Sweet Spot or Sour Spot?


Actually, the issue is which lot has the largest increase in eFTP.

You would have to think hard about the control variables, as well as training type and hours.

1 Like

There’s a huge thread on Polarized training on this forum, I started reading through it but my eyes crossed, lol.

It’s so hard to keep up with this stuff.

One thing to notice when you look at the literature on Seiler’s Polarized model (where they retroactively looked at elite athletes over time to see what approach the athlete’s utilized…you’ll see that elite cyclists did not use a Polarized approach).

In this study; The training intensity distribution among well-trained and elite endurance athletes

Look at Figure 1, Table D. You’ll notice that the seasonal analysis shows that in retroactive analysis rowers and cyclists utilize a Pyramid approach to their training intensity distribution. So…not polarized. And Seiler noted a polarized intensity distribution in x-country skiers and biatheltes.

My takeaway. Elite endurance athletes don’t all follow a polarized approach. In particular cyclists do not. Elite cyclists. X-Country skiers seem to follow a polarized approach in retrospective analysis.

You and I are decidedly non-elite, and we’re cyclists.

Polarized might be a good experiment for you personally. Hopefully you will report back to us how it went.


From a mental perspective you’re spot on. Finding a training approach that you enjoy and will therefore stick at consistently for the long term is essential.

From a physiological perspective I suspect the optimal training approach for a given event, age and volume doesn’t vary that much by individual. So think the big data analysis is hugely useful to help hone in on that approach. The trick then is how best to adapt it into something that you will enjoy and stick at. E.g. I know I get bored sitting on the trainer more than a couple of times a week, and I get a lot of motivation from group rides, so I try to find group rides that either are a good replacement for the training I need to be doing (e.g. a hilly group ride where I can do the hills at SS or threshold) or can fit in around key sessions (e.g. a z1-2 coffee ride that doesn’t take too much out of me).

1 Like

One would think right? But…

First, it is not important. Cancer research is important. Cardiovascular disease is important. Infectious disease is important. Making weekend bike racers slightly faster by doing well controlled, expensive physiology experiments is NOT important. This is stuff which is of interest to a very very very very small number of people who represent a very very very very small subpopulation of an already small group.

Second, if folks honestly think it is “not too hard” to do these types of studies they probably lack the appropriate background understanding of what needs to be done to get a valid study. And then replicate that study and then actually use the findings in a population.

*** I do think this area is absolutely ready for a crowd sourced experiment. Either using ML/AI to look at a ton of data or direct experimental design. Or more likely, both.

For those that want to read more on a Sunday morning:

Put in context of “how should I train”, the person who has popularized POL cannot even describe to a recreational athlete, or hobbyist, how that hobbyist should implement POL over a long duration nor what to expect from it. That is a fundamental problem with retrospective analysis of a very small group (elites). It may define what they did. It does not define what someone else should do. I love the work. I love the discussion. But it’s an observation of a very biased dataset and not a prescription on how to train best. Note the author doesn’t suggest it is, but others are trying to get there so it is a fair discussion.

There is an axion in strength training that: “The best program is the one you are not on”. It sounds funny but it is mostly accurate. Doing a new program will almost always give a new adaptation. That gain is interpreted by the athlete, and trumpeted by the coach or seller of the program as great success and then exclamation that the program is amazing. Doing the same program over and over and over the athlete will eventually plateau. Then there is sadness and exclamation that the program stinks. Insert Einstein quote here…

Note 1: Proper, all in structured training is hard. It often not possible for an athlete to do the same programs over and over. Sometimes you need to HTFU, sometimes you need to take a breather. That is very individualized and most programs don’t account for personalization.

What you don’t know are two things: When you will plateau on a program and what your actual potential is.

Nobody here knows if their program has extracted 80%, 90% or 99.9% of their potential. Because nobody knows their potential. All they can know is if they are better or worse today than some period before. However, given that they almost always change other factors in life, it is very hard to know how much credit, or blame, to ascribe to any one component of life (or environment or ecosytem for more science oriented folks).

