So what's my correct training pace?

Something that I never really grasped is what pace to train at. There are of course so many training types, I'm doing short intervals, long intervals, fartleks, tempo workouts, hill repeats, easy runs and long runs. Each of these has a purpose and must therefore  be done purposefully and at the right effort. I guess the easiest way is to drop the pace constraint and to go by effort. Take that a step further and use heart rate. But I hate running with a heart rate monitor. Scratch that detest it. There's something about having that darn belt around my chest, it irritates me, I'm constantly adjusting it...anyway I digress.

My solution was to just go to a running pace calculator, like this great one from Runner's World that even does race predictions! So I've just run a race, recently the Jackie Mekler 10km in 42:38 and it gives me the following output under training pace:

• 5:16 min/km Easy run training pace
• 4:23 min/km Tempo run training pace
• 3:57 min/km Maximum oxygen training pace
• 3:39 min/km Speed form training pace
• 5:16-5:57 min/km Long run training pace
• 3:15 min/800m Yasso 800s training pace

Great! So on Monday, Tuesday on whenever I head out again I do my tempo at 4:23 min/km, my easy run at 5:16 and my LSD no faster than 5:16 but no slower than 5:57 min/km. Then the following week I do some 400m reps on the track at 87s per 400. Easy?

But what am I actually doing? I never really thought that all I'm doing is maintaining 42:38 form but surely that's not what I want? But if I'm training at paces computed from that as my most recent race time surely that's what I'm doing? It's like the training pace here is the end goal and not a journey when using the race pace method and that seems a bit wrong. I understand the logic that it's a good measure of what I'm capable of, particularly when shooting for a new distance. I'm training for my first marathon, so a good estimation of what I'm capable could be what my times now predict. So using my 42:38 PB run a month ago, I get an estimation of 3:16:06. Using my half-marathon PB of 1:35:01 ran four days before that 10km, I get a marathon prediction of 3:18:06. Two weeks later I ran a 5km, albeit one as flat as a pancake in 19:25, and that predicts 3:05:53!

Here's the thing though, surely this method predicts current capability with a few assumptions, the most obvious to me being you realistically plan to run the predicted distance on a route profile similar to that of the input race result? My 5km time is a useless barometer for Soweto Marathon, but might be a good indicator for a race like Slow Mag, which is also reasonably flat. However the other assumption is that you are fit and ready to run the predicted race NOW. I can tell you now I know I can't run a 32km race at whatever pace my 10km PB would extrapolate to, never mind a marathon. Heck I think I would blow out if trying to do 25km at a hard effort.

This has been weighing on my mind as my different PBs run within two weeks of each other predict times 13 minutes wide for the marathon, but also because I feel I ought to train with a goal in mind and my effort in training should reflect that. I'm am heading into the great unknown so how I do logically plan for that. I have a half marathon scheduled for 29th September, roughly a month before Soweto and that will obviously be a good gauge of what I will be able to achieve. But here's my thing, if I train at 42:38 for a 10km determined pace, am I selling myself short. I have over 5 months to go till Soweto. A lot can change and I will inevitably, judging by my progress in the first half of 2012, be fitter and faster come 4 November. Surely I want to get faster, not maintain the status quo?

When I decided to train for a half marathon I consulted a few people on what I should aim for. The first question was always, 'What have you run?'. At the time it was 48 flat for 10km. I thought of going for 1:40 but guys said that a 1:45 would be more realistically based on my 10km PB. I did a 10km, the Deloitte race in the first week of my program to get a feel for race pace since my 10km PB was from November 2011, and out of nowhere ran 45:44. It was hard, I walked a lot, I was spent, butsuddenly I was in 1:40 territory. As a relative novice, I had no control of pace and actually ended up inadvertently running by feel, ie getting the workout wrong 90% of the time. So easy for me was simply being able to hold a conversation. Tempo I just ran hard, intervals I went all out. As the weeks progressed, I was on a 10 week program, I noticed that interval and tempo workouts, at the same rate of perceived effort were getting faster with no extra work on my part. I ran 1:38:42 and then two weeks later ran 1:35:01. That second half-marathon was at a faster pace than I managed over my 45:44 10km.

Now surely if I had restricted myself to training paces for a 45:44 10km, I never would have managed a 1:35 half marathon? Or a 42:38 10km? Or a 19:25 5km? I would have condition myself for a 1:40:54 half-marathon and a 21:57 5km, hypothetically speaking.

I'm definitely leaning towards a feel based approach with a bit of mathematical prediction thrown in. For example I have four races I am focusing on leading up to Soweto, a 10km each in June, July and August and the half in September. For the 10km races I hope to take a minute off my PB each time, route, weather, health, and the alignment of the stars permitting, which would put me into sub 40 minute territory (crikey!) and a realistical shot at a sub 90 minute half marathon, in turn an opportunity for a fast debut marathon, 3:05 to 3:10 maybe, if everything is perfect on the day (I'll take a 3:30 any day!)

So how do I train for that? I could repeat the exercise above and put in a 40:00 target 10km. But the speed form pace is a blinding 3:26 and I know that is probably to hot for me right now to hit consistently over 4,5,6 up to 10 reps. I can go all out and nail it first time but I would be bushed! What I think is logical and something I'm trying to do, is to have an end goal in mind for training pace I should hit. For example I like that the Yasso 800 is tried and tested so I should be able to do 10 of those at whatever pace I aim for, and if I have done my tempos, fartleks and long run, that together should work. But the goal should be just that, the end goal. The 10x Yasso 800s are usually done just before the taper. My current pace should be a springboard to that end goal, which leads into the foot race. So right now a 4:23 tempo and 3:39 speed form should be at my 80% and 85-90% effort, but like how we increase mileage weekly to improve stamina, I should be able to push a little harder in later weeks so that 80% becomes 4:20, 4:17, 4:14 and down to 4:07 by October and equally 85-90% becomes 3:35, 3:31 down to 3:26.

So a bottom end based on current ability and a top end based on desired, but realistic ability, bridged by running on feel and listening to the body? Well that's the theory anyway.