I personally don't think you need to obsess over data (even though I'd admit I enjoy gathering and analyzing it). I have used a lactate meter, heart rate monitor, and plain old sensory data (i.e. effort relative to the distance being raced or workout being run).
I don't know if it was sirpoc or another poster that mentioned once you get a familiarity with the effort you are putting out in a workout and how that correlates to your blood lactate readings, you can use your sensory data to execute the workouts to a pretty optimal level.
For example, I live in Florida and know how my expectations and training during June through September will need to be modified. I do less volume in the workouts and not as long of a long run, yet inevitably feel more of an internal training load stress because of the heat and humidity.
I have seen rTSS discussed and understand it to be a metric used to quantify training stress/stimulus, which I think is great. Daniels, in his second addition (and possibly updated ones) included a table which assigned a point value based on duration of the run and intensity. I also think just learned experience and an objective assessment of your training and performance is equally viable. You know when you're pushing beyond what is sustainable and repeatable, it's just a matter of discipling yourself and having confidence in the approach to see it successfully through!
Training Stress? wrote:
sirpoc and others,
what tools to you guys use to monitor training load/stress? it sounds like sirpoc uses training peaks, what other ones are out there and any good?
Training peaks is the only one that official can use TSS, although golden cheetah did, I don't know if they still do. Strava score is pretty much the same, as well as intervals icu. That has it as fitness score. It's all basically the same calculation and equations Coggan came up with alongside hunter Allen, from memory, .But you must keep your zones updated. Or it's useless, even individualised to you it'll be worthless. This is where I run parkrun every 3-4 weeks as it gives me a chance to update paces, zones etc.
Curious1! wrote:
When doing this kind of training would you still have a medium long run between workout days? And would the long run be easy then so you can recover? And on what days would you do strides?
Nothing here is original, all structures that plenty of coaches and runners have used before. Coaches like Tinman, Rubio, Wetmore, Hudson us a pretty standard outline as follows:
Sun - Easy w/ strides
Mon - Workout
Tue - Medium Long Run
Wed - Easy w/ strides
Thu - Workout
Fri - Easy
Sat - Long Run
If that's a structure that appeals to you, you can use that template and just make both your workouts threshold sessions. If you're wanting to follow the "Norwegian" outline:
Sun - Easy
Mon - Tempo
Tue - Easy
Wed - Tempo
Thu - Easy
Fri- Repetitions
Sat - Long Run
If in your warm up for the two tempo sessions you do 8*100m strides, that's two sessions of strides there. The repetition workout is essentially a stride workout, so far as pace goes (assuming it is done at approximately 1500m pace) although if the repetitions are longer and/or the total volume is high, the physiological impact can be different than a basic leg speed session.
I think a lot has to do with how you define these different training philosophies. For example, if you just think of the Norwegian System as emphasizing threshold workouts then whether you use the first template or the second, you're training using the Norwegian System. This is nothing new to be honest as runners have been using these outlines for years. I think what gives the Norwegian System it's novelty is 1) the regular utilization and analysis of lactate readings and using lactate values (or efforts) that are lower than the traditional LT as popularized by Daniels and others here in the US as 60 minute race pace or 4.0 mmol and 2) the redundancy in the workouts used. This even goes against some coaches suggestions on never repeating the same workout or always progressing the workout so as to modulate the training load. I personally think consistency over many weeks, say 18-24, and then just repeated for each build-up is better than the really intense workouts that are progressed toward at the end of most runners' training plans in order to illicit a big stimulus and "peak" them. Hope that was helpful and gives you some ideas.
Curious1! wrote:
When doing this kind of training would you still have a medium long run between workout days? And would the long run be easy then so you can recover? And on what days would you do strides?
I literally do the same thing every week.
3 sub threshold sessions (all last about 65 mins with a warm up and down) 3 easy runs of 50 mins and a long run of 75 mins. All are the same pace roughly. 65% of MAS (well current estimated based on my 5k time, but will get you within a few seconds per km of it).
No strides. Nothing. The way I do it is so simple. The only thing I would/ might change is maybe add on 5 mins to the easy days and long run. The only rationale for this is there's a point where I saw Jacobs hobby jogger brother increase volume a bit, the odd extra rep here and there and he kept the overall rough 80/20 ish (more like 75/25) time in zone to the same ratio by then just making the easy days slightly longer. So it slowly moved him up to where I am now (just under 7 a week) to up to around 8 hours on average. For him, 10x1k has slowly become 12 etc. 5x2k has become usually 6.
Shirtboy made a really good point. This is pretty boring. Pretty robotic. But it's by far the best jump in performance I've had after quite a while being around the high 18s. I really have changed nothing else, not weight, time training , literally nothing. Except the make up of the actual training load of the week.
Follow up question wrote:
Probably a dumb question given the intricacies of the info but can someone clarify what you mean by sub threshold? For instance my watch estimated threshold HR is 164, with my threshold zone being 157-167 bpm. Would that mean it’s more optimal to shoot for closer to 157 heart rate while doing the reps or even slower? Thanks in advance for the help!
If you know for sure (within a beat or 2) that your threshold is 164, then you have a starting point. It won't be perfect, but you can certainly so the 1k and up intervals probably by heart rate. I've never looked at heart rate during, as I do it on paced but this was married up to my lactate testing I did, which as a result I know will almost always keep my HR under threshold.
But you could aim for between 153-163 (absolute max) for longer reps, say 5x2k. If you start out the reps and try and keep the pace the same and after the first one you are nearer 153-155 by the end, you probably will keep the whole lot under. I've not really tried to do this, but you could experiment. Just looking at my runs, my LTHR is 175, I'm usually in the mid to high 160s by the end of the first couple, and the low 170s by the end of the 5th. Definitely don't start out too hard, what you don't want to do is go too far over your LTHR. It defeats the purpose. I posted way back on another page, but I purposely once went over by upping the pace in the last rep, spent almost 6 mins which doesn't seem like a lot, but I was trashed the next day compared to what I usually am. I couldn't have done that 3 times that week. So it's a fine balancing act, very controlled, very measured, as others have said, very boring.
Side note: you just wouldn't be able to control anything under 1km by heart rate. I've never tried but I don't see how you could really. Someone might be able to correct me on that. Even at 1k it might be tricky.
HR q wrote:
I usually run 8:20 pace too. I've inadvertently done this sort of threshold work in the past, as my speed is poor even in workouts. For example my 400 repeats would usually be 6-6:05 pace and my 5K 18:07, yet I have also run 36:39 for 10K and a 1:20 half. Is the secret to do threshold 3 times a week rather than 1? I probably can't do it without getting injured.
Everything is run at sub threshold state, the pace is going to be whatever pace it takes you to get there. You wouldn't be able to do 3 true threshold runs a week, even 1 + a vo2 max session is asking a lot. But these take less out of you, but you need to know you are running in the right area.
Let's say the 18:07 is your most recent, you are running you 400s a little faster than I would, as you are right bang on about CV. If you've run that 36:39 and that's more like your most recent fitness, then for you it might just require a slower % than me of CV for the 400s (which for me let's say 98% of CV). You are likely going to fast I would say and probably at least the last 5-6 reps are pretty rough? I'm just guessing as that's all we can do without measuring lactate.
Perhaps try 10x1k or a couple less if appropriate to your mileage. But run them around 6:15-20 per mile. See how that feels.
This is where once you get going with this, as a few others have said it becomes easier as you can adjust based on the repeatability of your workouts. The racing helps as well so you actually know where you are at, not guesswork, or what you might have done a year ago (not saying that's necessarily the case for you).
I feel like what this thread is really uncovering is getting to the bottom of what training really is. We should all be individualizing more. There are certain shortcuts or benchmarks people use as a guide for how to train which are really quite false and bad but are very common:
1.) Using a race performance to modify training. It should be done in certain ways but not in other ways. Where this goes wrong is when the race performance is good and people use this as a reason to modify their training. Isn't that just crazy? Good races are often used as justification to increase training intensity or volume, like "I need to update my training paces according to this new performance" or "I have achieved all the improvement I can get out of this training volume and now I need to increase it if I want to continue to improve." Both ideas are insane! But they are so common. If you have a good race, that is an indication that your training is effective as it is! Don't change it! It's only when you have a bad race that you might say "I'm overtraining" or if you have plateaued then you might say "I'm undertraining." If you have a good race, your training is effective, don't significantly change it! Run the same volume at the same perceived effort. Paces and mileage will naturally increase without conscious effort. Keep doing that until improvements stop coming, and only then do you intervene.
