mgopack42 

Ahem... you mean "on the gripping hand" me thinks!

Carbonfiberboy 

My first outdoor ride with the Powertap Pro+ was RAMROD, 153 miles and 9400'. I've done this ride many times, so I kinda know what to expect and how to apportion my limited strength. I used my Edge 800 as the receiver. Having power was a big help. I noticed my long climb average power numbers from Strava, using only HR, were about 131w. I figured I could ride a bit harder than that and see what happened. I held power between 130w and 146w on all the climbs and came up with ~135w averages. At the top of the last climb, I had trouble holding 130, so I knew I’d done it about right.

Some observations: A big help from power was with gradient changes. It’s easy to shift or change cadence too early or too late. One doesn’t really notice the slight changes in pedal effort, breathing or HR, but the meter notices. Makes it easy to correct immediately and hold power steady. It was also nice to be able to monitor strong efforts, like bridging or holding the wheel of a faster rider through small rollers. Easy to see when it’s worth it and when not. I set an upper limit for efforts at 250w.

I had the Garmin set to 3 second averaging, smart recording.

Before I uploaded to TrainingPeaks, I set my FTP to 180. Using that estimate, TP said I had 8 minutes between 189w and 200w and then 23' between 200w and 540w, which makes no sense. I also uploaded to Strava. I don't have Summit for my single bike Strava account, but Elevate works. It has reasonable numbers. With the same FTP, Elevate showed 9', 5', and 3' for zones, 5, 6, and 7. Do you know why this difference between software?

I see lots of tiny spikes on both the TrainingPeaks and Strava graphs, though many more on the TP graph. I assume the TP spikes are the reason for the crazy Z7 numbers. The Strava spikes look like they happen after short periods of coasting, then resuming pedaling.
.
The meter kept varying on the climbs, between like 106w and what I was supposed to be doing, then hold steady for a while, then drop out again. I'm really sure I was holding a steady pedal pressure. Anything to do with the spikes? Batteries?

Thanks for all the help.

rubiksoval

Accumulated minutes in an 11 watt range would certainly be smaller than the amount of time spent at 200+.

11 watts is a really small window, even more so when you consider you're only seeing 3 sec averages. 23 mins above threshold isn't really uncommon or abnormal, especially on way long rides. I've done 20-30 mins above threshold in just an hour long crit before.

Hermes

I get similar results. It is very common for me to have 20 minutes of threshold and 8 minutes of VO2 and 3 minutes of ANO2 in a two hour ride especially when riding with a training partner.

jadocs

What graph are you talking about in the non premium Strava account? The speed graph?

Power spikes are normal. Using a PM is helpful to smooth out your application of power. You’ll get better at it, which will make you more efficient. As an aside, the best way to do that is to take it off 3 second average and go with real time power.

How come your not referencing the power analysis in the Elemnt Bolt app? It’s light years better than Strava.

Carbonfiberboy 

Of course a 10 hour ride which one is riding as a TT, almost entirely solo, has a much different strategy than a 1 hour crit, which is the reason I was a bit surprised to see those higher numbers.

Weird about those TP zone numbers which are contradicted by Elevate.

I think you're saying that that intentional higher power numbers just happen. Understood. Like I rest my butt every 10 minutes on climbs. It's more comfortable for me to pick up speed out of the saddle and I don't look at the PM when I'm doing that because it's unimportant. I'm listening to my legs. I did watch my power while making a few efforts, nothing over 275. So yeah, I'll have higher power numbers here and there and they add up.

I understand that either TrainingPeaks or Strava ignores zero power numbers, The other one doesn't ignore zeros and counts them in the average, believing that coasting enables one to go harder when one starts pedaling again. I don't know. One does get a little rest on descents but not that much. A fast descent is quite a strain, really.

Carbonfiberboy 

If one clicks on Analyze in free Strava, one sees graphs of speed, power, heart rate, cadence, and temperature. The spikes I see in Strava are very short, some over 300w, and preceded and followed by zeros. I know that I was not coasting, then briefly sprinting, then coasting again. I see similar spikes in TrainingPeaks. During climbs, I saw very brief power dropouts while I was pedaling very consistently. My guess is that it's the PT's batteries. I'll change them. Or possibly because my Garmin was set to Smart Recording. I've changed it to 1 second intervals.

