Polarized Training vs. Sweet Spot (Dylan Johnson video)

Looks like the one! Thanks!

This puts me in mind of a post from OBNYD somewhere above which caught my eye - a deceptively simple statement to the effect that perhaps polarised training should be viewed more as an outcome than a prescription per se.

1 Like

Because that’s how the term ”polarized” was born, i.e. Seiler looking at the e.g. african marathon runners and noticing that out of all the km they ran, 80% was easy and 20% hard. :smiley:

(Or maybe they were 40-50 years ahead of time…)

Hmm, if Seiler would look at Kenya runners there would be a lot of tempo in polarized training:


Both good points.

What’s interesting is so many of the threads on this forum and similar forums are looking at TID of the elites to try and establish the best way for age-groupers to train. I rarely see many discussions looking at the TID of the successful age-groupers training the same kind of volume that we all are. Surely that would be more applicable to us all. I’m guessing we’d be talking about POL a lot less.
(To be clear - I’m not talking about age-groupers like Dan Plews who are basically training pro type hours).

Anecdotally, the best couple of runners I know aren’t doing POL. They mainly train at tempo type pace and do regular track sessions. They don’t own heart rate monitors and have never been near a lactate test. What they have in common - they’ve both trained consistently for a number of years and reliously do track sessions every week. Could they have been much better if they’d established their LT1 and done the majority of their training time under that intensity? I’m not sure, but I’m guessing not.

Recreational triathletes:

and then we have the two studies on time crunched runners (referenced previously in this thread … actually three but I discount the one on 3hrs/week).

There is some discussion/data on non-elites …

1 Like

Maybe. But that is not from where he ”came up” with the polarized training method/term.

Here is the “source”, at around 8.40:

Yup, Seiler started from the assumption that how elite Norwegian XC skiers train represents “best practices”. In doing so, he overlooked many things, including genetics, time available, historical biases, sport specific demands, cause-and-effect, etc. Then when others surveyed elite athletes, it turned out that most in fact do NOT train in a polarized fashion. Really, it’s all just a big house of cards, based on a questionable starting premise. It’s therefore really sad that the issue of TID has gotten so much focus, but people want simple answers and there are always some willing to provide them.


To your second sentence I would also add ”nature of the sport in question”, i.e. what is required to win. IMO a polarized approach suits quite well for xc sking (you need a really high vo2max) and marathon running (high aerobic threshold). But now we are slipping a bit OT, so no need to reply or further comment on this :stuck_out_tongue_winking_eye:

I didn’t say there weren’t any studies based on random recreational athletes. I was referring to looking at the guys finishing on the podium of age-group races… what does their TID look like?

Catching up with this thread, your approach is exactly how I interpreted Seiler on the various podcasts. Incorporate a “hard” day as often as you can with priority given to recovery, make the “hard” days very “hard,” and otherwise ride Seiler Z1 with a loooong ride when you can manage one (hopefully at least once/week). Simple stuff.

And to muddy the waters further, that 2mmol measure that he references (which may or may not be an individual’s the aerobic threshold…never mind the whole scientific debate about whether “thresholds” exist and what to do with them, from a training standpoint) can be up there in terms of intensity, for elite athletes especially.

Rowing calls UT1 “low intensity,” and Seiler categorizes it as such in looking at rowers. UT1 is, uh, what cyclists call power zone 3.

Kenyan runners? Some of them are cranking some pretty fast “zone 1” in Seiler’s terms.

Over on Wattage, Coggan noted a few years ago that he sees Seiler’s 2mmol at 90% of FTP in most of the well-trained USCF category punters that he’s tested over the years.

So much of this “polarized” vs. “sweet spot” is spin to sell programs.

1 Like

To your point about different sports, polarized vs sweet spot is a very different debate for other endurance sports besides cycling. Running at moderate and above volumes almost HAS to be polarized by itself- spending significant time running at SS will almost certainly lead to injuries because of the high impact nature of the sport, even if the cardiovascular system can sustain it. So if you can’t spend much time near threshold and above, you’re only option to build aerobic endurance performing that sport is to add in a lot of easier, less demanding training. On the flipside, swimmers do a ton of HIIT training and spend a LOT of time near or above threshold, even for aerobic events (which I would consider a 400 or longer). Swimming is the ultimate low-impact sport, and the only risk if from repetitive use injuries in the shoulder (which come from total volume and form moreso than intensity). It’s critical to have a high VO2 max and maintain good form for swimming, which is why their training sets are split up into short intervals with static rest.

Cycling doesn’t really have any limiters in terms of injury risk and is less taxing on the body then XC skiing or swimming, so you can do either approach with success. Measuring outcomes is also so diverse in cycling that it’s hard to quantify which produced better outcomes- meaning that even if sweet spot hypothetically produces larger FTP gains, that won’t be indicative of performance for a short track MTB or crit racers. On the flipside, I don’t think a long-course triathlete or gran fondo rider would care about peak 5 min power or anaerobic repeatability.


