Modifying the Norwegian approach to lower mileage

Totally agree. Why not just stick to one protocol and use that? CV is absolutely useless in the spirit of this excellent thread. I mean, by the time you have ran 4-5 races or time trials to determine CV, you've probably got fitter, right? So you will just have to start all over again. Just run a 5k or 10k and I've been using the rough guide someone posted up in this thread summary some pages back and it's just about right for me with lactate testing to check. 15k pace for 1k reps gets me nice to where I want. Basically just run a 5 or 10k race and just calculate it from that. Same with my 3k reps. Just run HM pace or a touch slower. I mean really why to try over complicate yet another arbitratory debatable physiological point? I will search for the Strava group, sounds interesting. Thank you everyone.

Hitting the right intensity/pace is key for the Norwegian approach. You see exactly here in this thread, several questions about the correct pace. So this is an integral part of this thread. The method i provided is relatively simple if you get used to it. With EXCEL it is a peace of cake. User questions can be answered immediately. CV has also a relation to science, which is important to make it universal. HM pace, as an example, is not universal as one does it in 2h the other in 1h. CV is a clear language.

In the thread attached above (i guess on page 5) you find several ways to calculate CV. CV is not a fixed value, i do not define it with 30minutes, and it moves up or down based on fitness level and other things. You can use also a single race, 5k is best, to calculate it.

If you train below CV you have a buffer (distance to CV) and combined with a lacate meter and/or HR or perceived effort you have a good indicator.

CV is related to MLSS, it is a little bit higher (faster) as MLSS, with the original scientific definition. So basically CV is a threshold which can replace MLSS without using any lactate meter and measurement. How cool is that?

Do you genuinely think most people should run 4-5 time trials, in a short time span, to get enough data points and then perform a linear regression just to calculate their CV? If statistics suggest, after reviewing the results of the accrued data from the research literature, that CV is approximately 30 minute race pace, then you can simply run a 5km race and, based on well established methods, approximate your 30 minute race pace.

I like data, math and science as much as anyone else in the field, but I find the idea so impractical. It's almost laughable that it's even proposed. Which to be fair, I know it likely has it's roots in cycling so they can recover from an all out effort slightly better than a runner I imagine due to negligible impact forces. Still, it leaves me feeling like physiologists missed the forest for the trees (the 4-5 of them specifically haha) on this application.

I read that thread when you started it as I was genuinely curious about the topic, I even used 4 race times from 800m-5000m and plotted out the hyperbolic function for fun, then took it a step further and compared the prediction to what I would have got if I'd used just a 1600m or 5000m to estimate my CV. I've got those saved somewhere haha.

My point though is that if you use the hyperbolic function and interpret it strictly mathematically, you can theoretically maintain CV well beyond 30 minutes, even beyond 60 minutes. In practice, that's not even close. It does generally fall around 30 minutes +/-. That's another issue, you can not only get an estimate of CV but also get an estimate of races that exceed 30 minutes (which since most of us are distance runners would be helpful) by using a function with two exponentialy decaying terms where the asymptote is likely closer to your aerobic threshold.

Sorry data man, thats nonsense. If you can hold your CV for 60minutes, you did something wrong :) The pace you can hold for 30 minutes has to be higher than the pace you can hold for 60 minutes. Otherwise people would run a HM with the same pace as 5k. -> ridiculous and not true.

Something is wrong in your data assessment. I have not the feeling you are into this CV topic to be honest.

Which in any case is not this thread's topic.

Sorry, to clarify, I’m not suggesting my result indicated a pace I could maintain for 60 minutes or longer (it indicated 29 minutes to be exact). I’m saying that if you treat the model purely mathematically, it does suggest you can run at the CV pace obtained for 60 minutes or longer, even though in practice it is observed and recognized that you can’t.

My question, acknowledging that, is why not just define it as 30 minute race pace? And I don’t mean you specifically but the idea in general. Save the 4-5 time trials and hyperbolic function for sprint and middle distance athletes and just use a single race effort to approximate CV. Also, from that race effort, you can obtain classic race distance predictions via standard methods. The hyperbolic function can’t provide that (using the 2-20 minute race effort protocol) because it asymptotes at approximately 30 minute race pace, in practice.

On that last topic, I read a paper that Andy Jones published where he used 1500m - 15km times of 12 elite athletes to approximate CV and compare it to marathon race pace. That’s obviously slightly different than the 2-20 minute race time protocol but the results he obtained were still interesting. Of course, that interested me (as you say, I’m data man), so I pulled the data for the top 200 times run in the 1500m, 3000m, 5000m, 10000m, half, and marathon. Using just the 1500m - 10000m I fit the model to those 200 performances in each category, so it is an approximate average. The CV for that dataset is 360.26 meters per minute, which if you multiple by the average time for the 10km of the group works out to 9,748.64 meters, so slightly slower than 10km pace for a group that averaged 27:04 for 10km (369.55 m/min).

