Got a power meter - first ride... Observations and questions

CptKernow wrote:

However, what was the biggest surprise was how it felt easier than expected to knock out higher watts as soon as the road turned up, but on the other hand it felt incredibly difficult to keep the watts up on the flat.
On the hills I would regularly be hitting 5w/kg without thinking about it. On the flats maintaining 3w/kg took serious concentration.

I read a bit about what the causes this discrepancy, and apparently I'm not alone. Seems strange though as I never had problems knocking out this kind of power on the turbo.

So, would this be something to target, i.e. sustained flat power?

This is a pretty normal revelation for those seeing their actual power output for the first time. Most don't realise how much they crystal crank the minor descents nor how hard they punch a rise.

It's why when we talk about good pacing in time trials often the immediate perceived sensation of intensity is misleading when riding on variable gradient terrain.

That said, other than for purposes of learning to pace effort for optimising speed (e.g. in a TT, or managing yourself in a lone breakaway), there is no specific need to be concerned with evening out the effort a great deal. Races don't work that way in any case, and neither does any group ride on the planet.

That said, some days you may want to even out the effort a bit more, while on other days you may want to attack them more. Depends on what it is you are looking to achieve for that day.

CptKernow wrote:

Also, any other suggestions for training with a power meter outdoors would be appreciated. Thanks

As to suggestions, firstly just ride and collect data. Don't change what you are doing right away - it's helpful to gather data on what you've been doing, Then you can review and begin to make some assessment as to whether or not some changes are in order.

Avoid the temptation to stare at the numbers lest you might crash or injury someone else.

Learn the features of your meter and learn how to use it correctly, learn about calibration and torque zero and whatever else may be required to ensure your data is as good a quality as it can be.

Eventually you might look to do some specific test efforts to get a sense of your capabilities: short duration sprint efforts, hard maximal aerobic efforts of a few minutes, longer hard aerobic efforts of 20+ minutes.

Perhaps get your hands on the book "Training and racing with a Power Meter" by Allen & Coggan. Full of all sorts of useful information, advice and ideas about how to get much more out of your power data.

Ask questions.

The main thing I noticed was that I'd subconsciously ease off on the flats, or I used to have a problem where i'd go out too hard and then fade a lot when riding in general, as I went off perceived exertion before i had much riding experience, but that basically boiled down to the fact that higher wattage can feel easy at the start, then suddenly gets harder. Then once you are tired, doing 100w less can feel just the same. Pacing and discipline is definitely something that comes with tim

Though one main thing i will chip in regarding actual averages is that I tend to find over intervals on the flats, the actual average can come out a fair bit lower than you might anticipate.

For example, you could be constantly riding to say 250w on your display, but then when you end your ride and do your analysis, the average comes out at like 235w for that block.
I tend to find that this is the result of "micro-rests", i.e. momentary external factors that can cause you to stop pedalling for a few seconds over the course of an interval, e.g. going around a corner, slowing over a pothole.

It is rather annoying as these can make your averages feel lower than that on the turbo for a pretty much equal effort, given that 1 second here and there doesn't give you any time to recover. Mathematically speaking, it's just down to averages, so for example, 10s of not pedalling across a 10 minute interval will yield 98.3% (590/600) of your "target power", i.e. which will knock your 250w average down to 245w. Which can feel pretty significant for a random second here or there. (You may look down constantly at your head unit and see 250w, then be surprised when you see the average being lower)

It tends to happen far less up hill as you are both moving slower and thus have less ground to cover and less chance of needing to avoid something, or negotiate a turn.

The reason I mention this is because it can be a very valid explanation for why power numbers can be lower outside, or even why there can be a pure power discrepancy between riding up hill and on the flats. Most people don't consider micro "rests" to actually impact the average a lot, but they do. Equally, if you try to compensate the live interval average, you end up having to ride above your target power which can cause you to fatigue if you aren't careful.

