Training WITH a PM ...

[MichaelB]

Following on from the thread re buying a 1st PM, I've started using the PM and following some of the great advice given in the above thread.

As part of understanding all of what is available resulting from better info of the HRM and PM, I've done the free 30 day trial for Strava Premium, signed up for the free trial of TrainingPeaks and also loaded StavaStix to the home Strava account.

Been also doing some reading and getting a better understanding of the Form, Freshness & Fitness (or whatever each site calls it) and how to maximise and plan around it, but the thing that is confusing me, is that whilst they use the same data, the curves look similar, but the actual numbers are different ??





and so you can read the Training peaks numbers ...



It seems that the methodology is the same (Stravastix mentions Friel and TrainingPeaks is all based on Friel), yet their numbers are different.

I guess it doesn't matter what the actual numbers are, but I'll stick to one (likely StravaStix as it presents the data better), but still curious.

Brave new world .....

EDIT : Looking at the graphs, it seems that the data base is maybe different and that Training peaks doesn't have some rides from mid November ? There were some rides where I was trialling the Garmin 1030 and I think that data is missing from the TP data set, but that wasn't much ?

Anyway, a couple of decent rides this weekend and then taper a bit for the TdU BUPA ride on Friday

[queequeg]

TrainingPeaks is similar to Today's Plan (which is what I use), with a couple of things to note (direct from Today's Plan)

Threshold Test: A threshold test is a way that we determine your lactate threshold – either using a power meter or via heart rate data. In theory, this is the maximum output which you can sustain for an hour. We use a 20 minute test to predict what your 1 hour threshold value would be. Once we calculate this value all your training rides are scored in comparison / relative to this value. For power thresholds, this value becomes a goal to increase. Assuming you stay at the same weight, a higher threshold means you can sustain a higher power output for longer.

T-Score: This value is a calculated score used to objectively measure rides completed with power data. The formula takes into account both the ride intensity and duration. If you are working to a plan, then a session will have a target T-Score – in which case you need to ride to both the specified duration and intensity to reach the desired goal. Generally speaking, a T-Score of 30 would be an easy hour recovery ride, 120 would be a solid 2 hour training ride, and 300 would be a massive Sunday epic.

ATL: Acute training load is a way of objectively gauging your current training load. It is the short term (7 days), daily T-Score average of your rides. ie your daily T-Score average over a short period of time. When compared with your CTL, you can see if you are currently riding above or below your long term average. By comparing your ATL and CTL you can track if you are building or recovering, and the comparison between the two values gives your TSB or freshness score.

CTL: Chronic training load is a way of objectively measuring your general fitness. It is the long term, daily T-Score average of your rides. ie your daily T-Score average over a long period of time. To increase your CTL you are doing (on average) rides with a higher T-Score – this can be performed through either longer rides or more intense rides. CTL is a great way of tracking your overall fitness.

Intensity: Intensity is the ratio of the power range compared to your power threshold. In theory, a 1 hour time-trial race would be a value of 1.0. A recovery ride will be around 0.6, a tempo ride around 0.8. Some power uses like to ride their sessions to intensity numbers and others prefer the raw power values. In the intervals table we show both for power users.


In Today's Plan, the CTL defaults to a 42 day rolling average, but it can be changed. I don't know what other sites use.

I have also highlighted the comment in the Threshold Test, because the first thing you need in order to get a half decent T-Score is to have the correct baseline, otherwise anything after that point is rubbish.
For example, I have been off the bike for coming up to 10 months. My pre-injury FTP was 350W, so everything in Today's Plan is still based on that Threshold. If I went out and did a 1 hour ride, I might come back totally shattered because I have been off for so long, but Today's Plan would score that effort as if my power output was still the same, and it would score it pretty low. Most likely it would think I did a recovery ride (based on my power output).

The other point, is that unless you have done no riding recently (as in the last couple of months), your starting ATL/CTL is not actually 0 so you will need to make a guess as to where it should start. That will be easy for me now..it will be zero after 10 months off the bike. When I first switched to power though, I had to make a wild guess.

Bottom line, first is to understand what the numbers are telling you. Regardless of the site you use, they should all show the same trend, but stick with one of them for your training. For me, the source of truth is Today's Plan because it goes so much further on the analytics than anything else, and you can do training plans in there that hook straight into Zwift, TrainerRoad etc for your workouts, which are all tailored to your current fitness level.

[g-boaf]

Following on from the thread re buying a 1st PM, I've started using the PM and following some of the great advice given in the above thread.

As part of understanding all of what is available resulting from better info of the HRM and PM, I've done the free 30 day trial for Strava Premium, signed up for the free trial of TrainingPeaks and also loaded StavaStix to the home Strava account.

