Saturday, 21 June 2014

Too many notes

Most runners don't see the sheer possibilities inside a training schedule.

Consider a weekly training block, one which the runner is going about 80 miles per week. This leaves plenty of room to play. Here's a theoretical week for such a runner broken into mornings and evenings:

You can see the running is semi-distributed, with chunks of big miles followed by rest days. But let's change things a little, moving miles here and there, and modifying workouts a little to spread the miles even more:

And again let's return to lumpier mileage, but still different from the first:

Or for that matter let's try having only single-run days

This could go on all day. Clearly there are thousands (millions, really) combinations to be had while holding mileage constant. I didn't even touch on the rest intervals of the workouts, their intensity, or even switched the day they were on. Plus there's could be extra variables like changes in terrain, strength work, and easy running pace. Ok, let's go with just one more:

If you think all of these workout weeks are the 'same' just because they add up to similar mileage then you're a damn fool shortsighted. You could run the same total 40, 60, or 80 miles a week for the rest of your life and never do the same week twice before you die. Decrease the mileage too far (say < 20 miles a week) and repetition does have its way. Increase mileage way too much (say 200 mi/wk) and there is no room for a satisfying pause. While none of the purple-hued schedules are inherently 'better' or 'worse', they are certainly different. And depending on the runner, their background, the type of race to peak for, one could be more appropriate than another. For myself, I prefer a little bit of lumpiness in my training; I don't want the exact same run twice in a row else I get bored. Others like oatmeal-level consistency and that's fine too. As a season progresses, a competitive runner will find one schedule more suitable early, another beneficial later.

Let it be clear: all the above schedules are different. The opposite view is absurd. Imagine someone arguing that all songs with the same number of notes are equal. Looking at the sheet music below, what kind of person would evaluate the following musical score by counting the total number of notes?

This would be crazy, but it's equivalent to what a lot of runners do. The appropriate way to acknowledge the above music is that it's possibly a complex piece of music. But that is in the eye of the beholder. Others may see this piece as simple repetition.

To continue the musical analogy, it is perhaps even more appropriate since both running and music share a common thread: the fugue.

In music a fugue is described as "a contrapuntal compositional technique in two or more voices, built on a subject (theme) that is introduced at the beginning in imitation (repetition at different pitches) and recurs frequently in the course of the composition"

As with music, running can be described as a mixture of elements ranging from strength, endurance, short intervals, mid-distance speed, long intervals, and -of course- rest. Each element has it's own schedule, not fully independent of the others, but rhythmically intertwined. Here is how I image, in simplified terms, six weeks of training cycle:

Running fugue

Each day a vertical slice of time, in which you have time to press certain keys, else pause and let the previous day's efforts take their desired effect. Like with music, periods within a training block/stanza are similar yet slightly altered. In the above diagram, all the brackets are on a weekly loop. But certain elements like strength take longer, while sprint-type intervals for pure speed work may take less. Or endurance building, which takes perhaps years to master. They must each be matched for on another.

Actual fugue
As each element has its own rhythm, or wavelength if you will, the challenge is to find a time period in which they reach crescendos and/or become resonant. The more in-tune your various rhythms are the more frequently these bright spots emerge. Less coherent means such peaks will dissipate rapidly before and after a brief period in space and time. Hence on a long-term basis one can be said to 'cycle' their training in order to master the existence of such peaks and troughs.

It is patently obvious the arrangements in a single week are near-infinite, and for multiple weeks the combinations lie beyond imagination. Like with music, it's easy to invent meaningless combinations of notes, but hard to find particular frequencies that mesh properly together.

So why do so many runners effectively focus on the most meaningless element of the music: the total number of keys pressed? The true complexity of training comes from a very different place: renewed stimulation of body and mind by joining, cooperatively, the many elements together. Person X bragging they run more than person Y is akin to mashing all the keys on a piano and declaring a superior composition.

We all have certain rhythms and themes in our training and in ourselves. As certain habits become solidified, other habits crystallize around these core habits until a schedule becomes so inflexible that it could be described as a case of rigor mortis. An athlete who is 'stuck' in a particular training pattern is like having the same tune forever stuck in one's head.

What then happens when such an athlete decides to break the mould and 'take it to the next level'? Since only one variable is left to be tinkered with, mileage, the athlete simply adds more of the same stuff to the pot. The ingredients don't change, nor the flavour. The aspiring runner hopes the body will say "yes, you had it all right before, only there just wasn't enough of it". If only it were that simple. It should be clear by now why this simple 'more is better' concept is seldom the answer. New changes to training are meant to cause new stimulations, which can often be the byproduct of re-arranging the elements in a creative fashion.