I'm not using an Elemnt Bolt because I use a Garmin Edge. I upload and display my data on my desktop.

Seattle Forrest

Can't, the 800 won't show instantaneous power. There are a lot of data fields for power, but that isn't one of them.

I thought I was the only one who liked real time power output.

Seattle Forrest

Smart recording means only write a new point you the file when your speed or direction changes, eg when a new point is required to show this. It came from the days when disc space was expensive. It's still useful for sending less days over the wire. (I read that smart recording hits the battery harder because of the processing that needs to happen every second.

Every second, it's obvious that that means.

Your PM sends your head unit your power every second.

None of us have access to Garmin's code, so no one knows for sure. But it definitely sounds like smart recording and power measurement don't go well together.

Also, it's considered good practice not to use auto pause.

jadocs

Ok, I gotcha... I've been so used to only using the mobile app. Didn't really pay attention to the desktop site which does have a power graph. As you ride more with the PM, you will see that it does take a conscious effort to smooth out the power where it's not jumping in 80w intervals as an example.

A personal observation on my part is when I do a threshold effort for 1hr. I setup my screen to just have power and speed displayed. With speed displayed, I find myself doing what I call "micro recoveries". By this I mean, the effort is so hard that if I know the speed is up there....my mind subconsciously takes a break from pushing and I can see the momentary drop in power even though I am still pedaling.

Carbonfiberboy 

Thanks. That's helpful. The biggest reason I've always wanted to use power instead of HR is to get a more accurate TSS and thus a more accurate TSB. Auto-pause was useful with HR for getting a more accurate hrTSS, but I can see that it wouldn't affect power TSS, though it will affect average power.

I'll get back on my bike sometime to test all this. My butt's still sore and my TSB is still a -5, having done nothing since RR. I might go for a brief run today.

rubiksoval

Yeah. Strava doesn't ignore zeros. It does use a different algorithm for NP (they call it weighted average power) and TSS, though, not weighing higher power numbers as much as WKO/Garmin NP.

Seattle Forrest

Zeros while you're coasting belong in the average if they're what happened. I agree that they're a period of rest and allow you to put out more power subsequently than if you had been pedaling all along. But a simpler reason to include them is that they're the truth.

Carbonfiberboy 

Thanks. I was wondering about that terminology. My free Strava gives 2 WAP numbers: the Coggan calculation (131w for my RAMROD ride) and Strava's WAP of 126w. Does it matter what I look at?

TP also gives a NP of 131w FWIW.

Carbonfiberboy 

Got it.

OBoile

I think the difference isn't necessarily when coasting, but when you've stopped and the activity is auto-paused. At least, that's my guess from what I've seen in my results.

Carbonfiberboy 

Thanks. I'll turn it off, that should then eliminate the difference . . .

Heathpack 

Why?

I use auto pause because the WKO algorithm as I understand it continues to rack up TSS when you are stopped but not paused. If you do a long ride with multiple regroups and a coffee stop, all that non-moving time inflates your TSS and imo you’re therefore better off with the pause.

Whats your rationale behind stating that use of auto pause is not good practice?

Seattle Forrest

@Heathpack

I would manually pause for that. 🙂

There are two reasons auto pause is considered unhelpful with power.

TSS should accumulate while paused per definition and not all software understands the pause. I don't personally agree with this one, I didn't make it up, I can't defend it. I can only say that's part of the received wisdom.

More importantly: auto pause isn't perfect, it doesn't kick in until after you've started to move, and that means you don't record the power it takes to accelerate. Which is usually a harder effort than cruising along at speed. For a ride with 200 stop lights, that can miss a significant bit of work, all stuff that would be weighted heavily.

pdlamb

That sounds like a ride from heck, wouldn't you rather ride rollers all day?

Or it that what it takes to ride out of downtown Seattle?

Heathpack 

Interesting, but definitely not compelling enough reasons for me to discontinue the use of auto pause. Back in the days when I’d manually pause my Garmin at mid-ride stops, I’d forget semi-regularly to unpause which was never a huge issue but still was annoying. I definitely don’t do enough rides with a zillion stop lights to worry about missing data due to that.

