Trainer Road FTP test results (Page 2)

I'll throw in my answer to this question as this subject's very interesting.

1:26 standalone HM, 37min standalone 10k, 41min Oly 10k.

I *suspect* that I am nowhere near 4.0w/kg, but am much higher than where I was 1 - 1.5 years ago.  Does that mean that continuing to gain power until I reach "equilibrium" with my run pace should be easier than, say, if my 10k was currently at 45min?

I think the bike FTP = run FTP seems pretty accurate. I'm definitely a stronger biker than runner, but working on improving the run a bit. Current FTP is around 4.5 w/kg (340 watts @ 75kg) and open half is 1:23, but aiming to get down around 1:20 for a race at the end of may. My coach says my goal target is a 1:25 HIM run in mid June. Let's not relate run or bike to the swim, as I definitely need some help there!

I'd say that's pretty light.

The reason I asked the question about the 40min Oly run, is because I suspect anyone running sub 1:30 for HM, should be able to run 40min or under for an Oly.  In your case, and I suspect yazmaster, I would guess that you over cooked the bike, or swim, and faltered on the run.  The drop from an open 10K of 37min to 41min in an Oly seems too great a gap IMO.

You're right, it is too large of a gap.  Interesting thing though is that in that particular race my bike placing vs AG was noticeably worse than my run place.  It was extremely hot that day.

I suspect that if I ran an oly now my run would be in the 39-40 min range.

Are we talking about a TT position FTP?  When my cycling fitness is highest, my w/KG is above 4.0, but I'm still far from a 1:30 half marathon even if I were to transition to a run focus.  Probably closer to 1:35-1:37.  But I put out way more power on my road bike.  I'd say I'm closer to 3.7-3.8 w/kg in my TT position.

Well I guess the thread is totally off topic now, so here we go in another direction-

I don't think i'm going to buy a power meter any time soon.  I honestly don't think I would use it for the majority of my rides anyhow, other than to have another number to check once I download my Garmin.

On the other hand, doing a FTP test a few times a year would surely be beneficial.  There are some bike shops with Computrainers that I could potentially use.  Would a CT be just as accurate as a regular power meter?  Also, what would you guys recommend as the test protocol?  I'm thinking I don't want to do a whole hour test, so maybe 10 min warmup, 20 min test, 10 min cool down?

I wouldn't expect that a 1:30 open HM should yield a sub-40 Oly run split. My best HM is 1:23, and I have yet to break 40 in an Oly run, though I could probably just about do it on the right course. A general guideline would be that Oly 10k pace should be around the same as open HM pace. That would call for a < 1:24 open HM time in order to expect a < 40 Oly split.

As for the earlier question: will it be easier for somebody to reach 4.0 w/kg if s/he is already at the "run equivalent" of 9 miles/hour? I think it would depend on the cause of the imbalance. A person in that situation already has some of the elements needed to pull off 4.0 w/kg, but we don't know why they lack the others. If it's because the other components are for some reason less adaptable in that person, then the run speed confers little advantage. If it's because the other components simply haven't been trained adequately yet, then I think we could be more confident in that person's ability to reach 4.0 w/kg.

No, I definitely disagree with your assumption.

A 1:30 open HM does NOT translate to a 37-40min triathlon Oly 10k. In fact, plug that into Mcmillan, and you get an OPEN 10k of 40:22.

This means for an Oly tri, you're looking at 41-42, which is exactly what my triathlon Oly 10k split is. 

Mcmillan's estimate for an 38:xx open 10k is a 1:24:43 HM, which is almost exactly the numbers I had as a pure runner a bunch of years ago. (I'm a bit slower now as a triathlete on the run!)

Yeah, got derailed a bit

I think computrainers are pretty accurate, at least more so than TR and a trainer. Most important is to have the information be consistent over time. Using the same CT every time would be best.

The traditional test is warm-up, 5 minutes all out, 10 minute recovery, 20 minute test, cool down. This removes the anaerobic power from the test (in theory).

And 64 Kg may be light, but it's not light for a biker

I didn't think so either at first, but that's about what my HM would have been if I'd run one near the time of the Oly. The McMillan calculations also match up fairly well to those points in time after factoring in a few things like general fatigue and course difficulty. Now think about the guys running 32 or faster. o.0

What kind of biker? That sounds close to the GC mountain guys and they tend to be smaller. I'm like Hincapie at 80-82kg.

... but without the juice

I am reeling this thread back in for a few minutes...............

