The Keys to Marathon Training: Modern changes to Renato Canova’s elite marathon training methods

While researching my blog post on Renato Canova’s marathon training book, I came across a lecture that Canova gave at a coaching conference put on by Spanish marathoner and coach Antonio Serrano in 2017. The talk, called The Keys to Marathon Training^[1] was held in conjunction with the 2017 Valencia Marathon.

This lecture directly answers one of the questions I had when writing up my analysis of Canova’s book–what’s changed since 1999? From his answers in a 2011 interview, I knew that Canova believed some important things had changed, but that video didn’t go into too much detail. Canova’s talk at this conference goes into much more depth, so I wanted to do a more formal write-up on it.

In addition to addressing the new elements in training since 1999, like increased specific volume and more emphasis on building the ability to remove and oxidize lactate, Canova’s lecture provides more details about some of the types of training that were only briefly mentioned in the 1999 book, including his circuit training routines.

To make this talk more accessible, I used OpenAI’s state-of-the-art Whisper model to get an initial transcription of the lecture, then I manually corrected any errors and annotated the transcript with useful information or context. I’ve made a few very minor idiomatic changes to the text to make it more readable, but any more significant changes are [in brackets like this]. I also added emphasis where appropriate. The topical subheadings in the text are my additions as well. I’ll present my main takeaways up front, then provide the transcript and footnotes afterwards.

My key takeaways on modern Canova-style marathon training 🔗

I noticed seven themes in Canova’s talk that are worth keeping in mind. Here’s a quick summary; below, I’ll go in detail on each point.

• Mileage is lower for top marathon runners today.
• Increasing “modulation,” or day-to-day variability in training volume and intensity, is critical for doing workouts that enable top performance.
• The way to improve your training is more about adding new elements of training, versus replacing one type of workout with another.
• Marathon-specific workouts are higher in volume today.
• There is a greater focus on workouts to oxidize lactate aerobically.
• Starting from speed is easier than starting from endurance.
• You want to touch on many different speeds in training.

In-depth analysis of modern Canova-style marathon training 🔗

Here’s a deeper dive into each of the seven key themes from Canova’s talk, with some focus on what has changed in elite training methods for the marathon in the past 20-25 years, and some focus on the overall points of emphasis.

Mileage is lower for top marathon runners today. Often, top elites today “only” run 100-120 miles per week, as opposed to the massive (180+) mileage levels mentioned in Marathon Training - A Scientific Approach.

This lower mileage allows top runners both to do more high-volume, marathon-specific training, since when you’re running mega-mileage it’s difficult to be fresh for the longest, fastest workouts, and difficult to recover well afterwards.

More controlled mileage levels also help enable a longer career: Canova notes that many athletes who did mind-bogglingly high levels of mileage wound up destroying their body within a year or two. A more moderated and periodized approach to mileage allows for better career longevity, something that’s evident today both in Kenyans like Eliud Kipchoge and Americans like Keira D'Amato.

Increasing “modulation,” or day-to-day variability in training volume and intensity, is critical for doing workouts that enable top performance. Modulation is mentioned in Canova’s 1999 book in passing, but today, Canova argues that it is not possible to achieve very high quality in training if all of your running is in a narrow range of speeds.

To paraphrase something Canova says in this talk, if you never run very slow in training, you can never run very fast either.

Increasing modulation to achieve higher workout quality goes hand-in-hand with lower overall volume. When mileage in between workouts is lower, and the pace is slower, it’s much easier to arrive at your next workout feeling fresh, ready to run fast for a long time.

The way to improve your training is more about adding new elements of training, versus replacing one type of workout with another. Canova highlights his experience with contemporary European runners Arne Gabius and Sondre Moen to show that adding in ingredients that are missing in training can lead to huge improvements in marathon performance.

This has been my own experience as a coach, too–often when I start working with runners who are reasonably fit, they’ve already found some training ingredients that work for them. What they need to do is add new, missing components of training, versus completely scrapping what they’re doing already.

Marathon-specific workouts are higher volume. In the ‘80s and ‘90s, Canova says he would use 20-25 km of marathon-specific volume in workouts with top Italians like Gelindo Bordin. Today, with top Kenyans, it is often 30-32 km. These higher-volume workouts come with the requirement for greater recovery, hence the point above on greater modulation.

There is greater focus on workouts to oxidize lactate aerobically. Canova puts much more emphasis now on workouts that oxidize lactate: these sessions involve fast repeats with fast recovery, e.g. alternating sets of 1 km around 105% of marathon pace, and 1k only a little bit slower than MP. The 1999 book mentions these sessions, but Canova suggests these types of workouts are emphasized to a much greater degree today.

Starting from speed is easier than starting from endurance. Canova points to the contrast between his current approach and traditional marathon training methods that start from very high volume running, then gradually add more intensity.

This point makes me think about classic American-style marathon workouts, like 20-milers with short blocks of MP in the middle or progressive finishes. These workouts employ what Canova calls ‘qualifying the extension’–starting from a 20 mi easy to moderate run, you might add a few miles at MP, or do the second half progressive, adding more quality to the same distance as you progress in training.

Canova’s approach, instead, focuses on ‘extension of quality’–starting right away with fast, continuous running at marathon pace ±10% or so, and extending the duration that you run at that speed. So, with this approach you may not get up to 20 mi until you’re doing it at 90-95% MP!

