How accurate are Heart Monitors?
 

I started using a polar f4 for the first time today. If this is accurate I'm in worse shape then I thought.

They are moderatly accurate at tracking your heart-rate... you wouldn't see one in a hospital though.

In terms of shape, heart rate data can only be used in conjuction with other data like watts or perceived exertion.

It's about as accurate as you can get without some really expensive equipment. Definitely close enough for tracking purposes; way closer than those online estimators.

I think they are very accurate. IMO, the most important factor with heart rate is the recovery rate, when you stop exercising, how fast does your hr come back down.

Polars are about +/- 1%. Close enough. Make sure you're drawing the right conclusions, though. HR isn't perfect for training, but it will work if you know how to use it.

I doubt it.. maybe over a larger period, but not a minute. There are also various inefficiency in the contact area and transmission can cause gaps. The best is an electrode cemented on, with a wire. Certainly something a polar meter is something you only want to rely on in "general" and not obsess about 3-4bpm over one minute.

Yes, the true heart rate is just that, it's a rate. You need a certain period to establish a rate and this period can screw data up like your hearts response time to a sudden increase in load. There are other issues to think about here - like cardiac drift.

HR training will work.. don't get me wrong but you need to combine it with something like a wattage meter and most definitely using a perceived exertion scale.

Just quoting the manual for mine. And even 5 bpm is close enough for exercise. Heart rate varies so much from day to day that the absolute numbers aren't so critical.

This study (first link, pdf available, discussion on page 7) concludes that they're good enough to be valid for measuring R-R intervals at rest. The issue doesn't seem to be as much with the accuracy of the equipment, but with wearing the strap correctly and it getting moved around during exercise.

For cyclists the biggest problem would probably be the jersey flapping around in the wind, giving false high readings. That and the occasional high-voltage power line, which can interfere and give a zero reading.

Polar HRMs are very accurate, IME. Pretty much 1-beat accuracy. It tells you what you need to know. Believe it, except when it jumps up to like 220 during a ride. That's jersey flap. Rewet the transmitter with spit and that'll usually go away. My wife restarts hers after every rest stop, though I don't have to do that.

Even with exact measurement of the heart beat, heart rate is an interpretation of the data. I do not know what algorithm Polar uses. There are all sorts of ways to come up with a heart rate. The easiest would be to take the last 2 beats, measure the interval between and lift the rate from that. I don't think that kind of method would be the most accurate. It might be best to do a moving average of the last 5 heart beats intervals or better yet a weighted average with more weight with the more recent heart rate.

But what the cyclist REALLY wants to know is their heart rate at this instant in time. If your heart rate is speeding up or slowing down, the monitor will be a bit inaccurate because it has to use the past data for its estimation. One could come up with a method of estimating the instantaneous heart rate by the rate of change in the recent intervals but that seems like a bit too much.

Doesn't really matter, because HR takes a while to respond to changes in body function. So whether the monitor lags a bit is immaterial. It lags a lot less than the heart itself. That's the reason that short high intensity intervals are done w/o regard to HR.

This is my problem - the sampling rate and the period over which it is taken... the device must be accurate to track this over a wide range. There is data to be gained by tracking these changes in athletes and those with cardiac problems. The standard HRM will average out any small perturbations to be safe. That is why i don't think they're accurate in the medical sense.

What is it showing you that makes you think you're in bad shape?

Here's a simple test:
1) Put on heart rate monitor.
2) Use fingers to check pulse over 10-60 seconds.
3) Write down average from heart rate monitor during same time period.
4) Repeat for many different heart rates over the range from your min to max.
5) Make a scatter plot; plot measured heart rate in step 2 on the abscissa (x-axis) and that from step 3 on the ordinate (y-axis).
6) Use an ordinary least squares regression to calculate the slope and intercept. The slope of the line should be 1. The intercept should be 0.
7) Plot regression line on scatter plot.
8) Post to BF!

I was wondering the same thing. Maybe a high HR for a slow/medium speed?

You forgot:

9) ???
10) Profit!

;)

Just for kicks, I put on two HRMs at the same time. A Polar FT7 (with matching wrist computer) and a Bontrager Strap (feeding a Node 2 computer). They moved almost precisely in concert. Almost eerily so. Step for step, never varying by even one BPM and even updating at the same time.

I wondered if both computers were somehow feeding off of the same strap, but my understanding is that Polar and ANT+ are totally incompatible data transfer protocols and that this wounldn't be possible. I should have shot some video, it was pretty amazing to me and made me much more confident in both meters.

Now the total calories burned that they show... that's another story entirely... let's just say that I WISH I could eat an extra Ribeye and baked potato every time I workout and still lose weight as the polar would seem to indicate...

This only makes me more suspicious! Haha.