Good to know you are on a Peloton, are you using the Peloton for you power source or another power measuring tool such as power pedals?
IMO if you are doing your workouts in the Peloton universe then I would stick with their 20 min test. Then you would have consistent source and would be able monitor how you progress.
Yes, by all means mine all the threads, FTP debates generally don’t stay healthy, but that is the internet…
Hot take… People should just train based off their power curve. After all, the best predictor of performance is performance. 
Did you do the 5 minute blowout first? Just curious since so many skip this.
and RPE, presuming you’re fresh enough for the type of workout. “Alls you can do is alls you can do.”
I do the KM test these days a few times a year.
My problem with the 20 minute test was that it’s brutal - the pacing is brutal. Either you go at a constant 5% above threshold for the whole 20 minutes or you start slow and then do a 5 minute vo2max interval at the end. Or, you really screw it up and go out too hard.
The best thing about the KM test is that you end up learning how to feel your FTP by RPE. It slowly ramps up from sweetspot. When you cross over the line to unsustainable you should feel it. You can even go up and down during the test looking for the a confirmation of the feeling.
Once you get in tune with the KM test you can figure out your FTP in 20-30 minutes. You also have to be honest with yourself - like “I can do this pace for another 20 minutes if I wanted to”. No maybes.
Yes. Though it was a little less power followed by a little less rest between than prescribed due to the parcours
Chased By a Bear FTP test - you may be on to something. My guess is this turns out to be the most accurate predictor of FTP.
*Most accurate predictor of hour power 
Very accurate!
My highest 20 minute test was 1000 watts average and my hour record is exactly 950.
I’ve only done a 20 min test twice, both times on the trainer. On the latter I averaged 404W for 20min. I haven’t done a lot of real threshold efforts in recent months (more sweetspot or then significantly over threshold) but recently I did 20+25min with around 10min rest at 379W. So if I take that 379W as my threshold, the 20min effort tracks very accurately with that given that it’s 25W over my threshold for 20 minutes or precisely 30kJ work over threshold, and my W’ battery is 30-33kJ in 2-5min max efforts.
I’m going to add a small point to this - you need to “feed” your power curve relevant results and workouts for this approach to work.
As of right now I haven’t done any real 100% threshold work for 3.5 months, but that doesn’t mean my threshold has dropped - in fact I’ve put on small gains since then. (I shift down the power curve throughout my winter, spring and summer taking a block based approach VO2 → Threshold O/U → Threshold Steady State → Sweet Spot → Tempo and Fartlek work)
By the time I get to my most fit, my power curve isn’t a great predictor of FTP anymore like it was when I was doing a lot of threshold work.
My test protocol - variation of Kolie Moore, or more often just go by feel on 2x20, 3x20 100% workouts. If you’ve done enough, you can dial your FTP in by RPE and Heart Rate just doing relevant workouts
At bit out of left field, but personally, the test protocol I like the best, by far, is the Systm Full Frontal 20 minute test. Even better for me is to do their 1 week test, which includes the Half Monte FTP best followed a few days later by the FF.
FTP is a metric to set training zone. Only the cycling gods know your mystical FTP. Your mystical FTP is what you could do if everything was perfect. You slept well and were not suffering from training fatigue. You were perfectly fueled. You paced your effort perfectly. You were highly motivated.
All the FTP tests are approximations of that number. All the tests have flaws and biases. The longer tests require you to “know the answer” for pacing before you start the test. Ramp tests may be biased as they can be influenced interference from anaerobic systems.
What FTP has become, as the only widely available objective metric of cycling, is a ranking system of cyclist. That’s why people get their knickers in a twist if a Ramp test or AIFTP has their riding buddy’s FTP higher than their own. FTP has become the metric people can compare. TR even has a banner on this page “Want to Raise Your FTP?”
The workouts for my training plan have given me progressively harder workouts. Just a few watts higher than the last six week average. So if my AIFTP or tested FTP were higher or lower than my mystical FTP, it doesn’t matter.
Remember it’s just to set your training zones. Only the cycling gods know the real number.
One could argue using a single number to set all of your zones falls flat, especially at intensities >FTP.
This is the argument Wahoo used for their 4DP test measuring 20 min, 5 min, 1 min and 5 sec peak power.
You dare argue with a GOD!?!
All the caveats aside, I’d argue the “testing as training” aspect of TR is the big value. Progression through your workouts should imply that you’re getting stronger in those systems and “FTP” becomes secondary.
If I’m on a threshold workout at level 3, and progress in levels at a similar RPE, I’m getting stronger, and to the OP, I don’t think the testing really matters. The test becomes your ability to complete harder workouts and your actual performances.
The problem in the TR context now is that their FTP is a bit squirrely and the PL system is in flux (?) too. Nate’s described in the past having more absolute levels for workouts, so you could then compare across time and across cohorts. If I understood correctly, a level 5 workout, for example, would be a set wattage for set intervals, say 5 x 5m @ 300w. That would be an absolute, so relatively easy for some, and a peak for others. You’d have a levels for each energy system, like PLs, but I think they’d have a much greater utility.
I don’t think this is possible, or what they would have meant, you’d need to expand the levels 10-fold or more.
Think about it, you have people on this board with a sub-200W FTP, and people with over 400W. 300W for me is Sweet Spot, and I can go for 2+ hours at that pace. Someone with over 400 - that’s mid-high endurance pace and they could go all day. Take someone with a 175W FTP - they might be able to hold that for a minute? (Over 170%)
I think AIFTP is a little squirrely, but it works pretty well. And if you don’t like it, just do a test of some sort, pick a manual FTP, and move on. That works really well already with just a little more work. (I do think they should just get rid of FTP prediction - quite honestly I think they do more harm than good and are rarely right)