Again, in my own coaching I’ve found that this ‘extension of quality’ approach is more effective at building the ability to run long and fast, compared to trying to gradually add more and more speed to runs that are already very long. I still use long easy runs, but less often, and I see them as strictly a "time on feet" endeavor.

You want to touch on many different speeds in training. Canova says the typical “polarized” approach to training is wrong: it leaves too many gaps between adjacent speeds, which makes it difficult to build your ability to support race-specific fitness:

So we normally speak about easy run, maybe 3:40/km. After this, something that is in the middle, 3:15/km or something. After that, we go for 400 meters very fast on track. No. This is to create a plan with big stairs that are not connected one with another. We need to have many small stairs and we need to use all these stairs in speed for having a connection.

This is a very different approach than a Jack Daniels or Pete Pfitzinger-style approach to marathon training, which employs exactly this kind of polarized, “big stair” approach. Canova, instead, recommends longer micro-cycles of training that incorporate many adjacent speeds to build fitness.

Video and transcript of ‘Keys to Marathon Training: The Methods of Renato Canova’ 🔗

You’ll find the video, embedded below, on Antonio Serrano’s YouTube channel. Huge thanks to him for uploading it (and putting on the conference in the first place!). Make sure to subscribe to his channel if you enjoyed this talk.

You can follow my transcript below as you watch the video if you’d like. It has some useful annotations that provide context or more details when necessary.

Introduction 🔗

(Introductory remarks, translated from Spanish by OpenAI Whisper, with some minor corrections by me)

…Among the best known athletes he has trained are Kenenisa Bekele, Wilson Kipsang, Florence Kiplagat, Abel Kirui, the current world champion of marathon, and Geoffrey Kirui. He has collaborated with the federations of Kenya, China, Qatar. He is a world-class coach but above all a person with many experiences and with many knowledge.

He lives right now halfway between Italy and Kenya, where he spends most of his time training the best athletes in the world in the famous Rift Valley. It is a great pleasure to talk with him: Renato Canova.

(Canova speaks from here on, in English)

I prefer to speak in English because I have a very, very good translator. And we speak about the philosophy of marathon training for the top athletes and the something that we changed in the last 15 years.

The first thing we will speak about is the training system and practical examples.

This is the last start from my athlete after winning the World Championship this year after winning Boston, Geoffrey Kirui and is not yet in good to training because his volume is very, very low and we hope that we can improve this.

The important the points to follow are these ones: The athletes of international level must produce quality. So, when you run many kilometers in a small range [of speeds], you never run very slow, you never run very fast. We need to have quality.

Quality means high modulation in the type of training. That is exactly the opposite of what we ask of amateurs, because for them it is not possible to produce high quality. So they need to have another type of system.

But when we speak about the top athletes, top quality is the key. In order to produce quality, we need to use big modulation between high intensity and recovery. And specific quality is the key for running at the best possible level for everybody.

If you want to run a fast marathon, what is specific? The marathon is 42 kilometers. Everything that is shorter than the half marathon is not specific.

Everything that is slower than 10% of marathon pace is not specific. So running very long and very slow is not specific but is general.

Phases of marathon training 🔗

Antonio [Serrano] spoke before about periodization with two big stages. One is general and the second is the specific [stage]. We are a little bit more precise, but the system is the same.

‘Transition’ is what we have to do after a marathon. So we need to do something for recovering. Recovering is not only to wait and to rest.

Every time we run a marathon, we go to lose something on the side of strength–muscle strength. So during the transition period, the first idea is to start to rebuild strength. We use circuit, we use the gym in this period, for about four weeks.

The general period is also four weeks. So together, the transition and general period are the [same] eight weeks that Antonio spoke about.

After this, we have the fundamental period that is the base for everything and the finally the specific period that is, of course, the most important.

Elements of marathon training during the transitional period

The transition period lasts four weeks after the marathon. What do we do in this period?

• Easy run, not longer than one hour
• Recovery for muscle strength
• Reactivity exercises for running technique

The general period again lasts four weeks

General resistance: it means that we start to run longer.

Training strength, endurance and ability to recruit the faster fibers: What is this one? This one is because we start to put shorter sprints up here, 10-12 seconds at the maximal speed.

Don't look at shorter sprints as something useless because you run a marathon. The final goal for this is a physiological goal. We need to be able to have the connection between the brain and the fast fibers. When we never use fast fibers, we lose the possibility to work with fast fibers.

And what about the energy, about the fuel? We have about 3-5% of the fuel that is inside the fast fibers. But if we don't have the key of the fuel store for opening the door, we can never use this type of fuel.

So this type of training at the end [of training] can produce the possibility to last and to still have fuel in the last 3-4-5 kilometers. Nothing to do with full speed [sprinting]. It has a physiological reason.

Elements of marathon training in the general period

In the general period, we also have:

• A long run at an even pace, 80% of marathon pace.
• Long run with variation
• Fartlek
• Long continuous run, also uphill, we use normally 10-12 kilometers, maybe 3-4% of gradient.

In Kenya, we have one course of 27 kilometers where we have 2,000 meters of difference in altitude. And sometimes we use it two times [in training]. It's more psychological than a real way of working. Because at the end, for Wilson Kipsang, for Eliud Kipchoge, the best in the world, the final average is about 4:15-4:20 per kilometer. So, okay, fatigue! Mentally, when you finish, you tell to yourself, I am very ‘hard’ to do this, but really, under the physiological point of view, it is not so much interesting.