I worry some might read the above and think I'm advocating for a 'less is more' policy. This is missing the point. The point being that high mileage is good if and only if it can be used in the same creative ways a complex fugue can exist outside a specific note count.

Consider the following music by Bach, visually illustrated. Now imagine each colour represents a difference training element, the pitch is the intensity, and the length of dashes equals the volume (accomplished on that particular day). Only a true master can obtain coherence with so many elements at play, while others -bent on imitation- will just make a giant mess of it. Can you imagine where in this training song you might fit your desired race?

I apologize for using some mixed metaphors between soup, crystals, photons, running, and music. But I used these analogies to help demystify why so many runners seem obsessed, however wrongly, with adding pure volume to their routine. Runners mix up cause and effect. One should strive for coherence AND complexity, whereby volume is the result. An running 'ingredient', no matter how useful, must be modified with the hole. Volume is not such an ingredient, no more than a 1400-page count is essential to War and Peace being a good novel (there are no epic 15-page novels, but 900 pages might do). Without proper consideration of all elements at play such runners will never rise above mediocrity.

******* THE END? *********

A small aside: Using myself as an example, I did not actually start running faster by running more. Instead I discovered these things called intervals, and tempo. Up until my mid 20s I had never really experienced anything other than fartleks and easy runs. These were solid foundations on which  to build as I was in no danger of burning out, but it wasn't enough. Soon I was running 400m to 1000m repeats, hills, drills, and a few spills. All together it boosted my running speed quite a bit. During that same time I did not much increase my total miles run, although ironically this was all I could focus on. The double irony was that increasing my mileage was the path to future success, but it was not an end unto itself. Volume is merely the means to elaborate -with more complexity- based on what I had done before. This crucial point I woefully misunderstood at the time.

Too many notes

Not enough notes


  1. Ok, hopefully these comments will remain.

    First, I really like this post because it gets at a very important training element: variety. Variety is so important because that's what actually makes the body improve. If you have been doing nothing but lots of easy miles, and then you start doing less but with speed work, you'll probably improve, just like if you were doing only speed work, with very little mileage, and you increased the amount of easy running you did by a significant amount, then you'd also probably improve. The best training plan depends entirely on what your previous training was.

    The second thing about this post, though, is that I think your music analogy is a little off. The goal of training is to achieve a performance in a race. The music is a performance in and of itself. You wouldn't expect to be able to perform a piece of music well without rehearsing it, but also, you need to have other, unrelated technical abilities. You can't just take the most complicated piece of music, give it to a beginner piano player and expect him or her to play it well just off repeating it over and over again (or playing the same notes in a variety of orders). I suppose one could argue that the training could very well be the end goal (i.e. the performance in the musical analogy), but if that is the case, if you are happy doing the same thing week-in, week-out, then good for you. I play the same three or four songs on my guitar (rarely). It does it for me. It doesn't make me a musician, really. So, while you could argue that all that is needed is contained in the sample weeks above, and that the simple variety of that is enough to sustain, I think there are probably other things that need to be done with respect to each element. Some things have to remain constant (i.e. running at race pace), while others need to change (the amount of running at race pace, the amount of easy running, running at other paces).

    As for why runners fixate on volume, other than the fact that it is well-known that the amount of total running that one does is in fact related to how fast one is going to go (on the whole, there will always be exceptions), I think it is a well-founded fixation. I think once you get to a point where adding more volume is not going to get you any faster (the point of diminishing returns) then it is useful to look at a variety of permutations as the way to achieve progression. But the body needs that progression to improve, and for most people, they are not running enough for this to matter: just work at running more easy miles and improvement will come. When there is no more improvement, then things can get more complicated.

    I will email you a very interesting presentation outline that has two elements I think you will really like. 1) a list of workout variables to manipulate (I think you may have done a similar list). I can past it here: ¨ Speed
    ¤ Within reps, workout
    ¨ Recovery
    ¤ Length (between reps, between sets)
    ¤ Standing/jogging/steady/with “stuff”
    ¨ Rep length
    ¨ Terrain
    ¤ Hill, soft, hard, variable, etc.
    ¨ Volume (total/sets)
    ¨ Density
    ¨ “Stuff”
    ¤ Aerobic/clearing
    ¤ Sprint ¤ Strength
    ¨ Surges
    ¨ Feedback manipulators
    ¤ Knowledge of splits, reps, distance of reps, total workout, terrain, etc.

    2) A chart that describes how long it takes (in hours) to recover from a variety of types of running. e.g. easy running 8-12 hours, tempos 24-30 hours, sprints 30-40 hours.