2.) Comparing your training to someone else's. This is also crazy, especially when you tie it in to comparing your races and your training to someone else's races and training. Our ability to train is way too individualized, just like our ability to race. One untrained person can run a 17:00 5k while another runs a 23:00 5k. One untrained person can do training load X while another untrained person can do training load 2*X. It can really vary by that much such that it's not even close. And the two things (our natural ability to race and our natural ability to train) aren't related. The 17:00 5k runner might be that much faster in the 5k but they cannot handle anywhere near as big a training load as a slower 5k runner. In fact I'd guess that they're probably inversely related: the faster you can run with less training, the lighter your training should be and the more cautious you should be about increasing it.
3a.) Key workouts. People do "indicator workouts" or "race simulator workouts" which are totally unnecessary and completely break the optimization of training (in the sense that the amount of time they take to recover from is not worth the benefit they're providing). RACES are the thing which we try our hardest in and break the optimization of training to do. If you want to do hard efforts, then do a race! Don't race in training! People need to have confidence that their training is effective. Jakob has talked about this -- he doesn't need to prove anything to himself in a workout. He's 100% sure that the training he's doing is effective. His hardest efforts are reserved for races, where they matter. All the rest of his effort is used for optimal training, which doesn't involve maximum efforts.
3b.) Peak weeks and "down weeks". This is so high risk and is there really a reward? You're essentially overtraining on purpose. What if you don't fully recover? Your race performance will be ruined. What if you would've gained more fitness by just having two normal weeks? Marathoners will increase and increase their mileage and then have a huge marathon pace long run "on tired legs" for the peak of their training. Maybe, MAYBE, designing your training cycle with this kind of thing is optimal. But SO many people mess it up and it is absolutely not necessary. 99.9% of runners do not need to mess with this concept at all. Not every training week should be identical in nature (there's some training that's better done far away from your race and some that's better done closer to your race) but no one should be overtraining on purpose, even if they're scheduling extra recovery.
My main takeaway from this thread is that having 3 quality days a week and running really truly easy on the other 4 days is a very effective foolproof template that everyone should try to follow. And in order to follow it, the quality days have to be easy enough to recover from that you actually recover and can sustain the training. And in order to do that, the easiest type of workout that you can do which is still effective is these easy threshold intervals, starting with as small a volume as you need to sustain it and building up to what Jakob does as a maximum volume. These sessions are what you can do on the days which precede 1 easy day. What you do for the 3rd quality day of the week, which precedes 2 easy days, can be a little more difficult and should cover a different aspect of physiological development (like doing hills / speed, or for a marathoner, a long run with some kind of tempo / fartlek).
So as someone else mentioned, trial-and-error is very likely the best thing an individual can do. Doing some kind of fitness test, max heart rate test, monitoring heart rate or monitoring lactate, etc etc, should just give someone a starting point. And then it should be individual trial-and-error: can I do my 3 sessions a week indefinitely? Am I really staying on top of my recovery? That should be the top priority. Make it easy enough so that's a confident "yes".
After that, you occasionally race. If your performances are improving, then change nothing (except what naturally changes by keeping the same perceived effort). If they plateau, then think about how you want to increase the stimulus so you can continue to improve: maybe your paces have slacked off a bit and should increase, or maybe you should increase volume a bit. Either way, make a small incremental change, make sure the training is still indefinitely sustainable, and race again in a few months.
There's NO NEED to continuously compare your journey to anyone else's or to refer to a chart or any physiological measurements or anything. I understand that doing so has been massively helpful for Jakob and Gjert says it's essential, but they're attempting to reach absolutely 100% of Jakob's potential, setting literal world records. And copying them might not even work for our own physiologies. They've made no attempt to provide a universal blueprint that will work for any physiology. We should guide our own journeys. Start somewhere -- err on the side of the training being too easy -- and see if you improve or not. Get an idea for what you can handle and increase it only if necessary.
Anyway, that's my takeaway. I personally have been the guy who runs relatively fast on little training, who has then taken cues from other peoples training (and not just on my own in some misguided and incorrect fashion -- I've received individualized training from coaches who have coached elites, and from elites who have gotten into coaching) in an attempt to get more serious and perhaps graduate from "sub-elite" to "elite", and then just ended up overtraining over and over and over again. If that rings a bell, I think this is a good lesson to learn. If that's not you, maybe my perspective isn't helpful. I'm too old now to ever be elite (well, there's always the masters category) but finding good training late in life is still a great gift, better late than never.
I respect the people who are keen on comparing training and races in order to build general knowledge and understanding. It's an essential part of all of this. It's a worthy project and could eventually lead to comparisons actually being more useful than they are now. But for an individual just managing their own training, my strong opinion as someone who has been led astray so many times, is to use it as a starting point and then live in your own individual world.
PS: I have a pet theory that a lot of sub-elites and low tier elites have been overtraining their whole lives, but not so bad that they don't improve, so they just keep doing it their whole careers. Then at some point they stop improving and they cannot possibly train any harder, so they think "Well I hit my genetic peak. This is it." But they're wrong -- they could've hit a higher peak by training lighter along the way. Some of us can't improve while overtraining but I think a lot of people can and then they develop a totally warped idea about the relationship between training and performance, and that trickles down and permeates to everyone else, creating all kinds of false ideas about what kind of training is necessary to achieve a certain performance. I base this on the fact that (like I said earlier) people keep increasing their training loads before they plateau. They have so much fear about "wasting" a season of training by undertraining that they constantly err on the side of overtraining, but not so much that it's catastrophic, and then they end up spending their whole careers slightly overtraining, never actually training optimally. They burn out and/or plateau before they accumulate as many years of effective training (before getting too old and declining) as they could have.
Wow! This thread has really taken off again in the last two days, awesome! There are some interesting details being shared and ideas floated! I don’t want to turn this into a training manifesto (LOL), but I have a concept to pose to the group.
The idea of blood lactate readings vs intensity and duration. If LT1 and LT2 are defined as a pace for a specified duration, those durations can be fairly accurately estimated using a race result. I’d even propose using a time trial as that will probably not result in as fast a time as a race but could actually prove more useful for training. Sirpoc has said, and I agree, the fatigue you carry during training usually results in a small percent reduction in training paces when compared to what you actually race at, assuming you are running consistently. It’s not that hard to find a consistent route or hit the track and do a 3K-10K time trial. Nut up folks! LOL
I brought up Daniels and mentioned how I found his T pace sightly too intense when doing regular tempo interval sessions, so I ran at or slightly slower than half marathon pace instead. Well, for me, a few seconds slower per mile than my half marathon pace is a pace I could hold for 90 minutes. I mention this because the pace also correlates to about a 3.5 mmol blood lactate value for me. Which, when I was using a lactate meter, is where I settled into for my tempo interval sessions, 3.0-3.5 mmol.
Tinman on a thread from YEARS ago (perhaps before I was born LOL), provided a conversion table for tempo training as follows:
4.0 mmol = 1.07*(5K_Race_Pace)
[3.0] mmol = 1.10*(5K_Race_Pace)
2.5 mmol = 1.13*(5K_Race_Pace)
Tinman knows there can be variations in those readings, but I believe he provided them as a general guide (this may have been before he created his calculator even). If memory serves me correct, he said the paces would approximately work out to LT, half marathon, and marathon, respectively.
If you read Marius Bakken’s training page, not only does he detail his training and experience, he provides the general outline of the Ingebrigsten’s training and a link to an interview with Kristian Blummenfelt’s coach, Arild Tveiton. They all reference plenty of threshold work in the 2.5-3.5 mmol range. Brad Hudson in his book, Run Faster, references using 60, 90, and 150 minute race pace for threshold sessions.
Guess what, if you do the math (and I did LOL), the multiples Tinman gives, the lactate values we see being cited by the Norwegians, the Zone 2 range commonly currently referenced… all of those work out to 60, 90, and 150 minute race pace. To associate those times with the current language, [150] minute race pace is LT1 and [60] minute race pace is LT2, with 90 minute race pace falling in between of course.
Now, I tend to think most runners are not going to buy a lactate meter, lancets, and test strips. Whether it is the cost, practicality of use, or general lack of interest in being that technical. However, the associated paces are close enough. In fact, I’d call them speed barriers. Meaning, don’t see them as paces you have to hit, but paces to not exceed!
Bringing this back to Jack. In 1979, Jack Daniels and Jimmy Gilbert published Oxygen Power. In it, he provides the regression equation he uses to derive his race prediction times. From that equation, you can calculate exactly your 60, 90 and 150 minute race pace if you know your vVO2Max. vVO2Max can be obtained by running a time trial (or race) that takes 8-10 minutes. It’s an estimate of course, but one that gives you a quantifiable starting point to gauge your sessions.