Long continuous run in progression of speed.

The example [in Antonio Serrano’s talk] was the 30 kilometers of Javier Guerra.^[2] 15 kilometers plus the last 15 at the 3:03/km or something like this.

Long run at even pace, we have two sessions per week that are easy. 1:30-2 hours in this period. Two sessions are at 80% [of marathon pace], no? And a little bit shorter. One session is faster. And five sessions per week of regeneration.

You can see how many sessions, because we start with the idea that these top athletes always have practically, in this period, two sessions per day. Normally, we work with 13 sessions per week with this type of athlete, because there is not yet something of very high intensity that needs only one session in one day for recovering.

Fartlek with fast recovery: a new ingredient in training 🔗

This is practically the normal fartlek that we use: 20 times one minute faster and alternating with one minute more slow, is a fartlek for the marathon.

What is the difference between the fartlek for maybe 5,000 meters, 10,000 meters, also short distances, and the fartlek for the marathon? In the fartlek for the marathon, we don't jog after the [fast segment]. We run. We run at a good speed.

For giving an idea, in this final type of training, that is one hour, we have normally the athlete running 19.2, 19.3 kilometers at 2,300 meters of altitude.

So, it's possible to do this only if you run fast during the recovery. And to run fast during the recovery is the key, because we changed the system of training from many years ago.

Why? Because many years ago, the idea that we had about every type of lactic workout for the marathon was that it was a mistake. And we tried every time to avoid every lactic situation for this reason.

Even with Italians, winning Olympics, like Gelindo Bordin. We started to change something with Stefano Baldini, practically, in the last period [of training], because of the experience that we had with Kenyans.

And we had experience because they were very strong. So, with this type of athlete, we could try something different. So, what is different?

The difference is that Kenyans use in almost every training session, big variations of speed. And we saw that something could happen in the body of these athletes. What happened? You see in some competition, five kilometers in 14:10, 14:05, not the last [kilometer], but for example between 25 and 30 km. And after that, they are able to recover again.^[3] So, it was something strange for us coaches.

And we started to put this in a more “scientific plan” in training. And after, we discovered that with the variation [i.e. the fartlek with fast recoveries], something happens inside the body.

What happens is the permeability of the cell membrane becomes better. So, practically, with this type of training, we can make faster the possibility to eliminate lactate that we produce.

So, if we eliminate [lactate] in short time, we can go in this direction [pointing upward].

Because you know that, following the physiology, about 20% of lactate can be energy.^[4]

 So, now we use the “lactic situation,” that is aerobic power already over the [lactate] threshold, like something giving more endurance and more possibility to run faster for the athletes.

This is possible because now, you can produce more lactate because–using variations [with fast recovery]–you can eliminate more lactate in a short time. So, this is the real reason. And we use with all the Kenyans a lot of this.

And not only with the Kenyans because I have now two or three Europeans. And one guy was Arne Gabius, another guy is now Sondre Moen, who is from Norway.^[5]

And using this, they improve very much under this point of view. So, this is something that has a physiological effect. And the physiological effect is for everybody at this level, of course [not just Kenyans].

Long run, nothing new to see. Okay. Long continuous run uphill, you can see, is not a big problem.

Canova’s circuit training for strength endurance in marathon runners 🔗

What becomes interesting is what we do with the modified circuits. Modified circuits is the idea to create and to enhance the strength endurance, not the strength. We don't need high strength for a marathon runner. We need strength endurance.

We need to be able to last for two hours at a percentage of the maximum strength that is very high. And for doing this, we use circuits.

We can have a lot of different types of circuits. Depends. We use hills, but also it is possible to use a track. For example, here [in Valencia], you don't have hills, you can use a track.

Canova’s circuit training on the track

We can see, for example, an example of modified circuit on the track. We run 400 meters at marathon pace. And without stopping, immediately after this, we have one session of exercise: 30 seconds.

Maybe skipping, bounding, heels to buttocks, jumping in place, like a squat jump, or with blocked knees, only feet [i.e. a calf hop].

So, in this situation, we run one set: 2400 meters running, plus all these exercises.

And we have five, six minutes of interval of rest, because it's demanding.

And we go from, at the beginning, three, four times. At the end of the [training] period, we arrive at seven times for this. And seven times are already 2400 meters, so are already 16.8 kilometers on track, plus exercises. This is one situation.

Canova’s circuit training on hills

Another thing that personally I use in Iten, because we have a track that is at the top of a hill. If there is not this, it is not possible, no?

I use, for example, three times 800 meters or 600 or 1K, about marathon pace, with one minute recovery. After, they go down [the hill], and they go sprinting six times 60 meters, maximum speed [up the hill]. And they repeat this for five times, for example.

So, mixing the maximum sprint with something that is about aerobic power, we increase the ability of the strength endurance. That is the base of the endurance for every type of distance.

Because the problem is that we start to lose reactivity and strength with the distance when we become tired. We have a circuit on hilly course, maybe possible, no? And this is the number of repetitions.

And the goal is to increase the ability to recruit a high percentage of fast fibers when already we are not fresh. When we have already in our muscles, a high level of lactate in the muscle fibers.