Note 2: A lot of people come to structured training later in life. That is a problem because aging is a huge factor. If you start a program at say age 40 you may be able to do it well with high compliance and see good gains. That same approach at age 50 may not be doable. Even if it is, you may no longer see gains. Age is a huge variable and over long periods of time is very hard to control for. 30 year old me completed programs I can’t even think of doing 20 years on in my 50s. But 50 year old me would like to improve still… hmmm Bummer :frowning:

There are reasons much of science trying to understand biology and physiology is done in model organisms (nematode, fishes, flies, rodents, et al). Setting aside cost, the #1 reason is that many more variables can be controlled. Unfortunately, the model systems very often don’t translate to humans. Hmmmm Bummer :frowning:

So if we don’t know the unknown, what do we know?? We know that for humans to perform well in athletics there are a few good things to do. A few key factors which can guide the Weekend Warrior ™:

Number One is Compliance to a structured program. If you train regularly and include different stimuli you will perform better than if you just go out and do the activity.

Number Two is sleep / rest / recovery and general stress reduction. Most of us are frazzled from life. We aren’t pros riding, eating, resting. We have complicated lives yo!!! But get some sleep.

Number Three is diet. A lot of otherwise healthy people eat like crap. I used to. Sometimes I still do.

Somewhere under those three is whatever structured training program you choose. Some are better than others, some make more sense, some are simply more fun (which assists with compliance). But as long as it has some hard work above threshold, some work at threshold, and some more work to drive endurance you will do fine. Maybe just good, maybe even great. The athlete shouldn’t blindly follow a program as each athlete has different goals. But structure works over the long haul so have some and you’ll do better.


Stop looking for magic programs. If you train regularly and do the other stuff you’ll get better. There is no perfect program and there isn’t enough data to tell you one is better than another except in very broad strokes. We are making this stuff far more complicated than it needs to be. For the average Jack and Jill. Pick a program and stick to it for a while. When you eventually stagnate, or simply want a change, you’ll be better than before. Then pick a new program, change it up and keep going.



Definitely. Likely will ask for help when time comes to create a concrete plan.

1 Like

Ha, I was wondering the same

One thing that is generally missing in the debate about polarized vs sweet spot is a hypothesis based approach to figuring out the answer.

By this I mean, developing some POV on the situations is which polarized is better and why, and when and why SS is better. With a hypothesis based approach you could then start to look through existing data I.e. a natural experiment of sorts, to confirm/refute the elements of the hypothesis. And do targeted studies to fill in gaps.

For example, why do x-country skiers do polarized but not most pro cyclists? Do some pro cyclists do polarized but others do pyramid? Is polarized better for some athletes based on their physiology (eg % slow twitch) while sweet spot better for others? You could develop a hypothesis on each of these questions (and more), and then go test the pieces.

Course no one seems to have the time/interest/incentive/ability to really figure it out.

FWIW, the INSCYD approach is the best that I’ve seen on tailoring training based on the athlete physiology and goals. So maybe that’s the way forward, and training modes are a continuum, with no longer any clear distinction between “polarized” and “sweet spot”?

Until then:


This particular individual who wrote the article seems less interested in a debate and more interested in “getting back” at TR because they hurt his feelings/ego:

I don’t like the way Nate Pearson calls me “this youtube guy” without using my name or qualifications.

Interesting how he attempts to smear TR for their focus on Sweet Spot, yet barely even mentions FasCat who’s business model is also built upon SST.


Pro cyclists end up pyramidal because they do 4-5-7 hour races and a lot of racing is in tempo or sweet spot range. If you look at pro cyclists on Strava you can definitely see that their base season is often 6 hour rides day after day in zone 1.

Cross country skiers do 1-2 hour races. Rowers do very short races.

I do think there is more known than is easily available to the average recreational cyclists. The work of Sebastian Weber was being done before him by Jan Olbrecht and before him by Mader.


Doing INSCYD testing is probably the easiest method though it costs around $800/year (4 tests).

All that said, structured training is trying to ring the last 10% out of the body and surf the peaks for race performance. None of this crap we try to do is going to turn any of us into a World Tour Pro. There is no untapped potential locked inside us that we could release if we only found the right training method.

Take Phil Gaimon. he started bike racing late, joined his college team, maybe won his first race, and then upgraded to cat 1 in a very short amount of time. You often hear this kind of story about pros. Did Gaimon do some super secret training program that unlocked his potential? NO. He was just born with it. He went out and rode a lot and ended up way better than all his friends in short order.

Back to SST - To me SST is partly a hack to make up for a lack of volume. You might do 20 minutes of SST to replace a 2 hour ride outside. SST can also be a race intensity simulation and you can often get this doing group rides.

