Heavy rider (120kg) with somewhat high peak power having difficulty finding accurate fitness profile

I started indoor training/cycling 1 month ago and have mostly been training weights (on and off) for the past 13 years. I’ve tried out a few other platforms and enjoy the metrics Xert provides the most, but I’m having a hard time getting a good idea of where my fitness is at.

I’ve done a few different kinds of workouts over the past few weeks and each time my fitness signature updates the numbers seem to be all over the place. I’m able to put out fairly decent power levels but I think my TP, LTP, and HIE are possibly wrong.

Here are some numbers I’ve gotten in the past 24 hours.

After a 90 minute endurance ride with a small Fuego Flats Zwift sprint:

Before it:


After enabling estimation of Xert metrics using Heart Rate and manually specifying my resting/max HR (59 / ~180bpm):

After letting xert auto-estimate them (it thinks 65bpm is my resting and 172bpm is my max, does this include the ~10bpm you’re supposed to subtract from your cycling max HR when compared to running?):

Do I just need to continue training so that xert has more data to work with or are there workouts I should focus on to make it more accurate (e.g. the breakthrough v2/v3 workout)? Should I play around with the (PP-TP)/HIE numbers until my values fall into the 28-32 range? The numbers I calculate tend to still be in the outer yellow band of that HIE graph.

I’ve used both the Kickr v5 and SB20 so the only thing I know for sure is that my peak power is between 1300W and 1500W, that I’m apparently able to do 1200W for more than 10 seconds in a sprint, and 830W average over 36 seconds (according to Strava’s segments). Yesterday’s 90 minute endurance ride @ 164W average (LTP) felt like it’s a good estimate of my LTP so the estimate of my LTP being 118W feels way off, as I feel like I could sustain that power indefinitely once I get used to sitting on a seat for that long.

I’ve read the various pages & forum threads on these topics and feel like I’m at the stage where I need to ask for help, so if anyone could please suggest additional resources then I’ll be grateful.

p.s: The freshness indicator seems like it might be a little bit pessimistic? It thinks I’ll be tired from now until next Wednesday with no workouts scheduled but I feel like I could ride today. Most videos I’ve seen have this left at the default 0 setting. Should I play around with this too or will it adjust itself over time?

Thanks for the help!

I forgot to add this one. All 5 are in the last 24 hour period:

I don’t really know which to believe. Thanks again for any clarification or links to resources I can use to help me dial in my metrics and train better.

Do a proper breakthrough effort - the ones you suggest are fine, but do it in slope mode rather than auto / erg (despite what some descriptions say) and truly go to failure even if the workout interval time runs out. I would choose one that takes around 10 min to draw down MPA (look at the chart… v3 does that) to get a good idea of TP and HIE given that’s where your questions seem to be.

The max HR should be your cycling max. You’ll probably get close to that if you do a breakthrough effort. That won’t directly affect your signature (no extraction from hr only rides), but will affect training load which then drives signature. Usually that’s not so material since most people load in a longer history, but given you only have a month of data, it will be very sensitive. You need 60 days for training load to stabilize (it’s a 60 day weighted average). I think that’s why your numbers jump around so much when changing HR parameters. It may mean you get odd moves in the next month too, but should then settle down

Yellow / tired status doesn’t mean you shouldn’t ride, it just means you shouldn’t ride high intensity. It’s the Xert way of getting some polarization in training by suggesting you don’t ride too hard too often.

As an aside, if you do want a polarised plan with high intensity, you should consider changing your athlete type to one with shorter focus e.g. rouleur

Thanks for clarifying. What’s considered to be a proper breakthrough effort?
I did a Zwift ramp test a few weeks back and it gave me a 240 FTP (I got up to around 340W in the test before my legs gave out). I’ve since done a few other breakthrough efforts (going all out for different durations with Xert registering it as a breakthrough) that also involved going to failure. Do you mean specifically one of the following?

