"There's only one hard and fast rule in running: sometimes you have to run one hard and fast."

Wednesday, August 4, 2010

Steve vs the Experts #10: Daniels

The number of problems I have with the way Jack Daniels presents himself and his ideas are seemingly infinite, starting with the title of his book, "Daniels' Running Formula" (1998, 2005); it suggests that he's come up with a scientific system that allows one to plug in one's goals and then following his plans, your success is guaranteed. I'll try to limit my comments to the validity of the basis of his plans and whether or not the Plan A for the marathon works.

Daniels started with a book, "Oxygen Power," which no one will ever read, but attempted to equate performances at different distances, as did Gardner and Purdy in "Computerized Running Training Programs." (1970) The Gardner/Purdy charts work well for populations of runners, but not particularly well for any one runner. The Daniels charts (Table 3.1 in 2nd edition of DRF) work well for the less than 2% of the population who are natural marathoners or ultramarathoners.

Daniels' theory of training falls along the line of the one presented by Scott (see earlier post), but divides training by percentage of one's maximal oxygen uptake and then tries to equate that with specific paces. He posits that one should train only at those paces which are maximally effective for gains in one component of running ability: (for elite runners) 74% for developing cardiovascular system (easy running), 88% for improving endurance (anaerobic threshold), 100% for stressing aerobic power (VO2max intervals).

There are multiple problems with this theory. First, tests of maximal oxygen uptake rarely have runners reach their maximum; there's some extrapolation. Three years ago, I could get my heart rate to 178. With heavy training at as close to maximum as possible, I have increased that number to 184, even though one cannot actually increase that maximum; I can only force myself to that point for at most three seconds, when his interval training would require me to hold it for a number of minutes. Second, the test has an athlete run on an inclined treadmill for about 12 minutes and is only a reliable test for events like that: it's a reliable predictor for 5K's for elite runners or 3K for back of the pack runners, because those races are about 12 minutes, but not for a marathon; it's no coincidence that the runner with the highest recorded VO2max (Matt Carpenter, 90 ml/kg/min) is most competitive in uphill races, like the Pike's Peak Marathon, as the treadmill tests one running uphill.

Training is specific, but not as specific as Daniels tries to make it sound (I hope to later show Brad Hudson makes the opposite mistake) and he ends up continuously patching his ideas until there's nothing left. First, he states that a threshold run is 20 minutes done at a pace one could race for one hour. Then he makes it a range of paces to be run up to an hour, where it overlaps with marathon pace running. Marathon pace running doesn't fit any of his categories, so he created another one for it. Then, beyond interval training, he had repetition runs, but he needed another pace beyond that, which he called "Fast Running" (cf. his 800m training schedules). Finally, in the 2nd edition, in Table 2.2, he adds 10K pace running, which now has every pace above recovery or ultramarathon pace. In other words, he admits all paces are valid, not just the specific ones he states.

But does it work?

The "Plan A" marathon schedule, first presented in Runner's World in 1996 under the title "One Size Fits All" is popular and has a lot going for it, if one ignores all the bad science he claims it's based upon. I won't infringe upon the copyright, but here's someone who did (schedule).

After a 6 week mileage build-up, he starts with intervals, then threshold runs, then marathon paced runs. This is another "top down" method, starting with speed, then extending the length of the fast runs while decreasing their pace and is the opposite of plans such as Glover's, which has one peak by running faster toward the end of the schedule. This makes complete sense for most runners, who are moving from 5K or 10K races up to the marathon; it is the length of time that one has to run fast that is the challenge of the marathon.

He does not give mileage totals, but states percentages of one's maximum one should do each week. He does not explain how he arrived at these numbers, but one assumes that as the individual workouts get harder, the mileage decreases. One of the things easy to miss in the book (and with which I disagree) is that he feels that, if you run more than 50 miles per week, you should run twice a day.

Hidden in the footnotes of his plan in the 2nd edition, he has runners doing 6-8 x 20-30 second strides twice per week. He doesn't explain why, though I think it is critical to the plan working - and it is remarkably the same as every single plan I've described thus far in this one detail!

He has runners do two very hard workouts per week. For example, in week 18, one runs a 19 miler with 15 miles at marathon pace (or 120 minutes, for slower runners). Every three weeks, he has one run 22 miles (or 2.5 hours or 25% of one's weekly mileage). He has a lot of runs that have up to 8 miles run at threshold pace, interspersed through a long run; again, not explaining the division of the fast sections. These are each decent runs, but they are each very difficult and probably too difficult to do twice a week for most runners. I expect most runners using this schedule find themselves failing in a workout and the fatigue carrying over to the next workout, making them fail at that as well; there is no "wiggle room" - you either do the workout as described, or fail.

The most telling thing to me is that he does not have his own runners follow this schedule, but an "Elite Marathon Training Plan" (included in the book). In this, he no longer has phases of training, but one does all different types of running throughout the season. And that's my method, though I doubt I could do any of the workouts he lists.


