uh oh, this was a very interesting thread about VO2max, and I've inadvertantly hijacked it to talk about math ed. One last comment, and I'm through!
Indian math ed is generally very poor. In India, millions of kids work very hard, but the education methodology is too focused on memorization over understanding, and so most of the kids don't truly learn math.
Meanwhile, the Indian kids in the US/Canada kick butt - they take tremendous advantage of the opportunities in the US (and usually their being in the US means that their parents were successful professionals, often with degrees in math/physics/engineering, etc.). Check out the Olympiad/Putnam performances of Bjorn Poonen, Ravi Vakil, Kiran Kedlaya, or the general studliness of guys like Shamit Kachru and Rahul Pandharipande.
Of course in every country there are a handful of kids who learn a tremendous amount, usually because a parent or uncle/aunt is a great mathematician/physicist, and guides that kid. A few choice books handed to a kid at 13 can relieve the nightmare war of attrition towards calculus that is the US secondary school math ed system.
Russia does an AMAZING job teaching their best math kids, and seems to do a decent job with all of their kids who study technical subjects. Every year, Harvard accepts one student from Moscow State University into their PhD program. Invariably that incoming grad student is 22 years old, and already knows tons more math than the BEST American grad student at Harvard. There's no comparison.
The Russian kid gets his/her PhD in ONE year, and becomes a Harvard Junior Fellow. Over the next few years, they establish themselves as the best new mathematican around, and within a year or two they have multiple offers for tenure-track jobs at the top universities.
Very rarely does any American kid come close to that kind of performance, though recently their have been a handful of bright young stars.
Meanwhile, US math ed in general will always suck as long as everyone who understands any math can get any number of jobs that are much more rewarding financially and logistically than teaching HS math is.
Math Ed policy in the US is dominated by people who actually don't know mathematics, but are mathematics education folk. Of course I'm generalizing, but the psychology in the US is that, if you actually know math, than you were "naturally gifted" at it, and so you are precisely the WRONG person to teach my kid, because you won't understand or tolerate their difficulties.
This is supremely wrongheaded thinking. Unfortunately, teaching itself is a skill that must be taught, so it's often true that people who are good at math (but weren't taught to teach well) fulfill on this stereotype.
Of course a truly good math teacher needs to know math deeply AND know how to teach. The American Mathematical Society has been pressing these points for years - these days there are lots of great mathematicians who are serious about education. But they lack the sheer numbers to impact education at a national level...
***********************
OK, end of rant. Back to VO2max.
We know from Geb's training that he does all of these tremendous workouts year-long, and this is in sharp contrast to the Lydiard system of periodization (though I'm sure Geb does his own type of periodization, I've never heard anyone describe how his training varies over the season).
The question is: is this optimal for us, and if so, how do we get to that point? We all believe that building a strong aerobic base is key, so certainly a beginning runner shouldn't jump into VO2max sessions when they start (as Lydiard and everyone else says, your body isn't ready for this work yet).
But Geb is supremely aerobically fit, presumably due to the years of running in his youth, at altitude...
As I see it, there are two (obviously related) questions: one question about what "we" should be doing (assuming that "we" are still relative aerobic novices) and one question about what Meb (and other top US runners) might NOT be doing...
Question if I'm a relative beginner/young athlete: Do I develop my aerobic base like Lydiard says, and then (after a year or two of Lydiard-type cycles) eventually transition to a schedule where I'm constantly working on all my systems (as trackhead suggests)? And, if so, is their a kind of micro-periodization that goes on (Canova's schedules show very specific types of periodization for marathon training...)?
Or do I reject Lydiard, and start doing some kind of VO2max work from the beginning? (We've all learned from hard experience that it's hard to avoid injury this way!)
Question 2: if we accept (if only for argument's sake, I'm not here to second-guess Meb's training) the theory that Meb hasn't reached his 5K potential because he hasn't developed his vVO2max the way Geb has, than what should he do to address this?
Here are some ideas based on what we know of Geb's training, and that of others like him, but I certainly don't know the answer to this question:
1) Canova-style short hill bursts. Canova says that these recruit the max number of muscle fibers, and increase biomechanical efficiency (so, this may not improve your VO2max, but it may improve your vVO2max).
2) A combination of intense track sessions in the mornings, followed by incredibly slow, easy recovery runs in the afternoon, to lower the blood pH back down. Someone here said that there are days when Geb runs 8-10 min pace(!) on his afternoon run. That'd be like a 36min 10Ker running 12-14 min pace on their easy runs, no?!?!
3) some strong AnT runs (an hour at marathon pace), at least once a week, to maintain the aerobic fitness.
Anyone have any insight on either of these two questions? Of course, Malmo will say we're all experiments of one, trial-and-error and steady work,... :)