As you go up the continuum of adding more training hours in the attempt to get faster, you have to add more time in Z1/2 and become more polarized. Most people can’t absorb hours and hours of SST. And there are benefits to the long slow rides.

The fallacy is that people tend to have the “more is better” mindset - like more intervals must make me faster, more SST will make me faster, etc. You get the point where you don’t need more intensity, you need more time in the saddle.


And therein is the problem for so many of us ‘time crunched’ non-pro, spouses, parents, etc.

I’m there now. An hour a day, bashing my brains to the stem of my bike is about all I’ve got time for, sadly. I’d love many long, slow, rides of 5+ hours… but it’s simply not in the cards at the moment.



It’s Sweet Spot BASE. Not Sweet Spot YEAR ROUND.


There is a middle ground. You don’t have to bash your head with 7 rides a week with intensity nor do you have to do 5 hour rides. Seiler has said that he’s seen 6 hour a week amateurs improve their lactate curve in six weeks with polarized training.

1 Like

and what came after the 6 weeks? What when the body has adapted to riding 0.9x6h of LIT? At some point you have to apply more stimulus or you stagnate. And with LIT this comes basically from adding volume.


Another academic paper saying 6 weeks of HIT increased FTP 0.02% more than LIT

(don’t mind me - jaded and sarcastic ex academic researcher - LOL)


This is what seiler(and others) has repeated over and over

We can argue about this until the cows come home but I think we all agree 1/ that if you train 20 hours/week or more like the pros then you have to go polarised as you can’t do threshold/SS for that long 2/ Any well structured programme will make us weekend warriors faster initially. The tricky bit is what happens when we have been riding a while and have picked up all the easy gains…I have about 7-10 hours/week and try to do a group ride for about 3.5 hours and a combination of SS/over under/threshold/VO2 max for 2 turbo sessions plus a couple of aerobic rides. That is Tue/Wed/Thr plus weekend. I do the hard stuff Tue/Sat as I need a day off if I want to smash it. If I had 20 hours a week I would go polarised…but I don’t - I should imagine most of us are in a similar boat.


I did this experiment on myself last year. After about 8 weeks of Seiler style base I was flying, breaking all my PRs, and my FTP tested 20 points higher. Probably the biggest benefit to a polarized base was my fatigue resistance improved substantially. Three hour group rides got a lot easier. They no longer left me in gutter wanting to do nothing more than sit on the couch for the rest of the day.

My FTP over the year went something like this:

polarized base
added a little tempo and SST but stayed mostly polarized
275 end of the year test (Nov)
For end of Nov/Dec I did a lot of easy rides and rode around 6-8 hours per week.
Two weeks off around Xmas and New years.
Started base miles again and a few weeks later I was flying, breaking PRs, getting Strava KOMs that I had no business getting.
WKO5 predicted my FTP at 291. Though I didn’t test again I believe the number from the results. I was also climbing with a better class of rider in my bike club which I had never done before.

My yearly average for 2019 was 7 hours a week. My biggest week was probably 13 hours and there were only a few of those.

I think this idea that you have to ride 15-20 hours per week to successfully ride polarized is false.

BTW, this is not an anti-SST post. Some SST and tempo can be integrated into a largely polarized approach.

Training is about stimulus and response. If you always pound yourself on the trainer and never let up to allow for adaptation you’ll be stuck in the middle and/or constantly tired or burned out.


I’ve been training ~20h/week for the last 4 years or so. Last winter I followed a strict polarized approach from Nov-March. 2019 was my worst racing season ever, clearly lacked the turbo diesel and the high intensity work led me nowhere. I guess we can find success and failure annecdotes with all training regimes.


I’ve experimented with polarized blocks over the last 18 months, and my current conclusion is that in the future I’ll do a 9 week polarized block only after doing a more “traditional” early base with strength training focus (Aug/Sept) followed by sweet spot base (Oct-Jan) to focus on muscular endurance and build out training load (CTL). For pyramidal distribution I’m a fan of at least one longer 3-8 hour aerobic endurance rides on the weekend.

I’ve seen similar although on 5-8 hours/week. That said polarized does make me feel fresher during the week, although same is true if I do more pyramidal 2 days/week of sweet spot and a true 3-6 hour aerobic endurance ride on the weekend.

1 Like

So what does your 20 hour/week look like?