What's polarised: power or HR?

What’s more important for polarised training, heart rate or power?

E.g. these workout examples could be eather zone 2 or zone 3 sessions depending on the definition:

A) I could do ronnestad style (short on-off) intervals that are 100% zone 1 and zone 3 power, but haveme sitting 100% at a zone 2 heart rate. Here MPA would stay relatively high.

B) I could do XSSR type intervals that could give lots of time at zone 3 heart rate, but mostly zone 2 power. Here MPA stays relatively low, and difficulty would be high.

I would expect maybe heart rate zone is more important than power zone?

Isn’t it a case of it depends on the focus of the workout? Easy workouts (as in Seiler zone 1) should be done with HR and then the hard workouts (Seiler zone 3) by power, but with a HR cap of 90-92% HR Max

OK I’ve found this on google where seiler talks about zone analysis on his polar training:

Stephen Seiler (@StephenSeiler) Tweeted:
@HeinerMe Well, they are of course connected, but HR (and BLa) is my calibration of «stress» and power adjustments x time fine tune the stress, so that is the «directionality» in training, for aerobic zones. For anaerobic capacity training, HR goes out the window and Power is king. https://twitter.com/StephenSeiler/status/968925381402222592?s=20

That suggests to me that HR is the intended parameter to monitor in the 3 zone polarised training model.

I can train at two different zones depending on whether I use HR or power. I tend to focus on power for Zone 2 and 3 but Hr for zone 1. Which aligns with what Seiler is saying I think

I think you are right, at least initially, as in one Podcast Seiler mentions he needed a common thing to monitor (or something like that) and all he had was HR between cycling, running, XC skiing and swimming as not all sports used power (at that time).

1 Like

I would say that you’re not doing the work intervals hard enough if your heart rate isn’t getting out of zone 2, but I get the point you’re making. In this case it would be more appropriate to look at the average power for the set. There’s a good chance it’ll be below threshold.

If you’re doing XSSR intervals at greater than 100 XSSR then these are in your power zone 3 as they tend towards threshold but never get there.

One thing that Seiler says time and time again is that the power and heart rate zones need to be calibrated against each other so that you know that, ignoring drift, when you’re in zone 2 in one you are in the same zone in the other. Don’t mix and match systems.


yes I see that quote, it is a couple of years old (like 2018) and as with all hings Seiler (or anyone in research) may have shifted his thinking slightly. I think if I were to interpret the exchange in that quote Seiler is says that maybe HR is non linear and in zone 1 and 2 is usable but in zone three where you are approaching anaerobic that HR is out the window and power is the metric. A more recent tweet was of 3 buckets:

1 Like

My confusion is what happens during times of transient power and HR, and how that gets allocated to theoretical stress buckets. Sure, power and HR zones correlate well when looking at efforts that reach a steady-state heart rate. But thats not true when power changes and HR is rising or falling as a (delayed) response.

E.g. from sitting on the sofa, to jumping on my bike and pedalling, i can hit zone 3 power (say 110% TP?) in a few seconds. Now while holding that power, it takes my cardio system a few minutes, starting from resting HR, to finally reach a steady state value that is what i would classify as my true zone 3 HR.

Now, from a stress/stimulus point of view, my legs and muscles feel every second spent at that zone 3 power (and maybe more for no warmup!). However, my cardio system doesnt feel the same level of stress (and i suspect training stimulus?) until the HR is up in the (steady-state) Zone 3 HR zone.

So under transient power - and with the intent of training your aerobic system - it surely has to be HR that reflects stress and training stimulus the best. And therefore HR is your best guide if you are trying to minimise zone 2, and maximise, zone 1 + 3 time (for transient workloads)?

