Just to share some responses from Tymewear support:
BP (Balance Point) is the point between VT1 and VT2 where VE/Power is at its minimum . Representing the best ventilatory efficiency (lowest ventilatory equivalent for power).
I didn’t know that. Steve Neal’s protocol was to do a ramp test to max, then pedal at half that power increasing it every 3 mins and take lactate at the end of each step. Lactate would fall with each step until at would start to rise. That was the Balance Point . Minimum VE/Power seems a lot simpler to me.
@ hdkiller
I did my first Tymewear Test on the bike today & so am posting my charts if anyone is interested. I am still waiting for the official Tymewear analysis which will likely be 2 days with it being a weekend. I did ask Coach Watts to have a look & so it might be interesting to compare his finding when the official Tymewear result comes through. I have got the tail end of a chest infection & so will need to repeat the test once full recovered.
This is the Gemini analysis using @Timur prompt
As an academic sports physiologist, I have reviewed the high-resolution ventilatory time-series data from the Tymewear FIT file and the accompanying dashboard visualization. The following analysis focuses on identifying your physiological thresholds (VT1, BP, VT2) and assessing your metabolic efficiency and ventilatory reserve.
The data was segmented into 3-minute stepwise stages. Analysis was performed on the steady-state period (last 60 seconds) of each stage to ensure heart rate (HR) and ventilation (VE) had reached a relative plateau, minimizing the impact of oxygen kinetics.
| Parameter | VT1 (Aerobic) | BP (Balance Point) | VT2 (Anaerobic) | VO₂max (Peak) |
|---|---|---|---|---|
| Power | 140 W | 120 W | 220 W | 220 W |
| HR_corr | 120 bpm | 115 bpm | 142 bpm | 145 bpm |
| VE (L/min) | 61.2 | 50.4 | 132.5 | 138.2 |
| BR (brpm) | 24.5 | 21.8 | 46.2 | 52.4 |
| VT (L) | 2.50 | 2.31 | 2.87 | 2.64 |
| VE_per_P | 0.437 | 0.420 | 0.602 | 0.628 |
| BreathingPower | 1,500 | 1,098 | 6,121 | 7,242 |
| %VO₂max | 63.6% | 54.5% | 100% | 100% |
| Drift (HR / VE) | 1.8% / 2.1% | 1.2% / 1.5% | 4.8% / 7.2% | N/A |
Based on an estimated age of 35, the predicted maximal ventilation is calculated.
Your Fractional Utilization of VT2 is exceptionally high (100% of peak power). This profile is typical of an endurance athlete with a very high “floor” but a missing “roof.” You can sustain high percentages of your max power, but you lack the high-end power to surge or respond to attacks.
The primary limitation is Cardiovascular/Metabolic. With a high Ventilatory Reserve, your lungs are moving air effectively; however, your $VT$ begins to decline at peak intensities, suggesting the mechanical cost of breathing (BreathingPower) is becoming a significant energy sink at 220W.
Finally Coach Watts again re the comparison of Ai v Ai
Yo Dave! 
Wow. This is absolutely fascinating. Seeing the “Academic Gemini” process your raw FIT file using Timur’s specific Python logic gives us a totally different lens than the standard “cycling coach” view.
Let’s look at the “AI vs. AI” head-to-head.
| Metric | Coach Watts (Manual) | Academic Gemini (Python) | Difference |
|---|---|---|---|
| VT1 (Aerobic) | 145W | 140W | Only 5W! |
| VT2 (Threshold) | 187W | 220W | 33W (The “Collapsed Zone”) |
| VO2 Max (Peak) | 253W (Max) | 220W (Completed Stage) | Logic diff |
Gemini’s analysis identified something critical that I want you to pay attention to: “RTI_VT2: -15.0%… your VT2 is at 100% of your peak power.”
In simple terms: Your “roof” has collapsed onto your “ceiling.” Because the Python logic only looks at fully completed 3-minute stages, it sees your VT2 and your Peak Power as the same thing (220W).
This confirms exactly what you said: “Lungs failing also… legs gave way.” Because you are fighting that chest infection, your body hit its Respiratory Compensation Point (VT2) and then basically had zero room to move above it. Usually, an athlete has a “Zone 5” buffer above their threshold. You currently have a “Zone 4.5.”
Gemini’s Ventilatory Reserve of 90.1% is interesting. It’s saying your lungs could move more air (theoretically), but the “Breathing Power” (the metabolic cost of moving that air) became a “significant energy sink” at 220W.
Basically: Your lungs weren’t too small; they were just too expensive to operate today because of the infection. They were stealing all the blood flow and energy from your legs.
Dave, between the Python script and my look at the raw heart rate:
How does it feel to see the “Academic” version confirm that your Zone 5 is currently missing? It’s a perfect physiological explanation for why you felt like you hit a brick wall! 
(Coach Watts has me with 5 HR zones but 7 Power zones.)
I’m on my phone at the moment and not had chance to read through everything but to my untrained eye VT1 @ 15m and VT2 @ 20m ?
I agree with VT1 around 15m, but VT2 is all but ‘easily’ identifiable. I would even think that VT2 has not been reached… or maybe just when the ramp was aborted.
All things taken into account, if VT2 was around 20m, then HR should plateau, which is not the case.
I see that too, but nothing beats doing a lactate threshold test to calibrate, as recommended by Timur
Doesn’t HR plateau at VO2 max?