Last point, and sorry for such a long post, even if YOUR lactate at those aforementioned race times varies from the standard BLa values, it is still a 60 minute, 90 minute, and 150 minute effort that is used to define the zones we see. Daniels amassed an incredible amount of research data. He states in his book while most runners will average a BLa of 4.0 mmol for their 60 minute pace, he’s measured runners with a range of, I believe, 2.8-7.2 mmol.
I hope this reads as clear and robust as it plays in my head. Mainly, I hope it gives a different perspective and sparks further debate.
P.S. Daniels derived a point system based on the intensity of your run and accumulated duration, so that you could standardize and track your training. That was published in his second edition in 2005. Just wanting to point out what a tremendous resource he’s been.
(This is data-nerd, logged in)
As I hope I have indicated, I am a complete beginner on this and am only reflecting what I have learned reading this thread and some side research in the meantime. So I write this with the utmost respect, to all but JS, and forgive me if I am repeating what everyone already knows. My collected thoughts are thus:
To bring it back to the 'Norwegian system', which for argument's sake let's say is that outlined by Bakken on his website. Bakken argues:
In theory, it sounds natural to train at a speed that you need to perform at when racing. However – the mechanical “speed” you are running will always at one point or the other be majorly be limited by the aerobic abilities, where the ability to run at a maximum speed at the anaerobic threshold is the main one. So specific training from a more physiological standpoint, where you optimize the internal cell processes the absolute most effective way is not at race speed, which can just be done limited. Rather it is a combination of larger amounts of lactate threshold work with a fair level of total running, which was in my case about 180 km weekly running.
So, the Bakken system involves two key characteristics: lots of LT training and lots of volume. The common refrain on this forum has been that the former goes hand-in-hand with the latter - hobby joggers, don't even bother.
What makes this thread so interesting is that, as pointed out by spoc initially, we observe significant progress from K Ingebrigtsen running pure sub-threshold with a modest amount of volume. Spoc and others have also observed this in themselves, with further (very interesting) corroborating evidence from cycling experiences. This suggests that the pure lactate development approach relies less on volume than implied by Bakken and the conventional wisdom, though I think we can forgive Bakken who had the elite athlete in mind. It seems the more important factor to reaching one's potential is spending as much time as possible in the 'lactate state', maximising capabilities at LT. Perhaps the hobby jogger with a job and kids has hope, after all?
So, in our fixed and limited time available, we do as much volume as we can. It's not 180km a week, but that's ok, we're not trying to win medals. The remaining, and more influential, variable is time spent in the 'lactate state'. How do we know we are in the desired lactate state? In my view the only direct way is measuring the amount of lactate in the blood, with reference to lactate concentration at the anaerobic threshold (which varies between individuals). Heart rate, pace and power at LT may be observed but are ultimately anchored to the lactate measurement.
What sticks out (to me) about this 'Norwegian system for hobby joggers', other than the apparent revelation that volume is not essential, is that it involves systematic observations of individual parameters. Tinman, Jack Daniels, Pfitzinger, all the tables and calculators are based on averages. Instead of assessing performance by 'did I hit the prescribed pace that would put the average 18:00 5K runner in the lactate state', its 'did I reach MY lactate state, and is this providing adequate stimulus over time?'
sub-sub threshold wrote:
So basically it comes down to lactate level, and how that lactate effects your body. That is easier to understand.
I think I'm also going a bit too slow then, because during my 30 or 60s breaks I'm not breathing too hard, and after the last rep I feel fresh and like I can do the whole interval session all over again.
Is there a percieved effort for sub-threshold in terms of breath and how your legs feel? Are you supposed to feel "tired" but in control? I find it really difficult to actually feel the lactate when I run, especially compared to biking, where my legs become two stiff burning logs.
I'm just trying to think how to explain by feel. It's difficult I think as it's relative. Let's use 10x1k as an example. As I do this weekly. The first one barely feels like an effort at all. Sometimes the second one, the same. The middle rep, I'm thinking, this isn't so bad - I can finish this no problem. The last 1-2 you think, OK that was a workout.
Compare that to a race, say a 5k, the first 1k feels OK, but by the middle you question can I even finish? And by the end I'm wondering if I can even reach the line that I can see without my legs falling off.
I don't know if that helps in the slightest, but that's almost how I feel or see it or however you want to put it. It you have a recent time to convert, have a look at the paces I put up and go on the slow end of them and see how it feels. I still think it's an excellent starting point. For example, try 10x1k at 15k pace , don't worry about looking at any of the data during. Just run it and have a look back at it all later.
Ok, this thread seems to have a bit of life back in it. So let's talk about the easy run and power.
So I've said before, I like pace. I live in a flat area and I'm looking at around 65% of MAS for these runs. This is for the 3x normal runs and the long run of the week. This will normally amount to around an average of 70% Max HR, probably slightly under.
I totally concede, pace is not the best metric for some people. Living in the hills etc. So I bought a Stryd, as you can see a few pages back. Previously, I didn't have much luck. But I've persisted. I'm having trouble (or was) with the sub threshold stuff, but I feel I have some really good data to feedback for those interested on the easy runs.
So I did 4x easy runs blind, I just did my usual pacing and noted down the results. Then on my 5th easy run today, I tried to run ONLY to power , I didn't look at anything else but aimed for average power of the 4 previous runs. The result? The pace of the run was a bit more up and down. However, the power curve was almost flat and the result a run bang on 65% of MAS on average for the 55 minute run and 68% of max HR. Incidentally, the HR was a lot flatter. Much flatter in fact, due more even power distribution. All 5 runs it turns out were within a few "watts" of each other (let's use watts like that, as it's not really watts). This was at 78% of my FTP.
This seems to make total sense. In the stryd zones, this is the upper end of zone 1. Much like it would be for pace and HR. So all matches up very nicely. Now I'm not saying that is the case for EVERYONE. But I know there are some guys out there interested in using power . And if you have a pretty decent idea of your FTP, I would suggest 78% is a really good starting point. This resulted in the top end of the MAS suggestion, for the easy run. So let's say 78% is the cap and work down from there.
jiggymeister wrote:
f wrote:
You can chose to believe that running your workouts at 6:00 pace versus 6:10 matters. But science can't answer that.
The small difference in pace (i.e. the few seconds you mention) absolutely matters and makes a noticeable difference.
I performed my LT stage test a few days ago to see how my lactate values changed since the last test which took place 7 weeks ago.
I will report back with a more detailed summary of the changes in a separate post, but I wanted to note here and now that a difference of 5 seconds per km between the last two stages made a huge impact. Going from 3:45/km to 3:40/km pace had my lactate shoot up from 2.4 to 3.8 mmol/L.
Hey jiggy. Would be really interested in this summary As it's something I have played around with. I've posted before about deliberately going over threshold, just so see if it would kill me 😂 obviously I didn't die ha ha but I did go about 7 seconds per km like you mentioned for the end of a 2k sessions and my lactate rocketed to well over 4+. I was otherwise in around the 2.5-3 ish range. What I really felt was it in my legs the next day, quite shockingly in fact for what was only a relatively short amount of time. The amount of fatigue is totally out of proportion to what seems like only a little bit of extra pace.
It's why I agree 10 seconds here makes a huge amount of difference. Once you get up in that high ish range just under LT state, it's a real balancing act to make sure you stay in that zone. It's why as boring (yet again) I sound, I would much prefer someone to run slightly too slow and pace on the cautious side, that to push the limits where the risk reward just isn't worth it. Even if you lactate is only 2.0 mmol or just above as a random example , you are still going to be getting a huge amount and % of the benefit as if you are at say 3.3 mmol, to pick another arbitratory point as the example. But the issue is say you are regularly hitting the 4+ range. There's no way anyone could handle this 3x a week on maybe 7 hours like me, for almost a year. The 4+ might be giving you a a slightly higher TSS per session , but the value just isn't there if you can only do it twice a week instead of three times.
This continues to be an interesting and thought provoking thread. I’ve enjoyed the serious discussion as well as the scattered troll insults. Some pretty clever ones a few pages back.
Some additional thoughts:
- We should keep in mind that “low volume” is basically the antithesis of the “Norwegian training philosophy”. Their goal is to train as much as possible. This applies to Jakob, the triathletes, and even Karsten Warholm. As Sirpoc writes though, this 3x per week threshold method does allow for about as much training as an adult with a family/job can handle.