Been also doing some reading and getting a better understanding of the Form, Freshness & Fitness (or whatever each site calls it) and how to maximise and plan around it, but the thing that is confusing me, is that whilst they use the same data, the curves look similar, but the actual numbers are different ??





and so you can read the Training peaks numbers ...



It seems that the methodology is the same (Stravastix mentions Friel and TrainingPeaks is all based on Friel), yet their numbers are different.

I guess it doesn't matter what the actual numbers are, but I'll stick to one (likely StravaStix as it presents the data better), but still curious.

Brave new world .....

EDIT : Looking at the graphs, it seems that the data base is maybe different and that Training peaks doesn't have some rides from mid November ? There were some rides where I was trialling the Garmin 1030 and I think that data is missing from the TP data set, but that wasn't much ?

Anyway, a couple of decent rides this weekend and then taper a bit for the TdU BUPA ride on Friday

Okay, so, what you really need is an accurate FTP to work off - this goes into the zones section on TP:

User name drop down > Account settings > Zones > Power

Enter the threshold value for your FTP. In Training Peaks, this is NOT your FTP. You must subtract 5% from your FTP figure and put the number in there. Work out your heart rate zones as well, you can do that off the threshold and use one of the auto calculation methods like from Andy Coggan or CTS. More details to that here:

https://help.trainingpeaks.com/hc/en-us ... your-Zones

More on the settings too and how to get the numbers:
https://help.trainingpeaks.com/hc/en-us ... te-or-Pace

That will give you a base to start from so that your CTL/ATL and TSB (training stress balance, or more easily, your form) are accurate.

TSB can go too low if the threshold power is too low (ie, you've become fitter and more powerful), so it's important to keep that fairly up to date. In general, when you see values in the very negative area (-30 or more) you should be feeling fatigued.

If you are going for certain events, put them in the training calendar as events and then put in the training you are planning to do on particular days or over the weeks leading up to your event and put in approximate TSS scores. You can sort of base this from existing workouts you've done. Then you can see, by looking at the performance management chart how you are going and you can work out too when you should taper off your training in order to recover and be fresh for your event.

[eeksll]

ATL: Acute training load is a way of objectively gauging your current training load. It is the short term (7 days), daily T-Score average of your rides. ie your daily T-Score average over a short period of time. When compared with your CTL, you can see if you are currently riding above or below your long term average. By comparing your ATL and CTL you can track if you are building or recovering, and the comparison between the two values gives your TSB or freshness score.

does anyone know how well these work when one does other sports like running and swimming and gym weights?

[MichaelB]

I'm sorry I asked !!!! My head hurts !!

The FTP (estimated from my ride I did up Greenhill Rd (30min climb) in early Jan and from Strava (was est as 291W at the time from the Strava power curve - had gone up to 300 and now back down to 294W) has been used as a base across all three, and the MHR of 195bpm has also been used (although, think I need to check the setup in Stravastix better).

I need to digest the replies a bit better so that I can understand them and maybe I can understand it a bit more.

Still doesn't explain why the numbers are that different though.

One thing I really do like with the graphs on the Stravastix is that the 'Form' is shown below the other two, so it's much easier to read.

Also, I don't have 'races' or other big events that I'll do during the year, so targets are not that important (atm anyway ....), but I wil;l use it to judge when I am pushing to hard.

A good example is the riding I did in Qld - did 426km and 4,500m of ascent over a week. And the increase in fatigue was huge and probably the reason why I came down with a cold over Xmas.

Anyway, lots more to think about. Ta

[kb]

Numbers will be different. Strava’s weighted average is generally lower than NP. Strava defaults to using HR when no power - not sure about Training Peaks and whether they would use the same stress (TRIMP?). Strava also calculates using your current profile settings for weight and FTP rather than using historical data for historical rides.

[andrewjcw]

I'll hopefully be getting into some consistent training with a PM this year too. Then again I thought my 2 week chrissy break was going to be prime riding time but here we are on the last day and I've done 2 rides in 2 weeks and put on 3kgs so what do I know. So much for the festive 500.

I'm not a big one for all the acronyms and theory and calculations. I do have a pm though and like using it just for some motivation/consistency and progress checking. My 'pm training' so far basically comes down to checking my power curve and selecting a segment to set a power best per time on.

Basically the plan at the moment is to ride as much and as hard as I can and set a new ftp every 4-6 weeks. I think my mental energy/motivation wanes long before the body does so I'm not to worried about fatigue/getting enough rest of whatever (beyond the common sense things like taking a day off every week or two, don't do continuous days of smashing yourself or long days etc...). Risks will be the usual work/family stress but we do what we can.