I have been thinking about the Seiler 3 bucket idea and it seems to me there are an infinite number of ways o get the 3 buckets allocated and that not all are equivalent. This is both on a micro as well as macro scale. What I mean by that is if you are looking at a ride and even within a ride that is what I term micro scale. Within a ride you can have many short high intensity periods of varying duration such as the stochastic situation of an outdoor ride. This may end up with the 80-20 (where 20 = x in mid and y in red) but how much goes into 2 and 3 is infinitely variable, as well the duration at each interval of time in either 2 or 3 is variable so this does not describe a clear mode of allocation. Then there is the macro scale, ie weekly, monthly yearly. Seiler’s initial hypothesis came from looking at the annual allocation of efforts and found the relationship of 80-20. So again within any of these time periods one has infinite ways to allocate. ie days where the whole effort is 1 or days that are short but 50:50 2 and 3 with very little zone 1… and all sorts of things in between. The training stress will not be the same nor the adaptation for all options.

So in my feeble minded way of looking at this polarized approach I have to think in terms of 2 areas, the micro environment, ie the workout, and that means structured and the macro that basically is the week and again structured. Ultimately you want to achieve the allocations in 80-20 but in a way that allows you to progress. Seiler also indicates that Heart Rate is more useful as a target metric in the zone 1 and somewhat in zone 2 but not in zone 3. In each one has to find out what that that is. That is where I get a bit lost. I have asked on other Forums what people think and the opinions seem to be as many as stars in the sky. This I think is where a good discussion still needs to be held on how to find these zones on an individual level. And in answer to the topic in the title, it is both HR and Power.

1 Like

I think this is straight on point!

I was hoping my question was a simple one, and that I was an idiot and missed some key article or discussion somewhere which everyone already agreed on…

But i think you’re right; the more I read, the more opinions seem to vary and contradict one another, and that will take much time for new research to bring clarity!

on this we agree. i asked on another forum when someone said zone two work, which zone system, the answer was amusing, and they listed at least 5 systems… so without some reference point then some way to understand what is meant by the concept then we need to dig a bit deeper and so yes the research is not to that point yet. I am really curious what others think and in particular what @ManofSteele and @xertedbrain have to say on this, they have been noticeably absent from this thread.

1 Like

While I do think Seiler’s work has been very enlightening, both in terms of polarization as well as the work on 4x4, 8, 16, it does oversimplify the training process and becomes harder to apply unless all you do is do workouts that fit into the buckets cleanly.

We’ve take a more broader approach and say that polarization works because some systems need recovery while other may continue to be trained. There is no concept of this on Seiler’s or other methods out there to my awareness.

We also don’t say that workouts need to fit into buckets. We just say that every workout or activity applies strain into three systems and that individual strain improves the fitness of that system and applies a recovery demand on that system. Low intensity system can handle more load and recovers quickly. High/peak is the opposite. Hence we have “Tired” status which means your high/peak need rest and you can do Endurance work and “Very tired” meaning your low system needs rest too, i.e. total rest or active recovery.

This generally follows an 80/20 pattern but the pattern isn’t the goal in Xert. Everyone is different and at different points in their training and different amounts of time available to train. The key is adhering to the letting the high/peak recover when yellow but continue to build the low system and stay off the bike / go very easy when red. When blue/green status, fire away with hard efforts. We don’t say 4x8’s. We say add high and peak strain mixed in a way that is similar to what you will be racing with (or build towards these ratios over time).

Hope this makes things clearer.

The xert strain system makes alot of intuitive sense, based on feels - but the only bit I struggle to grasp is this bit

But Xert Low energy system includes the sweet spot powers (sieler zone 2?), and some research suggests that power range isnt good for recovering your high energy systems?

However, if xert had 2(or more) different low energy recovery rates split at LTP, then I could easily see how it agrees with the anti-sweet spot research.

There is some merit to this argument and we are working on ways to potentially improve that allocation of strain across the Zone 2 range. (Potentially because we haven’t yet confirmed that this should be reallocated).

Within the current model there is some inherent bias toward LTP work because you don’t accumulate the same difficulty level. At LTP, you roughly max out at about 2.5 diamonds. Hence, XATA will look for 2.5 diamond workouts when your status is Tired. This works to deprioritize Zone 2 since they will increase difficulty.