An elite races a 10k in circa, say 27 minutes. Would it not be better for the recreational runner to determine their, say pace for running 27 minutes all out, and apply the percentage to this pace?
Again I don’t disagree that most people are probably doing their easy days too fast. But I think you’re overfitting the principles you’ve taken from a small sample size of people who, by definition, are exceptional outliers. You also more or less glossed over the fact that the women were running at a higher percentage of their race paces and provided no explanation. I think mechanical load is an important part of this puzzle and goes some way to explaining why elites would run so very slowly on their easy days. Have you considered sampling sub-elites and successful age-groupers?
To your more detailed points: I don’t need to assume that 3:30-45/km is a significantly higher mechanical load than 4:30/km — it simply is! Mechanical load scales unevenly with an increase in pace, meaning the risk increases nonlinearly as faster runners approach their LT1. Your caveat about weight is something I already included… but if you insist on assuming that amateurs are heavier, then it is you that is making assumptions, not me. Indeed, a lot of dedicated amateurs are already in a similar weight category as elites and I doubt that differences in running economy would lead to significantly different tendon loading.
Again — because you started restating arguments that I already said I fundamentally agree with — I think most amateurs could do with running their easy days easier. I do not see any evidence in what you’ve laid out, however, that supports simple scaling from one extreme (elites) to another (amateurs racing 10km slower than 50min).
I disagree with this point. You'd have to provide evidence that this is the case.
I don't disagree that a 3:30-45/km is a greater mechanical stress than a 4:30/km, but I disagree that an athlete's ability to tolerate paces (in proportion to their LT1) worsens as they get faster.
If an elite trained exactly like a recreational runner (with all the runs at the same proportion of LT1, but only doing 60kms per week), they would be able to do it. But by slowing down their easy days they're able to do triple the volume. In contrast, a recreational runner would not be able to perform triple the volume if they kept their easy days at such a high proprotion of LT1.
In other words, the mechanical stress of running close to LT1 is proportional for both elites and recreational runners, but recreational runners can get away with it (sometimes) because they aren't doing it in very high quantities. However, it's also the thing preventing them from be able to run higher quantities.
This post was edited 20 minutes after it was posted.
An elite races a 10k in circa, say 27 minutes. Would it not be better for the recreational runner to determine their, say pace for running 27 minutes all out, and apply the percentage to this pace?
From earlier in the thread:
"The only problem with 30-minute race pace is there isn't elite data available for that precise metric (and I wanted the conversion to be data-based - so as to avoid being accused of simply plucking figures out of thin air). 10k race pace was selected because it's a true reflection of an athlete's absolute in-competition capability.
However, I take your point, and I agree that if your 5k personal best is far closer to 27 minutes than 13, it might be better to use that as the basis of the 63-70% conversion. There'd likely still be a big overlap in the range. I did try to avoid inserting my own opinions when interpreting the data. The conversions are what they are, because that's what elites are doing - and I didn't want to stray from this premise."
Strava is now full of people using the Runna app training plans. Runna assigns a pace not to be exceeded during each session, and this narrative appears in the Strava heading.
Guess what? Most runners, I see, are exceeding the recommended pace during their easy runs.
That's a surprise to me, I've only used the Runna app once but the easy run paces were pretty aggressive. It's not like they set the cap at 6:00/km for most hobby joggers.
To your more detailed points: I don’t need to assume that 3:30-45/km is a significantly higher mechanical load than 4:30/km — it simply is! Mechanical load scales unevenly with an increase in pace, meaning the risk increases nonlinearly as faster runners approach their LT1.
I disagree with this point. You'd have to provide evidence that this is the case.
I don't disagree that a 3:30-45/km is a greater mechanical stress than a 4:30/km, but I disagree that an athlete's ability to tolerate paces (in proportion to their LT1) worsens as they get faster.
