I came across in one of the interviews that Attia did with ISM, the two of them converging on Z2 being somewhere between 70-80% of your max HR, and/or using the talk test to confirm you’re still in Z2. Maybe consider less focus on FTP as a guide.
Anyone have a fundamental disagreement with this?
In my personal experience I start to fail the talk test as I exceed 80%.
But we know the deficits of the million other %’s used for zone definitions. Why do all this testing to end up resorting to a population average that doesn’t apply to any individual.
One of the virtues of ISM is his reliance in constant and consistent testing….never mind the Norwegians in tri.
Talking in a sentence or two, and then grabbing a breath? Or saying the entire alphabet? I’ve seen both and they aren’t the same for me.
Talk test is difficult for me, I can’t use it. However I can sense two changes in breathing, one within 5+ minutes of starting a light warmup, and the second, presumably VT1 and related to LT1, around 78-80% HRmax (actually I reference mainly to power). Everyone is different, hope that helps FWIW.
He does do that. Believe me: BIG TIME!! 
Steve’s is very big on testing for individual riders. And often. Learn your body, etc. For some riders, almost to a fault.
The percentages are just to get you in some sort of ballpark prior to doing all the testing. He understands that not every MAMIL comes to him wanting to go through a battery of tests (usually an irrational and uninformed opinion about lactate). That is part of batwood’s point. Use the assessment he outlined above to get into the ballpark, make adjustments from there.
Because he’s the coach of world tour pros! Of course he can. If he tells them to jump in the lake for recovery they will. All good coaches (not the least of which is Steve Neal) rely heavily on “do some training, test, adjust the training, test” cycle. It’s literally the entire philosophy and business model of WKO. LOL
I get sense the WKO crowd wants to push for the elimination of physiological testing and over reliance in power curve modeling.
The WKO human performance model is pretty good, when fed good data. And it is far more practical than testing lactate and hooking yourself up to a metabolic cart for gas exchange.
Can’t speak for everyone else, but WKO outputs information to help with coaching decisions.
I agree. Having said that, I wouldn’t throw the baby out with the bath water. It’s an extremely useful and flexible tool.
Is it more practical? To properly feed the model you need to give it maximal efforts.
This is key!. Otherwise you get tangled in the duality of power as input and output.
I agree, however we are starting to see diminishing returns and what most people want and need is objective/reliable measures of internal strain, which we will get once the economic incentives are there. Perhaps I’ll be too old to finde them useful 
Cycling still has a huge advantage over many other sports. We are constantly “testing” if we ride around with a power meter all the time. Runners use timing and pacing which requires a lot of going by feel. WKO5 just adds up all the power data and models it out. It’s pretty nifty if you give it some efforts which shouldn’t be a problem for any racer.
If you just want to be a recreational rider and never do max efforts then you probably don’t need a power meter or WKO5.
Has been for me 
You can use WKO and get value even when the model is not fed.
FWIW I don’t keep the model fed the entire year. My coach doesn’t yell at me when I feed it max efforts. Those efforts haven’t derailed training. Here are the 3 tests (short/med/long) that WKO is asking me to do:
Today I’m doing a 10-min pacing effort for an upcoming 20-30 minute field test and I’m targeting 280W-290W, that one effort is going to have the biggest impact on the model. Based on current curve, and my own experience using WKO, the short and medium targets will have virtually no impact on the model.
WKO max efforts has not been an issue for me. And I’m the guy telling everyone on this thread to get in touch with feelings of recovery/fatigue, and use that to balance power of steady state endurance with your interval work.
The problem is that these are not real tests from the methodological point of view, because “you can’t step in the same river twice”
What is your overall point? That nobody should use a power meter and a power duration curve?
That the power data side of the is robust enough, and technology has to make progress on the internal load side of things to keep things moving forward. Looks gimmicky, but the Norwegians are on the right path.
Has somebody done an independent analysis of the predictive ability of WKO model?
If you can’t step in the same river twice, then all testing is flawed? 
Functional field testing with power is here to stay, and it is far better than RPE or HR.
I think that @oldandfast ‘s point is more nuanced than this. I think he’s saying - our power (load) data is really good. But our other data (specifically strain -or how much it costs us to produce that load) isn’t as good right now. Generally - and I think you’ve mentioned this @WindWarrior (let me know if I get this wrong) - some athletes are using a combination of power data (load) and heart rate or RPE data (strain) to gauge how their fitness is changing.
The point is that the power side is much further along than the HR side, so it would be good to see the pendulum swing to focus on improving the quality of strain data (such as HR).
This is my take on what he said as well. We’re not as far along in the strain side of things for sure. HR is problematic, RPE is problematic, HRV is problematic, athlete feedback is super problematic, breath/respiration has potential but the gizmos are prohibitively expensive. We are further along than most (all?) other endurance sports with regard to measurement but the precision of power meters spoils us for more and gives us a false sense that we can treat human organism like a giant computer.
and lactate is problematic. IMHO strain is ‘problematic’ in the sense it is about trends and not precise measures in time. I’d hazard to say in the era of smartphones, some expect instant and definitive answers when the answer to strain is “its fuzzy” and “its about trends.” And you can fumble any strain measure, so lactate testing can become super problematic. Fumble HRV and it becomes super problematic, etc., etc.
For myself, HR is the only one that is consistent enough to compare against power achieved and trend to monitor changes in fitness. And then trends in Garmin’s (post-ride) performance condition, pre/post-ride feelings, and RHR.
For all those reasons my suggestion is to get in touch with your feelings and learn to look at strain trends. And keep in mind that what happens off-the-bike can, for some, be more important than what you are doing on the bike.
FWIW.
Really good point. And I think that’s what makes this hard for everybody - we don’t have a good way to gauge or measure what’s happening off the bike - and yet that’s exactly the most important factor that affects strain measurements.