- On that note, if the goal is to improve, your goal should be to always be doing more. It is about what you can consistently sustain, so the progression will be extremely slow. One way to think about it: over a few months it might look like you aren’t progressing or changing the training at all, but look a year later and you should see a difference.
- In running, as you hit high speeds the load on the muscles/tendons is exponentially higher. I recently tried to add a hill day as my 3rd workout and a few weeks later ended up with an achilles injury. I think that as you get older, the impact of high intensity workouts on recovery is greater. And I think Jakob thinks the same thing — this is why he says it is harder to be good at 1500 compared to 5k at older age.
- For those commenting that you can’t race a good 5k without ever running fast in training: first, note that Sirpoc is running a lot of 5k races. My personal experience is that it can be a little hard to push myself after a few months of this training (only did it once). Recently, I did a 5k and in the lead up did a workout where I’d do 5 x 1k, starting at the usual mile repeat threshold pace with 1 minute rest, and after the first 2 reps, increased the pace to the 3k-5k pace range, extending the rest before 3rd/4th/5th rep. I felt that it helped me to work into the pace (being not used to it), but also prepare to hit the slightly higher lactate levels. I’ll probably use that again in the future (when the achilles is better).
Some observations from KI strava:
- It is remarkable how much volume a person can progress to doing on this system. KI is doing 36-42k of work per week at this reasonably high intensity. That is massive.
Note how he is constantly doing more. Anyone who says Jakob is just doing 2x 10k threshold sessions on Tues/Thurs is forgetting that the main principle is to always do more. There is no way he is doing the same volume as he did a few years ago. (Nordas implied this in an interview as well)
- The focus seems to be on the pace, much more than the exact lactate. It seems like he checks lactate at the end to calibrate where he is, and also to decide what to do next session. But the pace is very tightly controlled (he even uses wave light, which is remarkable in my mind).
- He ran a 10k at 3:20 pace back when he never went below 3:30 on any intervals. Race pace may be over rated.
- When he was first making his massive improvements, his interval pace was actually slower than the HM paces he eventually would hit. This was even for 1k-2k reps. And back then he typically would just run the same pace for 1k, 2k, 3k reps. Now he seems to adjust more, but I thought that was interesting.
- They pay attention to everything to ensure progress. I believe he previously was running easy days a bit faster (7:20/mile or like 4:35/k), but now is more often closer to 8:00/mile or 5min/k pace. He tried a period of slightly lower volume of shorter, faster reps (10x1k) but felt banged up and didn’t improve much, so reverted to 12+k of longer reps. Initially long runs were shorter but in the last several months has started doing ~2 hours or 26k long runs.
- While the training started at about 1 hour per day, he’s doing quite a bit more than that now. So us (likely less talented) regular hobby joggers should not expect indefinite improvements if we aren’t also gradually progressing the training volume.
- People reference Tinman a lot on this thread. He did seem to have some understanding of this, but I’ve recently been thinking how it is crazy that a hobby jogger like KI can be training at what seems like significantly higher training stress than a Tinman athlete like Drew Hunter. I know Drew is no longer with Tinman, but workouts like 8 x 1k + some 200s seem somewhat unimpressive when KI is doing 14 x 1k on a Saturday. Makes you wonder if Drew was coached by Gjert how much better he would have been.
Well that was quite the novel. I guess I’ve been thinking about this a lot.
sirpoc 8/22/2023:
pace clarification wrote:
That's interesting. I do that pace usually for 3-5 miles continuously and label it as a tempo run. Usually not every week though as I get tired/lazy. Did the improvement come from breaking it down into intervals, or simply running that pace more frequently? Is there a certain amount you aim for each week, like 1 hour?
That's the whole point of these. You can do a lot more than straight tempo by breaking it up with even very short rest. I would struggle to probably do this 3x a week even in the 25-30 min range straight. But I can do 10x1, 6x1600 and 5x2 quite comfortably on short rest for about 35-36 mins total each session. The goal is basically to cram as much as we can in without breaking down. This sub threshold range with the rest seems to be the sweetspot, if you want to call it that. I do about 105 minutes a week out of my 7 hours I've allocated for training as of right now. You get fitter , simply because this seems to be the highest range of lactate you can reach pretty much every other day without fatigue setting in. I'm on about a year straight of this with a tiny break got COVID in between and running every day.
I know some people like the idea of a day off. But I feel worse when I do so just have got used to running everyday. I think this is highly individual. Listen to your body. If you really know you can't run everyday, don't. It's also convenient for me, I can pretty much so the same stuff, on the same days per week to fit it into my lifestyle.
I've also noticed as pointed out above , KI has really ramped it up. He's still on singles and the same system basically, just pushing the envelope. The really long runs and stuff like 14x1k maybe suggests he's training for a marathon? Something longer than the half he was training for before? There's quite often some comments in the chat under his runs , but it's usually in Norwegian.
The lack have race pace has never ever worried me. As I've said many times, I was training on the bike for years and I never did any race paced specific stuff for Time Trialling and this was never a problem. I've seen zero issues either at working like this and suddenly 5k pace being a problem or a shock if I've gone a month or even more with no parking.
I think one of the reasons this has been such an insightful discussion for many is that this method produces fantastic results in a commonsense way, YET flies in the face of foundational training philosophies. There have been several in disbelief of the results achieved. I think behind their troll appearance may be genuine questions of WHY this works.
Like Summary shared, here was a great post around page 60 summarizing what this training methodology IS, and I think it would also be fruitful to summarize what this is NOT and WHY. Below are common training ideas that are challenged in this approach. Note that these comments are for 5k-HM not the Marathon, which is a different beast.
Periodization: No up weeks or down weeks. No weeks with shifting focus. No peaking.
Just settle into your maximum sustainable training load and repeat. This not only works for hobby joggers from this thread, but also for the elites coach by Grete Koens. Perhaps periodization is an engineered way to handle a training system that inherently tears you down more than it builds you up?
Progressive Overload: The workouts do not need to progress for you to progress. There will be a ceiling to this approach, but it will take a surprisingly long time and will take you much higher than other approaches. There is a progression, but it looks more like season over season rather than week over week.
Variety: Variety is overrated (and helps coaches keep their jobs). Running is simple. A 5k is 95-99% an aerobic event. Variety works in fields like investing because there is no way to know for certain what will pay off. But in running, we know what trains the aerobic system with little cost: easy running and sub threshold running. In this case, the best investment strategy is going all in what we know works FOR THE LEAST COST.
Specificity: This is a big one. Perhaps a good way of looking at this methodology is that the quality of a session is determined by how little recovery time it costs you, rather than how strong the stimulus is in preparing you for the event. Adaptations take a long time,
so any stimulus will need to be highly repeatable for it to have any effect. To stimulate the demands of an event there will need to be ample recover, resulting in lower overall training load due to needed recovery time. Better to keep aerobic load high and humming along, and just freshen up as needed before events. Grete Koens does this approach for her elites as well. She says that the 1 specific session before a major event are more to convince the athlete they can perform, rather than any underlying physiology. Training at 5k
pace consistently WILL make you more efficient at 5k pace—however, it will cost
you. And it’s the cost that matters in the long term (injury/ overall training load).
Also, racing fairly frequently helps with performance. Often, a 5k race is less taxing than a 5k specific session!
Speed Training: I love strides. I’m surprised that KI and Spiroc don’t do
them. But it makes sense, because the likelihood that speed is the limiting
factor for hobbyjoggers like me is very low. We are all aerobically undertrained, and no amount of short speed will make up for that.
I hope this helps summarize what this training is NOT. I'd love to hear y'alls thoughts!
I would say that Spiroc doesn't come from primarily a physiological perspective, not that he isn't scientifically oriented. He certainly has experience making hypothesis and testing the results and in that way he is scientifically oriented.
This is what I was trying to get at in the above post about CTL and TSS. It's a very different approach to training than most runners (myself included!) are accustomed to. That's also why it's so fascinating. It's different and it works.
As runners we've been conditioned to think primarily physiologically. Any training argument that Jack Daniels, Steve Magness, or Pete Pfitzinger makes, for example, is primarily about physiology and NOT load accumulation.
When Spiroc is comparing 3 sub threshold a week to a Jack Daniels plan, his argument isn't about vLT, VO2 max, Aerobic threshold, economy or anything physiologically oriented. It's basically that a JD plan beats you up too much for a lower load. In the long term you can accumulate more load more consistently with 3 sub threshold runs. If you're trying to PR in 8 weeks, a JD plan will probably be better. But in 8 months, the argument goes, 3 sub threshold is way better because you create much more load and you can do it over and over again (more on that later).