All pretty much common sense, eh? If only Seiler had thought things thru this way, perhaps the world wouldn’t be in the mess it it.

Seems many are moving to a similar position.

I don’t think the time in zones needs to be so extreme, meaning VT1 and V02 etc.

I’ve simply taken the polarized sessions advice, that seems to be quite effective, for me at least. Do a majority of your training at a lower enough intensity that you can recover and also execute your harder days.

This lower intensity could be endurance, tempo, hell, for some maybe it is sweet spot.

The art seems to be working out what is a fairly optimum endurance intensity, then working out what duration you can deal with. The blunt instrument version of this, I guess is TSS. Hypothetically, you’re trying to do the absolute maximum endurance/tempo TSS daily, that you can just recover from.

You’re then sprinkling in the high intensity days you need. That depends on another million factors. Obviously, younger athletes can tolerate more high intensity. Maybe, an 18y junior can adapt well on 3 hard days a week. Equally, maybe this crushes a 60y veteran in weeks.

The whole game is working out YOUR optimum ratio. What’s certain, is it’s not the same across the board. Not by a long shot. So by very definition, a non individualized training prescription is far from optimum.

This is the interesting part for me. Trying to work out my own unique training distribution.

So far, I’m barely scratching the surface. I have made a ton of observations, on hundreds of amateur riders. From that I noticed, most, if not all, are leaving a lot on the table. Myself included.

I’d be very interested in seeing others training breakdowns on what is working for them.

Mine seems to be this currently. It could change an hour from now. It’s what I do as ‘base’, basically all the time until quite near a race. I’ll then add in race specific intervals for no more than 3 weeks. Normally, just two weeks.

Monday - Full rest
Tuesday - 3hrs Z2, maybe a bit of tempo, sometimes a hard day
Wednesday - Same Z2 3hr, maybe a single sprint
Thursday - 3hrs Z2, maybe a very short 10min tempo/threshold effort
Friday - Same, but possibly on the trainer
Sat - 4hr hard group ride, all zones, max effort on a chosen duration
Sunday - 3/4hr easy Z1/Z2

16-20hrs a week
800-1000 TSS
One hard day, sometimes 2.
46y Male
4.2 w/kg ish FTP (if there’s even such a thing)
Sprinter phenotype
6 year training history
690hrs riding in 2020
MTB and Road Racing

This load seems to be providing me with semi continued progression. Obviously, that will end soon. At that point, I’ll change it.


I read half of this post, but I am still confused on how a polarized approach is fit into the base phase? Surely doing VO2/Anaerobic intervals 2 or 3 times a week in the base phase (3-4+ months away from race) is too much intensity. Do you just skip the Z3 (Seiler) and stick to only Z1(Seiler) with some “stuff” (tempo/SS)? If so, that is not polarized. Wondering for a non-time crunched cyclist.

1 Like

for inspiration (from Solli et al 2019):


Thanks for sharing. To reinforce that point, here is an interesting comparison between my first full season (322 hours) on a road bike and last full season (310 hours), using a HR view:

Same Friel zones as used by Intervals.icu and simplified into 3 zones:

  • low: z1+z2
  • medium: z3+z4 (well, z4 is actually 150-158bpm so not quite all z4)
  • high: z5+ (and some z4)

and the detailed view, same as Intervals but turned 90 degrees:

Like many others, your HR for z1+z2 power is a LOT lower than mine.

Well looking at 16-17 season thru that lens its no wonder my season was interrupted by so many unplanned recovery days!


One confounding factor for me is that I am capable of pushing to much higher HRs in outside rides than I am inside. Particularly, in races when I am fully motivated I see heart rates I am incapable of touching elsewhere. This may confuse some of the max HR algorithms.

That said - I’ve been doing a significant amount of time at 55% of FTP in recent weeks and my HR there is remarkably consistent - I can chill at 200 watts and 104 BPM for hours and hours apparently.

Here is some of the data from last year (no racing)

All that base resulted in some pretty dramatic increases in my power:HR relationship compared to prior years, but you can see that things got a little wobbly at the higher end since I didn’t collect a ton of data there and didn’t spend a lot of time working on that


Honestly, I’m amazed at how hard so many people are working in a ton of these posts. So much intensity. It’s amazing they aren’t burning out faster and more often


I definitely have a lower HR on inside z2 workouts than on outside z2 workouts.

What has been most surprising is how the shift to doing more z2 work has improved my 2-20 minute power bests. Here is a custom chart I made in WKO, showing all my power bests since starting structured training with TR in December 2017:

In Fall 2019 I did TR traditional base (MV), experimented briefly with polarized Nov/Dec 2019, and then Feb 2020 got frustrated with SSB and switched to FasCat SSB (more z2, less intervals). Then hired one of their coaches in August. I haven’t done any real interval work (sweet spot / threshold) since June - my off-season is June-August. The training distribution shows pyramidal and not polarized, averaging something around 8 hours/week.