  • A Pain That I’m Used To
  • Xert Fitness Test For Breakthrough V2/3
  • Break On Through
  • Under Pressure

Or do you just mean to give all out efforts in some of my rides? (which I’ve been doing)
I know that Xert’s supposed to remove the need to do FTP tests etc. Would those workouts be considered to be a form of FTP test? A friend has since suggested that I do one of the Zwift 20-minute FTP tests but my limited understanding of xert is that I don’t need to do these anymore and that it’s more about my overall fitness signature & power profile.

Re: HR. Thanks! I’ve definitely hit my HRmax_cycling in the breakthroughs. Should I leave HR disabled for now?

Thanks for clarifying about tired status!

I had my rider type set to Power Sprinter initially but figured that my endurance probably sucks so chose this athlete type as an in-between as I’ve never done any base training and most of the training cardio I’ve done over the past 13 years has been short intervals.

I just tried doing some bad algebra with the PP/TP/HIE formula and failed miserably:

These settings must be getting used for the fitness signature as I get wildly different results just by toggling them on/off and having it regenerate the signature:

Different former activities also alternate between being bronze, silver, gold, or near breakthroughs. It seems like there’s an element of randomness to the fitness signature generation where it doesn’t give the same results given the same input parameters?

== A: Initial settings

== B: A few minutes later with the same settings (seems like it dropped my max HR)

== C: A slight change in max HR

== D: Is it non-deterministically changing these each time I visit the HR page?

I guess I also need to know what RHR is being used. My HR drops to 49bpm while sleeping, Garmin Venu thinks it’s 57-59, and the lowest it’s been during an activity (Polar H10) must be 65bpm if that’s what keeps getting chosen by xert.

How many rides have you loaded into Xert? My guess is not many given how recently you started, so each ride is having a big impact on your training load (rolling average over 60 days). If you only started a month ago, I assume you don’t have the recommended minimum of three months data? See here under ‘Establish your fitness signature and goals’

If you don’t have the data, you may actually better off coming back when you do. Or sticking with it and just free riding in Zwift, RGT, Rouvy or whatever you like in the meantime… and starting to apply structure via Xert after that.

I suspect the reason HR affects is signature in your case is that Xert’s signature moves due to:

  1. breakthroughs (step changes)
  2. increases in training load (incremental changes, but possibly massive if you’ve not much data - going from 4 rides to 5 is a massive % increase in training load hence unusual results… going from 29 to 30, not so much)
  3. signature decay (unless set to ‘no decay’)

So when you change your HR parameters you are changing your training load… lowering max HR makes Xert think you are closer to max and increases TL for each of those rides… that then increases signature… and especially materially in case you don’t have a lot of data. Practically, I’d probably turn that feature off until you have a lot more data - put in a manual estimates for those rides instead e.g. 50*hours with endurance focus unless you know it was high intensity. And ideally HR based rides should be the minority of your rides

Re breakthroughs, v3 is probably better than v2 for you, as it takes longer to drawdown MPA… but if you have already got some true all out efforts with long MPA drawdown, you don’t necessarily need to do another one. No need for a structured test. You do need to be fresh in order to consider it a true maximum effort though. And ideally it’s not just from a sprint or 1 minute effort.

Not sure I’d worry too much about the formula given you are just starting out. You will have lower endurance given your training background, so working on that (base training) would make sense as a priority.

I think you are over thinking this. :wink:
All you need is a ballpark signature to start with and move forward from there.
HRDM is for when you have sufficient power, cadence, and HR data on file to estimate XSS and Focus for rides that lack power but include HR and cadence.
If you lack the required volume of historical data (months worth) you should disable HRDM.
Power is king and that is what you should be recording for the majority of your activities indoors and outdoors along with cadence and HR if you want HRDM to work well if needed in the future.

You will know if your signature is “off” if recommended workouts seem too easy or too hard based on their difficulty level and intensity. When you have sufficient data on file you can also request a signature review with support@xertonline.com if you still doubt the numbers.
Any of the BT workouts will suffice to test your signature periodically. You will be warned if no BT events in more than 3 weeks. Try them all over time and see which are your favorites.
Otherwise just ride and accumulate more data while XATA learns about you and you learn about Xert.

