Power numbers from a TT to estimate FTP?

bigpikle

Did my first TT last night - a sporting 11.4 miler - and GC gave me a new CP power figure as I set some new PB's all the way from 11 to 30 minutes

I know its said not to use power numbers from races to determine ftp/zones etc but does a TT count in that? It was a sporting course, so not a steady effort as it had peaks for various hills, so doesnt equate exactly to a flat 20 min steady paced interval, but does this really matter? Is it a good representation of power for determining progress?

I guess it does break an obvious rule in that its not the same course, conditions etc as my previous 20 min power test, so maybe its not ideal....?

jibberjim

From a 30minute solidly paced effort then that is a pretty good test and can be used for measuring FTP. The test is relatively short so not your best marker for an FTP test but if you've only been doing it for 20 minutes previously then it's likely better than that.

doyler78

What was the VI? Were there are any long uphills/downhills?

If you were using 20 min power figures and you have beaten that as you suggest then your FTP is clearly going to be set too low now (assuming you have the relationship between FTP and 20 min power licked) and how low wil be determined by how well you paced the TT. If it was well paced ie VI between 1.00 and 1.02 (assuming that was even possible on that course) then just go with your standard formula however if it was more variable than that then your FTP will probably be a wee bit higher.

BTW properly paced maximal effort TT's (though admittedly the longer 25mile/40km TT) are actually the recommended approach to determing FTP so no issue with using that.

Race data can be very useful in determing FTP also but you have to careful about how you use it. If you done a 1 hour crit and rode it very hard then your Normalised Power for that ride would give you a very big hint about where your FTP lies. If, however, you done 2 hour race, even though it was hard, it probably will not be that great at helping determine FTP as your effort will be too long most likely and thus 1 hour NP will probably be underestimating FTP. It's about context.

Good stuff in any case in beating all your records from 10 mins onwards. It's a great feeling when you do that

Alex_Simmons/RST

Bigpikle wrote:

I guess it does break an obvious rule in that its not the same course, conditions etc as my previous 20 min power test, so maybe its not ideal....?

Provided meter is correct, then more power is more power, no matter the course.

If you produce less power on a given course, then it's not your physiology that's holding you back.

Pokerface

doyler78 wrote:

Race data can be very useful in determing FTP also but you have to careful about how you use it. If you done a 1 hour crit and rode it very hard then your Normalised Power for that ride would give you a very big hint about where your FTP lies. If, however, you done 2 hour race, even though it was hard, it probably will not be that great at helping determine FTP as your effort will be too long most likely and thus 1 hour NP will probably be underestimating FTP. It's about context.

Just curious about this comment.

I have power data for 10 and 25 mile TT's, plus NP power from a 2-hour RR yesterday.

Using the 10 mile TT figure (at around 95%) gives me the highest FTP - but closely matches my best 1-hour NP block from the RR. These figures are both higher than my best 25-mile TT (just over 1 hour) wattage. All within 10w of each other though.

Which (if any) should I base my FTP on then? I assume the 1 hour TT figure.

Anonymous

I'd go for the 1hr TT...

Alex_Simmons/RST

Pokerface wrote:

I have power data for 10 and 25 mile TT's, plus NP power from a 2-hour RR yesterday.

Using the 10 mile TT figure (at around 95%) gives me the highest FTP - but closely matches my best 1-hour NP block from the RR. These figures are both higher than my best 25-mile TT (just over 1 hour) wattage. All within 10w of each other though.

Which (if any) should I base my FTP on then? I assume the 1 hour TT figure.

Note that:

From a hard 1-hr ride:
AP <= FTP <= NP
where:
- FTP will be closer to NP than AP
- For a steady state type effort, then AP approaches NP anyway, so it's a moot point
- presumes AP & NP are being calculated correctly (some hardware and / or software combinations don't)

Ratio of FTP to 10-mile or 20-min power varies considerably (rarely is it 95%, more likely less), so better to use 1-hour power and note the current ratio for future reference.

You could also perform another maximal test along with the 10-mile / 20-min TT data to estimate FTP. Adding a maximal test of ~ 5-min within a week or so of the 20-25-min effort would do the trick, and then can use the critical power model to estimate FTP. That's helpful if 1-hr long stretches of road for quality testing are hard to come by.

PS - nice work at the World Cup. Hope the drug testing wasn't too much hassle.
I was pretty happy with my power in the TT.

Alex_Simmons/RST

Pokerface wrote:

doyler78 wrote:

Race data can be very useful in determing FTP also but you have to careful about how you use it. If you done a 1 hour crit and rode it very hard then your Normalised Power for that ride would give you a very big hint about where your FTP lies. If, however, you done 2 hour race, even though it was hard, it probably will not be that great at helping determine FTP as your effort will be too long most likely and thus 1 hour NP will probably be underestimating FTP. It's about context.

Just curious about this comment.

As for the comment about picking data out of a longer race, it's about context. You can still get good data from such races for the purpose of assessing changes in fitness, it just depends on how the race pans out as to whether the data is valid for the purpose of assessing changes in fitness.

e.g. there might an explosive opening which you survive, then everyone left settles down for a steady effort, before the finale when all hell breaks loose. Or there might be a cracking long climb in a race that will be a selection point and up to then you haven't taxed yourself that hard.

Depending on how long such sections are determines whether or not such data is useful for assessing FTP.

