Polarized Training vs. Sweet Spot (Dylan Johnson video)

Hi, I’m wondering if other people here have watched Dylan Johnson’s recent video comparing polarized training to sweet spot in which he presents a lot of data suggesting that polarized training is better.

Do you think there’s something he’s missing or is this a compelling argument that we should be doing polarized training instead?


Polarized (POL) has been discussed at great length in a number of threads (see search results below):


If you watch the video, the key thing to expand in this thread is the benefits of a Pyramidal approach. SST vs POL is a false dichotomy. Cheers!


Wouldn’t doing SSB result in a pyramidal TID?

IDK, I don’t follow that plan…you tell me. H

For me it would if I didn’t have so much z1 and 2 volume. It bumps me into the Threshold training bucket, but I think you might be right.

I’m sorry, but this study https://journals.humankinetics.com/view/journals/ijspp/9/2/article-p332.xml is some kind of crap, in my mind. We’re gonna take guys and see how they do in competition based on their training? As opposed to measuring physiological changes after training. I don’t have access to anything but the abstract, but on the surface this doesn’t seem like a particularly robust study


The one thing that confused me was he said SS was Z2. My understanding was it was something between Z3 and Z4, mostly 3. Therefore he said Z2 was too hard, but I equate Z2 to recovery. Is my definition of Z2 easy enough or is he saying you need to go below 55% of FTP on my easy days?

They’re using a 3 zone system, so zone 2 in this instance would encompass tempo-threshold


If you have lots of Z1 and Z2, wouldn’t you end up pyramidal, not threshold? Or are you referring to Seiler’s Z2, and you end up doing more of it than Z1?

I don’t follow it either, I’m just assuming that Z1 makes up the most of any program, so how Z2 vs. Z3 you do dictates whether you end up pyramidal or polarized.

1 Like

Isn’t threshold in Seiler’s Z3?

For Seiler, threshold is the limit between z2 and z3


If you look at the difference between Threshold and Pyramidal, Threshold has a larger percentage of Z1+2. My extra Pettits and Whorls drive that Z1+2 up.

1 Like

I think this assumption is wrong, specially when it comes to TR SSB…If I had to bet, is probably 35-40% z1, and the rest z2 and above. Assuming z1 is 75% ftp and below.

Hopefully someone with actual data can contribute…

1 Like

there’s something wrong with that classification…you are basically 75/25, but it’s called “Threshold”…don’t get it. H


I don’t know where that graphic comes from, but what is shown as “threshold” training is actually pyramidal. The orange bar would have to be the longest of all for it to be threshold.

1 Like

It very well could be that my assumption is incorrect, especially for folks who spend most of their time indoors.

It comes from intervals.icu

I guess if you watch the video that explains their zones, they say Threshold can be pyramidal in nature, so perhaps it’s a subset.

He snagged it from his Intervals.icu site. They have analysis that classes each week above. The problem is that the large one is the “selected” training style for the selected week and it can be misleading.

See the pic from one of my weeks, that is about Base, and see how the Threshold preview looks very different from the one Russel shared. It exposes a potential issue leading to confusion, because the preview is altered to show the actual distribution of the given training week. As such, that preview may be a less than perfect representation than the “ideal” that is shown in the non-selected types.

My “Base” week, with the typical display for the other modalities:

Russell’s “Threshold”, that is misleading IMO:

Note how much larger the orange Z3+4 is in the default “Threshold” than the one shown in Russell’s. So I would argue that the “fit” applied to Russell’s that shoehorns his week into “Threshold” is a bit misleading. Something for @davidtinker to review, maybe?

A suggestion would be to show the default for all the modalities, along with a separate one for the week in review, so you can see if there are any real differences between your week and the “ideal” distribution for any given mode, especially the one that is “associated” with it via the black magic behind the scenes.

Edited to add suggested layout (specifically grabbed Russell’s Threshold because it shows the issue I see pretty well):

I’d think “Base” is more appropriate in this instance, or “Pyramidal” maybe, but “Threshold” seems just plain wrong compare to the other two options, IMO.