Losing faith in HR monitor for calories burnt
ConicalFern
Posts: 121 Member
This morning I went a long bike ride (66.7 miles; 3h47m) and was looking forward to a nice big number of calories burned to eat into throughout the day. However, my Garmin gives me a pretty feeble 1678 calorie burn estimate. In contrast, Strava gives me a much more rewarding 2219 calories (so a 25% difference). The difference between the two is that Garmin uses HR data to estimate calories burnt, whilst Strava uses the elevation profile to estimate power required to maintain a given speed.
So why the difference? I suspect that because it was pretty cold (34F/1C), my HR was pretty low, thus the Garmin thinks I wasn't really exerting myself - but I was - honest!
I don't think it's just that Garmin's estimates are just lower. Normally I find Garmin and Strava estimates to be pretty similar. Furthermore, I looked at a similar ride in the summer (77F/25C)and calculated the calories per mile at 35, whilst today's was 25, despite being of similar topography and speed (today's was actually both more hilly and faster). My average HR for the warm ride was 158 whilst today it was 131.
So the question? Partly just sharing my wisdom - going out for runs in the cold might give you a lower calorie estimate than the 'true value'. Also, has anyone had similar problems? Should I just use the Strava estimate and not worry that I might be eating into my deficit for the day?
Potentially niche topic I know, apologies.
So why the difference? I suspect that because it was pretty cold (34F/1C), my HR was pretty low, thus the Garmin thinks I wasn't really exerting myself - but I was - honest!
I don't think it's just that Garmin's estimates are just lower. Normally I find Garmin and Strava estimates to be pretty similar. Furthermore, I looked at a similar ride in the summer (77F/25C)and calculated the calories per mile at 35, whilst today's was 25, despite being of similar topography and speed (today's was actually both more hilly and faster). My average HR for the warm ride was 158 whilst today it was 131.
So the question? Partly just sharing my wisdom - going out for runs in the cold might give you a lower calorie estimate than the 'true value'. Also, has anyone had similar problems? Should I just use the Strava estimate and not worry that I might be eating into my deficit for the day?
Potentially niche topic I know, apologies.
0
Replies
-
Power to calories burned depends on efficiency curve which is far from constant. AS it was colder, your energy transfer to the outside WAS easier. Try to run/ride on a hot humid day, it is actually harder than on a cool dry day.
So it could be either way.
http://www.runnersworld.com/sweat-science/how-does-heat-affect-fat-burning
1 -
HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.0
-
rileysowner wrote: »HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.
Sorry no - newer HR monitors using variance in HR analysis are as closer to energy use for non-stready state cardio including lifestyle activity. See the FirstBeat research for this. They report a maximal error of 7-10%.1 -
I've used Strava, Garmin, Runkeeper and different Polar HRMs.
They all give different calorie counts - sometimes wildly different.
The closest I get to consistency is with a Polar HRM that has been calibrated with my tested VO2 max and max HR settings. Then I get very similar figures across the FT60 HRM, a Wattbike Pro (power meter trainer), Concept2 rower power outputs/calories and ARC Cybex cross trainer power outputs/calories under suitable steady state conditions.
Still don't know if it's accurate of course!
My HR is very affected by temperature - indoor training especially. Much higher pulse for same power output when I get hot. First half hour of a fast ride on a trainer might see both Polar and the power meter telling me I've burned around 400 cals, second half hour the Polar will be 100 cals ahead despite keeping a constant power output.
There will be some extra burn as my body is pumping extra blood to the skin surface trying to keep temperature down but can't quantify that.
Garmin estimates I do find low generally, but I use it for very long rides for convenience, plus on long rides I can't eat enough anyway so it's a moot point.
Strava doesn't know I have appalling aero so I discount it.
Runkeeper seems reasonable for general non-exercise rides but too generous for fast rides.
In the end accuracy isn't as important as consistency and adjusting overall calorie balance based on results over time.0 -
EvgeniZyntx wrote: »rileysowner wrote: »HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.