If you do a 2-hr ride/race with an NP > 1.0x, then it's pretty likely FTP is underestimated.

bigpikle

thanks for the help guys...

From GC:

xpower = 244w
av power = 238

so I see the VI as 1.02. As I said, it was a sporting course, and I'm a newbie at pacing TT's, so the power profile is fairly undulating as well. GC has given me a new CP of 231w, so I'm thinking I'll use that number for calculating power levels from now on in GC and get out and try a 20 min test on my standard flat course soon and see whats changed.

Best news is GC has bumped up my CP figure from 210w at end of Jan to 231w so I'm doing something right and seem to be making reasonable gains.

Enjoyed the TT as well so will do some more of those, but sadly my work means huge amounts of travel, so actually being able to make my local events is a real challenge during the week

amaferanga

Bigpikle wrote:

thanks for the help guys...

From GC:

xpower = 244w
av power = 238

so I see the VI as 1.02. As I said, it was a sporting course, and I'm a newbie at pacing TT's, so the power profile is fairly undulating as well. GC has given me a new CP of 231w, so I'm thinking I'll use that number for calculating power levels from now on in GC and get out and try a 20 min test on my standard flat course soon and see whats changed.

Best news is GC has bumped up my CP figure from 210w at end of Jan to 231w so I'm doing something right and seem to be making reasonable gains.

Enjoyed the TT as well so will do some more of those, but sadly my work means huge amounts of travel, so actually being able to make my local events is a real challenge during the week

Couple of points. Firstly, my experience with GC v3 (which has NP, TSS, CTL, etc.) is that NP is almost always greater than xPower, sometimes quite considerably higher. So when people talk about NP it doesn't follow that the same goes for xPower. Secondly, if you are letting GC calculate your critical power, do you know which shorter duration effort it is using and when you did that effort? It'd be pretty meaningless if it was taking a 2min effort from last year for example (not sure if GC picks recent maximal efforts only or if it picks season best or even all time best). You might be best to calculate your critical power manually in GC by selecting recent maximal efforts of suitable durations.

bigpikle

^^ good question....

I dont have an answer on how GC calculates it, but I only have data from the last few months, so it wont be too far away, and as it has all improved significantly recently, all the better numbers are very recent anyway.

I dont have the issue of cherry picking historical numbers as I dont have any, and I was even slower across the board last year As time goes on though, it will be an important thing to bear in mind so I need to make sure I do it manually for sure.

Interesting that NP seems lower than xP though, as I was surprised the VI was so close to 1 given the course and the numbers I saw out there. I guess that makes te VI low as well....

Oh well, maybe time for a good 20 min test later in the week on my standard course!

doyler78

Just download v3 of Golden Cheetah and see what your VI is under the Coggan metrics (NP/AP) then you don't have to wonder

You can get it here in case you didn't know:

http://goldencheetah.stand2surf.net

You can run both GC2 and GC3 side by side and it's easy to import all your GC2 files into GC3.

jibberjim

Bigpikle wrote:

I dont have an answer on how GC calculates it, but I only have data from the last few months, so it wont be too far away, and as it has all improved significantly recently, all the better numbers are very recent anyway.

It picks it from the entire period of the graph - obviously that's not so good.

Go to the metric tab though, and create a season (shift click and drag) for a recent week or whatever where you did maximal efforts in 3-5 and 15-30 region. And then the CP curve will be calculated just from those when you view that season.

Alex_Simmons/RST

Data for CP calculations should be from maximal efforts within about a week or so of each other.

Cherry picking from over many weeks/months is an example of what I call sin of sin #8:
http://alex-cycle.blogspot.com/2009/07/ ... ftp-2.html

bigpikle

missed that v3 was available for the mac...just d/l'ed it but havent found how to see NP, TSS, IF & VI in the standard ride data screen? I have added it to the interval metrics but cant get it to display it for the entire ride?

Surprised how different NP seems to be to xP though....

doyler78

You need to go into the Tools -> Options -> Interval Metrics (or hit the settings button if it is available) and add TSS, VI, NP from the left panel to the right so that these metrics appear in the interval summary that you will get on the ride plot. Once you have saved that then go

Ride Plot and set your interval for the TT effort (if you haven't already) (Shift + left click and drag to highlight to set interval. Look on right and go into the interval section and down the bottom part of the panel you will see the interval details which will include those metrics you added.

I'm using windows so I hope the Mac screen is the same but I think I recall there are differences and from what I've seen on the GC forum I think they have been struggling with the Mac build but that bit should work.

bigpikle

cheers - lots of gaps in the mac build it seems so far, and I'm struggling, but can make it work at a basic level. Will be great to align terminology with TP and Coggans stuff but I think its a little while away yet, and hope hope hope it can be applied retrospectively to ride data so I can rebuild the PM using TSS etc.

doyler78

That's a pity about GC3 on Mac but I suspected as much from what I had read.

Just had a thought re: Mac. Are you using a powertap? If so you can use the poweragent 7 software to set and calculate intervals. I haven't used it myself but worth a shot.

doyler78

Failing that there are excel spreadsheets out there that will calculate NP, TSS & IF.

bigpikle

its not a problem as I simply use Bikescore for the PM and will continue to track xpower and IF etc using that.

As Alex said above, power is power, so I can still measure progress based on 20 min test intervals and best intervals on longer regular rides. GC is a great piece of software that provides everything a sportive rider, new tester and 2nd year CX vet might need to plan, measure and track training