If an elite trained exactly like a recreational runner (with all the runs at the same proportion of LT1, but only doing 60kms per week), they would be able to do it. But by slowing down their easy days they're able to do triple the volume. In contrast, a recreational runner would not be able to perform triple the volume if they kept their easy days at such a high proprotion of LT1.
In other words, the mechanical stress of running close to LT1 is proportional for both elites and recreational runners, but recreational runners can get away with it (sometimes) because they aren't doing it in very high quantities. However, it's also the thing preventing them from be able to run higher quantities.
I’m not saying it’s automatically the case for all athletes, but as a general rule, an athlete with a higher LT1 would, if controlling for other variables like weight etc., be exerting more mechanical force at a defined percentage of their LT1. While I obviously can’t provide direct experimental evidence to back this up, it does just follow logically that running just below LT1 carries greater forces, and thus greater potential injury risk, the faster an athlete is. Thus why we see the faster men running at a lower avg percentage of their race pace than the women.
A compounding factor, as you correctly point out, is the trainability of resilience. An athlete doing greater quantities of easier volume is finding that sweetspot of mechanical stress (which is not purely “bad”, but indeed essential to train resilience!) while still providing a solid aerobic training signal. The core of my argument is that this sweetspot for easy volume is not just a function of their aerobic fitness, but also of their resilience and thus the mechanical load. While aerobic fitness means the cardiovascular response to 4:30/km between two athletes of the same weight can differ wildly, with athlete A below LT1 and athlete B above LT2, the forces exerted will be similar. I’m not suggesting that aerobic fitness and resilience don’t correlate, as running training will affect both. But to simply ignore the role that mechanical stresses could be playing in elite athletes’ easy pace selection means that it’s not really valid to scale everything down to vastly slower athletes. Particularly if someone has several years of running under their belt and thus a decent resilience built up, but only 30-60 min to do an easy run on a regular basis, they might really be leaving fitness on the table if they’re plodding out 6:30+/km. They’d possibly also find it difficult to join others on easy runs at, for example, what would otherwise still be a reasonable 5:45/km for them.
To be clear, I think this consideration cuts both ways. An athlete that is very fit from cross-training needs to be particularly careful about the mechanical load of increasing running volume, because their aerobic fitness is more likely to be decoupled from their resilience. And I still agree with the general conclusion you’re reaching. But considering how thorough you are, it seems to me to be a big oversight to leave out mechanical load and resilience entirely from your considerations. And, as I said, drawing your numbers from elite athletes and applying them linearly down to athletes racing at half the speed would appear to me to be overfitting your observations into unsupported prescriptions.
You're making the assumption that a 3:30-45/km is a far greater mechanical stress (proportionally) for an elite runner (who's likely running 30 seconds+ below LT1, who likely weighs less than 60kgs, who likely has far better running economy and who isn't working a full-time job) than a 4:30/km is for an amateur runner (who's likely running much closer to LT1, who likely weighs 70kgs+, who likely has relatively poor running economy, and who works a full-time job on top of training).
You're also making the assumption that one's ability to tolerate mechanical load doesn't scale with the amount of training one does each and every week.
But let's ignore all that.
This, in a nutshell, is the problem with the approach. If the only thing I knew about training were what these 24 athletes did, and my assignment were to get fast, I would start by doing exactly what they're doing. No question. You have to use what's available.
But why would should we ignore everything else we know?
Whenever someone points out reasons that your thesis might not hold, you accuse them of making unfounded assumptions, even though you are invariably assuming the opposite, without evidence. Instead you just nakedly assert that your assumption is more plausible.
The exchange above is a good example:
It's a fact that connective tissue responds to training more slowly than anything else. It takes years and years for bone density to increase and for tendons to become stiffer. That also suggests that even if recreational athletes should emulate elites, they probably can't do so in the short-to-medium term because they don't have the same bodies. Your data is about people with many, many years of high level training. It doesn't tell us how to get there. Because different adaptations have different time horizons, it's reasonable to think that the proper mix of training will be different at different stages of an athlete's career. You'd probably call this an "assumption," and it is, but it's grounded in science. It's certainly no less reasonable than your assumption that the training should be identical.