My starting point is below



Not a complete picture but a pretty good idea of the starting point from the few weeks I've had a PM for at the end of 2017. As you can see I've put a goal down of 350w for 20 mins (332.5w FTP) for the end of May. Considering current is 307 that sounds like a pretty safe target, but really the goal would be to do that while losing 5 or more kg which would overall be a pretty substantial w/kg increase. Current is ~3.6w/kg and -5kg +25w would increase to ~4.2w/kg so that's sounding more respectable.

As I said before my post christmas and start of 2018 'riding' has been abysmal but I'll be forced to starting commuting again starting tomorrow and hopefully some training/calorie discipline mojo will flow on from that and we'll be on track for a good year.

[macca33]

I may have missed it, but have you done an actual FTP rest with your PM yet?

If so, you'll need to extrapolate that result into all of your applications and they'll set your power zones - I'd take the power zones from Training Peaks, or Today's Plan, etc, over the Strava designated zones.

Then, when you start a training plan, it will dictate your power intervals, etc, towards whichever goal you have chosen to achieve.

I've found the Strava F & F to be similar, but subtly different to Today's Plan & TP...

cheers

[g-boaf]

I'm sorry I asked !!!! My head hurts !!

The FTP (estimated from my ride I did up Greenhill Rd (30min climb) in early Jan and from Strava (was est as 291W at the time from the Strava power curve - had gone up to 300 and now back down to 294W) has been used as a base across all three, and the MHR of 195bpm has also been used (although, think I need to check the setup in Stravastix better).

I need to digest the replies a bit better so that I can understand them and maybe I can understand it a bit more.

Still doesn't explain why the numbers are that different though.

One thing I really do like with the graphs on the Stravastix is that the 'Form' is shown below the other two, so it's much easier to read.

Also, I don't have 'races' or other big events that I'll do during the year, so targets are not that important (atm anyway ....), but I wil;l use it to judge when I am pushing to hard.

A good example is the riding I did in Qld - did 426km and 4,500m of ascent over a week. And the increase in fatigue was huge and probably the reason why I came down with a cold over Xmas.

Anyway, lots more to think about. Ta

It's not too difficult, you'll get it. TP is a very good tool.

I would be wary of strava power figures. Upload the same ride in TP and then select the portion of your ride that had the climb and provided you had this power data in that ride, you should see what you did over 20min. In fact, TP will tell you that anyway. When you look at the ride in expanded details screen, you can look at power levels for different durations, eg, 1 minute, 5min, 10min, 20min, 60min, etc.

That will give you the number you want. Then just take that 20min power figure and multiply it by 0.95, and that's it.

To the person above asking about running and swimming, that's not my area of expertise, so hopefully one of the triathletes if there are any here will know more about how that is in Training Peaks.

[Alex Simmons/RST]

but the thing that is confusing me, is that whilst they use the same data, the curves look similar, but the actual numbers are different ??

I see you have opened that can of worms. That's to be expected given all the options out there.

In the beginning, we had one consistent means of reporting such things based on power meter data. A system based on sound physiological principles and which simplified the overly complex and significantly less practical Impulse-Response model described by Bannister as well as provided a means to use a power based stress score as the input.

Ever since then a whole stack of people/organisations started ripping this rather good idea off and in order to make it appear as if their implementation was novel, interesting or original they made slight changes. Some just changed the names of things, others tinkered with the formula (with no sound physiological rationale for doing so or in one case a change that was physiologically reasonable but as the author realised it made no improvement*), while others made changes to the inputs used. Some made/make a complete hash of it. Worse, many started to read much more into the charts than they really should and implied a level of insight beyond the plausible (and marketing this "unique insight").

Whether and when these differences matter is something that's not quickly explainable.

In the broadest sense what matters most are the overall patterns they reveal and on that front most such systems will provide similar feedback. By and large these systems are fairly insensitive* to such changes when it comes to overall patterns.

The problems arise when you seek to be somewhat more specific in how to interpret and use the data, and of course as you have discovered, in what absolute values and rate change values make sense relative to your individual training experience.


* part of the reason for this is these training load pattern curves have specific limitations since they are all based on a training stress measure or one kind or another.

What would provide additional insight would be a clearer way to objectively measure training response. That quite hard nut has been reasonably cracked but is not public domain and may not be for some time (if ever) due in part to the unethical behaviour of some of the people/organisations referenced earlier.

[andrewjcw]

I may have missed it, but have you done an actual FTP rest with your PM yet?

