True or False, Salazar said

I spoke with a coach who was timing KK at the Oly training center in Chula Vista during a few of the 20 mile timed training sessions. These were done a few weeks before the Tokyo 1/2. Total time for the measured 20 milers was 1:35 and KK did this twice within the two week he was down there. As the coach said himself "I wouldn't have believed it myself if I didn't see it with my own damn eyes." KK also did 25 x 400 at 62 on a 2 min cycle during the 2 week stint in Chula Vista. Based on the 25 x 400 workout you would think 62 pace for a 5k, so that's roughly 13:00 or so there. But the main thing here is the ability to maintain his marathon race pace for 20 miles in a workout setting consistently. Talk about event specificity. The combination of the two workouts should indicate it's really a combination of factors that lead to success at the event.

As much as I respect Salazar, my opinion is that we are placing the cart in front of the wagon regarding statements such as this one. My opinion is that if we had more people training at a truly international level specifically for the marathon, then we would get those times requirements as a byproduct of this type of training. To say you need those times as a pre-requisite to running a fast marathon isn't necessarily true. There are just too, too many examples of guys being less than stellar at the 5k or 10k and beating the living heck out of "better" runners at the marathon distance. From personal expereince it has everything to do with event specific training, and some guys are just better suited to that sort of training and benefit from a marathon program more so than say a 10k program with a long weekend run attached.

My own feeling about the recent lack of success Americans have had in the marathon event has stemed from the assumption that marathon success lies in training like a 10k runner with an easy 2 hour run on the weekend. Sure some can do this (such as Salazar), but so can some people train like a 1500 runner and have success in the 10k. To be most successful at any event requires a large amount of time spent preparing specifically for the demands of that event. I think if many of the 29-30 minute 10k guys in this country started training at the levels of a true national class marathoner from the 70's, they would see the times in their shorter events improve dramatically as well as their marathon times.

The marathon requires specific training that American men today are not readily willing to do and the surprising fact is it's the guys at 30-31 minute 10k who would benefit most by having an Oly Trials experience who cry the loudest about the work needed to run decently fast. They explain that they try real hard but can't make a sub 2:22 time for some strange reason. When pressed they explain they are running 80 or so a week on 10 a day and 18 on the weekend. That ain't gonna get you squat.

When I am asked for workouts by prospective marathoners and I mention things like 30k time trials at marathon pace or consistent training days of 20+ miles that include long tempo work or 1k's or mile rep days such as 6 miles in the am, then a pm session of 6 x 1600 at 10k pace with 400 jog, plus a 4 mile warmup and a 4 mile cooldown and add a 8 hour workday sandwiched between the sessions as an example, they respond with "are you serious, that's 20+ miles in a single day and work too?!" It ain't rocket science, it's simply realizing that you can do 105 MPW a numer of ways. You can do it as 15 a day or you can do it as 3 days of 22, 1 of 18, 2 at 8 and 1 at 5. Which week do you think will get you better prepared for a race lasting 26 miles?

Salazar is right, take a sub 13:20 guy and feed him a bit of marathon work and you'll get some results, no doubt about it. But there are a ton of examples of guys going the other route of running a ton of weeks foscusing on the marathon, then dropping down to run some very, very good times at the shorter distances. The fact is, we have too few guys currently looking at the old school, work your tail off training for a marathon route and too many 30 flat guys thinking their "speed" will get them results. Look at the number of guys running sub 2:20 and realize this speed route ain't the way to get it done for the vast majority of folks.

Joe

Jersey_guy,

Good data. And I agree somewhat with what you say (as I will get into below), but just to take one runner you mention; Gert Thys and his 2.06.33.

You say he has never run sub-28/10k, and I believe you. But consider this, Geb can run a 10k , 33 seconds per mile faster than his marathon pace,. You don't think that if he really aimed for it Gert could not run a 10k a mere 19 secs/mile faster than his marathon PR? (and thereby break 28.00?)

