Yes, this is not a shoe review (obviously). Get over it.
Going into my first marathon last April, I had just come off an injury and a 1:23:27 half marathon so was not confident of running super fast but pretty sure of a sub-3 clocking. Well, the first half went quite smoothly, passing the half way point around 1:28: right on schedule. However, I was already feeling the burn. I tried and failed to stay with someone going backwards as I was but even he was too fast. As the race went on, I kept readjusting my aims: 'Come on, you've missed 2:55 but surely you can do sub-3!', then after a few more miles and another recalculation: 'Oh man, no sub-3, but just get under 3:05 and kill it at London next year' (3:05 is the London Marathon good for age qualifying time). I had to keep stopping to stretch out a spasming hamstring. I might add that stopping is a good way to get spasms and cramp so it is something of a vicious cycle that you might want to avoid. Then after readjusting to 3:10 and 3:15 and realising that that wasn't going to happen it was just a matter of getting to the finish. 3:17:10. I waited for about a month before adding it to Strava to stop people seeing it in their feed. OK, my half marathon was also disappointing and could have warned me not to try for 2:55. It is important to note that the often repeated formula of working out goal race time (t2) from previous race time (t1), goal race distance (d2) and previous race distance (d1) as such:
t2=t1*(d2/d1)^1.06
predicts 2:53:59 marathon from 1:23:27 half marathon.
We'll get onto that formula later. It was often said to me that half marathon to marathon jump is much larger than from 10k to half marathon. In my hubris I rejected it but now fully accept it. The main difference is the training. You can do a quick half marathon off 10k training. My half marathon pb at the time, 1:20:05, was without pacing and a build up race for a 10k. You cannot however, run a fast marathon off half marathon training. This was probably my downfall. For marathon, you really have to put in the miles and the desired training benefit is more about learning to deal with exhaustion rather than anything specific like improve VO2max or lactate threshold. Half marathon training is much more based around speed work.
Coming off that disappointment, I decided to give the marathon a rest and try to run under 80 minutes for the half. I started off training very slowly, with a couple of just easy running weeks. Then slowly mixing in some tempo runs which were very conservatively paced (around 19 minutes for a 5k run). I added in some very intense speed work. It became apparent slowly that I was hitting form so decided I should do a marathon after the half. It's important to have workouts to show you where you're at. It was after two workouts of 5*kilometre on a 5 minute cycles under 3:20 average that I started to realise I was hitting form. I mixed in some marathon training: long runs up to 35 km but more importantly running 6 days a week with perhaps 2 workouts based on marathon or threshold pace. Racing short distances for me is a great training tool. It has 3 major benefits, two of which are pretty unique to racing: 1) It improves you're VO2max (5k race pace is VO2max pace). 2) It makes you accustomed to racing pain. 3) It lets you know where you're fitness is. One of the races I did was a 5 miler in 27:55. This translates to 1:17:30 for the half marathon using the formula. Come half marathon day, I set off at just under 80 minute pace and feel good at 7 miles. Too good. I ran away from the group of 3 I was with almost accidentally, taking the racing line around corners. Anyone who has been in this situation knows it is very hard to slow down to let them catch back up. Anyway, there was a group 100m ahead which we were catching anyway so I decide to chase them. It felt tough but nothing too bad. Looking at the data after, it looks like I gained about a minute ahead of my goal splits over the course of 3 miles. At 10 miles I was over a minute ahead but couldn't breath properly and dropped off the back of the group I'd tried so hard to catch. At 12 miles I was only 13s up. A quick calculation shows that if you keep at the same pace, you'll miss it... Somehow I held on to come in 14s under my goal: 1:19:38.
I then signed up for the Abingdon Marathon, an event which was full up about 9 months ago but luckily I was able to buy a number from an injured runner. Wary of the doubts I had about the marathon predictor I decided to stick to sub 2:56 plan. I had 6 weeks to fit in last minute training. Common wisdom states that you should leave 4 weeks for tapering without a long run (I took long run to mean over 25k). That gave me 2 weeks of hard training. I managed 80km and lots of bike riding in those weeks with a 30k fastish long run and a 35k slowish long run. The taper was filled with lots of marathon pace and threshold pace. Two weeks out I did the Cardiff Half planning to do marathon pace which was what my brother was aiming for (so aiming between 1:27 and 1:28) and ended up being mid 1:26s (I had to beat him) which was not too bad. It was great practice for gels (one at 6 and 12 miles) and a great confidence booster. Come marathon day, I followed my plan: gel at 6, 12, 18 (although I did need some jelly babies at around 21 to keep myself going) and going through half way in 1:28 as before but much more relaxed. I had planned to speed up after half way to get a cushion but ultimately I wasn't feeling it. I did speed up, but it came much later at around 18 miles. I came home in 2:53:50 so a negative split which is not bad. I was convinced that was the fastest I could have done it.
So all well and good except there is still that pesky formula saying I should be running 2:46 based on half marathon. I am not the only one to have a slower than predicted marathon time. I know a 1:20-2:56 guy and a 1:21-2:54 guy. I don't know anyone who is not elite who ran as fast as the predictions say. I read an article after the race here: http://www.fetcheveryone.com/cms-37 which advocates for a power of 1.15 when it comes to half to full conversion instead of 1.06 based off experienced long distance runners' times. There is also an ability factor too (i.e. elites will run relatively better in the marathon than non elites). Interestingly the page predicts my time to be 2:53:45- 5s off!
