Holding marathon pace

800 - I would submit the opposite hypothesis. That the runner who believes he can only run a marathon as fast as he can run 20mi in training is in fact the one who is underperforming.

Negative splitting is certainly a noble aspiration in a marathon. But did that negative split happen because they started at a pace that they thought was sustainable because it's what was sustainable for 20 miles in training? Or because of some physiological adaptation that occurred from running that far that fast in training? Is it possible that they could have even split an even faster time if they had a little hutzpah?

Obviously this discussion loses some context without info on the surrounding training, but I consider the 20mi marathon pace run the tool of the 4hr/40mpw training crowd. One hero workout and a week to rest up for the next.

going faster miles an hour wrote:

Since Don Alfredo and Coach Jeff brought up Canova, it seems useful to note that Canova, who emphasizes the importance of building specific endurance with fast long runs as much as just about anybody, DOESN'T appear to use long runs at MP with his athletes. Instead, he has them do their fastest long runs just a little slower than MP, generally about 95% of MP. (Most of the work that Canova's guys do at MP during the final period of training comes in the form of long intervals.)

Marathon Training: A Scientific Approach:

[C]ontinuous runs at marathon speed (or MARATHON RHYTHM) and at a slightly higher or slightly lower speed, usually between 97 and 103% of marathon speed become the most important training means as the date of the competition approaches.

And the specific workouts listed include 30-35km runs at 98-100% of MP.

95% is never marathon specific, though it is sometimes the upper speed limit on a recovery interval for MP reps.

It depends what you mean by underperforming. I don't doubt that the vast majority of marathoners will run their marathon faster than their fastest 20-miler. My point is that these people are STILL underperforming because they have the talent to do more, which is usually born out by looking at their shorter PRs. I have met very few runners who can run a marathon at a time that's close to "equivalent" (via VDOT or whatever) to their shorter race performances. That, to me, is underperformance.

I'm not advocating negative splitting. I'm advocating even splitting as a more likely indication of a race run to one's capacity. The hutzpah is in trying to do 26.2 at MP when you know how brutally hard it was to do 20 at MP.

It's actually a tool of the 2:05 marathon crowd. For a four hour marathoner, MP isn't really a relevant concept.

Smoove wrote:

I would submit to you that there are various ways to get to that point. As you know, I'm a pretty big Daniels advocate. He would suggest that we extend our ability to run at marathon pace by the dressing other specific limiting factors (in particular improving our lactate threshold so that we can stay aerobic at faster and faster paces). I think that taking alternative approaches and not trying to run 18 or 20 miles at marathon pace can achieve the desired goal with less stress on your body.

At bottom, my issue is that I don't really think it makes sense to talk about achieving the desired goal with less stress on the body. The desired goal IS stress on the body. You may be able to elicit some of the adaptations that you need by doing other workouts, but whatever it is that you're avoiding is likely going to be the thing that's missing when the chips are down.

I would certainly agree that long MP runs are a very bad way to train if your objective is simply to maximize your aerobic capabilities and get a high LT, which is why no 10k runners do long MP runs. But there's a lot more to running a marathon than having a good LT. Most runners have far more aerobic fitness than they need to achieve their marathon goals, but their legs can't take it.

So what did the average sub3 runner do for his long runs in the 70s, 80s and 90s? Where they just steady efforts?[/quote]

That's right, people just ran back then and didn't worry about pace very much and guess what? The majority ran faster and the depth was incredible. For a 3 hour marathoner the long run tended to be between 2-2 and a half hours once week, the rest of the week was 8-12 mile steady runs. They did race a lot too at other distances, 10K to half marathon. A lot of times these were hard but not all out efforts. The elites trained in a similar fashion but usually had a mid week longer run (15 miles or so), more overall miles (130-140 a week), 5-7 miles of intervals once a week at around 10k pace and, of course ,racing more often and at various distances.

All of this marathon pace training has made for slower American marathoners, all the way up to the elite level. The drop off in marathon performance in the US since the invention of the Garmin and the obsession with training at marathon pace is astounding. Here are some numbers:

1983 -- 23 under 2:13, 32 under 2:14, 44 under 2:15, 193 under 2:20

2018--1 Under 2:07, 2 under 2:16, 7 under 2:19:36 (American women's record), 10 under 2:20.

I would love to see someone or a some training group go back to the training of Rodgers, Meyer, Durden and literally hundreds of others and see what happens. We really don't have much to lose.

So what did the average sub3 runner do for his long runs in the 70s, 80s and 90s? Where they just steady efforts?[/quote]

Per my previous post, there were quite a few sub 3 hour marathoners back then too...

In 1978, 3,872 runners (mostly men at that time) were recognized in the published results for finishing the Boston Marathon (which in 1978 meant finishing within four hours). Of those finishers, 2,047 broke three hours.