It's also just simple physics that the mechanical stress on a runner's legs increases exponentially with pace. Elites can't really stress themselves aerobically unless they put a ton of stress on their legs. Does that prove that elites need more pace variation than recreational runners? Of course not. But, again, it's a plausible hypothesis grounded in science.
(Incidentally, it's not an "assumption" that elites are under more mechanical stress at any given percentage of race pace. It's just math. Kinetic energy is mass times velocity squared, so a 15% difference in mass between a recreational and elite runner is going to be dwarfed by the difference in their paces when it comes to how hard they hit the ground. Running economy is irrelevant. That's a measure of how much oxygen you consume at a given pace, but impact forces are unaffected because of Newton's third law. There's no way to hold a pace other than to hit the ground hard enough.)
An elite races a 10k in circa, say 27 minutes. Would it not be better for the recreational runner to determine their, say pace for running 27 minutes all out, and apply the percentage to this pace?
Mistake #2: treating the intensity of running effort as a function of distance instead of duration in time.
If an elite trained exactly like a recreational runner (with all the runs at the same proportion of LT1, but only doing 60kms per week), they would be able to do it. But by slowing down their easy days they're able to do triple the volume. In contrast, a recreational runner would not be able to perform triple the volume if they kept their easy days at such a high proprotion of LT1.
Can recreational runners triple their volume if they slowed down their easy runs to your recommended pace?
Well, here is the problem. Suppose I am running 6 miles in 60min on my easy days. If I slow it down to 11min/mile, that would take 66min. But if the purpose of slowing down is to increase the volume, then I will need 77min to run 7 miles at the pace. To increase my volume by 50%, I would have to run 99min instead of 60min.
39min on on day may not sound much. But if you do that 3-4 times a week, that's 117-156 min per week. Finding extra 2 hrs for one week may not be too hard. But finding extra 2 hrs every week is much harder.
And it may actually take even more time. I have discovered that I need more sleep when I am running more volume. So in addition to running a few more hours per week, I have to sleep a few more hours per week. Pros don't have to worry about time constraint because they don't have a day job.
If an elite trained exactly like a recreational runner (with all the runs at the same proportion of LT1, but only doing 60kms per week), they would be able to do it. But by slowing down their easy days they're able to do triple the volume. In contrast, a recreational runner would not be able to perform triple the volume if they kept their easy days at such a high proprotion of LT1.
Can recreational runners triple their volume if they slowed down their easy runs to your recommended pace?
Well, here is the problem. Suppose I am running 6 miles in 60min on my easy days. If I slow it down to 11min/mile, that would take 66min. But if the purpose of slowing down is to increase the volume, then I will need 77min to run 7 miles at the pace. To increase my volume by 50%, I would have to run 99min instead of 60min.
39min on on day may not sound much. But if you do that 3-4 times a week, that's 117-156 min per week. Finding extra 2 hrs for one week may not be too hard. But finding extra 2 hrs every week is much harder.
And it may actually take even more time. I have discovered that I need more sleep when I am running more volume. So in addition to running a few more hours per week, I have to sleep a few more hours per week. Pros don't have to worry about time constraint because they don't have a day job.
There would come a point where the extra duration would be a cause for extra fatigue and injury risk.
Elites max out at around 10 hours per week of running. Could every recreational runner build up to this? Assuming they have the time, possibly, but unlikely. The slower runners would still only be building up to 50-60 mpw running in 10 hours per week. However, their running would very likely improve, if they didn’t get injured.
I’m not saying it’s automatically the case for all athletes, but as a general rule, an athlete with a higher LT1 would, if controlling for other variables like weight etc., be exerting more mechanical force at a defined percentage of their LT1. While I obviously can’t provide direct experimental evidence to back this up, it does just follow logically that running just below LT1 carries greater forces, and thus greater potential injury risk, the faster an athlete is. Thus why we see the faster men running at a lower avg percentage of their race pace than the women.