(There is some additional info on circuit training later in the talk, click this link to jump directly to that section)

The fundamental training period for marathon runners 🔗

We go in the next period (fundamental), lasting six weeks, practically. And these are all the different speeds that we use with different goals, of course. Nothing new. Everybody uses the same situation here. No?

Long run, about 80% of marathon pace. In this period, this type of long run is duration. You need to think that when we speak about marathon, we have two different parts of preparation.

We need to prepare the body structure. Because, for example, if an athlete comes from something short and runs not very, very fast, maybe an athlete of 14:00 in 5000m, for example, it's not a big problem for him running at a full marathon in 3:15. No? Like, also without big preparation.

What is, in many cases, the limiting factor? The body. Because after one hour and thirty minutes you start to have a problem in the knee, the tendon, because there is not volume. You are not trained to stay a long time on your legs.

So, for this type of preparation, we use duration. Duration is something that you run not at a high level for preparing the structure and also the mind of your body. No?

And this doesn't have any direct influence on the performance. But it is something for preparing your body for something more specific. And what is more specific? This is another type of training.

Long continuous run at even pace. Run fast. 87-93% [MP], starting from there. And the idea is to arrive to have more distance.^[6] So, we call “duration” something connected with time, and “distance” something connected with the goal of speed for our marathon.

If I want an athlete to run a marathon at 3:10/km, for example, I put from this period something at 3:10/km. Five kilometers in 15:50. And I don't need to run faster! I need every time to run longer.

So, the physiology, the philosophy, is to extend the intensity, not to qualify the volume. Understand? So, this is what we do.

The Kenyan vs. European/American marathon mentality 🔗

I want to tell you the different mentality that we have now in Europe comparing with the African.

[Let’s say that] I am a crazy coach. And I have two athletes. One is African, one is European or American. And I tell them, your training tomorrow is running 20 kilometers in one hour at 3:00/km. And both of them are not able to do it.

So, the European starts to think, “I am not able to run one hour for 20 kilometers. So, what do I have to do? Maybe I am able to run at 3:20/km.” And he starts running 3:20/km [from the beginning of the workout]. At the end, 18 kilometers. Because for him it is very much important to finish one hour.

For the African, instead, the idea is: “I start to run at 3:00/km. When I finish the fuel, I stop.”

So, for the African, everything starts from the intensity. And practically, like a methodology, we need to extend the intensity.

Instead, for European normally, American, the idea of “general volume, a lot of easy something,” it becomes very difficult, when this is the normal level of training, to put a real intensity in this.

To qualify the volume is very much more difficult if you want to reach your top. Not if you want to finish a marathon the first time. It is very much more difficult than to extend the intensity. Because when you normally “frequent” the intensity, for you, to a period of one hour, 1:20, 1:30, at a high level of heart rate, for example, becomes normal. And you understand that if your normal base in any moment of the season is 85, 90% of your maximum intensity, it is not so difficult to move from 85 to 100%.

But if you are at 60% every time, to go from 60 to 100% is very difficult. And after two marathons, you are no more able to improve.

Because what is training? Training is [not] the proposal to your body. It is the answer to the proposal. This is training. So, training is to give to the body some stimulus. The answer of your body to the stimulus is training. So, when there is no stimulus, there is no training!

Lower mileage for top marathoners today 🔗

So I ask you one thing: If you are not already able to run a full marathon at 3:00/km, like for example in the past Rey [Julio Rey, a 2:06:52 Spanish marathoner] or Ríos [José Ríos, a 2:07:42 Spanish marathoner], these people–three minutes [per km].

Which type of reason there is for running maybe 10 or 15 kilometers at 4:00/km? Tell me. Which type of stimulus there is?

It is very much better that he can go with a bike, or can sleep all day! Because this type of training is not training. It has no effect on the body, but has an effect for making more kilometers and to increase the consumption of the body. It is like to use a Formula One [car] for 20,000 kilometers at 100 kilometers of speed. Which is the reason? What do you do with this? Nothing!

So many years ago, we had athletes running, top athletes, about 300 kilometers per week. In Italy, Gelindo Bordin ran some time at 320. In Japan, Toshiko Seko ran 480 kilometers in one week! And they had two times in the preparation, one training of 100 kilometers, calling it “mentalisation.”

The only result was that no Japanese athletes were able to last more than two years, because this volume destroyed completely the body.

Now, the top athletes run maximum 200 kilometers [per week] and are between 170 and 180.

What is the key for improving your personal best in your career? At any level, also for an amateur, it is to increase the volume of the intensity. So, not the volume, and not the intensity, but the volume of intensity.

So, when you are young and you start to prepare your body, you have maybe 80% of training that is general before arriving at 20% of specific.

When you have 5, 8, 10 years of career, you need to have 50% of specific [training]. And the support is only 50% general. And you cut the general [training] because it's useless or damageous. Understand? So this is important!

Adding what you didn’t do to improve marathon performance 🔗

Also, another important thing: the only mistake that we have in training is not what we do, it is what we don't do.

So, when an athlete comes to me, in Kenya–but not only Kenya–take the case of this guy from Norway. He comes to me with 2:12 in marathon, 28:25 in 10k, 13:33 in 5000 meters. And [in] the last three years, [he’s] not able to improve. So, they are not bad, these times.