I imagine you have already reviewed this thread but in case you haven’t –
Beginner questions - Support - Xert Community Forum (xertonline.com)

Related: Huge difference between SYSTM/Garmin & Xert - #20 by ridgerider2

@wescaine 24 rides loaded into Xert and dozens more historical runs/walks with HR data from Garmin Connect, so yeah, not 3 months worth of rides with power data.

I’d like to ride with structure between now and 2 months from now and to do workouts that help xert develop a better fitness profile of me. I’ve checked out Sufferfest/SYSTM and Rouvy but they’re not for me, so I either train in Zwift using a Xert workout or just in Xert’s remote player with youtube videos/podcasts playing.

Should I disable the signature decay for the time being or leave it on default?

I can disable the HR data for the time being but I thought that Xert might be better able to figure out my fitness signature if it had my historical HR data from runs.

Thanks Wes!

@ridgerider2 Yeah I’d love to, I just need a good idea of where to train for my endurance rides. I’ve read here that below LTP is where I should be training for endurance, but when it could be anything between 107W and 168W it’s hard for me to know if I’m undertraining. I suspect it’s closer to 168W than the former.

I’ve read some of @ManofSteele’s PezCycling articles on DFA-alpha1 and it being close to LTP, so I was wanting to possibly use data from that with my H10 to figure out where to train. It doesn’t look like Xert lets you set LTP in options though. I tried out https://play.google.com/store/apps/details?id=com.asma.hrvlogger last night but wasn’t able to get it to display real-time DFA-alpha1 data so am still looking for an alternative app that does.

I only train indoors right now with power, cadence, and HR data.

Most of the Xert workouts (HIIT ones) have been reasonably spot on (as in if sets lasted 2 seconds longer I’d be cooked) as far as difficulty. I’d love to know how to train in a way that helps Xert narrow in on what my correct metrics are so that all of the workouts are at the right level. With the LTP moving around a lot, some of the endurance workouts have been really boring but others have been enjoyable which makes me wonder if I’ve been accidentally sweetspot training.

Yeah I’ve read the Beginner questions thread and the past several months worth of relevant threads trying to understand XATA better. HIIT workouts have been great but endurance workouts are still hit or miss due to the fluctuating LTP.
I’ll disable HRDM for now, keep training with the remote player or Zwift, try to find a good app that gives me a DFA-alpha1 reading and then do some endurance workouts where the average power is slightly below whatever the value ends up being. As soon as I’m fresh I’ll try one of the long MPA drawdown breakthrough workouts. It seems like there’s not a lot else I can do without more data.

Current LTP/TP difference after disabling HRDM. I was able to easily do 168W for 90 minutes so I think that the LTP is way too low, but there’s no way to change it in options.

Thanks for the advice!

Understand the thinking re HRDM and it will help with load once you’ve more data, but for now it’s good to leave it off.

I would leave it on optimal decay while things are settling (increases chance of a BT, and each one hopefully improves accuracy) assuming you will keep doing the odd really hard effort every few weeks.

Assuming you want to follow a polarised approach I.e. keep doing some high intensity each week, some simple options for endurance would be:

  • Keep HR around 70% of (cycling) max hr
  • Ride at an intensity that you can still speak comfortably (aka the talk test)

If you really want a power target in the interim you can do an FTP test in Zwift and stay around 70% of the resulting FTP for endurance rides. Not as personalized as Xert but ok for a ballpark for a couple of months.

I’ve not had useful results from DFA alpha 1, but many seem to have. The HRV logger app should work - was the issue perhaps that you had connected your H10 to Xert or Zwift already? There are two bluetooth and one ANT+ channels for the h10 but you need to enable them in the polar app, else you only have one bluetooth available, meaning you can only connect to one app. (And it resets when you change the battery for info)

You can’t directly set LTP - it’s the result of other parameters… but if you increase TP and decrease HIE you’ll also increase LTP. You can do that in the advanced MPA tab of an activity and then save / lock. Not sure I’d bother, but up to you.

