Forums >General Running>McMillian Calculator
Duh. I corrected myself earlier and said that didn't work for me...if you bothered to read the 2nd post.
OP already stated that here mileage is in the low to medium range (35-40 MPW) for optimal marathon performance. Your suggestion seems to be a recipe for disaster.
"Shut up Legs!" Jens Voigt
My curve falls off faster than the McMillan calculator at all distances going up, I believe. That is, my 10K PR is a bit slower than McMillan predicts as equivalent to my 5K, and so forth all the way up to the marathon. I score better against my age group competition at shorter races, so I suspect that my curve is just more favorable to shorter races.
My curve falls off faster than the McMillan calculator at all distances going up, I believe. That is, my 10K PR is a bit slower than McMillan predicts as equivalent to my 5K, and so forth all the way up to the marathon.
I score better against my age group competition at shorter races, so I suspect that my curve is just more favorable to shorter races.
+1. Those McMillan times sitting out there have been the bane of my existence when it comes to marathoning. Although I am grudgingly coming to accept that either I have not properly trained for the longer distances, or I am just not built to meet his formula. I imagine that regardless of training, not everyone's natural race time dependence curve is the same. The RA tool of course lets you predict from 2 races rather than 1, and also lets you enter the exponent to create your own formula.
Dave
Feeling the growl again
McMillan works well if you are a slow-twitch or very high mileage (i.e. real marathon specialist) runner. During my peak, it actually predicted slower than realistic times for my marathon and predicted blazing fast 5K-and-under times I couldn't have run with a gun to my head, based on 10K/HM inputs.
The fact is, most runners are under trained for long races. Therefore, the predictions will be aggressive.
"If you want to be a bad a$s, then do what a bad a$s does. There's your pep talk for today. Go Run." -- Slo_Hand
I am spaniel - Crusher of Treadmills
Options,Account, Forums
McMillan works well if you are a slow-twitch or very high mileage (i.e. real marathon specialist) runner. During my peak, it actually predicted slower than realistic times for my marathon and predicted blazing fast 5K-and-under times I couldn't have run with a gun to my head, based on 10K/HM inputs. The fact is, most runners are under trained for long races. Therefore, the predictions will be aggressive.
(Rewriting, as I have thought about this further, and think I was conflating the general with the specific.)
(I am aware of how it panned out for your case, but am not sure how typical you really are; your performances seem biased for better at long-distance, right? So I wonder if that means higher mileage is more beneficial for you even in mid-distance racing, as some corollary.)
For my personal case, the problem for me is that I think I am not willing to actually try out mid-to-high mileage training (let alone high mileage) and it is difficult to otherwise find out how well it would work, I suspect.
I am not sure what the right numbers are. I arbitrarily picked in my own mind
70mpw = mid
85mpw = mid-to-high
90+mwp = high
It's a 5k. It hurt like hell...then I tried to pick it up. The end.
Since the post under discussion in this set of two responses is mine, are you suggesting that the fact that my 10K is slower than McMillan's prediction (from 5K) means that my training was insufficiently high mileage for 10K? And similarly for mile to 5K, that my training was insufficient mileage for 5K? (I am neither agreeing nor disagreeing, just asking if this is what you logically meant.)
Since the post under discussion in this set of two responses is mine, are you suggesting that the fact that my 10K is slower than McMillan's prediction (from 5K) means that my training was insufficiently high mileage for 10K? And similarly for mile to 5K, that my training was insufficient mileage for 5K?
(I am neither agreeing nor disagreeing, just asking if this is what you logically meant.)
1) If your actual performance at longer distances based of off McMillan predication from any shorter distance doesn't measure up...
a) Mileage is a convenient but not exclusive surrogate for "training". The algorithm assumes appropriate training for the distance. This is a compilation of mileage and quality. For most people, mileage is the issue. You are a bit of a difference case, and the balance of quality and quantity may play more of a role than average.
b) There are different types of athletes. If you are a shorter distance guy, your longer distance predictions will suffer no matter what. I'm the opposite end of the extreme, I excelled at the longer distances but no about of training or dedication would have gotten me to run the 5K/mile/1500/800 times that my marathon/HM/10K predicted.