    I think a lot of these elements are not new and are "built-in" to the traditions we have in distance running. But it is always good to see them expressed in new ways.

  2. [my re-edited reply to John via email]

    I like to picture training as a multidimensional entity in which we observe only its shadows. The music shadow is one of many, and a rather crude one at that. I’m not really a big fan of piano, so it’s not even my favorite. Like you said, it does reflect the multi-tasking aspect of training.

    I was worried my analogy wouldn't come off clearly. It's not one I've seen before so there's not much basis for comparison. You wrote that “You wouldn't expect to be able to perform a piece of music well without rehearsing it, but also, you need to have other, unrelated technical abilities. You can't just take the most complicated piece of music, give it to a beginner piano player and expect him or her to play it well just off repeating it over and over again (or playing the same notes in a variety of orders). Hence adding volume should be associated with adding complexity, not simply “more of the same”, which sadly too many people misinterpret this nuance.

    If you replace "piece of music" with "training plan" you have a reasonably similar argument. Here's my attempt at paraphrasing:

    You wouldn't expect to be able to execute an advanced training program well without rehearsing first at a lower level, but also, you need to have other, unrelated technical abilities [peripheral strength, neuromuscular coordination, mental attitude etc]. You can't just take the most complicated plan, give it to a novice runner and expect him or her to do it well just off repeating it over and over again (or re-arranging the same elements in a variety of orders).

    Which was my point, that the analogies hold in you can't give a novice runner an advanced training plan any more than a novice piano player can perform an advanced piece of music. Both require mastering the elements individually before expertly combining them, in addition to other skill considerations. Nevertheless, there are millions of combinations of training plans beginner, intermediate, and advance programs.

    It would be wasteful to reuse the same few songs/plans over and over. I recall Wetmore (in Running with the Buffaloes) would always change his seasons' training plans no matter how well they worked that year.

    I had to question when you said, "Some things have to remain constant (i.e. running at race pace)". It’s trivial to say that “race pace” changes with distance. But we also know race pace changes with the season, and the number of years training, and also pace changes within a race (even among pros). If race pace changes within a race, between races, between years, and distance, I’m not sure if it can be called “constant” (More importantly can one achieve a 'breakthrough' performance if they constantly train at their present pace? Personally I’ve stopped training myself this way).

    I'm against mistakenly turning training into its own performance, but one must recognize good training reaches a crescendo just before the best races are executed. An advanced near-competition workout will include many different training elements (say for a 31:00 10k race you might run 3 miles in 15:00 followed by 3x1km in 2:50 and 6x400 in 63s). Or here’s Bekele's workout just before a 10k race:

    1. Here is my email reply. In case anyone is actually reading. ;)

      I guess nothing should every remain constant forever, but in a given training block, I would usually keep goal race pace constant. So if someone is building up to try to run 17:30, then we would run workouts like 25x200 at 42, 13x400 at 84, 9x600 at 2:06, 6x800 at 2:48, 5x1000 at 3:30, 4x1200 at 4:12, 3x1600 at 5:36... But the next year that goal pace might be the same. And certainly the reality of those workouts will be that they won't all be run exactly at that pace anyway. Of course you also do other workouts at other paces...

      That presupposes training for a specific time goal, however, and maybe, as you say, getting runner from point A to point B just means, general improvement. Do we need a Bowerman-style Goal Pace to do that? Or can we just run at 3k pace (which is faster than current 5k pace) and current 5k pace and 10k pace among other things?

      What's funny is that in coaching courses (I've been to many) they teach us first you make your plan. And so you have to create this year-long thing with every detail planned out. Then, they acknowledge that very soon into this plan you will be throwing the entire thing out the window. So while I do have a plan, I just go week-to-week now: what did we do last week? How far along the plan did we get? Then the next week is based on that. There are certain touchstones we want to hit, but whether they happen on Monday of this week or Wednesday of next, is not all that relevant. Even more so with marathon training as the bulk of it is such that I've told people that there are no "workout days" really. You run a specific marathon workout when you are ready.

      A somewhat related idea is that while traditionally, we would do a "peaking" workout about 10 days before the main event (like you were saying, the crescendo), I recently came across an idea that for 5k and 10k, it might be more useful to have that sharpness further away from the main race, then go back in for a 3-4 week block of aerobic work, like base season, then race to peak. This is what the pros do when they race a lot in Europe. Because the sharpness comes at the expense of aerobic fitness, but aerobic fitness lasts longer (similar to Olbrecht's chart, but perhaps an inverse? How long does it stay in the system), they use racing as workouts to stay sharp, then recover with a few weeks at altitude and easy/tempo running, then back down to race again. I always wondered how they managed to race so often at such a high level. Perhaps the reason is better aerobic maintenance.