Sorry no - newer HR monitors using variance in HR analysis are as closer to energy use for non-stready state cardio including lifestyle activity. See the FirstBeat research for this. They report a maximal error of 7-10%.
Also, wasn't the issue with non-steady state cardio that you would get overinflated estimates due to the HR going up and down?
OP, I'd hardly call 1678 feeble.
Personally, I err on the side of caution and go with the lower one. If you really want, split the difference.0 -
I've used Strava, Garmin, Runkeeper and different Polar HRMs.
They all give different calorie counts - sometimes wildly different.
The closest I get to consistency is with a Polar HRM that has been calibrated with my tested VO2 max and max HR settings. Then I get very similar figures across the FT60 HRM, a Wattbike Pro (power meter trainer), Concept2 rower power outputs/calories and ARC Cybex cross trainer power outputs/calories under suitable steady state conditions.
Still don't know if it's accurate of course!
My HR is very affected by temperature - indoor training especially. Much higher pulse for same power output when I get hot. First half hour of a fast ride on a trainer might see both Polar and the power meter telling me I've burned around 400 cals, second half hour the Polar will be 100 cals ahead despite keeping a constant power output.
There will be some extra burn as my body is pumping extra blood to the skin surface trying to keep temperature down but can't quantify that.
Garmin estimates I do find low generally, but I use it for very long rides for convenience, plus on long rides I can't eat enough anyway so it's a moot point.
Strava doesn't know I have appalling aero so I discount it.
Runkeeper seems reasonable for general non-exercise rides but too generous for fast rides.
In the end accuracy isn't as important as consistency and adjusting overall calorie balance based on results over time.
Agreed 1000xs.
I just started using Strava so I haven't really compared.0 -
-
EvgeniZyntx wrote: »rileysowner wrote: »HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.
Sorry no - newer HR monitors using variance in HR analysis are as closer to energy use for non-stready state cardio including lifestyle activity. See the FirstBeat research for this. They report a maximal error of 7-10%.
I was unaware of this. Thanks for the info. Off to read!0 -
3dogsrunning wrote: »EvgeniZyntx wrote: »rileysowner wrote: »HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.
Sorry no - newer HR monitors using variance in HR analysis are as closer to energy use for non-stready state cardio including lifestyle activity. See the FirstBeat research for this. They report a maximal error of 7-10%.
Also, wasn't the issue with non-steady state cardio that you would get overinflated estimates due to the HR going up and down?
OP, I don't really think it is a huge difference for that period of time. I'd hardly call 1678 feeble.
Personally, I err on the side of caution and go with the lower one. If you really want, split the difference.
Yep, you are right.
I tend to go with the lower burn also.0 -
tincanonastring wrote: »EvgeniZyntx wrote: »rileysowner wrote: »HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.
Sorry no - newer HR monitors using variance in HR analysis are as closer to energy use for non-stready state cardio including lifestyle activity. See the FirstBeat research for this. They report a maximal error of 7-10%.
I was unaware of this. Thanks for the info. Off to read!
Here is the Firstbeat white paper section: https://www.firstbeat.com/science-and-physiology/white-papers-and-publications/0 -
Why are you concerned with how many calories you've burned?
Exercise is for fitness, not weight loss.
Weight loss happens by eating less.
If you've eaten the calories you should (the amount to support your weight - 100 to 500)
and you're really hungry at the end of the day, have a snack of a couple hundred calories, maybe
1/4 to 1/3 of the most conservative estimate of what you've burned that day.
If you're consistently losing weight faster than is healthy, adjust your calorie goal slightly.0 -
Check your HR graph to see if it dropped out any during your ride. In warm weather your sweat helps the sensor pickup your hr. In cold weather you don't sweat as much plus the air is drier, which causes the skin under the sensors to be drier. This can make it harder to detect your hr.