The reason that people underperform in the marathon is usually either inadequate volume or, more likely, because the marathon is a long race, and in long races, there are more variables (hydration, nutrition) and small problems (heat, wind, ending up in no man's land, cramping) become big problems.

The expectations for longer races established by shorter races assume the same conditions as in shorter races. But the magnification of the ill effects of those variables means that is rarely true.

I agree that I could not pick up the pace in the last 10K of my PR race at Chicago 2017 due to leg (specifically) calf cramps. I ended up with a +10 second near even split. I did not feel like I had problems with excessive heart rate or shortness of breath in the last few miles. I have tried to remedy this by doing more lower body strength training and increasing my training volume. In addition of MP (6:15-6:30) and Min-MP (6:40- 7:30) long runs.. I will do Intervals like 10x800, 8x1000, 6x1, 4x1.5, 3x2, 2x5000 at VO2Max/anaerobic (5:00-5:30 pace) and 7-10 mile LT (5:45-6:05) tempo runs as my other quality workouts. My point is I don’t only do a MP long run as my only quality workout and I do a lot of easy/aerobic Miles as well (7:15-8:00 Pace). On the other hand, it is true I don’t see too many other Strava friends doing MP long runs, even guys who run sub-2:40’s. And to Smoove, I do seem to recover fine after these MP long runs as long as I do a day of slow recovery running the day after.

If you’re running 18-20 miles at marathon pace and you’re not completely ruined for a week or two after, then it’s probably not your marathon pace.

Six weeks prior to my first marathon (in the foggy mists of 2007), I set out to do a 20 mile race at goal marathon pace. I ran 12s/mile faster than that, and was pretty wrecked the following week. But even that was slower than my eventual marathon pace, which turned out to be another 11s/mile faster (in a very oddly paced effort). There were a good number of other 18-20 milers that I thought were at M pace that turned out to be 20-25s/mile slower than what I would run (that’s why they worked).

The upshot is that if you think you’re doing long runs at M pace, and recovering in a reasonable timeframe, you might be doing Canova 95% runs. If you’re not recovering quickly, and it really is at M pace, it’s probably not a good idea, and is likely keeping you from accomplishing other work you could be doing when you’re stuck recovering.

Far Out West wrote:

So what did the average sub3 runner do for his long runs in the 70s, 80s and 90s? Where they just steady efforts?

That's right, people just ran back then and didn't worry about pace very much and guess what? The majority ran faster and the depth was incredible. For a 3 hour marathoner the long run tended to be between 2-2 and a half hours once week, the rest of the week was 8-12 mile steady runs. They did race a lot too at other distances, 10K to half marathon. A lot of times these were hard but not all out efforts. The elites trained in a similar fashion but usually had a mid week longer run (15 miles or so), more overall miles (130-140 a week), 5-7 miles of intervals once a week at around 10k pace and, of course ,racing more often and at various distances.

All of this marathon pace training has made for slower American marathoners, all the way up to the elite level. The drop off in marathon performance in the US since the invention of the Garmin and the obsession with training at marathon pace is astounding. Here are some numbers:

1983 -- 23 under 2:13, 32 under 2:14, 44 under 2:15, 193 under 2:20

2018--1 Under 2:07, 2 under 2:16, 7 under 2:19:36 (American women's record), 10 under 2:20.

I would love to see someone or a some training group go back to the training of Rodgers, Meyer, Durden and literally hundreds of others and see what happens. We really don't have much to lose.[/quote]

You're comparing an entire year of running to 5.5 months of running, which includes a once in a century storm at Boston where many people usually run well and the USATF marathon championship was just last December. You can make the glory days argument if you want but do it with real numbers. At CIM alone there were 44 at 2:19:36 or less. Most of them haven't run a race since, or they ran Boston and were flooded out.

You're comparing an entire year of running to 5.5 months of running, which includes a once in a century storm at Boston where many people usually run well and the USATF marathon championship was just last December. You can make the glory days argument if you want but do it with real numbers. At CIM alone there were 44 at 2:19:36 or less. Most of them haven't run a race since, or they ran Boston and were flooded out.[/quote]

Fine, I'll give you all of 2017. 40 American men ran sub 2:20, 38 under the current women's AR (that's American Record). I'll take 193 v 40. Keep racing your training runs and keep under performing. The U.S. Should have more than 10 sub 2:20 runners by January 15th ( let alone June 25th) on any given year despite hurricanes, tornadoes, floods, typhoons or any other natural disasters.

44 under 2:15 in 1983 v 40 under 2:20 in 2017.

It was a different culture back in the early 80s, when US runners were world class marathon runners.

The culture has changed in Africa too and the super fast guys are doing long fast runs at altitude on dirt roads.

Asphalt is much harder on the legs than trails, although it is faster. I did both.