That doesn't mean Spiroc isn't scientifically minded, it just means he's thinking about it from a different angle. This is much more natural for cyclists because power is such a reliable metric. This is another reason why pace based TSS is important. Although it isn't perfect, at least it's load measure that's relative to performance, where as heart rate based TSS (physiologically oriented) is an indirect measure that will have much less correlation to performance. Spiroc and cyclists in this camp seem to be thinking more in terms of work in, work out, numbers in numbers out, not "what physiological system am I working on." Do more work, smarter, and you get better.
In this approach, there isn't a lot of the other stuff we commonly see in programs: taxing long runs, up weeks and down weeks, peaking, large tapers, strides, race pace focus, periodization etc. The reasoning why this program doesn't include those basically boils down to how much load those methods create, in how much time, requiring how much recovery.
This doesn't mean that physiology isn't important! I would be very interested in all of the tests mentioned above to see what's happening 'under the hood' with this approach. This approach uses physiology to control training intensity. Rather than use physiology as a tool to discern what workout uses what physiological system, this approach just uses it to ensure we're not overdoing it.
Easy runs at <70% HR Max and Sub T under 4. mm (generalizing here) is more about the recovery needed than the stimulus provided. Threshold runs at 4.2 will provide negligible load compared to 3.8. But over weeks and weeks of tipping over the threshold, fatigue accumulates and it's no longer sustainable. The lactate testing and easy runs are about ensuring there is a positive load to recovery needed ratio. Stay on the positive side of that for a while and the fitness will come.
Alfie wrote:
Thanks for your post.
My only query, if you are finishing just under your LTHR, should it feel as easy as this?
I would have expected it to be getting towards the comfortably hard, at the end of each rep.
I can't speak for anyone else, but there's been some discussion on this lately on Strava as well. This is how I see it. The sessions are 4-6 for me on the basic RPE scale your Garmin gives you .
I had a 7/10 recently, that was a Saturday leading up to my 10k where I pushed the boat out a bit and finished in like a 3:10 or something like that? That was just to see what that pace felt like, as it was roughly the target for my 10k. I think that's the hardest session I've done in well over 18 months and it felt like any other session for the first 8 reps. So it shows you even 6 minutes pushing the boat out at the end can really change how a session feels.
Anyway, back to the point. I would never say it feels easy. But it doesn't ever feel hard. When I get to 6/10 on those days, I feel like I could easily do at least another 4-5km even, no problem. I've had days where it's felt 4/10 and in my head my feeling is "wow I could do all of that again!". It's usually more like a 5/10 or 6/10 though, even if the last rep when looking back after you might see you touched very close to LTHR.
7 for me is. I could definitely do another rep. Maybe one more rep on top of that if you offered me financial reqaed. When I did this the other week, I actually felt it for a good few days after and it was really in my legs for my long run the next day.
8/10 is I probably could do another rep if you paid me more than before and another after that maybe, if you held my family hostage (this is where I used to do most of my workouts 1.5+ years ago). But even with that jeopardy I don't like my families chances.
9/10 is a badly paced race where I probably ended up with too much in the tank, I think I've done one session of 8x800 a couple of years ago that sat here. I think that's my only ever training workout that hard and I genuinely thought I was going to die. 10/10 is something you will never do in running training and I don't think is possible. It's basically the end of any 5-10k I've ever done where I've got the pacing spot on. I just don't ever imagine I could push myself that hard running. (note I've done workouts on the bike that are probably 10/10, where it's way easier to fight through a brick wall)
I know it's not quite what you were asking, but feel it might add some context to the thread. Also, someone like Jiggy for instance, his perception due to climate might be totally different. It's all very personal I think who considers something easy and who considers something hard.
Once again some people are getting lost in the weeds here, this time with LTHR [lactate threshold heart rate].
Keep in mind that LTHR is nowhere near a clean cut physiological variable that some of ya'll are trying to make it -it's a proxy of a proxy for the thing we actually care about, and there are a variety of different definitions and methods for determining it.
Will LTHR change much in response to training? That depends on a few things. How developed is the athlete? How are they determining LTHR? With a relatively untrained athlete it could change a lot. With a highly trained athlete probably not so much. It may also "change" as a reflection of getting better at executing the method used to determine it. I'm a fairly experienced runner of reasonably high ability, but I've never done a Friel LTHR test, so if you had me go out and do that right now there's decent odds I don't get it perfect the first couple attempts -out of context it may look like I improved my LTHR, but really I probably would just poorly pace the first couple attempts.
What I'm trying to get at is that speaking to this concept broadly across individuals is impossible and useless. The accuracy of a metric like LTHR is entirely specific to each individual and how they are able to refine their use of that metric with experience. The effectiveness of a training method to raise LTHR will depend largely on the context of the individual using it, and ultimately doesn't matter much because we care about race performance not physiological estimations.
What actually matters
However you want to determine pace/effort just make sure it's a sensible protocol for your current ability and something you can repeat a few times throughout the year.
Within reason control training variables so there isn't a ton of day-to-day variability. When in doubt err on the side of caution.
Most important: get out and actually run, see how you respond, refine, repeat.
With sirpoc (or any other anecdote of success) the utility of analyzing their numbers is only to understand the framework such that you can take that and run your own experiment, you probably don't want to get too caught up the numbers themselves. The framework here with how it relates to LTHR is that you should mostly stay under your estimation of strict LT so you can do more of it throughout the week and repeat that week after week. This is accomplished through a balance of taking it slightly easier and the interval scheme itself simply reducing the accumulation of stress. Certain interval schemes with allow for an instantaneous effort that is harder than strict LT pace/effort, but the breaks keep the cumulative stress across the workout below strict LT.
I carried out a 6 week experiment this summer, testing lactate in the heat and humidity. I’ll link the full document, which includes graphs, and just provide a brief summary here.
Avg Temperature = 79.3 F / 26.3 C
Avg Humidity = 96%
Avg Weight Loss = 5.1 lbs / 2.3 kg
Every data point in the sample is from a 6 x 1600m session, performed outdoors, at the same location and same morning start time. I started the block with a 6 x1600m lactate test. From that data, I defined my speed at LT (vLT) and my heart rate at LT (LTHR).
In my specific case, I could be as high as 110% of LTHR while running at a speed less than vLT and reading lactate around 1.5-2.5 mmol/l. Where as, if I was in that same heart rate range as a result of running faster than vLT, my lactate varied from about 2.0 to 5.0 mmol/l.
The takeaway, using heart rate as a proxy for lactate would be very inconsistent in the heat and humidity, in my situation.
Hopefully the document provides some interesting data for those curious about this.
https://docs.google.com/document/d/1dPm4smu8PJnzZoQrAzBLBn5ohh0TjTbsX8KiMWK6Njk/edit?tab=t.0
ah so you don\'t understand .... wrote:
It's actually you who doesn't understand. There is NO SUB THRESHOLD PACE. It doesn't exist. The posts here and that everyone has talked about are a guide. To try and get you in to the sub threshold state. Look at sirpoc or KI runs. We have lots of clear data for them. Sirpoc might do 3:16 one week, other weeks we see 3:23 say the week before, all for the SAME DISTANCE. The conditions and a whole number of factors effect the running pace. But he or KI do not care about the pace in reality, just that they are probably reaching around or just below LTHR as a rough rule of thumb, by say the end of the last rep.
If people don't understand this then firstly, don't come here and try and act clever when you aren't. Secondly, understanding this is absolutely fundamental to making 3x sub threshold sessions a week work. Where you are pushing the limits but never exceeding it.
The "paces" are just a rough guide to get you started. That aren't set in stone as there is technically no actual definitive sub threshold pace itself.
My experience is very much this when it comes to paces. Eventually you just learn to accept you are going for effort (and using pace/HR/LT as a proxy) and the results is going to be highly dependent on the day and what else is going on. Mostly I try to just get going and settle into an effort level for the rep distance, then after about 5 to 7 minutes of subT work I'll note the pace and have a pretty good idea of how the rest of the workout is going to go that day.
Some days I look at the pace after that first period and think, alright we're moving pretty good! Other days I look at it, shake my head, and think oh well, that's just the way it's gonna be today, let's do the work and move on.
Took a couple of months to figure it out but realizing that the goal isn't to hit the same or better paces than the previous week, but to get the work in an make sure your good to go again in 2 days. If the end of my workouts has me at or preferably just below LTHR, I'm normally pretty happy, but that's normally just a consequence of getting the effort level right across the entirety of the workout. Pace could easily be 10-20 seconds slower than the previous week for any number of reasons but over time it'll all come out in the wash over time. It's always nice to have a quick day though.