If your hr is dropping out early in your workout, then the simple fix is to put some water on the skin under the sensors. If it is dropping out later in the workout, then you might need to use a hr gel.0 -
Also you do burn more calories in cold than when it's warmer due to the fact that the body is trying to keep warm. Sweating is an indication of exertion, however it's primary is to cool the body down.
A.C.E. Certified Personal and Group Fitness Trainer
IDEA Fitness member
Kickboxing Certified Instructor
Been in fitness for 30 years and have studied kinesiology and nutrition
0 -
Why are you concerned with how many calories you've burned?
Exercise is for fitness, not weight loss.
Weight loss happens by eating less.
If you've eaten the calories you should (the amount to support your weight - 100 to 500)
and you're really hungry at the end of the day, have a snack of a couple hundred calories, maybe
1/4 to 1/3 of the most conservative estimate of what you've burned that day.
If you're consistently losing weight faster than is healthy, adjust your calorie goal slightly.
Seriously? The deficit should be from your TDEE, which includes those calories burned from exercise.0 -
EvgeniZyntx wrote: »Power to calories burned depends on efficiency curve which is far from constant. AS it was colder, your energy transfer to the outside WAS easier. Try to run/ride on a hot humid day, it is actually harder than on a cool dry day.
So it could be either way.
http://www.runnersworld.com/sweat-science/how-does-heat-affect-fat-burningAlso you do burn more calories in cold than when it's warmer due to the fact that the body is trying to keep warm. Sweating is an indication of exertion, however it's primary is to cool the body down.
A.C.E. Certified Personal and Group Fitness Trainer
IDEA Fitness member
Kickboxing Certified Instructor
Been in fitness for 30 years and have studied kinesiology and nutrition
That's very interesting; I hadn't considered that the relationship between watts and calories might change with temperature as well. I guess there are two important questions that it raises:
- Is the difference between 0 and 20 degrees the same as the difference 20 and 40? As Ninerbuff points out, just as the body has to work hard to maintain body temp at 40, should it not also have to work hard to maintain it at 0?
- Does the change in HR follow the same pattern as the change in the relationship between watts and calories? Maybe the decrease in HR is in response to lower metabolism, in which case continuing using HR would be the best option
0 -
tincanonastring wrote: »Seriously? The deficit should be from your TDEE, which includes those calories burned from exercise.
I was about to ask about the same thing, before MFP dumped me out of my comment.
I'm not an expert by any means, but if CICO is the basis of weight management, OP would need to net enough calories for health and meeting whatever weight goals. If I burned 1200 calories, I'd be netting 400 calories!
0 -
3dogsrunning wrote: »EvgeniZyntx wrote: »rileysowner wrote: »HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.
Sorry no - newer HR monitors using variance in HR analysis are as closer to energy use for non-stready state cardio including lifestyle activity. See the FirstBeat research for this. They report a maximal error of 7-10%.
Also, wasn't the issue with non-steady state cardio that you would get overinflated estimates due to the HR going up and down?
OP, I'd hardly call 1678 feeble.
Personally, I err on the side of caution and go with the lower one. If you really want, split the difference.
Using old data, I was conservatively estimating that I would burn 2000 calories, I was just a bit put out when the value so much lower than I was expecting!0 -
3dogsrunning wrote: »EvgeniZyntx wrote: »rileysowner wrote: »HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.
Sorry no - newer HR monitors using variance in HR analysis are as closer to energy use for non-stready state cardio including lifestyle activity. See the FirstBeat research for this. They report a maximal error of 7-10%.
Also, wasn't the issue with non-steady state cardio that you would get overinflated estimates due to the HR going up and down?
OP, I'd hardly call 1678 feeble.
Personally, I err on the side of caution and go with the lower one. If you really want, split the difference.
I was thinking this as well. That's about 450 calories per hour.
Then again, the METS value for cycling that fast is 12, which would be way more than 450 calories per hour.
Are you sure you went that far? That's 18mph for almost 4 hours.
0 -
Why are you concerned with how many calories you've burned?
