You sir, are a negator.

(10 points to anyone who gets that reference)

---

**************
Too f@ckin depressed from various injuries to care about having a signature line.

Sponsored by Blue Shield PPO.

While power meters have an error range of +/- 2% or so, they are at least consistent from one ride to the next, provided the rider zeroes out the baseline reading. So, when a guy says he improved by, say, 10 watts, there isn't a need for an error range of a material size. Even if the precision is +/- 2%, that's only a 0.2 watt range around the reported result.

As for an absolute power reading, what is the 95% confidence interval around a ride-long average of, say, 200 watts, when my power meter is +/- 2% accurate*? I'll leave that to you, the statistics expert. But, I'll give everyone else a hint: You can have great confidence in your power meter.

*Assuming that +/- 2% is two standard deviations around a mean of zero; normally distributed.

Indeed, it seems complicated. When in doubt, JFT!

---

---

Steve Fleck @stevefleck | Blog

I think he still has a problem that MIT turned him down when he applied.

---

________________________________________________

Pasadena Tri Club

You do realise that the power meter isn't the only source of error.

---

Swimming Workout of the Day:

Favourite Swim Sets:

2020 National Masters Champion - M50-54 - 50m Butterfly

please elucidate ...

OK. If you are reporting on the effectiveness of a particular training program, you might use normalised power output over a given course to show the improvement. i.e. do a TT on March 1, and another TT on April 1. Hopefully there has been an improvement. Assuming that there is no error in any of the measurement tools you use, there is still error in the results simply because the tests are being done on different days.

Or to put it another way, think of a javelin thrower. The power meter is analogous to the tape measure used by the officials on the field. There is essentially zero error in the tape measure, but different throws by the same individual travel different distances because of small changes in conditions, fatigue, and inconsistencies in how the thrower performs the skill.

i.e. error is more than just measurement error.

---

Swimming Workout of the Day:

Favourite Swim Sets:

2020 National Masters Champion - M50-54 - 50m Butterfly

okay as you're in Halifax and can throw rocks at me I won't complain too loudly :). But Ashburn is quite aware of that sort of thing (NP vs. AP). No one should properly compare the two.

Simple AP is AP we trust within the accuracy limits of one's power-meter. It doesn't matter about the date, the wind, the air pressure. Within measurement limits, non-adjusted power is just fine.

I won't argue the typical repeatability (or precision if you prefer) is THAT high (0.2 %) but I'm willing to venture it's better than 1% for my PT Pro (compared to SRM Pro and a CT).

all the best ....

Not debating AP vs NP, or anything like that. The PM is just a tool for measurement, error is introduced by the tool, but also from other sources.

another example, I tell you to ride a 10mile TT at a constant 250 watts with one helmet, and repeat with another helmet, and you do that once for each. say your AP is 250 watts and NP is 250 watts on the button each time, and your PM is so good that it measures to an accuracy of +/- 0.00000%. You were 5.00 seconds faster with helmet A than helmet B. All other equipment was identical. Wind speed was 10 km/h and direction was from the SSE on both trials. How much faster is helmet A than helmet B??

---

Swimming Workout of the Day:

Favourite Swim Sets:

2020 National Masters Champion - M50-54 - 50m Butterfly

I've been waiting for months now for someone with a Power Tap to finally tell us how many angels can dance on the head of a pin.
.

---

Bob C.

The "science" on any matter can never be settled until every possible variable is taken into account.

There is statistical error and systematic error. One may know nothing of the statistical error, which isn't always so bad, but saying nothing of systematic error is unacceptable.

I'm talking about the power meter ONLY. I fully understand the vagaries of field testing and drawing conclusions. Even the wind tunnel has it's limits.