The two sides of this VO2/specificity debate are mostly just talking past each other.
The details of any training scheme emerge from the application of general principles to specific constraints and goals. Part of understanding any training method is understanding the parameters of the athlete it is designed for. There is a lot of useless debate that stems from people not understanding or disregarding these parameters. People are trying to compare or modify pieces of this method to different methods that have a different set of constraints and goals -they are trying to plant different crops in the wrong growing conditions.
To the point ningkaiyang is making: this method is giving up some aspects of hypothetically optimal training in favor of the more practical reality that a lot of rec runners are operating within. It's building in a some common sense and risk mitigation into the method itself.
Yes if we properly execute and recover from more specificity/intensity we will see more improvement, but the reality is that as intensity increases it's easier to screw something up. The mechanical and metabolic strain scales non-linearly. We're looking at marginal gains vs shrinking a margin of error.
Expanding the crop analogy: A lot of training plans are asking people to grow a lot of different crops, including some plants that come from tropical environments regardless of the growing conditions they are working with. We'll have a more varied diet but theres a good chance some of your crops fail and we don't get as much food in the long run.
This method flips that and asks "how do we reliably grow the most food in the in the soil and climate we're already working with?" The diet might be more boring but at least we'll be well fed. There's still room for variety, maybe we add some native berries (strides), we can seasonly rotate crops (using shorter or longer threshold rep schemes to target different paces), but the point is that we fit to the ecosystem we're in. Okay not sure where I'm going with this but I think ya'll get the point.
I am curious though. What is the fastest results anyone here is aware of with strict adherence to this method or similar? Doesn't need to be the exact scheme but at least threshold-ish only workouts, no X-factor/VO2max/whatever, only specificity comes from racing. Strides are allowed.
I am new here so I am not allowed to post links. There is a big norwegian running club (by Norwegian standards) called "SK Vidar Lang" that have been using a very similar methodology as the norwegian singles for years now with great results. On their (norwegian) Wiki they have a video explaining the philosophy. I have made a transcription with wisper and translated it and asked notebookLM to explain why they do 3 thresholds per week and not other types of speed work:
The source explains the rationale behind prioritizing three threshold workouts per week, emphasizing volume and endurance. Other types of intensity work, such as hill sprints or shorter intervals, are viewed as less effective for building the necessary training volume.
Key points:
Volume is the primary goal: The training philosophy centers on building a high training volume to enhance endurance
Threshold training as a means to volume: Threshold training is the chosen method to achieve high training volume while maintaining sufficient intensity
Balancing intensity and volume: The aim is to maintain an intensity that is high enough to stress the aerobic system but not so high that it compromises the ability to accumulate volume
Longer intervals for volume: Longer, less flashy intervals are preferred because they allow for a greater overall volume of training compared to high-intensity, short-duration workouts like hill sprint
Tedium vs. Stability: While threshold training might become monotonous, it provides a stable foundation that can be adjusted without disrupting the overall program
Specificity of Endurance Training: The training is designed to improve aerobic energy release, which is critical for endurance performance in events like 3,000m to half-marathon races
Managing risk: Threshold training at a controlled intensity is considered a safe, repeatable method that minimizes the risk of overexertion and injury, allowing for consistent training week after week
The source contrasts threshold training with alternatives like 4x4 intervals or hill sprints, noting that while these can be valuable, they don't provide the same volume of training. The focus is on consistent, repeatable training that builds endurance over the long term, rather than short bursts of high intensity
training peaks for the win wrote:
CTL is just the average of your last 42 days training.
No this isn't true. I think this is where a lot of people are getting confused. It's a exponentially weighted average, about 90% over that time frame. But all the data you have ever put in, contributes to it, just the further you go back the less weight it carries.
It takes a long time, but eventually for the most part your CTL will just become your TSS average for the previous 42 days.
But imagine you started at zero, the way it works is let's say you did 100 TSS a day. After 42 days your CTL would be 63.7. It would take you 316 days to finally reach a rounded figure of 100 CTL. After that, your choices are do more intensity, do longer durations, or both. If you stick to 100 TSS at that point, in simplistic terms , then you have plataued.
I've mentioned before, this is why the second year I trained like this, the gains have been harder to get. You have to be creative and try and do more. But still, essentially every session ever still contributes, it's just the stuff i did 2 years ago is now minuscule in the calculation.
I will also add, it's probably only useful if you are trying to compare like for like. If you are really mixing up your training outside of what I am doing, rTSS probably isn't going to tell you the whole story anyway. The consistency of it (much like with power) seems to really work around the sub-threshold level. Of course, I have said before probably easy runs are over represented in rTSS. But again, assuming you roughly run them at the same % of intensity relative to your current fitness you input, it'll stay consistently over represented in time, so doesn't really matter if our intention is just to use it as a practical tool (which is all I care about). I'll happily leave the science to the much smarter guys than me.
It's likely why the graph I shared on Strava fit so neatly, in terms of load versus performance, because my sessions are still for the most part comparible in intensity going right back (with a few changes to sneak in extra load) but also a decent invease in overall training time.
In absolute simple terms , my daily average per session a year ago was 60 TSS per week and now it's 70. There's probably no real suprise that I'm fitter now, because I'm doing the same thing with a sprinkle of pushing the limits of sub threshold intensity, or just more or the same but longer. How you make up that increase is for you to work out, but that comes with experience if training like this.
Daniel's reformed truther 2/28/2025:
Exerciser attempts to hobby jog wrote:
Race load is interesting. First, I find load more believable comparing like-for-like (e.g. did I build or decrease over months, given I've been running the same program the entire time), than for comparing completely different workouts. But let's ignore that for a second.
I agree the race wrecks the legs more. But if you calculate the rTSS per hour, someone near Sirpoc's fitness might get ballpark ~116 for 5k pace and ~103-105 for 12-15k pace. Suppose he runs the 5k for 15:low and the reps for 31:mid. That comes out to ~85% more load for just the quality time on the 10x1k vs. the entire race. And if you have more easy running added on to the workout than the race your workout likely gives you more than double the rTSS of your 5k.
Load is presented, at least in intervals.icu, as tracking both the stress on the body and the expected fitness gain. The race beats you up more than rTSS predicts; do you improve more from it as well? Is the inaccuracy in both aspects of rTSS, or does it point to a divergence between the stress and the fitness gain? If the divergence is real it suggests a major design consideration for our workouts should be how we can accumulate more rTSS without increasing recovery demand. That would probably lead us to pack on as much very easy mileage as we have time for, and to use reps to keep our lactate controlled in our quality sessions, and to separate our quality from our long run.
Here's a longer cut of Sirpoc's very slow quote:
> 3 easy runs a week. All under 70% of max HR which is usually about 65% of MAS for those who work in paces as well. This will keep you definitely under LT1. Long run, I tend to just keep the same and by the end I'll be almost at the upper limit of that 70% which is the goal. I think the recent studies which was excellent, on the training characteristics of long distance elite runners (2022) had easy characterised as under 70% Max HR. That seems slow. It is slow. Very slow.
The legs being wrecked more is true in my experience, but doesn't necessarily mean it's a better training load. It's taken me a while to get my head around that. This comes down to what sirpoc was describing at the start, the hard impact and feeling of the body the real speed workouts bring, but in the bigger picture not really leading to much improvement or possibly stopping you hitting the sessions that do make the difference- note his stagnation of a reasonably decent amount of time and the boom and bust cycle others have mentioned here, before he started training like he cycled.
I'm coming well around to the idea that whilst the 5k might feel harder , the 10*3 mins workout is actually banking more overall benefit AND you remain fresher. I guess here for runners around 18-19 mins or under. I think if we are talking guys slower than that, it's a different story.
Also, because I bug sirpoc all the time on Strava I followed up the quote above about how 'easy' he means. The answer I got was easy means easy and is easy, in COMPARISON to other running systems daily/easy/long run days , or what might get called steady runs etc. So it was much easier than what he had been prescribed by running plans before, I wholeheartedly agree with this. In this sense he seemed to mention 80/20 or Daniels that he said were remarkably hard in comparison and would quite quickly lead to problems if he followed the medium to high end of those "easy" day paces or efforts.
At the end of the day there is obviously a huge amount in this and it clearly works for a vast number of people now, just look at the testimonials not just here but all over the internet. I would still ask why from a scientific point of view and also how remarkable it is the balance in general for most people for the original sirpoc lay out of 3 ST-3 EASY- 1 LONG seems to work.