So I asked him to know what he did in training. And I didn't change anything what he did. But I put in training what he didn't do!

So, I discovered that he didn't use shorter sprint. He didn't use variations. He didn't use a very long run also easy. A lot of things that he didn't use.

Incorporating many different speeds into marathon training 🔗

So, maintaining what he was able to do and adding what he didn't use, it was not possible to have a “micro cycle.” Because for putting everything together, we have a “micro cycle” of about 20 days.

Because when you go to top level, 5 seconds of difference [per km] in the pace for marathon is another type of training. The influence in the body is different.

So we normally speak about easy run, maybe 3:40/km. After this, something that is in the middle, 3:15/km or something. After that, we go for 400 meters very fast on track. No.

This is to create a plan with big stairs that are not connected one with another. We need to have many small stairs and we need to use all these stairs in speed for having a connection.

So a faster speed can be the base of speed for something longer. And the longer speed, the longer at good intensity is the possibility to improve practically your endures, specific endurance.

Otherwise, we are in something general.

Ready to run the marathon vs. ready to do marathon training 🔗

And the real problem that we have when we speak about the marathon, is that normally–also with Europeans, also with Americans–the athletes go to compete not at the end of the specific period, but they go to compete at the end of the fundamental period. Thinking that they are ready. No.

They, at the end of this period, are ready for starting the specific period. Because there is not the right intensity in the right extension if you want to do your best. So practically this was the big difference that we had.

How modern specific training is different 🔗

And for example, always speaking about the variation, we now, we use distance for a volume of 30-32 kilometers replacing the normal long run.

With Gelindo Bordin, for example, we started to use 4 x 5 km at marathon pace or a little bit faster with a one kilometer recovery, maybe at about 3:20/km. Totalling 23 kilometers. And never we went longer than 23 km. The last training before winning Seoul Olympic was 3 x 7 km at 21 minutes, 20:58, 21:03, depending on the road, with one kilometer recovery–maximum 23 km. So it was a specific type of training.

Now, we arrive at 32 kilometers with this system and [this] 32 km goes to replace in the specific period the normal long run with a different quality.

Maybe that the average [pace] is the same like before. But for example, when we go for 7-6-5-4-3-2 km, a little bit faster than marathon pace, and in the middle we use one kilometer in 3:30 [for recovery], for example, for top athletes, the average is the same. But we have, in 32 kilometers, 27 kilometers of specific speed, something that is completely different from the past.

Marathon specific track workouts for top marathoners 🔗

Another thing is on the track. We use distances between one and four kilometers, but the total volume when we are in the specific period is about 20 kilometers.

So one classic workout that we do is 2x3000m. For example, for I give you a situation like Wilson Kipsang or something like this, no, before the world record:

Two times 3000, 8:42, 8:45. Three times 2k, 5:42-5:45. Five times 1k, 2:45-2:48. Six times 500 meters in 1:20, the last in 1:17. Recovery between 3k and 2k, three minutes. Not jogging, three minutes–stopping, nothing, walking, waiting. And for every 1k, one minute 30 [recovery], and between every 500, one minute. [Total is] 20 kilometers on track. This is what happened in the last period of the preparation.

Long fast runs for top marathoners 🔗

And when we speak about long run, 40 days before running the world record in Berlin, 2:03:23, we had a test (only one time) that we used [with Wilson Kipsang]: 40 km starting at 2000m [of elevation above sea level], arriving at 2200m almost, and coming back.^[7] And normally in this test, the athletes run the same time of the marathon at sea level–not the same pace, the same time.

Aside: doing workouts at altitude on dirt roads

So you need to think that we consider a difference of six seconds [per km] for these people between this training in altitude at the sea level.

Three seconds for the difference [per km for the] altitude. Because for these people it's not more than three seconds, for Kenyans or for Ethiopians. They live there, they have adaptation.

Three seconds are for the different grip [of the road]–because they need to run with heavy training shoes on  rough road, dirty roads, and there is not the same grip.

If you run also here, and you have, for example, a pace of 3:20 per kilometer on tarmac, on asphalt, and after you have at the same pace one kilometer in a dirty road, no? Strade Bianche, no?^[8] The 3:20/km becomes 3:23-3:24/km because the difference of grip.

Long fast runs and marathon fitness

So this is what I consider. So 2:03:32 for 40 kilometers [on this dirt road in Kenya] is about 1:59:30 for 40 km at sea level on asphalt. It means a marathon in 2:06. A marathon in 2:06 was practically the last long run fast 40 days before the world record.

And with many others we have the possibility to see that they repeat in the marathon the same time that they are able to do in this course, practically.

So what changed with many years ago is the intensity of long run, because [in top races] it's no more progressive but it's a long run from the beginning. These people, if they want to run fast, they start at 2:52 to 2:55 per kilometers. You need to be ready not to run easy the first part [of the marathon] and after to go fast at the end. Also in training we need to do this.

The real problem: How to prepare for specific training 🔗

Don't think that is what we use every time. No, we need to prepare this possibility because the real problem in endurance (and in every type of athletic event) is not really what is specific–because everybody knows what is specific–but is how is possible to increase the possibility, the volume of specific training.

