NP, NZAP, and AP for Garmin users
On another thread, there's a discussion of NP (etc) coming to Garmin. I'm a Garmin user, and I've always made due with its available averaging...which is either with all data (i.e., including zeros, "AP") or not including zeros (NZAP).
We all know that the thing we are shooting for in pacing on a race is actually the NP (Normalized Power, sometimes also designated Pnorm). The question is how to focus on that number if you aren't using s LYC or Joule. I am talking about a normal EN style ride here, i.e., pretty steady, not a ride with lots of short intervals. In the latter case, we are not so concerned with the NP during the ride.
Here are my thoughts:
1) If you know what IF you ride typically, you can use normal AP, but lower your target by a factor of the VI. For example, I pretty consistently ride around VI = 1.04 on moderately hilly courses, so I could take my target NP of 161 and divide it by 1.04 and shoot for an average watts on the dial of 155. The disadvantage to this approach is that you want/need to have a pretty accurate guestimate of your VI. However, it is pretty conservative to assume a VI of 1.05 on all but the flattest courses, so that's a reasonable base point if you don't have enough data.
2) Since we typically coast several percent of the time, the NZAP is always higher than the AP. By coincidence, the factor for a steady ride is commonly about the same as the VI. But this is just a coincidence and neither perfect in practice nor rigorous in theory. (In fact, it is quite rigorously wrong). Nonetheless, it is often very close. So option 2 is to just use your target watts and set the Garmin to read NZAP.
3) The most conservative approach from a data-massaging point of view is just to watch the AP itself. But this has somewhat less information than you really want...and in fact most people that do this are really doing #1 because you know AP is a lower limit on NP.
I used to use option 2. However, I am switching to option 1 now, at least for steady rides. Below are some real data from a RR ride I just did. I did two out-and-back "loops" of a 28 mi 1-way course, so lap 6 and 12 are short, and laps 1,6; 2, 7; etc are more or less identical terrain.
[I have added one more column, which is the "xPower" calculated by Golden Cheetah or RaceDay. It is a very similar marker to NP, but it handles the data averaging differently. (It still scales power the same way.) XP always comes out a bit lower than AP because it tends to dampen spikes a bit more.]
Lap AP NZAP NP XP VI
1 143 148 147 145 1.03
2 159 162 161 159 1.01
3 161 169 168 165 1.06
4 163 167 168 166 1.03
5 158 164 162 159 1.03
6 150 164 162 159 1.08
7 159 164 162 160 1.02
8 152 163 156 153 1.03
9 157 167 163 160 1.04
10 154 167 163 160 1.06
11 155 167 161 159 1.04
12 150 165 158 154 1.05
First half in entirety 156 161 162 159 1.04
Second half in entirety 155 166 161 159 1.04
The bottom line here is that there is no perfect solution. Terrain dictates (to a degree) how "flat" you can ride your power. As a result, even for intervals of as long as 30 min, the VI is different on my ride. Maybe that's because I'm a little indisciplined, but I suspect I am not too far from the norm.
As a result, my plan for my upcoming race is to take my target watts from the calculator (e.g., gear 2 = 161) and adjust them down by 1.04 (in this case 155), and set my system to measure AP (with zeros), keeping that 1.04-adusted number as my goal average.
I'd be curious to hear anyone else's take on the matter.
Comments
I thought mine were that close - and they are better for longer rides in general - but I've found enough variance that I'm now experimenting with #1.
I'm a little spoiled as I pretty much know my IF ahead of time as long as I have road the course before. In a hilly race i ride 1.03 - 1.04 and in a flat race i'm at 1.01 - 1.02.
I just adjust my target down accordingly and then watch for that number. In otherwards, I am targeting 211 normailized watts at IMWI and I know I will ride with a VI of 1.03. Therefore I know that as long as my Average Power (with zeros) is 205 my NP will be 211. If i didn't know the course I would use 1.04 instead of 1.03 to convert.
So far, I think I'm going to use NZAP for IMOO. Mostly because that's what I've used in training and in my experience is leads to underachieving my target IF, which is ok for me at this point. I thought about switching to AP, or even switching to a joule, but I'm going to leave those things for "next time" and keep with what I'm used to for now. I think the safe play is to target NZAP and try not to coast. I should also probably spend more time caring about current power, and less about lap average, in an effort to stay "in the box"