HR plateau is a classic sign of VO₂ max — the point where the body can no longer increase oxygen delivery or utilization, and HR approaches its maximum. But, in some cases, can start to increase at a slower rate at VT2. especially in well-trained individuals with VT2 close to VO₂ max during ramp test .
Yes, they did remove the option to manually set threshold values. Unfortunately.
However, at the seminar they said this feature would return.
For me personally, it is clearly better to have the ability to adjust these values, including based on lactate measurements and RPE.
In any case, the values in my bike computer and watch and the same ones entered in Intervals - differ from the values in Tyme
AI can certainly help analyze data and highlight certain variables, but I would not fully rely on that analysis. For each test, it still requires quite a lot of effort to refine the prompt. I haven’t managed to create a universal one.
The most accurate values, in my view, come from visual analysis and your RPE. If that is supported by lactate data, it becomes ideal.
Looking at your graph, VT1/LT1 is most likely around the 15-minute mark, between 145-150 watts. From my perspective, that should be the top of Zone 2.
It is harder for me to clearly identify VT2 here (which is surprising, as VT2 is usually more distinct). In this case, I would prefer to be conservative and place VT2 somewhere around 21-22 minutes, which corresponds to approximately 200-210 watts.
It will be interesting to see what the Tyme calculations show.
To be fair it was not an ideal test as I still have a chest infection. But I bought the Tymewear about 3 weeks ago & so wanted to get some sort of approximate result. I will post the Tymewear findings probably in the next 2 days when they arrive. Realistically I will not be doing Lactate testing as it is just too fiddly & my fitness standard/goals does not honestly justify it. I used to use DFAa1 (with AI Endurance for a while) but with arrhythmia it just did not seem that reliable. Though to be fair that seemed to be on the high side of my expectations as this test feels to me currently (though I am currently ill & so not really menaingful). I am happy to go 5-10% below thresholds anyway for wiggle room so absolute precision is not especially important. Often I am a litlle fatigued when exercising & so the measured threshold would likely be too high for a fatigued state. That was partly also why the compromised test was not such a major issue for me personally as it is probably closer to my reality than one done at peak fitness/health. But I will redo it when at full fitness & probably try another DFAa1 analysis also.
Anyway thanks for providing the Prompt as it was interesting comparing Coach Watts (Gemini 3.0 Pro) with my full activity history & profile against Gemini Free (Deep Thinking) with no personalistion/history other than this test.
For context I am 56yrs old doing riding, running, swimming, Skierg for general aerobic/metabolic fitness rather than aiming for competitive fitness level; no races for years. I have asthma but normally well controlled. The damp UK is currently challenging that slightly though hence the prolonged chest infection I suspect (3 weeks). But I do get PVC arrhythmias intermittantly, though from what I can tell mainly at low HRs rather than during exercise when I have used CameraHRV to monitor. But these have settled over the last 3 months with the 90% Z1/2 I have been aiming for.
By the way, have you tried my “PVCs PSVCs Blocks” finder Custom Activity Field?
My concern is that Tymewear’s VT1 is now balance point and thus too high for appropriate Z2 work if you go anywhere near it. Or have I missed something?
Overall, I would probably agree, but for me Zone 2 (the top of Zone 2) has never been just an easy spin. It is often quite a demanding session, and I have always done it as intervals - just longer ones than threshold work.
Now the distribution of threshold anchors and the training zones in Tyme looks more logical to me. However, what raises questions is the recommended time in zones. At the moment, out of 8 hours 07 minutes, I am supposed to train 7 hours 30 minutes in Zone 2. That is simply unrealistic.
Previously, the recommendation was ~6 hours in VT1 and ~1.5 hours in BP, which looked much more realistic in terms of time spent at intensity.
In short, I use the anchor values, but I rely on my own threshold settings and my own distribution of time across intensity zones
I have added & will keep an eye on them. Presumably it will need a Polar H10 or HRM that has RR measured? I think the Tymewear one does but for yesterday showing ? on the 3 fields after re-analysing. So unsure if not present or not working. To be fair I probably rarely if ever have them during an activity, to my knowledge at least. They are generally asymptomatic so it is usually chance if I detect them e.g. Garmin low HR warning. If my watch can detect them then the cold plunge or sauna might catch them as I have a ConnectIQ app that records HR during those activities.
You need to enable it. In a Garmin watch: Settings>>System>>Data Recording>>Log Hrv.
Polar H10 is the best, for that, Garmin HRM Pro/Plus great, others will work also.
Result should be something like
or
if you just want a number for Fitness Chart correlations
Many thanks. I have reprocessed a few of my Gamrin sync’d files & there is a screenshot of a selection below. Iwill have to experiment with what synchs will give a FIT containing the relevant metrics as I tend to use a few different origins though usually recorded with the same HRM.
Check also with “RR Raw Intervals”, and “PVC Detection Dashbord” Activity Charts, suggestion don’t use Timeline tab, create a new tab for that. Question the HRM recording quality for each activity, before analysing for PVCs. “PVC Detection Dashbord” is optimised for laptop screens, doesn’t work great with mobiles.
Agreed; sounds crazy to recommend that.
Anyone with limited time to train should look to maximise their high intensity (threshold and higher) first, plus one easy ride with the duration as long as possible. Then fill the rest with the appropriate duration and intensity.
It’s easy to say do the high intensity, and it depends on how stressful life is.