Example: if I have an athlete running 400 meters in 46.0 and in training is able to run 3x300 meters in 34.5 with a six minute recovery, everybody understands that if 34.5 becomes 33.5 with the same recovery, 46.0 becomes 45.0. Everybody understands that, you don’t have to be a genius.

The problem is what I have to do for arriving there. This is the problem.

So what do we have to do during the the fundamental period for preparing the possibility to have more specific volume–or in case of shorter distances, sometimes also, more specific speed? This is the real problem that we face [in non-marathon events].

In the marathon, it is the only event where we have an opposite problem. We need to understand what is specific.

Because the marathon is 42 kilometers, normally we don't go for some distance longer than this–although sometimes^[9] we use some times 45 kilometers in general training, two times [during the training period] for example, but normally it is not a part of training. Different from all the other events!

If we speak about the five or ten kilometers,  in a normal training, you go for something longer. Also on the track, you go for 12 or 15 kilometers, you go for something longer.

The marathon is the only event where we never go for something longer. So never we go really for something specific, fearing that if we go there we need to have, maybe two months for recovering. Nothing is more wrong!

Recovering after fast vs. competitive marathons 🔗

When a marathon is very fast, it is possible to recover in very short time. Because the problem is…why can I run a marathon so fast? Because the conditions of the weather are very good, otherwise it's not possible.

In the Osaka World Championship 2007, the winner ran 2:16–it was my athlete [Luke Kibet]. The second [place finisher] was also my athlete, 2:17:21 [Mubarak Hassan Shami]. That is ridiculous, because they ran 10 minutes slower than their personal best.

Two months later, they were not yet able to go in training! Because [at Osaka it was] 35 degrees [95° F], 98% humidity, the body was destroyed. When you go instead to Dubai and you run in 12 degrees [54° F], completely flat, good pace, 2:04-2:05, the recovery is very easy.

The problem is not the recovery of the body, the problem is the mental recovery. Because–and Antonio [Serrano, host of the event] was a marathon runner, so he knows [pointing at him]–you need, when preparing for a marathon, a lot of concentration. And when you arrive at the marathon, this is in any case the final point.

Mental vs. physical recovery and psychology of the marathon 🔗

If you are okay [with your performance] after, you are empty, like a nervous energy and you relax.

If you are not okay^[10] and you want to try, “hey I have a lot of training, so I had to stop after 21 kilometers for some reason–I don't want to waste all this training, we go to look for another marathon after two weeks.” Never this marathon can be okay! Never! Because the problem is not the energy in the body, but the mental energy.

Example: Moses Mosop ran his first marathon in 2:03:06 in Boston. And after this, immediately after this, we planned to attack the world record on track of 25 and 30 kilometers. It was exactly 40 days later.

And many told me, hey how is possible with a marathon in 2:03? After one week he was in training, because the mind was there.^[11] And after 30 days, he went for one of the best sessions of his life on the track: 18 x 1k with 1:20. The first 2:45, 16 in 2:42, the last in 2:36. And after this destroyed the world record.

So the problem is here [pointing at head]. If you go for a marathon, for example, that is in your plan, is the preparation of the final marathon. And you put this marathon four weeks before, something like this, you can run only two minutes or three minutes slower than the final goal. And after four days, you are in full training. Because in your mind, this was a passage, was a part of training, not the final goal.

So we need to think that the psychological approach is one of the most important in every type of athletic event, but especially when we ask full concentration in long fatigue. Because this is important.

Psychology of marathon workouts 🔗

For example, for Kenyans, I needed to learn to write the same workout in different way. I’ll explain–if I write “5x5k in 15 minutes,” for example, with 1 km recovery at 3:30, you know what they do? They run 15:00 and after the [recovery] kilometer become five minutes of jogging. Because for them recovery is recovery.

They don't look at 3:30 or something like this [as recovery]. So I started to write in a different way. I started to write “30 kilometers of continuous run, alternating five kilometers in 15:00 and one kilometer in 3:30.”

They start in their mind before training to think this is a continuous run and they are able to do it.

So the mental preparation is very much important when we speak about something of a “high personal level” in this way. And we need to take care about this.

It's not that we needed to have something strange [for workouts]. The problem is how to cook the ingredients. The ingredients are always the same.^[12]

Uphill circuits 🔗

This is an example of a modified circuit, but uphill.

So for example, we have a gradient of 4-7% and we go for:

• 200 meters running fast
• Skipping
• 300 meters running uphill, always at a good intensity
• Bounding
• Again running uphill something longer
• Again another exercise

And after we finish, we have something, for example, like 1500 meters [total]. And we do this in this period, the fundamental period.

(If you jumped down here from the previous section on hill circuits, click this link to jump back up)

Canova’s specific period of marathon training🔗

Now, we’ll try to see what I told you before, in the specific period. That is the real thing that is important. Antonio spoke about two months, nine weeks or something. Here, it is about 10 weeks. So, everybody has the [same] duration, the specific period is practically the same, no? So we need to train these qualities.

One thing that is important, the tension, which is the difference of training philosophy between the fundamental period and the specific period.

During the fundamental period, we work for improving the qualities that we need to use one by one. So we look for strength, we work for strength. We look for general endurance, we look for long run. We are not interested to put together these. We are buying the material for building the house.

[In the] specific period, I need to use the material that I already have for building the performance.