  3. Here's a Science of Running piece on tapering that seems to fit this theme:
    Too bad workshops tell people to teach training that way. I'm working on a recursive method of explaining training programs that sidesteps those problems.

    1. There are quite a few "alternative" tapering ideas out there. Another one is Scott Simmons, who basically just treats it like a "recovery week" but has the whole training phase growing throughout. Graphically it looks like a diamond.

      Something we are trying this year is to actually do the hardest workouts even further away from the goal race. So for provincials, which is happening in 17 days or so, we will do the most difficult workout tonight. It seems for 5k and below on the track the "tradition" is to do the last big workout 10 days out. Instead, we've done the last couple of weeks pretty hard, then go into a kind of aerobic maintenance mode, where the workouts are back on the edges again: tempo and under-distance, as opposed to goal race pace.

      As Magness says, the reaction is going to depend on the individual (some people are going to be doing things a bit differently in our group, too), but I think this is a good way to get the psychology right, as tempos and fast 200s usually feel good.

      The thinking physiologically is that during the last two weeks we've perhaps "neglected" (or reduced) the aerobic side of things, by lowering volume and doing more intense intervals (which apparently can break down mitochondria, thus weakening the aerobic system), so this allows, to use Magness's term, an "aerobic refresh" in a shorter time frame, while the muscle fibre, and VO2 and running economy benefits of the hard intervals are still being felt. We won't run "high" volume, but it will be slightly higher next week, then down to recovery week levels for provincials.

      Can't wait to hear about your method.

  4. Far out, Graydon. I always enjoy reading your posts, and this one is particularly stimulating.

    As you can probably imagine, the music metaphor hits close to home for me (and did you know that besides being the worst kind of high-mileage junkie, I am also a total fanatic for dense contrapuntal music - in fact, even occasionally a composer of it? Many of my runs have been powered by Bach.)

    Wanted to share that I often relate musical performance (rather than composition, as in your own neat analogy) to running, and I think I have similar shortcomings (and strengths) in both. With singing, ever since boychoir school, I've taken the attitude that one should be ready to perform at all times, because our performance schedule was so busy - however, since the majority of my performances were with an ensemble, we're not talking about the super focus demanded by a solo performance. Basically, I've always treated singing as something I should just be ready to do for many hours a day, any day it is required of me. And that has been very advantageous to me as a professional chorister, and taken away a lot of the things other singer friends of mine stress about (long warmups, being sick, etc.) However, now that I'm doing more and more important solo gigs, I've begun to wonder if perhaps I should place more importance on the arc of my preparations, on building to a real climax that occurs in the performance itself. I sing very well most days, but my training and my attitude towards performance has not prepared me to sing much better than I usually sing on particular days. Is that a bad thing? A good thing? I definitely used to think my way was cool, now I'm wondering (or just looking for more?)

    And naturally, I am the same way with running. In my self-directed "training" throughout my early 20s, what interested me most was being able to cover very great distances, at what I deemed a reasonably fast pace, in any circumstances and on consecutive days. Having pursued this agenda to its logical conclusion (and then taken a very long break from running and started again in earnest only this March), I've been trying to do the opposite, and it's very funny how my running and singing goals seem to be lining up. Interestingly, my decision at the beginning of March to train very intensely for a 5k which is still 6 weeks away (6 months to train for a 5k! The ridiculousness of it appealed to me) was more or less sparked by something a voice teacher said to me at Yale this year... he told me I shouldn't feel I need to be the "niche" guy, that if I devoted my energies to more mainstream repertoire (rather than obscure contrapuntal works), I might find I was as good as anybody else doing it.

    And that extremely kind and inspiring remark was, strangely, what made me want to run the fastest 5k ever. So it's really quite funny that we both had this mileage:complexity / speed:sophistication sort of analogy floating around in our heads. I've been sort of basing my philosophy for the coming year around it - get fast, do short conventional races in times I feel good about, sing well-known repertoire as best I can and try to make the performances count.

    Well, just thought I'd share...


    1. Edmund: what about taking 6 months to train for a 5k is ridiculous? Too long? Too short? I don't think there is a particular agreed upon length. I would also say that is true for a marathon (though we often see 12- or 16-week training plans). I tell my clients who are locked into this "xx-week plan" mindset that they should just be running all the time. What are they going to do between 12-week plans? Sit on the couch? That doesn't seem like a good idea. I guess we have the Running Room to thank for this.