KOM Informatics: Working With Humon Threshold Tests


Humon Support contains an article on how to perform an Incremental Threshold Test with the Hex. Use Case: Incremental Threshold Test


Here's what happened when I ran the test 8 times over the course of this past Spring.


The cycling protocol from the article is:

  1. Start cycling on a stationary cycle ergometer at a very low power (30W if following the Humon app protocol) for 3 minutes
  2. Increase the wattage by 30 W every 3 minutes
  3. Cycle until you absolutely can't push any further (when you reach voluntary exhaustion)

"When the athlete can no long push himself, he decreases the power and starts recovering. His threshold is considered to be when the red zone appears."


The first thing I noticed about the protocol is that it isn't necessary to go to voluntary exhaustion to get the threshold wattage because the wattage we're interested in corresponds to the wattage at the initial second when the Humon Zone goes to red. I know from experience that I can push for around 45 minutes during a threshold interval, and 14 minutes in a VO2Max interval while I'm in the red. Those sessions are metabolically very taxing though and need to be carefully integrated into a training plan. But if I'm interested only in the initial red second, the protocol outlined above (which I ended up modifying as you'll see shortly) becomes a pretty good way to get a gradual warmup without incurring additional training load over a non-structured warmup. I feel comfortable running the test this way for just about any trainer session.


I did make some modifications to the protocol to avoid problems I've had with running red at relatively low wattages and or heartrate. In one instance Humon support told me it could be the case that my muscles weren't primed for the effort and I needed to warmup longer. I've also gotten the "Make sure to warm-up longer to allow your muscles to recover better" warning many times in the Humon App. So I extended the 3 minute period in the protocol out to 4. I also went a little harder in the last 2 minutes of each period trying for a more gradual application of effort.


Here is a graph (Click on the thumbnail for full size) depicting the results from 8 tests I ran this spring. The graph is from my own cycling analytics application called KOM Informatics. (My app is just available to me in a test environment it isn't available to the public) Watts@FirstHumonRedSecond and BPM@FirstHumonRedSecond are 2 of about 40 ride based summary statistics that I can select and display on a graph. Some observations:



  • 4 of the rides had threshold watts in the 268-281 range, where 4 others were substantially lower, between 221 and 244.
  • The higher wattage rides were aligned with my expectations based on performance data from threshold intervals. The lower wattage rides were not. In the worst case I went to red while at Zone 2 wattage and heartrate.
  • Although Fitness (KOM Informatics uses a TRIMP derived system like several other cycling analytics applications) was around 17% higher in the last test as opposed to the first (70 vs 82) this wasn't reflected in the wattages derived from the tests (281 vs 279).
  • Form (which in a TRIMP based system denotes how well rested an athlete is) ranged from +6 to +17 across the tests. I expected to do worse in cases where form was lower, but this was mistaken. It doesn't seem to make a lot of difference for me, in fact the lowest wattage across all 8 tests was on the day with the highest form.

I did do some cross checking with the ride graphs from the ride pages of my system to make sure that the numbers I was coming up with for Watts@FirstHumonRedSecond were making sense. On one of the rides there was a situation where the Zone status went to red for 1 second before going back to green, which triggered the @FirstHumonRedSecond prematurely in my opinion for the purposes of the test. Here's a look at that Ride Graph:


I changed the triggering criteria for my tests utilizing an user configured averaging period which is used in a couple of different ways. The first way is that it insists that you stay red for the averaging period before completing the test (I used 3 seconds for mine). IOW a one second blip to red no longer counts. The other way is that it looks backward in time starting at the first red second for the averaging period and averages watts. This lessens the possibility that Watts@FirstHumonRedSecond could be skewed due to small sample size. If I wanted, for some reason to use the first red second by itself it would still be possible though setting the averaging period to 1 second.


Here's the same graph as above using the 3 second averaging period:




5 of the rides had threshold watts in the 274-279 range, where 3 others were substantially lower, between 213 and 237. So one of the rides jumped up to the higher watts group and the threshold ranges for both groups tightened up a little, which I think is a more realistic outcome.


For me running the tests has been a useful exercise in confirming performance data on what my Maximal Metabolic Steady State wattage is which turn helps me in setting target wattages for intervals. This despite the 3 tests which came back too low in my opinion. I'm just going to disregard those. Maybe in the future I'll extend step duration out to 5 minutes to see if this solves the issue.


One of the main benefits of the Humon Threshold test in my opinion is the ability to repeat the test on a regular basis. The protocol probably doesn't incur much, if any additional training load over a normal warmup provided you only pedal to a few seconds beyond the occurence of the first red second. This means you could do it any old day during trainer sessions. (I think staring fixedly at a Garmin during a road ride is a recipe for disaster.)


Another benefit of the Humon Threshold Test is it's cost ($0 after buying the Hex) weighed against the cost of lactate threshold testing at a Sports Medicine clinic. Around my neck of the woods it's $300, a nice chunk of change for data on something which apparently can move around a bit on a daily basis.


For those looking to use this test I recommend doing it a number of times (over weeks, not during the same session!). Running just one test runs the risk of firing red too early resulting in an understated threshold. It also couldn't hurt to "triangulate" your results against performance based FTP tests.


For additional information on how the Hex results mapped to lactate levels and a lab grade NIRS device check out the following article:


Validation of a novel wearable, wireless technology to estimate oxygen levels and lactate threshold power in the exercising muscle