I think most people would expect that he COULD do that than would believe that he COULD NOT do that.

I know I'm dragging you into a hypothetical situation here, and I don't like that, but you get my point. No need to agree, or disagree yet. Read on.

Joe Rubio,

Pretty much agree with you too, for the same reason. In fact if you waded thru my many posts on this topic you would have noted that I said that you need not have achieved the 5k/10k times required to pop the marathon time, but that you should be "capable of it" if you're going to pop a world class marathon.

Which is like I explained to Jersey_guy above; I would say that if you can break 2.07, it's pretty much a given that you can go back down, train seriously and run a sub-28.00 time.

Note that I also said you did not have to be able to do both of these concurrently (even within the same year!). In fact I pointed out that if you currently have the speed for 5k/10k there's a good chance you'll have compromised your current marathon potential.

I think in essence what Al Sal is doing is saying to himself, "Before I could run 2.08, I had been able to run XX for 10k, so I need to search for some guys in that kind of shape and move them up."

It's logical, although maybe not the only possibility.

You term this thinking "cart before the horse" and think runners should just train at international level for the marathon before they show some kind of credentials at 10k.

I think many runners today would see this commitment as a big leap of faith. In the absence of a sub-28 time, what is to convince me I've got what it takes? Have I to invest in a few solid years of 120mpw and then hope it's gonna be alright on the night? Tough sell.

You are bang on the money with the training required for marathon success, as I'm sure you know, but it is not so easy these days to get runners to commit to it. They tend to want more guarantees now that it's all gonna lead somewhere good (and who blames them?) And is an Olympic slot really a good enough carrot these days? That won't put a Range Rover Discovery in the driveway.

But all in all I pretty much agree with you. Salazar is just doing it one way that he knows was true for him.

More notes worth thinking about:

Geb's famed finishing kick and stellar 10k times meant that he had to slow up some 33 seconds/mile from 10k to marathon pace.

KK, with no great 10k time and no great final 200m kick only slowed up some 22 secs/mile.

Since kick seems to imply at least a certain percentage of fast fibres, which are not so necessary for the marathon, and in KK's case (and those of Jersey_Guy above) that lack of fast fibres (and more slow twitch) resulted in some form of greater economy at the marathon. Seems like what you really want to look for as marathon potentials are guys with only good (rather than excellent) 10k times and absolutely no finishing kick.

Think Paula Radcliffe.

Thanks for the input on KK's training. It's pretty interesting that he can knock off 20-milers in practice at 4:45 pace, but his half marathon PR is only 1:00:27 (4:37 pace), and his tanking in Sapporo is just unexplicable - his 13.1 split on that 20-miler would have been over 2 minutes faster than his Sapporo time (1:04:30-ish). He is just one tough cookie to crack.

Also, thanks for mentioning the alternative ("better") ways of training for the marathon. How often do you believe those 30k time trials should be done and how late before the goal race? What exactly do you mean by long tempo work? I don't think it's necessarily the lack of committment among the US distance runners that's the problem - it's often the lack of experienced coaches who would put their asses through that kind of training, sans altitude chamber or with it.

KK's 20-milers are no recent innovation. I recall watching Gelindo Bordin run in some low-key marathons in Italy and stop at around 20-miles.

He be invited. Do the talk. Press the flesh. (Genuinely nice guy, by the way. Unassuming, gentleman. If you didn't recognise his face, you wouldn't know who was the Olympic gold medallist in a room full of people.) Come race-morning, he would be cruising along in the second, or third pack. Maybe 30 secs off the lead. In that white cap he wore. Bothering nobody. Thinking nuthin'. Worryin' about nuthin'. And just stop at 32k where his helpers would be waiting with some drinks and a van to the finish.

This was around his European/Olympic period so would be '87-89.

No new ideas under the sun.

When I read of KK's running the same training, maybe two years ago, I remember realising that his 20-milers would be at a pace that would win 80% of the half marathons on the planet.