As an example, making up for that 10s at 0w would require you to make up 2500J worth of exertion, which can either be riding at 500w for 10s, or riding at 290w for a minute. Crazy huh?! (The gap gets more extreme at higher wattage intervals, given you need to make up for more work done, and can be a reason why it can perceivably take longer to boop your long interval averages up in the real world, and equally why TT's often yield a lower avg power than what you actually target)

I personally just employ the strategy of knowing that on average, these drops are going to happen most of the time anyway, so I tend to just stick to my target power and not go over it, knowing that over time, the averages will increase anyway. Trying to make up for a dropped average can mean training in the wrong zone and ruining the quality of future intervals, or tiring yourself out too early. (By this i mean if I want to target 250w lets say, i'll try to always get my power to that, and I don't display ride average,as this often causes me to surge to bringi it back up. I then just accept that after the ride, the actual average may come out a bit lower.)

I'll add that normalized power does tend to remove these anomalies and is more representative of the physiological cost, however you generally only see NP for the whole ride, rather than an individual interval or segment, and it is said that most NP algorithms do not work accurately for a duration of less than 15 minutes anyway. Another solution is to remove zeroes from the average, however you can then fall into the otherside where you don't consider valid rest enough. While 1-2 seconds may not be enough to get recovery, 5-10s is, and not including those zeroes would also be a disservice.

Real eye opener for me was to see how unbalanced my legs were, as in the power difference, taken me a while to train out the imbalance which was a good 17% between legs, its now under 5% which for me is liveable but on the plus side a smoother pedal makes you quicker as your no longer surging on one leg.

+1 for Xert. I'm using the Connect IQ app to do power based workouts out on the road and it's pretty cool overall. The big selling points for me are flexible training plans without the need for formal FTP testing. Its modelling of continuous Maximum Power Available (MPA) seems very innovative and accurate. Definitely worth checking it out with the 4 week free trial.

MishMash95 wrote:

Though one main thing i will chip in regarding actual averages is that I tend to find over intervals on the flats, the actual average can come out a fair bit lower than you might anticipate.

For example, you could be constantly riding to say 250w on your display, but then when you end your ride and do your analysis, the average comes out at like 235w for that block.
I tend to find that this is the result of "micro-rests", i.e. momentary external factors that can cause you to stop pedalling for a few seconds over the course of an interval, e.g. going around a corner, slowing over a pothole.

It is rather annoying as these can make your averages feel lower than that on the turbo for a pretty much equal effort, given that 1 second here and there doesn't give you any time to recover. Mathematically speaking, it's just down to averages, so for example, 10s of not pedalling across a 10 minute interval will yield 98.3% (590/600) of your "target power", i.e. which will knock your 250w average down to 245w. Which can feel pretty significant for a random second here or there. (You may look down constantly at your head unit and see 250w, then be surprised when you see the average being lower)

But are you monitoring 'lap power' as you ride? I've got that on a screen just for reps and hill climbs along with 3s and 10s power and gradient. Just hit the lap button and you're away. Then if desired you can just focus on that figure.

The real eye opener for me, was finding looking only at speed was a load of b*ll*cks. On one stretch of road I'd do 20 to 25 mph but according to the power meter, it wasnt me (or the Strava KOM) that was fast but the slight downhill gradient and the tail winds, the power output was quite low for that speed, hardly z1. So now I think looking at speed is irrelevant when its affected by so many outside variables, the power meter is the true gauge of how well your cycling. 200w is still a 200w effort, no matter what the wind or gradient is doing.

Interestingly one or two hills around here must be convex shaped, because I can see the power ease off as I crest the summit. Its good to have the power meter so I can keep the power on.

Every 6 weeks or so I do a 20 minute FTP, to calculate zones. Then just use these to gauge the effort of my cycling. Aiming for an endurance zone on long rides, tempo on normal rides or threshold / VO2 max on hard efforts.

ben@31 wrote:

The real eye opener for me, was finding looking only at speed was a load of b*ll*cks. On one stretch of road I'd do 20 to 25 mph but according to the power meter, it wasnt me (or the Strava KOM) that was fast but the slight downhill gradient and the tail winds, the power output was quite low for that speed, hardly z1.

Haha yep that was one of the first things I discovered too - there can be big difference between going out and riding to achieve a high average speed or a high average power! Plus holding a constant power for a whole ride is pretty much impossible.

CptKernow wrote:

Just got myself a Favero Assioma power meter.

Great choice, I've had a BePro and it's been faultless for 3 years.