The most common situation is people do well at shorter distances then under-perform at longer. If there are exceptions to this, mileage is not always the answer.
I think what Spaniel is saying is that you might be a fast twitched who does better at shorter races. For most others their mileage is the reason why calculators work reasonably well up to the HM distance, but not the marathon.
I kind of like the Slate marathon calculator. It seems much more realistic. It's giving 3:38:25 for the information you gave which would be a very comfortable BQ.
Oops, I did not see that you had already replied Spaniel, and tried to restructure my question to make it a bit more general.
For my own case, I have some theories, but I am not sure that I am willing to really test them.
I suspect I have weaknesses in specificity of training, and in intensity, and I think that that may hurt me at longer distances more than at shorter distances.
I had a pretty big PR in the mile after a season of hard 5K oriented training, aimed at a 10K goal race; the only actual breakthrough I had after that was in the mile, but it was a big PR. I suspect that was the only season where I had such high intensity training, and so another hypothesis is that mile happened to benefit from it more than the few other races I had at that time, for some other (unknown) reason.
There are just so many confounding variables in the experiment of one person.
jfa
Slate gives me (with 2 race times to compare) a 3:44 and RW a 3:31. Quite the difference.
After I ran a tune-up half of 1:41 last spring, McMillian put me at 3:33 for my then upcoming marathon.
On a cool and flat course I ran a 3:35.
Slate gives me (with 2 race times to compare) a 3:44 and RW a 3:31. Quite the difference. After I ran a tune-up half of 1:41 last spring, McMillian put me at 3:33 for my then upcoming marathon. On a cool and flat course I ran a 3:35.
You've got a good amount of experience with the marathon. And weren't you running 80 mile weeks? That seems like it'd put you closer to the RW/McMillan end of things. Not the typical marathoner. Congrats again on the BQ btw. What's the next goal?
I used the Slate calculator for my second marathon which I maybe averaged 35-40 miles a week and it was pretty dead on predicting 3:54 and I ran 3:56. A far cry from McMillan 3:45 or RW 3:42. Just saying Slate may be more realistic for the average relatively low mileage marathoner.
I always underperform the McMillian calculator in the marathon but the Slate calculator was equally off in the other direction. I probably always exceed that Slate calculator.
The slate calculator is probably a good thing for most newer marathoners because it keeps people who are still feeling out the distance from getting in way over their heads, which I believe McMillian may lead many to do.
I kind of like the Slate marathon calculator. It seems much more realistic.
Heh. I don't think there is a calculator that captures how much I suck at marathons. But this one moves it in the right direction.
FWIW, even more aggressive than McMillan is the Daniels calculator.
I always underperform the McMillian calculator in the marathon but the Slate calculator was equally off in the other direction. I probably always exceed that Slate calculator. The slate calculator is probably a good thing for most newer marathoners because it keeps people who are still feeling out the distance from getting in way over their heads, which I believe McMillian may lead many to do.
Your point about the Slate calculator is good. I think it totally saved me and made my second marathon a good strong race that I really really enjoyed. I was smiling and pretty ecstatic at the end. I got that opportunity to feel the distance out while pushing hard but not out of my range. I'm looking forward to the day I can just really go for it.
Heh. I don't think there is a calculator that captures how much I suck at marathons. But this one moves it in the right direction. FWIW, even more aggressive than McMillan is the Daniels calculator.
I don't know if you suck at marathons or not but I'm giving myself more time to figure it out. Some folks around here have suggested working on the 5k to get better at the marathon. That's my plan for the spring.
I train specifically with the goal of getting better at marathons; this gets me better at 5k's as a result. Some of the smarter folks here have said training for the two distances is pretty similar anyway until the last x weeks before the race. (I don't recall what x is, you can see here how much attention I have been paying.) I assume that continuing to improve my marathon will continue to improve my 5k, so I am skeptical as to whether I will ever match up to the McMillan formula. Which is fine.