Internal load vs. external load 🔗

So in the first period, we speak about internal load. What is internal load? Internal load is the level of effort that you use when there is a training.

So in this case, I give you an example. I ask to an athlete to run 10 x 1000m in 3:00, okay, with two minutes recovery.

We are in winter–very, very cold (that is not the situation of Valencia, practically!). Maybe windy, maybe some other problem.

The athlete starts with the level of effort that is the “right” level. He runs the first in 3:00, the second in 3:05, the third in 3:15, the fourth 3:20, and after quit telling, “Ah, it's not the day.”

No, this is wrong. If your level of effort is the same that you use running 3:00 when everything is perfect, you need to finish the training because the internal load is the same.

When we go in the specific period, there is no more internal load. There is external load. So training becomes mathematic.

If I have the goal for running 10 kilometers in 28:20, that is 2:50/km, and I put like a specific training, for example, 12 x 1k in 2:50, or 10 x 1k in 2:48, or something like this, and because the conditions are not good, I go at 3:00 or 3:05 because there is some situation that is out of my control, I need to stop because of this training becomes useless.

I need to have good conditions because there is the “study of the pace.” I need to understand exactly what I do. I need to prepare in a mathematical way exactly the performance.

So I don't want to see people that, when you ask, “what you can do, I run more kilometers, I had something good. We see how is the race.” Doesn't exist, this.

There is nothing “strange” about the marathon 🔗

I am against the idea that “marathon is something strange, and after 30 kilometers, there is ‘the wall.’”

It's something strange when there is not good training! But when you prepare exactly what you are able to do, you know. And only if you do some mistake [in the race], it becomes a problem for you.

But if you don't do any mistake, you know very well what happens until the end. Nothing strange. It's not different from other events.

Because you need to know exactly what is the pace that you can maintain for 42 kilometers, arriving after 42 without the fuel, not after 35 [and running out], and not after 32 with a lot of fuel, because there is some mistake.

So you need to know this exactly like the car of Formula One, where there is the study before and you put fuel enough for running at a speed that you know. So we know already for this. And which is the physiological point for changing this is to change the way of fueling.

Fueling the marathon with a mix of fats and carbohydrates 🔗

We have a “table,” okay, where we have one tank that is “super”^[13]–glycogen. It's not very big, this tank. We have on the other part of the table another tank, bigger, where there are the fatty acids. In the middle, we have a tank empty that we need to fill taking from here [the glycogen tank] and from here [the diesel tank].

So at the beginning [of training], what is the way for increasing the duration at the same speed?

When we prepare a speed, maybe 3:05 or 3:03/km, maybe the best speed for the athletes, where step by step for this speed we use less glycogen, less “super” and more fatty acid, more “diesel.”

Okay, so because this [the glycogen tank] is small, we can last every time more [i.e. longer]. How is possible to do this?

Example: the glycogen, normally we finish glycogen in one hour if we remain about the umbral^[14], the [lactate] threshold.

So of course we need to run a little bit slower than this if we want to last more. And for example, I want to prepare an athlete for running a full marathon at 3:06 per kilometer and I ask him to go for 30 km, maybe two months before the marathon at 3:06/km.

So he arrived till 1:33 for 30k, perfect [3:06/km avg]. And after I ask him to arrive till 33 km, yet three [more].

And the 31k becomes 3:20, the 32k becomes 3:35 and the 33k become 4:00. The real training are the last three K! The first 30 are for emptying the tank.

So at the end of this, when the body is no more able to find the fuel, you go to see in some part of your body where there is yet some fuel.

Next time, you know where you need to go and automatically you start to use less “super” for the same type of speed.

And we can have the control of this, for example, with the historical system of 6x2000m increasing in pace, with lactate, something like this.^[15] Here there was a different system in Spain, when there was Xavier Liviaz, a doctor in the Victoria University, but it was the same thing.

So we were able to control that at the same speed during the right preparation for the marathon, the athlete started every time to use less glycogen and the more fatty acid. And this means that he was able to “fill” this type of speed for all the marathon.

Training for the athlete vs. training for the event 🔗

So we needed to be able to prepare exactly the time that we want. For Kenyans, this is more difficult because of the courses.^[16] But with the top athletes, we know exactly if tomorrow in Berlin they want to go for 61:00 for the half marathon [for the first half of the race].

61 may be that the second part maximum they finish in 62:00, 61:40. The problem is that with the top three, there are ten other athletes that don't have any idea but go together.

So you start to see somebody that exploding after 25k, another after 28k, two that are not able to finish, because they are very aggressive, are very strong, but really don't have an individual plan.

And this is another important idea about the training. There doesn't exist the training for one event. There exists the principle of training for one event. But the real training is for the person.

So it is not the athlete following the training schedule. But it is the training plan following the athlete. It's different.

The principle of individualization of training is the winning principle when we speak about the top athletes.

Training groups with athletes of different abilities 🔗

I understand that with amateur or something like this, it is not possible, because it's more important to have a group. Individually, there are too many differences, for example, working, the time of the day, the possibility that there are, the family. So when it's possible to have something together is the best solution.

Maybe that you have six people running together. Somebody stops at 20km, somebody can be 25km, somebody 30km, [all at] the same pace. This is what I use also in Kenya.

For putting together people, I use the same speed with different distances. Depending on when they have the marathon and depending on when they have the value [i.e. their fitness level].