If so, you'll need to extrapolate that result into all of your applications and they'll set your power zones - I'd take the power zones from Training Peaks, or Today's Plan, etc, over the Strava designated zones.

Then, when you start a training plan, it will dictate your power intervals, etc, towards whichever goal you have chosen to achieve.

I've found the Strava F & F to be similar, but subtly different to Today's Plan & TP...

cheers

I did a hard 20 mins around a flat oval for my ftp test. That's where the 307 comes from, actual avg was 324 or something. I just like having the ftp number (or at least whatever you want to call a 0.95*watts/20min best number) as a way to measure progress. My training and riding will essentially be going for pr's on segments ranging from 1min to 15mins that I can ride to from my house mixed in with commuting and the odd longer ride.

I guess in 6 months if I'm stuck and frustrated maybe you'll get me reading books and writing up 12 week training block plans, but I just don't see it as necessary yet.

[MichaelB]

but the thing that is confusing me, is that whilst they use the same data, the curves look similar, but the actual numbers are different ??

I see you have opened that can of worms. That's to be expected given all the options out there.

And for the layman, its not easy to interpret and understand who is doing it right (or good enough) and underlines the importance of having someone like Alex in your corner assisting with the interpretation and planning rather than blindly trusting something, just because they say they do it rifght .

The problems arise when you seek to be somewhat more specific in how to interpret and use the data, and of course as you have discovered, in what absolute values and rate change values make sense relative to your individual training experience.

And this is the rub. For what I am doing/aiming (at this stage) is using the PM and a curve (whether it be from Strava/Stravastix/Training Peaks/Todays Plan or whoever) is to get a more informed measure of what I am doing and how I respond to it, and as some sort of pacing tool.

Actually, lets be honest. It's a toy that provides some numbers and some form of indication of how unfit I am and something to play with. There, that's out of the way now.

The aim is to get more knowledge and maybe use it better, and form a better 'training' plan, but at the moment, the cycling I do is more about general fitness (physical & mental).

Thanks once again Alex for a sensible and accurate summary. Much appreciated


On a side note, having had a real small play with a couple of programs, one thing that REALLY annoys me, is when you get one feature or display methodology in a web based program, that you don't get the same functionality in the app. Grrrr

[queequeg]

but the thing that is confusing me, is that whilst they use the same data, the curves look similar, but the actual numbers are different ??

I see you have opened that can of worms. That's to be expected given all the options out there.

And for the layman, its not easy to interpret and understand who is doing it right (or good enough) and underlines the importance of having someone like Alex in your corner assisting with the interpretation and planning rather than blindly trusting something, just because they say they do it rifght .

The problems arise when you seek to be somewhat more specific in how to interpret and use the data, and of course as you have discovered, in what absolute values and rate change values make sense relative to your individual training experience.

And this is the rub. For what I am doing/aiming (at this stage) is using the PM and a curve (whether it be from Strava/Stravastix/Training Peaks/Todays Plan or whoever) is to get a more informed measure of what I am doing and how I respond to it, and as some sort of pacing tool.

Actually, lets be honest. It's a toy that provides some numbers and some form of indication of how unfit I am and something to play with. There, that's out of the way now.

The aim is to get more knowledge and maybe use it better, and form a better 'training' plan, but at the moment, the cycling I do is more about general fitness (physical & mental).

Thanks once again Alex for a sensible and accurate summary. Much appreciated


On a side note, having had a real small play with a couple of programs, one thing that REALLY annoys me, is when you get one feature or display methodology in a web based program, that you don't get the same functionality in the app. Grrrr

You also have the option of getting an online coach from someone like FTP Training. Prior to my injury, this is what I was doing. The coach has access to your Today's Plan (or whatever they use), and can prescribe a training plan that meets whatever your goals are. That does two things. 1) It holds you accountable to something, and 2) You can just go ride and leave the mental work to someone who knows what the data means and can ensure you are not burning out.

[MichaelB]

Thanks queequeg - whilst that is an option, it means that rather than just ride and play it by ear sometimes, I would need to do specific 'things' and repeats and sometimes have to think about the ride and what I am doing rather than sometimes just ride.

I guess it is for that reason, and willing to admit, the PM is more of a toy and something to think I'm 'using'.

I've never been interested in doing repeats, racing, or following a plan (whether is a sensible and good idea).

BUT, in buying the PM, I've actually thought about how to ride better and learnt lots so far.

[Alex Simmons/RST]

Pithy Power Proverb: "The power meter keeps an accurate record of my decline" - Robert Chung

[queequeg]

Thanks queequeg - whilst that is an option, it means that rather than just ride and play it by ear sometimes, I would need to do specific 'things' and repeats and sometimes have to think about the ride and what I am doing rather than sometimes just ride.