I don't know exactly but the Strava group, here, Reddit, Facebook group I have seen dedicated to it, you have to be looking at 70-80% of people sticking to it improve over a decent amount of time as a conservative estimate. On top of that, some of the jumps from even seasoned or long time runners are almost remarkable. The bit take I can see is we are moving away from the ridiculous boom and bust cycles and allowing people to balance on the cautious side of where freshness lives which then seems to be allowing people to increase load beyond any other plan has ever allowed them before, whilst maintaining the 3 workouts which seeminglly whilst easier, aggregate in more than the equivalent of a 2Q week.
I will add I am incredibly fast twitch, I saw that debate - but I also have absolutely smashed through walls and boundaries in my second life as a runner. I've run 16:33 which is almost my best from college and I'm 37 with a family of 3 to look after and all the stresses that brings. I've totally ditched absolutely everything but the 3 sub Threshold sessions a week and the easy. I was pretty happy with my turkey trot 18:21 in my 30s but absolutely delighted with my 16:33 a year and a bit on. Shout-out again though to sirpoc himself, always happy to help and answer even though I'm sure he's fed up on Strava and pushed me onto the right path when I had some questions. TL;DR of this bit, whilst I maybe took me a few extra weeks to adapt than others have suggest. This is absolutely for older fast twitch hobby joggers like me as well.
Piocycling wrote:
Running power is proportional to pace though, right? There was a paper comparing power in W/kg to running speed in m/s and it was pretty close. I estimated your % of FTP for the repeats based on that, not on Stryd or w/e device available for measuring running power.
Btw, I have a question about lactate: do you have any experience or knowledge what happens when you go to too high lactate state too regularly? Is it going to be building fatigue of muscles making workouts difficult/impossible? If so can it last for days or is it easy to correct?
Power and pace don't really have much relationship, in my experience. It's good in the flat, with no wind. But the Garmin
wrist power is a mess and in the wind and the stryd doesn't make any sense. I can run in the wind and it'll give me 20-30w more for say a 10 min block because of how it overestimates power into the wind on the track (depite basically a total even effort). Nobody wants power to work probably more than me, but it's just too much of a headache for me to care about at this point.
It doesn't measure force. Until it does, it's just an algorithm really that is guesstimating a lot of things. If you run on a treadmill or in completely controlled conditions, it might work way better. I've never run on a treadmill in my life, so can't comment. I think a big issue is how windy it is where I am right on the sea and flat open spaces. The stryd really is comical with some of the stuff it comes up with.
In the early days I played around a lot with going right up to LT2 (if it exists, there's one to apart some debate lol) and beyond, testing on the meter to see what I could get away with. I spent a good number of weeks and months playing around with all combos before I settled on the amount of lactate I could generate and the pace proxys in place to replicate that (roughly) - and then feel OK to go again in 48 hours. None of this was done on a whim and a prayer, it was carefully planned out. But the big note, I only intended this for myself. I didn't intend, at any point to it blow up like this. Having said that, whilst some people have had to tweak things or start off with the paces slower, it's actually held up very well as a general one size fits all approach (obviously it's impossible to have a blanket approach for everyone).
In a microcosm, if I went deep past sub threshold the main impact was how much harder the next day easy run was and then carried over into the next workout day. Quite quickly, the hole gets dug and in simplistic terms it becomes unrealistic to then workout 3 times a week, for 2 years straight (not saying that is everyone's end goal , but that is pretty much what I have done). The Strava group has had a few people who have gone well beyond probably where they should on workout days , for whatever reason (deliberately or just making a mistake) and it never, ever ends well.
There's obviously a million ways to train. There might be a better one on 7 days a week and only an hour or so to dedicate to training in a day. But if there is, i'm yet to have tried it. I'm absolutely not the guardian of this system, despite what the internet seems to think. If there is another or better way to train and someone could show me that, I'll jump all over for it. After all I basically just stole this idea from my own training in another life as a cyclist, of which I just stole from others. All I've done really is lay it out or I guess communicate what I've done in a way people can interpret or understand. There's far smarter people than me in this thread who pick the bones out of the science of it.
I'm in the same boat as you, above.
Running at my 70% MHR is agonisingly slow, far far slower than what I perceive to be "easy", and also a lot slower than what my predicted easy paces should be. After plugging my recent rate times (5K around 19 flat) into the various calculators linked in this thread it would suggest I should be running my easy days no faster than 5:40/km, but to stay <70% MHR I'm running more like 6:30/km. I can very easily nose-breathe at 5:40/km, hold full conversations at 5:40/km, I don't feel any real fatigue after weeks and weeks of mileage of 5:40/km, but according to a crude HR calculation, thats not my easy pace.
I stuck it out for a few months just to see if my pace at those low HRs did improve over time, but the dial didn't move much. I eventually just clocked that % MHR method probably doesn't give good numbers for me - I have a particularly high MHR (203) and low resting (45), which may just mean I'm not suited for calculating zones in this way. As soon as I shifted to % HRR, things made a lot more sense. I still don't go out and push easy runs way up to the top of the % HRR zone, just in case, but I'm way more flexible with my easy run heart rate than I was at the start. I now go out at more like 6:00/km just to give that buffer, but also cos it allows me to run socially.
I gave 70% MHR a real go, but I don't think it fits every single person. The important part of these runs is that they are easy and you can rack up lots of mileage at that pace without compound fatigue over weeks/months - HR is only one way of estimating that. My experience was using % HRR worked fine effort-wise and (touch wood) I've had no indications that I'm pushing too hard by doing so. But I'm happy to be told I'm wrong if anyone has more experience.
Andrew Coggan 6/14/2025:
Charlesvdw wrote:
It's very clear Matt Fitzgerald doesn't understand what NSM is all about.
- he suggest trying it for four weeks : it's clear you need at least two months
On the latter note:
I don't believe that it takes that long for any physiological adaptations induced by lots of "sweetspot" intervals (which is what "NSW" is) to take place. That's simply not how the body works.
Rather, I believe that it is greater fatigue induced by the increase in training load (which is the point of the whole approach, whether or not you attempt to quantify the latter using, e.g., CTL) that "masks" the increase in fitness, preventing you from fully expressing your performance potential. As adaptations continue to take place/as you accommodate to the increased demands, the fatigue begins to dissipate, leading to what appears to be a sudden breakthrough.
Not sure that this matters from a practical perspective, but to me the distinction (or nuance) seems important.
It's worth taking a moment to think about how NSA is different from popular training programs like Daniels, Pfitzinger and others. There's a lot of things it consciously omits:
-Strides
-Hill repeats/hilly runs
-800/mile pace/Daniels R pace
-5K pace/Daniels I pace (even for runners frequently racing 5K!)
-Tinman CV pace
-Continuous tempo runs
-Midweek long runs
-Long runs over 80 minutes
-Periodization
-Multi-pace/alternation/complex workouts
-"Going to the well"/"Seeing God"-style workouts
-Down weeks/recovery weeks
-Doubles
-Double threshold workouts
-Race modeling/simulation workouts
Even with the special marathon prep block, NSA marathon preparation includes few or no:
-Steady MP runs
-MP segments in long runs
-20+ mile long runs (just 1 long run with time on feet equal to marathon goal time)
That's actually pretty revolutionary. Tons of conventional wisdom about what every 5K/10K/HM/marathon plan must include just gets set aside. Most of the training paces are Daniels' "gray zone/junk mileage" paces! The basic insight is that 99% of the battle is putting in the most work you can however you can, and the precise structure of the work only matters for maybe the last 1%, which just doesn't matter for us hobby joggers. Just compare NSA to a classic article like Pet Magills "Solving the 5K Puzzle" and you'll see how radical NSA is in its simplicity.
If you look at what NSA does include, it's as basic as it gets:
-3 workouts of (up to) 10K volume of intervals at 10K-HM pace (occasionally slower, to MP) per week, maximizing volume to fit recovery
-An 80-minute "long" run
-Easy runs
And that's it. Other plans aren't bad, and I've had success with a lot of approaches, but most of us don't need their complexity. You can complicate NSA by digging into TTS/CTL/lactic acid levels, but it's not necessary. You can just run the same set of workouts, week after week, gradually increasing your target paces, and still improve, without all the science and without all the complications. And succeeding based on monotonous, repetitive simplicity is pretty radical.
0/50 wrote:
Today I did 2 x 3 miles at 149 HR avg (80% max) & second chunk was 9 sec avg pace slower than first chunk so I know I have more work at 80% HR to do at 3 mile blocks before going up to 2 x 30 min blocks.