Smoove wrote:

The reason that people underperform in the marathon is usually either inadequate volume or, more likely, because the marathon is a long race, and in long races, there are more variables (hydration, nutrition) and small problems (heat, wind, ending up in no man's land, cramping) become big problems.

The expectations for longer races established by shorter races assume the same conditions as in shorter races. But the magnification of the ill effects of those variables means that is rarely true.

I pretty much agree, though I think it's hard to disentangle inadequate volume from inadequate marathon-specific workouts. The former supports the latter.

I also take your point about how it's harder for the stars to align in a marathon, but I don't think that's the main reason that most people perform badly in the marathon. You can throw out marathon races where there are obvious confounding factors like terrible weather, and you can throw out 10-year old track PRs, run at last chance meets, under the lights. Even comparing apples to apples, people still struggle to come anywhere near the kinds of marathon performances that their shorter distance races should indicate.

Something always changes on race day. This was a couple summers ago when I was out of shape. I remember reading to go half marathon effort for like 2/3's the race distance 9 days before. I ran 8.5 miles @ 6:25 pace and it felt pretty intense. I was kinda nervous about the race because I was definitely hurting.

9 days later I started the half marathon @ 6:25 pace, and ended up finishing my last 8 miles @ 6:10 - 5:50 pace no problem. I even felt like I had 3-4 miles left in my legs. Something happens @ the race and the rest leading up to it, even more so than in 800m - 5000m races.

I gotta say, I've heard a lot of theories about why US marathoning is worse than in the good old days, but "too much MP" is a new one. It's especially strange considering that "more MP" is also the typical explanation for 2:05 and below being so normal these days. I don't think macro-trends of amateur runners are particularly relevant when the question is how best to train. Much better to look at what the world's best athletes are doing--the laboratory of sport is a century ahead of scientists--and then make the compromises that are necessary for your own life.

Just anecdotally, I'll say that I've done 4 MP runs of 20 miles in my life, preceding 4 marathons. Two of them preceded PRs at exactly the pace I ran the workout at. One marathon was unseasonably warm, so I can't really judge how well it went. And in the other I went out 90 seconds faster than goal pace in the first 10k and blew up. I should probably add that all of these MP runs were safely under 2-hours; it's entirely possible that for slower runners, the longer time it takes to complete a 20-miler is what pushes it over the edge and makes it too much. One of the tricky things when we talk about marathon training in general is that the huge range of finishing times mean that we're really talking about different events for different kinds of runners. For a lot of people, the marathon should probably be approached more like an ultra.

Anyway, I didn't come on here to tell people they should do 18-20 mile MP runs. I posted because people were treating long MP runs as an unambiguously bad training tool that should never be used by people who know what they're doing. I wanted to point out that there is in fact no such consensus, that there's a strong case for longer MP work, and that many of the best athletes and coaches in the world use them. Whether they're right for you is a different question, and as I noted earlier, they're probably wrong for you unless you've fully adapted to pretty high mileage (120/week).

Key MP workouts:

4-6 mile ‘warm-up’ run at MP + 1-1:30

Then one of these:

3x 3 miles

3x 4 miles

2x 6 miles

3x 5 miles (2.5-3 weeks before race)

These are at MP or 5 sec faster

2:00 jog between reps

Okay, probably my last post on this one.

Sub 6:00 wrote:

If you’re running 18-20 miles at marathon pace and you’re not completely ruined for a week or two after, then it’s probably not your marathon pace.

I don't dispute your experience, but you really can't just turn your own experiment of one into a universal training law. It's pretty easy to demonstrate that this isn't always true. I think it's pretty well accepted that one's marathon potential is almost never "superior" to a recent 5k or 10k time using the accepted conversion tools. It's very rare for a runner to outperform their shorter distances, and even when they do, it's usually by a very small margin. So if you're doing 20 miles at an MP that you projected from a recent hard 5k or 10k, then you are, if anything, being pretty ambitious. Well, I can tell you that I recover pretty quickly from those runs. I'll do 12 miles easy the next day (in two runs) and 16 the day after that. I feel flat, like my legs don't want to bounce at all after each stance phase, but I have zero muscle soreness. By the time 6 days have passed, I'm better than 100% and my legs just want to fly.

Far Out West wrote:

Fine, I'll give you all of 2017. 40 American men ran sub 2:20, 38 under the current women's AR (that's American Record). I'll take 193 v 40. Keep racing your training runs and keep under performing. The U.S. Should have more than 10 sub 2:20 runners by January 15th ( let alone June 25th) on any given year despite hurricanes, tornadoes, floods, typhoons or any other natural disasters.

Where in the hell are you getting your numbers? There were 44 men with US hometowns running under 2:19:36 just at CIM 2017.

You're also confused with my training philosophy. You'll note above that I said doing a 20MP training run was wrong. "Super duper wrong" in fact.