Exercise is for fitness, not weight loss.
Weight loss happens by eating less.
If you've eaten the calories you should (the amount to support your weight - 100 to 500)
and you're really hungry at the end of the day, have a snack of a couple hundred calories, maybe
1/4 to 1/3 of the most conservative estimate of what you've burned that day.
If you're consistently losing weight faster than is healthy, adjust your calorie goal slightly.
You're aware the whole idea of this site is to ensure that, assuming you want to lose weight, calories in must be < calories out? Given that, it's important to know the calories out. I don't really want to have a huge deficit for any given day, so I would rather have some idea of how many calories I've burnt.
I'm always hungry so I find it hard to use this as a measure of how much food I should eat! Plus, I've read articles suggesting that cyclists tend to over estimate calories burnt on a ride and overeat, and I don't want to do that.0 -
Check your HR graph to see if it dropped out any during your ride. In warm weather your sweat helps the sensor pickup your hr. In cold weather you don't sweat as much plus the air is drier, which causes the skin under the sensors to be drier. This can make it harder to detect your hr.
If your hr is dropping out early in your workout, then the simple fix is to put some water on the skin under the sensors. If it is dropping out later in the workout, then you might need to use a hr gel.
Thanks, but no visible dropouts - plus on the ride itself I did notice that my HR was a bit lower than it normally is.0 -
ConicalFern wrote: »Why are you concerned with how many calories you've burned?
Exercise is for fitness, not weight loss.
Weight loss happens by eating less.
If you've eaten the calories you should (the amount to support your weight - 100 to 500)
and you're really hungry at the end of the day, have a snack of a couple hundred calories, maybe
1/4 to 1/3 of the most conservative estimate of what you've burned that day.
If you're consistently losing weight faster than is healthy, adjust your calorie goal slightly.
You're aware the whole idea of this site is to ensure that, assuming you want to lose weight, calories in must be < calories out? Given that, it's important to know the calories out. I don't really want to have a huge deficit for any given day, so I would rather have some idea of how many calories I've burnt.
I'm always hungry so I find it hard to use this as a measure of how much food I should eat! Plus, I've read articles suggesting that cyclists tend to over estimate calories burnt on a ride and overeat, and I don't want to do that.
+1 My broken sense of hunger is why I'm here in the first place!0 -
WalkingAlong wrote: »3dogsrunning wrote: »EvgeniZyntx wrote: »rileysowner wrote: »HR monitors are only accurate as an estimate for steady state cardio. While a bike ride could be steady state, I know in my rides there is a fair amount of variation due to the fact that there are hills, valleys, wind, and even changes in road surface. All of that makes it less true steady state. Add to that the inevitable HR drift that comes from long periods of time spent at an exercise activity which will skew the results, and the HR calorie estimate gets less and less accurate.
Sorry no - newer HR monitors using variance in HR analysis are as closer to energy use for non-stready state cardio including lifestyle activity. See the FirstBeat research for this. They report a maximal error of 7-10%.
Also, wasn't the issue with non-steady state cardio that you would get overinflated estimates due to the HR going up and down?
OP, I'd hardly call 1678 feeble.
Personally, I err on the side of caution and go with the lower one. If you really want, split the difference.
I was thinking this as well. That's about 450 calories per hour.
Then again, the METS value for cycling that fast is 12, which would be way more than 450 calories per hour.
Are you sure you went that far? That's 18mph for almost 4 hours.
Cycling is my main sport that I've done for most of my adult life, so.. yes
. Plus the Garmin has GPS for distance etc. Average speed for the ride was 17.6. 0 -
.0
-
Why are you concerned with how many calories you've burned?
Exercise is for fitness, not weight loss.
Weight loss happens by eating less.
If you've eaten the calories you should (the amount to support your weight - 100 to 500)
and you're really hungry at the end of the day, have a snack of a couple hundred calories, maybe
1/4 to 1/3 of the most conservative estimate of what you've burned that day.
