is one of the biggest determinants of marathon success and crucially that was noticeably higher in 2016, I had run an average of 52 miles per week in the 12 week period prior to the marathon, compared to 42 miles per week over the same period in 2014.

So nearly 2 weeks ago on Saturday 8th October in the evening I was agonising over the pacing strategy, and whether I should try and go out at my 2014 2:48 pace the next day. To non runners it will sound like splitting hairs, but a difference of only a few seconds per mile can be crucial, the marathon is a devilish challenge of fitness judgement. The optimal way to run a marathon is even pacing, i.e. choosing the pace which you can maintain for 26.2 miles and not a meter more. Its needs to be the pace which doesn’t cause you to blow up at 24 miles, but at the same time isn’t conservative enough to leave you anything in the tank at the end. Judging it right is a precarious calculation, you could go too fast by 2-3 seconds a mile in the first half and you wouldn’t know it was too quick until mile 22, by which time it would be too late, and you’d be losing minutes over the last few miles.

I decided to go conservative and chose 4:04 / km or 6:32 / mile, which would bring me home in 2:51:30. This was actually my 2014 York marathon pace, but that year I had the kind of finish marathon runners dream off, completing each mile from 19 onwards faster than the last and running mile 26 in a Strava segment record ( To this day Im not quite sure where that came from and 2016 it was probably arrogant to assume that kind of strength lay within me again. 4:04 and hold it for 26 miles, that was the race plan.

For 20 miles the plan went perfectly, glancing down at my GPS watch every few hundred meters the key data field was average pace and I kept it locked to 4:04. This meant compensating on uphills and downhills (even though the course is relatively flat), and avoiding the temptation to up the pace to catch little groups or to respond to pockets of support along the way. I passed 20 miles in 2:10:04, 3 seconds slower than my same split time in 2014. At this point in the York course, you turn off a bleak exposed A road and usually have a psychological boost as you drop down in a hill into a series of villages for 5 miles, I felt great coming down this hill, urging myself to hold the pace and stay relaxed, but as can happen so quickly in a marathon, things started to go wrong. I passed the 21 mile mark and found myself in no mans land, no one behind me and the runner in front a good 200 yards away, the power in my legs started to dissipate, aerobically I was doing fine, by the pain was ramping up, muscles gave way to black holes and the communication channel between my brain and my legs was becoming increasingly fractious. Km 34 was 4:18, Km 37 4:18 and Km 40 was 4:24. The legs were not responding and in total I gave away 2 minutes in that last 6 miles. I summoned a bit of a sprint finish for a finishing time of 2:53:25 and then promptly collapsed with cramp. I hobbled back to the car with the help of my family, never after any of the 6 marathons I have done have my legs hurt as much as they did that day.

For a few days afterwards I reflected extensively on the performance, and specifically why given the parity between my pre marathon form in 2016 and 2014 I couldn’t deliver the same sort of marathon performance? There are of course a multitude of factors that affect marathon day performance so Ill rule out the easy ones for anyone wondering, yes the course in 2016 was the same in 2014 and the conditions were the same, a cool autumn day within minimal wind. In terms of general health, I had no cough, cold or general grumbles, and no injuries either, I was fully fit on both days.

I was of course 2 years older in 2016, but age makes less of a difference than you might think. The longer the distance the slower the deterioration in performance by age. I found a study online which suggested marathon performances decline by 1 – 1.4% every year over 40, but that was based on a large sample size at the New York marathon. That would represent around 4 minutes on top of a 2:48 marathon, so might account for the extra 5 minutes it took me in 2016. But I refuse to let age be an excuse – I might expect that deterioration if I was currently at the top of my game – age would chip away at my fitness. Haile Gebrselassie lost about a minute per year on his marathon performances between his world record in 2008 and 2012, between ages 35 and 39, but I still have plenty of room for improvement. The age of the winner at York was 37, same age as me, and 12 of the top 30 finishers at York are all over 40.

I settled on two key reasons, and whilst I can’t be certain these are the factors that made the difference I have a strong hunch they are the biggest contributors. In June 2014 I completed a 100Km ultramarathon which is 2.5 marathons. In preparation for that I did two separate training runs of 35 and 29 miles. Frankly the ultramarathon was a bit dull for me (running mostly alone aside a canal for 11 hours) but I wonder how much those extreme distances helped me in the last 6 miles at York in 2014. The marathon distance of 26.2 miles is completely arbitrary, but fortuitously it is a distance which sits just outside the comfort zone for human beings. “The wall” is generally a phenomenon encountered by runners who run out of energy, the typical human can store glycogen in their muscles to run for around 20 miles, if you don’t train for longer or fuel up in the early stages, you may well hit it. If Pheidippides had run 34 miles to deliver news of battle victory in 440BC, perhaps we’d all be running 34 mile marathons, and doing our longest long runs of 28 miles in training. 26 miles would be a breeze. In 2014, perhaps I fared well in the last 6 miles because I had run the marathon distance twice in training and twice back to back in the ultramarathon, aswell as doing a few other standard 20 mile training runs? In 2016, although my overall training volumes were higher, my longest runs were 20 miles and I had done 3 of them. Perhaps I just wasn’t conditioned for the last 6.

Secondly, my weight. I was 72Kg on race day in 2014, compared to 76Kg in 2016. An online calculator ( ) suggests my 2:53 at 76Kg would be worth 2:45 had I been 72Kg. Whilst not perfectly accurate it’s encouraging to know that pound for pound, I may actually be fitter in 2016 than I was in 2014, I’m just slower because I am lugging more weight about.

I read a great book recently called Black Box thinking, which is all about the importance of learning from your mistakes to make improvements for the future, something the aviation industry is very good at, but unfortunately the NHS is not. I am a big fan of the underlying theory, if something doesn’t quite work, analyse why and make sure you make the changes to ensure it does next time. So I hereby commit my plan to writing, I want to run my next marathon in Spring 2017, and I’m hoping it will be London through a charity place, but if it isn’t I will have others to choose from. I want to run a personal best, and maybe something close to 2:45 – a milestone for marathon runners as it secures you an automatic place at London at the hallowed Championship start, reserved for the best runners in the country (although the elites still set off in front of the masses).

To achieve that I will:

1. Not get injured. Vital. If I do, all bets are off. 2. Do core training 3 times a week. Often forgotten and neglected by runners, but the core is the axis and engine from which the limbs operate. A shaky chassis and loose wheel bearings will stop a car travelling too far, and the same is true of the human body. Doing 2 will help with 1, and 3. 3. Lose minimum 4 KGs to get down to 72Kgs. This will be difficult through Christmas, but I need to keep an eye on my diet and avoid alcohol and temptation. 4. Increase mileage volume to hopefully 55 miles per week in the 12-week period before my chosen marathon. 5. Run at least 3 runs of 22 miles + to help build my resilience in the last 6 miles.

Hit the above plan and Im in with a fighting chance. You can follow my progress on Strava, ( and I promise a follow up blog after the marathon to report how it went.

Back to all news


This website uses cookies. You can read more information about why we do this, and what they are used for here.

Accept Decline