To illustrate my point about the mathematical implications of the model though, for that group, the hyperbolic function is

Time = 274.99 / (Speed - 360.26)

Say you run 3 meters per minute slower than CV, so 363.26 meters per minute. The result would suggest you could run for 91.66 minutes at that pace. Using the half marathon from that dataset shows an average time of 59:16, which is 356.01 meters per minute. So obviously slower than the speed predicted, meaning it wouldn’t be possible to maintain that speed for 60 minutes, in practice, let alone 91.66 minutes.

This is my point about the undesirable asymptotic behavior of the model. I hope that clarifies it. I was tempted to use more graphs and analysis but refrained… for now haha. Also, I hope this shows that I’m not being disingenuous in saying I’m interested in this concept. There are just some aspects of it that seem peculiar when going from theory to practice.

Now i see it: The hyperbolic two-parameter power–time (P–t) relation defines exercise tolerance within the severe exercise intensity domain. So using this approach below CV is not correct and therefore not accurate.

30minutes is arbitrary and is only on the edge related to the definition of CV.

For male elite runners, CV is basically 10k pace, so it is closer to 25minutes. However, for the hobby jogger who does the 10k in 50minutes, CV is faster as 10k pace.

Correct. I understand that, in physiology, it delineates the severe intensity domain from the high intensity domain. The speed I used in my example, 363.26 m/min, is faster than CV by 3 m/min. So I don't understand why you say I used this approach below CV?

Because you wrote that.

Maybe I should just give up on this or start a different thread haha. That’s not what I’m saying in that passage at all. Maybe I’m not being clear or maybe I’m really missing something obvious, but let me try one last time here.

Take the equation I provided which uses the data from IAAF for race times at 1500m, 3000m, 5000m, and 10000m. This produces the following hyperbolic function with 274.99 being the estimated D’ parameter and 360.26 being the estimate CV parameter:

Time = 274.99 / (Speed - 360.26)

For reference, 369.55 m/min for 27.06 min, and 356.01 m/min for 59.27 min are the speed and time for the 10km and half marathon, respectively.

As your speed approaches the critical velocity speed, using the above formula, this happens:

Speed = 370.26 m/min. Time = 27.50 min
Speed = 369.26 m/min. Time = 30.55 min
Speed = 368.26 m/min. Time = 34.37 min
Speed = 367.26 m/min. Time = 39.28 min
Speed = 366.26 m/min. Time = 45.83 min
Speed = 365.26 m/min. Time = 55.00 min
Speed = 364.26 m/min. Time = 68.75 min
Speed = 363.26 m/min. Time = 91.66 min

This is what I’m talking about when I say undesirable asymptotic behavior. As the speed approaches the critical velocity, the estimated time exceeds what is observed as sustainable in practice, but you are still technically running faster than your critical velocity and not violating the assumptions of the model, correct?

Ok, i understand you now. You identified the mathematical problem of a 2 parameter approach for a hyperbolic function. If you enter 360.26 as speed in your formula you get time=274.99/0 which is not defined mathematically.

The Norwegian model does not need D' or W', we can use the Norwegian model with just CV only, because we operate close below CV. I see CV as a marker which can replace MLSS, without doing any lactate measurement to identify MLSS, and a marker which you should not cross over it (running faster than that) for the purpose of this thread. The calculation of CV has its accuracy, so doing the reps below CV with a small distance to CV is recommended.

Why do you insist on replacing something successful (i.e. lactate in Norwegian system)?

It clearly works. Let it be.

I haven't been keeping up with this thread since around p. 30. I thought I'd share my experience with a version of this training method, though.

I started on June 27th after 3 years of injury issues. 2022 was the worst running year of my life and I made a commitment to run at least a mile a day this year regardless of health issues. All that to say that I jumped in with very little base.

I live in a very hilly area and refuse to drive the 15 miles to the nearest track, so I run hilly intervals.

I have no data on threshold levels, Max HR etc, so I ran a couple of hard solo 5kms (17:30/17:20) two weeks apart and based my interval paces/HR zones off of them (I know, very unscientific, but I'm 40 and had nothing to lose if this didn't work). I run more by feel and check data post workout to make sure perception is close to reality. I have averaged about 42km/ week these past two months + a 30-50km cycle biweekly.

The last two weeks of July I didn't run intervals due to time constraints.

I just ran a rainy track 10km in 34:45!

This training works for me.

The difficult part for me is holding back at the beginning of intervals and slowing down between reps.

If you're stuck in your training, not recovering between workouts, unable to stay healthy doing fast intervals, ... I would recommend trying this type of training.

I'm going to go for low 34 with super shoes in a few weeks.

*sub 32 10km pb is 16 years old

If someone reading this thread and would like to here test free live coaching on low mileage and maxed thresholds ( around half marathon pace) without measuring blood lactate you are welcome to do the test , We can also have the coaching on Strava for clear verification. You will be positive surprised how fast you will see significant improvement.

Just to clarify,it`s me the useless Swedish coach 👋🧙‍♂️🇸🇪

Just then let someone runner test it. 😉Are you afraid someone jumps on and make a fast great inprovement like all other runners I've coached?

I am too stupid hehe . I lie about all my coached runners improvement but all my verifiable runner got slower 😉 I am also too stupid,I don`t seem to understand I am not welcome here 🍆🧙‍♂️🇸🇪