We (Coach & I) also don’t care much what the TSS of a given ride is. TSS is tracked by WKO and I’m sure we both keep an eye on it (he never mentions it) but it’s just really one variable I use to come up with a ballpark idea of my “true” state of training stress, factoring in as well what else I’m doing athletically (which we generally track in WKO by me making data files with guessestimate TSS numbers), overall life/work stress, sleep quantity/quality, level of fitness/endurance and so on. I think we both see TSS as a pretty loose number and being off by a few points is no biggie compared to the convenience of having Garmin stop and start automatically.

I even autopause my Garmin for TTs. There’s enough going on in the final 15 seconds before a TT start that monkeying with my Garmin start button is not something I want to have to remember. So my 30 second person goes off, I roll up to the start line, hit “start” on my Garmin while the holder grabs my bike, then I can get clipped in and settled with nothing else to do before my start. Sure, it’s pretty obvious from my file that I miss a few seconds of pretty important data, but in the end the most vital piece of data is my official time and that’s not something I get from my Garmin anyway.

If I stop for a long time on a ride (like lunch or coffee), I just stop my Garmin entirely to avoid major TSS discrepancies on a ride. But in that case I take the Garmin off the bike and just put it in my helmet with my gloves and sunglasses. Somehow doing that, I never forget to put the Garmin back on the bike and hit “start” when I’m riding again.

I think deciding how to set up your Garmin isn’t really a matter of following some kind of common wisdom as understanding the pros and cons of different decisions you’d make and figuring out what makes the most sense for your paradigm. I get it that maybe I don’t care that much about TSS because I have a coach overseeing it all and he’s got an experienced eye on a lot of variables that aren’t captured in the TSS formula, but I don’t think *anybody* should take a TSS number for a given ride as a precise number. Which is to say that a TSS of 70 vs 76 really shouldn’t matter to anyone because the difference is within the margin of error for your “true” physiologic state anyway.

Heathpack 

Looking this up, it appears TSS is based on total elapsed time of the .FIT file and what you have displayed on your Garmin is irrelevant. Which I suppose is why Coach’s instructions are to turn the Garmin off entirely for any lunch-length ride pauses and why he’s never commented one way or the other on the use of autopause otherwise.

This does not address Seattle’s 200 stoplight paradigm but that’s not a scenario that applies to me so not of huge relevance for my situation.

Carbonfiberboy 

Get Out Of Seattle Eastward and Bicycle Under March's Partly-cloudy Skies. PM me.

Carbonfiberboy 

Strava and TrainingPeaks state that their algorithms ignore stopped time when AutoPause is in use. IOW it's like you didn't stop for the purposes of TSS calculation. However if you turn AutoPause off, the time spent not moving is included in the TSS calculation, basically NP*T. OTOH, your NP will be lower when time not moving is counted, so even though all time is then counted, maybe little/no difference in TSS either way?

I didn't look up WKO and other websites.

Heathpack 

See my later post, which I didn’t clearly state but is re: WKO (which is all that’s relevant to me). WKO bases TSS on the total elapsed time for the ride (or so I read). So, autopause or not, if I’m out for 3 hours that’s what WKO will use when calculating TSS. I think this is realistically only going to matter on an epic ride where you’re out all day and have maybe two long breaks.

Interesting that Strava ignores paused time when calculating TSS. I’m a Summit member on Strava (for the beacon feature) and can say WKO and Strava Summit correlate pretty well on TSS, TSB, and related metrics. So by that token, I’d say you get pretty much the same results whether your software counts the paused time (as WKO does) or not (as Strava does).

There’s other sources of discrepancy between Strava and WKO. Mostly it’s due to the fact that I don’t have a power meter on the mtb and Strava seems to base TSS in part on speed. It always assigns me a lower TSS for a mtb ride than I assign myself (and coach thinks I tend to underestimate my mtb TSS). I also don’t create Strava files for non-bike activities like I do in WKO. But still, even though the numbers don’t correlate precisely, you still get the important gist with both software algorithms- you can tell when I’m fit, or rested, etc. Over time you figure things out, like Strava not capturing mtb fatigue well for me, and can grasp the big picture.