If you're consistently losing weight faster than is healthy, adjust your calorie goal slightly.
@MKEgal
Not everyone on this site is massively overweight and inactive - or even seeking to lose weight.
Cyclists in particular can have very high calorie burns which they need to make allowance for.
0 -
Firstbeat is helping improve accuracy for non steady state activity but HRMs still are not accurate calorie counters.0
-
Wow, I'm surprised you need to worry about your intake at all with that sort of activity, assuming it's not a rare event!
If you want a third estimate take your BMR per minute X 12 METS times minutes of the ride. It's going to be higher than both of those, I think. Even with my puny BMR it'd be over 2200 calories.0 -
WalkingAlong wrote: »Wow, I'm surprised you need to worry about your intake at all with that sort of activity, assuming it's not a rare event!
If you want a third estimate take your BMR per minute X 12 METS times minutes of the ride. It's going to be higher than both of those, I think. Even with my puny BMR it'd be over 2200 calories.
I worry about my intake at that level of activity because I don't want to undereat.0 -
WalkingAlong wrote: »Wow, I'm surprised you need to worry about your intake at all with that sort of activity, assuming it's not a rare event!
If you want a third estimate take your BMR per minute X 12 METS times minutes of the ride. It's going to be higher than both of those, I think. Even with my puny BMR it'd be over 2200 calories.
I can eat a lot I guess... I ate 410 calories on the ride, plus a breakfast that was probably 300 calories bigger than I would normally have, plus 270 when I came in for recovery, which totals to 980 extra calories consumed already. If we're going with the lower estimate that 'only' gives me 698 calories for the ride, which isn't a great deal more than I might burn in a big gym session, but a bike ride makes me a lore more hungry, so I have to be careful.0 -
3dogsrunning wrote: »WalkingAlong wrote: »Wow, I'm surprised you need to worry about your intake at all with that sort of activity, assuming it's not a rare event!
If you want a third estimate take your BMR per minute X 12 METS times minutes of the ride. It's going to be higher than both of those, I think. Even with my puny BMR it'd be over 2200 calories.
I worry about my intake at that level of activity because I don't want to undereat.
The OP said he didn't want to overeat. I can see it both ways. It just seems like if you can do that sort of athletic feat, I figure you must be Superman, and Superman should be able to eat freely without calcing out his burns and measuring his food. Clearly I'm no Superman, and not around many!
0 -
WalkingAlong wrote: »3dogsrunning wrote: »WalkingAlong wrote: »Wow, I'm surprised you need to worry about your intake at all with that sort of activity, assuming it's not a rare event!
If you want a third estimate take your BMR per minute X 12 METS times minutes of the ride. It's going to be higher than both of those, I think. Even with my puny BMR it'd be over 2200 calories.
I worry about my intake at that level of activity because I don't want to undereat.
The OP said he didn't want to overeat. I can see it both ways. It just seems like if you can do that sort of athletic feat, I figure you must be Superman, and Superman should be able to eat freely without calcing out his burns and measuring his food. Clearly I'm no Superman, and not around many!
I also don't want to overeat but I want to eat enough to support training and to lose a bit of weight.
I haven't done too many 4 hour rides, but I still counted when I did.0
Categories
- All Categories
- 1.4M Health, Wellness and Goals
- 393.4K Introduce Yourself
- 43.8K Getting Started
- 260.2K Health and Weight Loss
- 175.9K Food and Nutrition
- 47.4K Recipes
- 232.5K Fitness and Exercise
- 427 Sleep, Mindfulness and Overall Wellness
- 6.5K Goal: Maintaining Weight
- 8.5K Goal: Gaining Weight and Body Building
- 153K Motivation and Support
- 8K Challenges
- 1.3K Debate Club
- 96.3K Chit-Chat
- 2.5K Fun and Games
- 3.7K MyFitnessPal Information
- 24 News and Announcements
- 1.1K Feature Suggestions and Ideas
- 2.6K MyFitnessPal Tech Support Questions