You mention high intensity workouts being spot as you are right on your limit… most Xert workouts (other than ones designed to cause a breakthrough) are not actually designed to be bringing you so close to your limit. If you do 4 diamond workouts, yes they get tough if you don’t have the training load, but if you follow Xert’s training advice it won’t be recommending those given you’re just starting out. Not to worry, just further evidence your signature needs time to settle with more data

Thanks again for the reply Wes!

I think it was likely user error with HRV Logger. I tried out a few different apps and settled on FatMaxxer (GitHub - IanPeake/FatMaxxer: Android Project to find FatMax in real time with a Polar H10), then had a look at the Zwift DFA alpha1 ramp test recommended by aiendurance.com but noticed that it went up in 10W/minute increments. I recall Bruce Rogers possibly saying something about spending a few minutes at particular power levels so just set Zwift on a 1hr warmup ramp from 120-250W, with it going up in smaller intervals. DFA alpha1 ended up hovering above/below 0.75 at 188-190W, then occasionally went above again a few times until around 220W. I’ve gotten alright at noticing when I start feeling lactate and I hadn’t felt any until around 188W, so I’ll use that as the lower bound and do repeated tests over the next month.

When the ramp test ended @ 250W I decided to restart it at 250W and turn it into a kind of makeshift MPA drawdown workout and see where I could get to. Lactate didn’t seem to be too much of an issue and my HR only shot up once I crossed 280W. I feel like I could hold 240-280W for an hour so I suspect that’s probably around where my FTP is. I’ll try the proper Zwift FTP test in 2 weeks.

I started overheating after ~310W so decided to call it quits probably earlier than I should have (we’re going into summer here in NZ and I’ve started trying to heat train without a fan).

Out of curiosity from this post MPA calculations (@skinner_ron) I decided to test if I could hit a peak-power sprint 80 minutes into the workout after already doing the ramp test. Anecdotally, it turns out that I can. I hit 1674W according to Zwift (I saw > 1500 pop up on the screen) but I recall that Xert strips out spikes above peak power if they’re under 5 seconds long. Damn, I was hoping for a peak power breakthrough.

I also decided to see if I could match my 10-second sprint power (~800W average). At least in my limited experience I think that @xertedbrain’s comment about hitting peak power is accurate (although admittedly 4 hours is longer than 90 minutes, I’m a complete beginner so don’t qualify as anywhere near a professional).

Here’s the graph of the ride.

Hopefully that shows that an MPA effort is still possible for an untrained person (although I do really wish it showed the 1674W peak, I’ll have to break that another time when fresh)

I know it’s just one person’s data but I found it interesting to see how relatively accurate MPA is at least for sprints.

These are my metrics after that ride. I feel like they’re far more accurate than they were before.

I’m looking forward to learning some really cool things about my training in the upcoming months. What an awesome platform!

Yes, that is a good sign you are pedaling slow enough. :slight_smile:
I read where one pro described low intensity early season rides as lifting the pedal on the backstroke and letting it fall on the downstroke. I guess they mean Z1 on the 7 zone scale.
I don’t find it that extreme but some do think it’s agonizing to go slow on long rides. Indoors you can watch a Netflix series or listen to podcasts or find something interesting on YT and create a solo session to watch it.

As to the benefits there are distinct adaptions at work that may not be readily apparent, but your TP can rise after weeks of nothing but low intensity activities.
Better base, better build. Batta bing batta boom. :smiley:

1 Like

Just to give an update for anyone else that comes across this issue when joining the platform, I ended up needing around 30 rides worth of data including a few breakthrough efforts before Xert started giving an accurate fitness signature. I also needed to increase my freshness and improvement rates as Xert kept thinking I’d be tired for over a week after each high intensity workout and would only suggest active recovery workouts. If you’ve got less than 30 rides then use the Xert EBC app and its difficulty buttons to adjust the difficulty during workouts as you train.

If you’ve got a well trained anaerobic system but less trained aerobic system you need to do some long boring rides for it to bring your LTP up. Using the FatMaxxer app to approximate DFAalpha1 can help with staying focused and ensuring you’re still fat burning on the longer rides.

Contacting support was also extremely helpful.