But I can't use the same type of training changing the speed. Because if I prepare somebody for 2:03 and somebody for 2:07 and somebody for 2:10, they can't run together the same distance. It's not possible. There is some mistake.

So maybe that with the top [athlete] we run 40k. With the other [athlete going] for 2:07 can go to 30k and with the other can go to 20k. The guy going to 20k maybe after two weeks has 40k at his pace.

But in this case, we can move in the groups, we can have the possibility to follow the group with the same type of speed (also on the track).

On the track, if you are not able to run 2000 meters in 5:40, you run 1600 meters–one lap less. And after you have 1:10 recovery more and you have another type of workout for you. And after this, for you, there is another day. The full workout with more volume.

So we need to, when we have a group, we need to be able to play a little bit with the situation. But when we are in the specific period, the most important thing is to have together people for speed, not for distance.

Marathon specific workouts and recovery 🔗

The long continuous run at even pace is fast. Maybe 90% [MP] but maybe also more sometimes.

When we go till 25k, something like this, we can have this [workout] maybe one time every two weeks or something (because it's very much demanding) near the pace of the race.^[17]

This is an example. 7-6-5-4-3-2 kilometers increasing in speed. The 7k is 98% of marathon pace. And after, the 2k is 103%. So, faster than marathon pace.

Alternative with one kilometer, 80%. If you are an athlete running 3:00/km, 10% is 18 seconds and 3:36 is 80%. So this is the idea. And at the end, you run 32 kilometers. So with this, we go to replace the normal long run at even pace.

You need to remember this is the way for producing quality. And after one session like this, we need to have easy, easy training and regeneration for four or five days. This is what we do.

So, when we increase the modulation for top athletes, forget the idea that we need to have three sessions per week with one day recovery in the middle.

It is not possible that in a real plan of high quality, the only thing that doesn't change is the recovery time. Because when we go for something more demanding in extension, of course, we need to “open” the recovery.

So this is the way. More quality in extension, more recovery. Otherwise, we go to kill somebody.

Instead, with the right balance between intensity, volume of intensity and recovery, we can grow and with the system, the new system, we can run two, three minutes faster than many years ago. This is what we do now with the top athletes.

(Talk ends)

Recap

Canova's 2017 Valencia Marathon lecture is an amazing resource if you want to know what's changed in the past few decades of marathon training. My biggest takeaways are the reduced mileage, increased marathon-specific workout quality, and the focus on extension of speed above all else.

If you enjoyed this article, please share it with your friends and on social media. I also have my own book on training for shorter distances if you want to see how I combine physiology with modern coaching approaches. Be sure to sign up for my email newsletter below so you can be among the first to find out when I publish new articles like this one.

Footnotes 🔗

---

Click the bracketed number next to each footnote to jump back to your spot in the text

^[1]  Technically, the talk was called ‘claves del entrenamiento de maratón,’ since the rest of the conference itself was delivered in Spanish!

^[2] Javier Guerra is a 2:07 marathoner from Spain who has placed in the top 10 at the London, Tokyo, and Madrid marathons.

^[3] Meaning they can recover from such a surge during the race and still finish strong without falling apart. This phenomenon suggests that Kenyans employing this Kenyan-style fartlek training were able to “absorb” the lactate and the metabolic byproducts they generated during this surge.

^[4] Research on the “lactate shuttle” cellular mechanism indicates that it might actually be as high as 75% of lactate that is oxidized during exercise.

^[5] Arne Gabius is a German runner who has run 2:08 in the marathon, and holds the German national record. Sondre Moen is a Norwegian runner who has run 2:05:48.

^[6] i.e. what you want to do is increase the distance of the run, not increase the speed.

^[7]  i.e. the run is an out-and-back 40 km run, ascending from about 6600’ to 7200’ above sea level on the 12.4 mi out, then turning around and descending on the way back.

^[8]  Strade Bianche is a famous cycling race in Tuscany, Italy with a substantial portion of the race on rough gravel.

^[9] In the talk, Canova says “also if”–he often uses this idiom to mean “even though” or “although sometimes.” I’ve changed it here for clarity.

^[10] Meaning “not okay with your race performance” i.e. you didn’t run well in your marathon.

^[11] In the sense that Moses Mosop was focused also on the upcoming world record attempt following the marathon, not just on Boston.

^[12] Canova often uses the metaphor of Minestrone soup–a basic Italian soup often made with beans, carrots, celery, onions, and tomatoes. Everyone knows the ingredients, yet depending on the preparation, the quality of the soup can vary quite a lot! Very much like your Grandma’s homemade chicken noodle soup versus what you find in a can at the store. Same ingredients, but very different outcome.

^[13] Analogous to high-octane racing fuel for a Formula One car

^[14] Umbral means “threshold” in Spanish. Alluding to the talk by Antonio Serrano from earlier at the conference which discussed lactate threshold training.

^[15] See the section in the article on Canova’s book regarding lactate testing for details on this 6x2k workout.

^[16] “Courses” in the sense of the rough dirt roads in Kenya, that have hills and poor footing, making it difficult to compare to running on flat pavement at sea level.

^[17] The slides are not visible in the video, but it appears that this 25k session is different from the long fast run at ~90% MP discussed immediately prior. This comment is most likely about workouts of 25k at 98-100% MP.

Related articles