I guess it is for that reason, and willing to admit, the PM is more of a toy and something to think I'm 'using'.

I've never been interested in doing repeats, racing, or following a plan (whether is a sensible and good idea).

BUT, in buying the PM, I've actually thought about how to ride better and learnt lots so far.

It does really depend on your goals. If you want to, for example, get better at climbing, then you won't magically just get better by "just riding". You need to put some focus on your climbing with some structured riding that lends itself to getting better at climbing. And you know what gets you better at climbing? Yep, doing repeats of climbs

It does work. Before my Injury I had been riding for 7 years without really much improvement. I had a coach for 12 weeks and I increased my FTP by 50W, and smashed out almost 1000 PRs on various segments. I was knocking minutes off my previous best climbing efforts.

[trailgumby]

Pithy Power Proverb: "The power meter keeps an accurate record of my decline" - Robert Chung

It's certainly been doing that for me the past 12 months. My PE kept me off the bike for months. And just as I was finishing Base 1 and starting to make progress again, I sustained a grade 3 ankle sprain on Christmas Eve. Another month off the bike, minimum.

[find_bruce]

It would build your strength

[Alex Simmons/RST]

looks a bit like Malcolm Turnbull with a cigarette

[g-boaf]

looks a bit like Malcolm Turnbull with a cigarette

It's a photoshop effort from a satire website.

The original image: https://image.shutterstock.com/z/stock- ... 877366.jpg

Sorry Bruce. Took about 2 seconds on Google.

[kb]

[...]
Then just take that 20min power figure and multiply it by 0.95, and that's it.

As a starting point. Or a bit more or a bit less. You’ll get to know yourself after a while. I seem to be more around .925 with my current mix/fitness and having a quick drop off in my mean maximal power curve.

[find_bruce]

looks a bit like Malcolm Turnbull with a cigarette

It's a photoshop effort from a satire website.

The original image: https://image.shutterstock.com/z/stock- ... 877366.jpg

Sorry Bruce. Took about 2 seconds on Google.

I hadn't expected to be taken seriously. ... I'll see myself out

[MichaelB]

It does work. Before my Injury I had been riding for 7 years without really much improvement. I had a coach for 12 weeks and I increased my FTP by 50W, and smashed out almost 1000 PRs on various segments. I was knocking minutes off my previous best climbing efforts.

I have no doubt it works, but PR's is not what is my driving force. I like getting them, especially when I've put in a solid effort, but it's not why I cycle.

At the moment its been only 2 weeks, so still truly in the data gathering, increasing understanding and just thinking stage. One thing I'm pretty sure of though, hill repeats are not on my list of "things to do".

[queequeg]

It does work. Before my Injury I had been riding for 7 years without really much improvement. I had a coach for 12 weeks and I increased my FTP by 50W, and smashed out almost 1000 PRs on various segments. I was knocking minutes off my previous best climbing efforts.

I have no doubt it works, but PR's is not what is my driving force. I like getting them, especially when I've put in a solid effort, but it's not why I cycle.

At the moment its been only 2 weeks, so still truly in the data gathering, increasing understanding and just thinking stage. One thing I'm pretty sure of though, hill repeats are not on my list of "things to do".

I forgot to add, I didn't actually go out and try to set PR's, nor did I have the PR's as my goal. They were just a consequence of my focus on building my FTP to prepare for Europe, which was Phase 1. Phase 2 was to use that as a launchpad to have a goal of being able to ride 3 x 1hr climbs per day for 21 days straight without blowing up.

In my initial rides pre-coach, I would be at peak effort pulling 300w avg up a climb. After the 12 weeks I was putting out 365W avg in a climb at 80% Heart Rate, and as a natural consequence of that I was smashing all my previous PR's without trying. I never once went out on a ride saying "I must ride this hill in 10min". It was always "I am going to ride this hill at 320W avg power"

Here's an example: https://www.strava.com/activities/861889873

You can then zoom in on the hill repeats and look at the power, and it a relatively flat line, allowing for changes in the terrain.

When I am out "just riding", I mostly don't bother looking at the power numbers as they mean nothing. On a long ride, I use the power on climbs to ensure I don't go deep into the red, and therefore I set a power target and stick to it. Outside of that, the PM just gathers data to upload later and give an overall TSS to track fatigue so that I know when I should be giving the body a rest, and making it a bit more predictable.

[MichaelB]

Now, if I was going to do 2 weeks in Europe again or an event like the Haute Route like others have mentioned above, I'd be getting a coach for sure.

I've already done my 2 week trip and as much as I'd love to go back, I don't think it will happen soon