A compounding factor, as you correctly point out, is the trainability of resilience. An athlete doing greater quantities of easier volume is finding that sweetspot of mechanical stress (which is not purely “bad”, but indeed essential to train resilience!) while still providing a solid aerobic training signal. The core of my argument is that this sweetspot for easy volume is not just a function of their aerobic fitness, but also of their resilience and thus the mechanical load. While aerobic fitness means the cardiovascular response to 4:30/km between two athletes of the same weight can differ wildly, with athlete A below LT1 and athlete B above LT2, the forces exerted will be similar. I’m not suggesting that aerobic fitness and resilience don’t correlate, as running training will affect both. But to simply ignore the role that mechanical stresses could be playing in elite athletes’ easy pace selection means that it’s not really valid to scale everything down to vastly slower athletes. Particularly if someone has several years of running under their belt and thus a decent resilience built up, but only 30-60 min to do an easy run on a regular basis, they might really be leaving fitness on the table if they’re plodding out 6:30+/km. They’d possibly also find it difficult to join others on easy runs at, for example, what would otherwise still be a reasonable 5:45/km for them.
To be clear, I think this consideration cuts both ways. An athlete that is very fit from cross-training needs to be particularly careful about the mechanical load of increasing running volume, because their aerobic fitness is more likely to be decoupled from their resilience. And I still agree with the general conclusion you’re reaching. But considering how thorough you are, it seems to me to be a big oversight to leave out mechanical load and resilience entirely from your considerations. And, as I said, drawing your numbers from elite athletes and applying them linearly down to athletes racing at half the speed would appear to me to be overfitting your observations into unsupported prescriptions.
I think this is the source of our disagreement and misunderstanding.
I say in the article that my argument isn't aimed at the time-crunched runner. If you only have 7 hours or so to run each week, then obviously building mileage isn't an option available to you (However, I still think there are more effective ways to spend that 7 hours than hammering your easy runs, such as the Norwegian Singles Approach, for example).
I'm trying to convince the people who are willing to put in the necessary time and who want to train in the best way possible. My argument is that in order to do this, they need to slow down their easy runs, so that they can run more; that elites have provided a blueprint on how to train as close to optimally as currently possible - which involves running a lot more, but in order to do this, a necessary step is slowing down, otherwise, you will not be able to do it. And so, I haven't ignored mechanical load - I'm saying that this is the best way to manage it.
And just to be clear, you could certainly scale down using this approach. As a demonstration, you might be doing 70kms a week while hammering your easy days. If you ran more slowly on your easy days, you might be able to run 80kms or more for the same training load - and you'd also be in a better position to build upon this. This is my point - there is a trade-off between intensity and volume - but the volume side (as evidenced by elites) has shown to have far greater upside.
Also, to address the argument on women running at faster proportional paces than men; there are many other variables to consider outside of them simply having an intuitive understanding of mechanical load. One might be; they do a lot of their easy runs with men who tend to run at slightly faster paces. This faster pace may then become the new normal.
This is a conversion from the sample of 24 elites. Their average easy pace range was between 63-70% of their 10k pace.
I have a hard time believing that this makes sense for slow runners.
At age 66, my 10K race pace is only about 9:15/mile. 63-70% of that speed is around 14 min/mile, which is walking. There's nothing wrong with walking, but it's not training for running. I'll keep doing my easy days at 10:30-11:00/mile, which is at least jogging.
Similar age and at 65, I found I could run reasonably well - 5k pace around 6:20 - by alternating Easy Interval Method Sessions (either 3x2000m; 6x1000m; 12x400; or 16x200m) with 3 or 4 mile recovery runs at around 8:00-8:15 per mile.
So maybe halfway between those e.g. 8:23 - 9:18/mile
Isn't running that slowly utterly destructive to mechanics? I'm 67 and can't imagine going out for a run at 9:26 per mile pace. I'd have about a 12 inch stride.