With NSM, I've often wondered if there's some low hanging fruit that could be squeezed out around the marathon pace area. I'm not a physiologist so perhaps as long as you're below threshold pace doesn't matter. Hadd training has a lot of similarities with NSM. No strides, no hills, nothing below 10k pace training in phase 1. Easy pace & LR pace is the same (though Hadd allowed up to 75% of HR max for EZ though, but encouraged more 70%.) Long runs were entirely easy, no speed mixed in. Whole idea was to squeeze tooth past from the back of the tube by starting with 80% work (he didn’t even feel comfortable calling them tempos). No racing allowed in phase 1 because that'd be above threshold. I've often wondered if NSM squeezes it more in the middle with the paces given in Lactrace which I've felt were maybe a little too fast for fast twitchers like myself especially for those who focus on the top end of the ranges given.
Excellent observation. I already wrote something similar many pages ago, but old timers were doing similar training to this, at least in principle, if not in specific application. Ron Clarke did daily progression runs, probably averaging around marathon pace for around 1 hour everyday. Did no intervals and broke many many world records (but could never peak probably due to not "sharpening"). Derek Clayton did similar training, and there were many others. Michael Musyoki and Norm Green ran around marathon pace continuous runs. And these guys were no joke: Norm Green ran a 2:27 something marathon when he was 52!
Hadd is similar but includes easy days and longer MP-ish pace "workouts". So instead of medium intensity runs with medium duration he polarizes it a bit with easy runs and longer medium intensity runs.
If you look at Lydiard's marathon conditioning he basically alternates 1.5h easy runs with 1h MP runs.
NSM is actually very similar to this but instead of solid MP runs it increases the pace while still staying under LT while introducing breaks (i.e. intervals). So the workload is probably similar to a 45 min MP run.
I just personally find running by feel around marathon pace more refreshing and natural, but I think the effect is similar to NSM. NSM controls the below lactate state more rigorously and includes higher paces albeit with many breaks, but the general idea remains the same.
To be honest this is not that novel, it's just a new version of what many high level runners naturally did in the 60s and 70s before polarized training took over, and medium pace got a bad rep.
... I actually think continuous tempos could be quite wise as a starting point, especially for the more FT and anaerobically gifted among us. There's a serious trap with interval work where you may inadvertently use your anaerobic system. Not only will it be more fatiguing (more damaging? hehe) but it's the wrong stimulus - not getting you as fit for 5k-marathon as the workout ought to be.
So a continuous tempo at a bit slower than MP - maybe 10-15sec/mi slower - could be a more foolproof way to achieve the stimulus that these subT workouts provide. And as you get fitter, this may improve to MP. But assuming we're concerned about aerobically undertrained people who are having trouble getting started on this method, then you'd want the pace to be quite conservative.
Furthermore, as discussed in the Steve Magness video, if you're having trouble recovering from subT workouts, you should probably LOWER the recovery between the intervals, not increase it, and definitely jog the recovery rather than stand or walk. Of course, if you don't slow down the intervals at all, this adjustment would make the workout even harder. But assuming you keep the RPE 5/10 to 6/10, the pace of the intervals slows down accordingly, and you're now ensuring that you're not using your anaerobic system more than you're supposed to. And while this doesn't turn the workout into a continuous tempo, it is more in that direction.
This idea of lowering the recovery could be why the 45/15 workout or 60/30 workout might be a good idea. It's not because the interval pace is faster and FT people are more comfortable at the faster interval pace. It's because the shorter recovery helps keep the workout aerobic. Because if you did make it an anaerobic workout with such short recoveries, it'd surely exceed 6/10 RPE. Naturally hitting your fastest paces on the last reps in a relaxed way, without increasing RPE, just wouldn't happen if you're too anaerobic. And remember Bakken starts the 45/15 at up to 30s/km (48s/mi) slower than the target pace. That's so slow. And then the target pace is hit on the final reps only if it comes naturally, which might mean he developed a feel for whether he's overly relying on his anaerobic system to hit the pace. So it's a workout that's done very cautiously to make sure it's targeting what it's supposed to target.
On a related note, a bad habit I've noticed from my own reps: I almost always start them too fast, like I'm starting them with a stride. I figured if I slow down in time before lactate builds up, it's not a problem. But I now think this is a really bad habit for the workout, using my anaerobic system. I cringe thinking about a time in my life when my spot for doing intervals started uphill... never again.
So, putting this "muscle damage" business aside, I think there is good reason for aerobically undertrained people starting on this method to begin with some continuous tempos rather than all intervals, never take more than 60s recovery, and always jog the recoveries. Maybe 4-6 months later, if things have gone well, then you'd be aerobically qualified to run the workouts as Sirpoc does them.
Alternatively, if we want to be stubborn about no continuous tempos, start the training with only the 3x10min reps at the slowest recommended pace with 60s jog recovery, three times a week. After a few months, mix in some 6-8min reps with 60s jog recovery. After a few more months, 3-4min reps with 60s jog recovery. Then finally consider standing/walking rest.
but why? wrote:
Am I right in thinking it's not just a marathon plan from scratch? As in more of a bolt onto this method.
Also, I'm genuinely curious, why do we think that most people who see success are the ones who have followed it closely? I'm not saying this isn't the case btw, just interested why it seems to fall apart quite quickly with tweaks.
Like most marathon build strategies it's pretty hard training so one needs to have a high level of fitness and a familiarity with the format to be able to properly execute the plan. I don't think there's much more to it than that.
As to why people who try to tweak the plan fail, I'd say it's simply because the adjustments they're making are not well informed. They're just randomly mashing together different plans and workouts with absolutely no consideration of the underlying logic that created those plans to begin with. They aren't properly personalizing the plan to their actual needs, rather just introducing a lot of randomness.
If you have a relatively normal life, are reasonably fit already, and are trying to optimize for time-efficient, sustainable aerobic development it's bound to look pretty close to the default Sirpoc plan. Given the constraints applied that's just how the training math works out. This training pattern was derived with the purpose of constantly riding the razors edge between overtraining and under-training, if one significantly alters the dosages, paces, or weekly structure they are likely to fall off the edge into either side of unproductive training.
If you need something different (i.e more specificity for an event outside of 5k-HM) you need to abide by the same logic that created the plan from the ground up rather than trying to add in stuff top-down.
For example a common mistake is with how people add in intensity -they do some sort of X-factor work that they copied from an external source without really understanding why. The role of X-factor work is just to bridge the gap between what the threshold and easy running is providing and what the goal event demands. A lot of people are so aerobically limited that practically there is no gap here. In cases where this is an actual gap one needs to properly identify what that deficiency might be, and then apply the minimum effective dose that remedies that deficiency (which often will not look like some random book workout or whatever they saw a pro athlete doing).
r0zina wrote:
I have always wondered if NSA definition for interval paces being distance based does actually scale to slower runners. If I remember correctly the 3 min interval pace is defined as 15K pace. But 15k represents a very different effort for someone doing 15 min 5Ks compared to someone doing 30 min 5Ks. So how come the prescribed pace is not defined in terms of equal effort on the body ie time based pace? If the pace was defined as I dunno, your 40 min pace, that would represent the same effort to slow and fast runners. Or am I missing something obvious here?
Your general premise is correct - a runner who runs a HM in 80 minutes will be working harder for their 3 minute reps at their 15km pace than a runner who runs a HM in 110 minutes. However, there's three reasons why it's not a huge deal, imo:
1) The slower runner (on average, obviously they could be a world beater in their age group but it's unlikely) probably should err a bit on the side of caution anyway as they are likely to have less lifetime mileage and be less attuned to their RPE, so it makes sense for them to go slower in both absolute and relative terms compared to the faster runner.
2) Most people don't know what their 90 minute race pace is unless they happen to run a HM in exactly 90 minutes, so giving paces based on distance PBs is just more practical.
3) The difference isn't exactly huge anyway. According to VDOT, the first runner's T pace is 3:45/km and their 15km pace is 3:42/km, the second runner's T pace is 5:03/km and their 15km pace is 5:06/km. So one would be running three seconds faster than T pace, one would be running three seconds slower than T pace. Six seconds isn't nothing, but as I already said in point 1, the slower runner shouldn't be too worried about going a bit too slow as they're getting used to the effort, and the faster runner should be mindful of not going out too fast but given that GPS might not be 100% accurate, and fatigue/temps/etc. also play a role, quibbling over three seconds seems like a waste of time to me. Just use the prescribed pace as a ceiling and get closer to it as you get fitter/faster/more confident in secondary metrics like HR (obviously lactate is ideal but I've never tested mine and I imagine the majority of people using this method now haven't either)