Anonymous said...

Are you still on the borrowed computer?

lovpowemn said...
This comment has been removed by a blog administrator.
Glaven Q. Heisenberg said...

Sweet Baby Jebus! How many "experts" are there? And do you intend to fight them all, Batman?

SteveQ said...

@G: There's a few yet to come (Pfitzinger, Beck, Higdon, Hudson). I know it's a long boring slog to the finish... but that's marathoning!

Colin said...

This is an interesting series; thanks Steve. I agree with most of your issues with the Daniels program. As a marathon plan, it just doesn't work for me. However for shorter distances (particularly 5K-10K), I find his tables helpful to indicate what sort of race shape I'm in. E.g. when I run one of his "hard" workouts and reverse-engineer a VDOT based on my interval times, the predicted race time is usually very close to what I can run. I find this quite helpful in setting realistic goals.

Joe Garland said...

What's a "natural marathoner"?

Joe Garland said...

Also, you write, " He has a lot of runs that have up to 8 miles run at threshold pace, interspersed through a long run; again, not explaining the division of the fast sections." Could you tell me which week and workout for Marathon A you're referring to?

In my experience, I've not found any of the described workouts, done at the pace indicated by the "Formula," "very hard" and have not had problems with recovery. But I have not followed his schedules slavishly, instead adapting them and interspersing races in the marathon build-up.

I've never thought the "formula" is intended to be a guide to running a "goal" race time. To the contrary, the formula is to give pace for the various types of workouts based not upon where I want to go but where I am at the time of the workout.

Finally, could you be specific as to what you mean by "the bad science he claims it's based upon"? And the basis for "The Daniels charts (Table 3.1 in 2nd edition of DRF) work well for the less than 2% of the population who are natural marathoners or ultramarathoners." and what "work well" means?

Colin said...


You have some good questions; I'll leave Steve to respond to those.

I've found the charts to work well for me up to the half marathon distance. Running the marathon time predicted by VDOT is tough, but I think that may have more to do with the difficulty of training for and running an optimal marathon. Shorter distance races have much more margin for error (and can be raced more often).

I do think some of his workouts are too tough, especially in his Elite Training Plan. I'm no elite, but on week 19 the plan calls for a primary workout of:

8M@E, 8M@MP, 1M@T, 4M@MP, 1M@T, 1M@MP -- no breaks, that's 23 miles of running, with 15 continuous miles at marathon pace or above.

If that doesn't kill you, the second workout that week will:

2M@E, 5M@T, 5:00 easy, 4M@T, 4:00 easy, 3M@T, 3:00 easy, 2M@T, 2:00 easy, 1M@T, 2M@E -- that's 20+ miles more of running, including 15 miles at your 10 mile to half marathon race pace!

Joe Garland said...


I'll take a look at those programs. My questions are intended to get clarification of some of the things in Steve's post before I respond more substantively.

In fairness, though, I'm not a big marathon fan (having done two 23 years apart) and have not followed DRF in specifics for a marathon. I've used it with some success for races of HM and shorter and think that the basic principles he articulates are a quite useful starting point for developing a training program.

I've heard Daniels say that the schedules are the least important part of the book, something done because the "market" demands them, and that it's the ideas and the formula that are what's important, applied individually. It's fair criticism to say "if you don't believe in the schedules why include them?"

As I say, I'll comment further when I see what else Steve has to say.

SteveQ said...

Joe, what I was thinking when I used the term "natural marathoner" was someone with 98% or more slow-twitch muscle fibers, someone who finds that they finish closer to the front of the pack the longer the race is and who finishes every race thinking, "if only it were longer, I'd do better."

In week 20, quality workout 1, he has a 2 mile warm-up, followed by 4x 5-6 minutes at T pace, then 10 miles or 80 minutes, then 4x 5-6 min. at T, then 2 cool-down. That's 22 miles with 8 miles at T pace for someone with VDOT of 59 (admittedly less for slower runners). There are other workouts with only slightly less at T pace.

You make a good point about the workouts being based upon where one is at the time, but if one's just finished a sub-40 10K and has a marathon on their schedule, they'll see VDOT 52 and 3:04:36 corresponds to that and the M pace runs are at 3:04 pace, so obviously, that's what their goal will be.

The chart in Table 3.1 I find can be used for predicting marathon finish times in just that way (i.e. "works well") for that subset of "natural marathoners," but not for most people.

As for "bad science," that was an unfair comment on my part. The science is sound enough (as a biochemist, I find most physiology to be "soft science"); the idea of using oxygen uptake as the one and only determinant of running success is highly questionable.

I can't go into every little detail of every book, but was trying to limit myself to the schedules for the marathon. I just couldn't resist making a dig at Daniels' hubris of thinking that running can be boiled down to a formula.

btw, looked up your second marathon. I haven't run that fast since 1985.