If I'm that tired I'd take a day off or cross train.
I looked back quickly and couldn’t find the reference to 4 hour marathons, but it’s worth noting that marathon pace has completely different meanings for fast runners and slower runners. Most marathoners run their race at the same pace as their easy runs or slower. I’d guess marathon pace pretty quickly converges with easy pace after 3 hours or so then gets significantly slower than easy pace as you get much over 4.
I can see that. My best distance was 3000m, and although I only did two marathons, my easy running pace (what I would run on non-workout days) when not marathon training turned out to be about the same as my marathon pace (in the 6:20s). Of course most of those runs were only five or six miles. I can't think that running 5 miles at 70 sec. per mile slower than 10k pace is very taxing. I certainly always felt recovered enough for track sessions.
This, in a nutshell, is the problem with the approach. If the only thing I knew about training were what these 24 athletes did, and my assignment were to get fast, I would start by doing exactly what they're doing. No question. You have to use what's available.
But why would should we ignore everything else we know?
Whenever someone points out reasons that your thesis might not hold, you accuse them of making unfounded assumptions, even though you are invariably assuming the opposite, without evidence. Instead you just nakedly assert that your assumption is more plausible.
The exchange above is a good example:
It's a fact that connective tissue responds to training more slowly than anything else. It takes years and years for bone density to increase and for tendons to become stiffer. That also suggests that even if recreational athletes should emulate elites, they probably can't do so in the short-to-medium term because they don't have the same bodies. Your data is about people with many, many years of high level training. It doesn't tell us how to get there. Because different adaptations have different time horizons, it's reasonable to think that the proper mix of training will be different at different stages of an athlete's career. You'd probably call this an "assumption," and it is, but it's grounded in science. It's certainly no less reasonable than your assumption that the training should be identical.
It's also just simple physics that the mechanical stress on a runner's legs increases exponentially with pace. Elites can't really stress themselves aerobically unless they put a ton of stress on their legs. Does that prove that elites need more pace variation than recreational runners? Of course not. But, again, it's a plausible hypothesis grounded in science.
(Incidentally, it's not an "assumption" that elites are under more mechanical stress at any given percentage of race pace. It's just math. Kinetic energy is mass times velocity squared, so a 15% difference in mass between a recreational and elite runner is going to be dwarfed by the difference in their paces when it comes to how hard they hit the ground. Running economy is irrelevant. That's a measure of how much oxygen you consume at a given pace, but impact forces are unaffected because of Newton's third law. There's no way to hold a pace other than to hit the ground hard enough.)
I've done my best to support every claim I make in the article with evidence .
If I'm to take counterclaims seriously, then I expect whoever makes them to do the same.
I don't see how the onus is then on me to disprove every unsubstantiated claim with further evidence in this thread - especially seeing as the claimant hasn't extended the same courtesy to me.
As for the first part of your final paragraph, you'll find that I've already conceded this point at least twice earlier in the thread.
But I've argued that an athlete's ability to tolerate greater load also increases as they train more. This should be completely obvious. How else would people get to such huge mileage figures?
And just as many recreational runners get away with running very close to LT1 for every easy run, an elite would too, if they were only doing 70kms a week. But elites have figured out that this isn't the best way to train. If they want to run, much, much more (which is a better way to train) - they have to slow down. I'm arguing that recreational runners should follow the same process. That's it.
Note: I also disagree that running economy is irrelevant, as biomechanics play a large role in running economy. If your muscles and tendons have to work much harder to achieve higher paces - then recovery time between efforts is going to be further increased (and therefore, being more conservative between harder efforts is all the more necessary).
This post was edited 1 minute after it was posted.
Also, to address the argument on women running at faster proportional paces than men; there are many other variables to consider outside of them simply having an intuitive understanding of mechanical load. One might be; they do a lot of their easy runs with men who tend to run at slightly faster paces. This faster pace may then become the new normal.