Don't recall him showing up at a whole slew of these for the paychecks in what would really be a "training run". Maybe make those who believe he is in it for the money think again.

Bordin was famed for his negative-split, come-from-behind racing technique. Aside from the physical aspects of such event-specific training, it's got to be a huge psychological boost to think, all the way to 30k+, that you've done this 4-5 times in your build-up. Think about it, 3.03/km is no big worry for 32km, when you've done it a number of times in training. Hell, it's training pace!

While all around you are guys who have not run 20 miles at this pace since their LAST marathon, six months ago.

I really admire the runners, such as Dan Browne who I KNOW can run under 2;10, but yes I too find Alberto's axiom about "you have to run under 28 first" to be total crap.

Atsushi Fujita has PR's of 14;09 for 5km and 28:20 for 10km, and his marathon PR is 2;06;51, course record at Fukuoka.

Takayuki Inubushi has a 10km pr of 28;09, and his marathon PR is 2;06;57.

On the women's side, there have been many Japanese women under 2;25, but no Japanese woman has ever run under 15;00 for 5km. However, Kayoko Fukushi should do it any day now.

I think that the Japanese have the full marathon down to an art form.

Jason

Jersey Guy,

Looks like you had a couple of specific questions here, namely how often to do the 30k time trial and my definition of long tempo work.

1. Since most of the people I work with tend not to have the background to handle a ton of these, I usually advise the following: 3 months from the target date a 8-10 miler at marathon pace. 2 months out a 12-14 miler and 3-5 weeks out, a 16-18 miler all at marathon pace. Not faster, not slower. I want it at pace. In Nor Cal they have a road series (PA's) that has a 30k 3-4 weeks before CIM, so I advise the athlete to take a few easy days before the 30k and prepare to run at marathon pace for the entire 18.6. Now given the fact the athlete doesn't have the benefit of a full taper, they usually have to run hard to hit the time. In my own personal example I always ran CIM a tick or two faster pacewise than the 30k. Still the 30k does achieve a few things. It tells the athlete if their time goal is legit. It also prepares the athlete mentally for the job ahead in the marathon. The thing is, after a 30k with no taper, the marathon is a tad easier. Not sure if it's a byproduct of the training effect or from mentally approaching what you'll need to do on the big day. It's probably a combination of the two. The other thing on these time trials is to run in the flats, socks and shorts you plan on using in the marathon to ensure that nothing goes wrong that's fully in your control. You also want to course to repilcate the marathon course you intend to do (ie downhill for Vegas or St George, rolling for CIM, pancake flat for Chicago). As much as possible do a course without stop signs, traffic lights or cross traffic that will make you stop. If you can have the drink or Gu or whatever you intend on using in the marathon available so you can see how it affects you the day of the time trial, that would be ideal. I ask for a few light days beforehand and no hard workouts for at least 4-5 days following. Sure get back on the miles as before the day after the time trial, but not hard workouts until later the following week.

2. My long tempos usually go as follows and I ask for one every other week. 30-50 minute AM or lunch run then the following after work: 4 months out a 50-60 minute run followed by 4 x 1600 at tempo/1/2 marathon pace (15-20 second per mile faster than marathon pace) with 60 second recovery and 1-2 mile cooldown. 3 months out a 50-60 minute run straight into 4 mile at 1/2 marathon/tempo pace with 1-2 mile cooldown. 2 months out a 40 minute run straight into 4 mile at marathon pace, 1 mile easy, 4 miles at half marathon pace, 1-2 mile cooldown. The last one is what I call a "marathon tempo". For those looking at sub 2:20 for the 26.2 mile distance, I would expect the first 4 to be done at 5:20-25 pace and then shoot for a sub 5:05 for the last 4. In '96, I ran my final marathon tempo finishing up the last 4 in 19:32 (average of 20:30 in the lead up months) and I ended up with a 2:20:30 on a hilly Charlotte course.