I tested on Trainer Road again last night, only doing the 20 minute test as some of you had suggested.  I have even bigger problems now,  I have a bigger power number.  My FTP went up to 339  after the test.  I think I'm going to make a call to Garmin this morning and let them know I'll heading to the Tour this year and they should reserve a spot for me.  And of course I'll be expecting my own set of needles and other doping equipment as well.

 New FTP: 339

I am planning on using the numbers to provide information for training and not worry about "real" numbers until I actually have something real to use.  As others have stated, these number can be just as productive when looked at from the right perspective as real numbers, so I'll take them in stride and keep working.

BTW, I am really try to get a grip on the discussion that you hijackers have brought into this discussion.  It is interesting to read of a correlation to cycling power and run speeds as well.  The last open HM I ran was in 2011 and I ran a 1:33:20 and my Oly 10k 43:01 with no stand alone 10k ever being run.  Since last HM, I have run 2 marathons and I'm 8 weeks out of Boston, so in the middle of more marathon training.  That said, I'm trying to grasp what is being discussed about running.

 Keep the discussion going as I'm not sure what questions to ask just yet.  And I'm out of time to reread for the morning before work.

I vaguely remember trainer road showing this somewhere and that it was at least fairly consistent from time. When compared to a powermeter, it read like 30 watts off, but was about that each time. The consistency in the setup matters noticeably, however.

So long as things like peanut butter, cookies, and ice cream don't make the list!

This is where my thinking came from.  I have a HM time of 1:33, but last year had an Oly time of 42:07 - and this was suffering with a bad stitch the first 2-3km, so I negative split the run.  Assuming (big assumption), I didn't stitch I would have expected my run time to be closer to 41 flat. 

At a 15-17s coastdown, you'll likely have less resistance than the KK power curve is based upon, so the numbers will be a bit high (by the way, I'm not sure that % difference in coastdown time amounts to a corresponding % difference in wattage for a given speed). From what I've read in various places, people find that around a 13s coastdown from 20mph will give you a pretty good approximation to the KK power curve.

Or, better yet, the new KK InRide device attaches to the trainer and measures coastdown time, adjusting the power curve based on that. DCRainmaker seemed to think that it was fairly reliable.

OP - all the relevant info has been said already.

- Your absolute number on virtualpower is not valuable for comparison to powercharts. Fuggedaboutit. 

- Virtualpower should be fine for training purposes as long as your trainer (which you have yet to still specify the model) is consistent. KK and CycleopsFluid2 are two well tested ones for consistency.

I don't think it's particularly useful to speculate about rolldowns or other means of calibrating short of checking your particular unit against a powermeter standard, or having your actual FTP corroborated on a Computrainer or accurate powermeter. 

Yeah its approx. as I only have 1 sec resolution and move my eyes from CPU to watch. I also have no tire slippage which is key in the setup.   I've read before that 13.4 sec is suppose to be dead on to the curve but when they say this it was at the start not on a warmed up unit.  KK states 15 sec coast down on the trainer webpage so maybe this is for a warmed up unit.  Who knows. 

The InRide looks nice but I'm currently Android not I-apple so its useless for me.  Can't wait until Google/Android gets their $hit together and provides OS API support for BLE.

I agree with your entire post.  There's no sense, in my mind, trying to get really close to my real power tap FTP numbers since I already know there are problems associated with the scheme  I am using right now.  I'll keep plugging away using the current plan and hope I can smoke someone in a race or two this season.

BTW, my trainer is a CycleOps Magneto.  Someday when the budget comittee convenes and give me the necessary approvals I'll jump on with a KK trainer.  But right now I have a lot of other more important stuff to buy.  I'm riding the bottom Tri bike with all stock equipment so a new trainer isn't high on my priority list right now.  I would like to get a goood set of race wheels before I splurge on a power tap or new trainer.

I think the problem is the curve for the Magneto. The curve that you are using was one of the first ones we added and we just have never had that much feedback on it. Looking at it compared to the data from this chart, it looks like our curve is quite a bit higher than the graph.  

http://www.cycleops.cz/files/trainers.pdf

We've seen discrepancies in some of the curves from that chart, too, but in this case, I'm guessing you'd see a more realistic number. I'll look into creating a new curve that matches the graph. To give you an idea, 20mph with our current curve is 368W. On the cycleops chart 20mph looks to be about 250w. 

If you have access to a power meter, we could create a new curve based on data from that, too.