Or maybe they are running relatively faster because they are slower runners.
I calculated the bivariate correlation between the estimated 10k pace and their easy run pace (as % of their 10k pace) for the 24 runners in your table.
The faster end of easy pace (as % of 10K pace) has 59% correlation with their 10k pace, and the slower end of easy pace has 58% correlation with their 10k pace.
I think this is the source of our disagreement and misunderstanding.
I say in the article that my argument isn't aimed at the time-crunched runner. If you only have 7 hours or so to run each week, then obviously building mileage isn't an option available to you (However, I still think there are more effective ways to spend that 7 hours than hammering your easy runs, such as the Norwegian Singles Approach, for example).
I'm trying to convince the people who are willing to put in the necessary time and who want to train in the best way possible. My argument is that in order to do this, they need to slow down their easy runs, so that they can run more; that elites have provided a blueprint on how to train as close to optimally as currently possible - which involves running a lot more, but in order to do this, a necessary step is slowing down, otherwise, you will not be able to do it. And so, I haven't ignored mechanical load - I'm saying that this is the best way to manage it.
And just to be clear, you could certainly scale down using this approach. As a demonstration, you might be doing 70kms a week while hammering your easy days. If you ran more slowly on your easy days, you might be able to run 80kms or more for the same training load - and you'd also be in a better position to build upon this. This is my point - there is a trade-off between intensity and volume - but the volume side (as evidenced by elites) has shown to have far greater upside.
Also, to address the argument on women running at faster proportional paces than men; there are many other variables to consider outside of them simply having an intuitive understanding of mechanical load. One might be; they do a lot of their easy runs with men who tend to run at slightly faster paces. This faster pace may then become the new normal.
I appreciate you taking the time to respond, but feel that you’re digging your heels in rather than accepting the limitations of the evidence you’re providing.
You‘re also resorting to strawmen by twice implying that I’m arguing for “hammering your easy runs”. Not only am I explicitly NOT arguing for this, repeating that I agree that most need to slow down on their easy runs, I would strongly disagree that a 44min 10km racer is “hammering” by running 5:45/km with some friends for an easy run rather than 6:30/km. I’d also be surprised if that runner, assuming that they’re reasonably fit, is pushing close to their LT1 to run that easy pace. To belabour the point — I’ve been training more or less with the NSA for the last 5 months, am in roughly 34min 10km shape, and tend to run my easy runs at ~5min/km. I obviously don’t believe in hammering easy runs.
You volunteer to concede this point when it comes to elite women being able to run their easy runs a bit faster to run with male colleagues! You do so without having any explanation as to why that’s physically acceptable, but reject the same argument (with a mischaracterisation as “hammering an easy run”) for the slower amateur athlete, despite the fact that the slower amateur will experience a comparatively smaller shift in mechanical stress due to the slight increase in easy pace.
I furthermore don’t see how you’re accounting for mechanical stress, other than by correlation. Simply following the logic of forces applied at various paces means you’re being exceedingly cautious for slower athletes. Again, you are taking a small sample of elite athletes that are by definition extreme outliers and linearly extracting their pace percentages out over people running at half the speed, while mechanical stresses are scaling differently. You can repeat your conclusions over and over, but your evidence is nonexistent that these paces are valid for people running at significantly slower paces. It seems odd to me to take so much time to write this article, analyse all the data, then overfit the curve away from the collected data points, and refuse to concede any ground whatsoever to well-founded criticisms. I for one would be fascinated to see what the trends are amongst successful and/or high volume age-groupers, etc.
Elites and recreational runners now race in super shoes. I don’t know if elites run easy in super shoes, but most recreational runners won’t because of the cost and durability. This should increase the differential between race pace and easy pace, for the recreational runner.
Help us build the best running shoe review site for a chance to win a LetsRun t-shirt.Help us build the best running shoe review site for a chance to win one of 10 LetsRun t-shirts.