Let me know if you have additional questions. Joe

Hello people. Let's buy a clue.

Khalid ran 27:44 on the road at Peachtree. That race is held on July 4th in Atlanta.

July and Atlanta together mean it's freaking hot with a lot of humidity. The average high is 89.4 and low is 70.6 for the month of July in Atlanta. I have no idea what it was like in 1998 but they weren't ideal.

Plus the race is on a road course that isn't totall flat either. All I'm trying to say is that Khalid is way better than a 27:44 guy. I'm tired of seeing people see someone's road times and then thinking that's as fast as they could run.

If you don't think Special k is way better of a 10k runner than say alan culpepper or dan browne or any other guy with a pr that fast or faster, then you are an idiot.

When he ran 27:44, he was winning virtually every road race in sight. Meb Keflizighi would be lucky to win one major road race and his pr is 27:13.

In a road race, you are running for one reason primarily and that's to win (unless there is CR bonus).

Great thanks for all the info, Joe. This year I am actually shooting for a half marathon at the end of September, but I will try to incorporate similar training in the last 2 months before the race (e.g like a 12-15K time trial at half marathon pace 3-4 weeks before the race). And those tempo workouts seem quite interesting too, maybe I should try them out.

Deena did the NYC marathon and set a road 5K world best and 10K track pr after her big 'thon debut last Nov. Culpepper and Meb ought to be interesting to watch at Chicago this fall given their heady 10K prs.

Yes, Deena has some good 5K - 10K times (OK, they're great times) but like Hadd pointed out, her kick ain't what gets her there. Who can forget the demoralizing sight of her getting hammered by Regina at the Olympic Trials after pulling her along all those laps?

I think there might be something to Hadd's fast twitch/slow twitch idea as far as marathon potential. Myself, I ran mostly 800/1500 in college and did OK once in a while in a sit and kick 5K. Post-college, I put together a few 10K road times that, on the damn charts, indicated some marathon potential. My marathons, however, were pathetic. So, either more evidence on the fast-twitch slow-twitch thing or, as my friends argue, evidence that I have almost zero mental toughness.

Slow twitch/no kick the key to marathon? No - Rod Dixon (3rd. in Olympic 1500) could handle marathon, as could 3:55 miler Geoff Smith, as could 4:00 1500 meter runner Grete Waitz. Greg Meyer ran sub 4. What the hell, what did those two marathoners named Tergat and Gebrasellasie have for 1500 pb's - 3:31 or something. KK is a good kicker, plenty of speed in that guy's legs - he's quick and fluid on or off a track.

Good to see the topic getting some discussion. Even I am interested in the input (positive or negative) of others.

Let me correct what may be one misconception. I am saying 10k runners with no kick may possess more marathon potential than other runners who may beat them over 10k by having a more lethal finishing kick.

I am not saying those no-kick athletes are slow by any means. I know well of Rod Dixon (the original shuffler if you saw his running style) and his New York duel with Geoff Smith and Gidamis Shahanga. Maybe one of the best New York marathons ever.

Sure, these no-kick runners can run fast at shorter distances, but only in a fast race. These are not athletes who get in tactical events and burn the last 400m and still get reasonable times. If the race is slow in the first 1200m for them, then the total time is slow, they cannot "fix" a slow early pace.

These are athletes who pretty much rely on getting in fast races to run fast times. Yes, they can break 4.00mins (as persepective has shown), but if we could check the splits we would see that it was pretty even-paced the whole way with no great pace injection in the last 200m.

To make it clearer. I am saying (proposing/positing) that if you take two or more runners with similar 10k PRs then the one who runs his PR without a kick will have more marathon potential than his adversary.

Grete Waitz was undoubtedly a great athlete, but her slower (at 1500/3000) and non-kick compatriot Ingrid Kristiansen achieved faster 5k, 10k and marathon times.

^^hot take