[00:02]
Loop was released. So, I decided to test it by wearing two Polar Loops for an entire week, one on my wrist and one on my biceps to see how the Polar Loop compares to the competition because a lot of competition has emerged over the
[00:16]
last year. Most prominently, I would say the Garmin Circa and also the Fitbit Air that I'm wearing right here and of course also the Whoop strap, the OG. And we also have the amazed Helio strap. So, a lot of other devices out there and
[00:32]
revisit all of these devices with new data. So, make sure to subscribe if you want to see those results. But today, I'll focus on the polar loop. Well, my two polar loops right here. By the way, for those of you that are new to the
[00:46]
post-doctoral scientist specializing in biological data analysis. And as always, that is the focus of this video, data and data science. But before showing you those results, I do want to highlight why I tested two Polar Loops. One on the
[01:02]
wrist and one on the biceps. Now, regular viewers will know this, but just to clarify, most devices get a much better heart rate signal on a more fleshy area like the biceps versus the more bony wrist. The wrist is actually
[01:15]
one of the worst locations to get good heart rate signals. It's just a things on the wrist and that's why manufacturers produce them for the wrist, but they're actually not the best location to get good signals. Polo
[01:28]
actually sells official biceps bands, so you can wear them on your upper arm. But instead of doing that, I use the two provided straps to make a single biceps band, but of course, probably the dedicated biceps band might be a bit
[01:40]
more comfortable. But with that out of the way, let's get to the results. And let's start with the things that the Polaro Loop did well. and let's then work our way down slowly to the things it let's say struggle with. And one of
[01:53]
the things it did quite well is measure my heart rate during the night. And in these overviews, I'll always show the wrist on the left and the biceps here on the right. And measuring heart rate during the night is one of the easiest
[02:06]
things to do in a way because you're not moving a lot, but you are working in a relatively narrow range of heart rate. So, we're going to compare it to a reference along the horizontal axis right here, which is an ECG chest strap
[02:18]
from Polar called the Polar H10. And then we have the polar loop here along the vertical axis. So, here we have the one on the wrist, and here we have the one on the biceps. And both of those show a nice agreement. So most points
[02:32]
are on or close to this black line right here, which would indicate perfect agreement where each of these single dots is a single 5minute average heart rate measured by either the Polar Loop or the ECG strap. So this looks quite
[02:45]
good. The Polar Loop does tend to detect slightly lower heart rates, maybe by one slightly lower heart rates, maybe by one to maybe 0.5 BPMs. This is true for both the biceps right here and also the wrist. The blue line is a little bit
[02:59]
lower than the black line. And we can see that even better if we take the five nights for which I have data with both the ECG chest trap and the polar loop over the night. So again, the reference here along the horizontal and the polar
[03:12]
loop along the vertical axis. And if we look at these points, we can indeed see that on average the polar loop detected lower heart rate, but if I had a higher rate. So there's sort of a standard offset by about 0.5 BPM. So very minor.
[03:30]
because you just want to know which nights that have a higher heart rate and which have a lower heart rate. And the same is true here on the biceps. So here I would say both on the wrist and the biceps it did a good job. And I wouldn't
[03:43]
say either is a lot better than the other. If we look at some example kn knight with the reference in red and the polar loop in blue green. You can see that as I fell asleep my heart rate dropped and both of them detected it
[03:57]
nicely. So very similar patterns also right here looking very similar. So this is the wrist. Then the same night for the biceps here. Another night first the wrist and then the biceps. And I would say these are looking very similar. It
[04:13]
could be that the biceps is ever so slightly better but not significantly in information from the biceps and the wrist. So here the Polar Loop did a good job. But what about measuring your heart
[04:26]
rate variability during the night? Well, [snorts] this was also quite okay. I [snorts] this was also quite okay. I would say here potentially the biceps on the right did a little bit better. So overall the points here are a little bit
[04:39]
close to the black line. I would say it is relatively minor, but the mean absolute error is smaller here than on the left right here. So this value at 8.7 is a bit higher than this value right here at 7 basically. But both
[04:54]
again are doing a good job. Though if we take a look at the average values over the night, I had a very limited variation here. But there we can see some systematic deviation where the polar loop always detects a higher HRV.
[05:09]
And we see that also right here for the one on the biceps. And also the patterns aren't super clear. So this knight right here should have then ended up right here if we wanted at least a nice correlation and ideally all points
[05:23]
should be on this black line right here. If we take a look at the individual knights over time we do see that they largely match always starting with the wrist right here. We see a generally good agreement and something very
[05:35]
similar for the biceps. Then here on the wrist generally good agreement but these were actually two different knights that I selected as examples. So, it's hard to say. Overall, again, I think they are generally very similar, though
[05:48]
potentially based on this, the one on the biceps is ever so slightly better. So, if you have a chance at night, I would recommend wearing it on the biceps based on these results, but it's a very minor difference. So, it really doesn't
[06:02]
matter that much. What could decide is how good the sleep stages are tracked at night by both the one worn on the wrist and the biceps. And I'm hoping that over time Polar has improved those results because before they were mediocre at
[06:16]
best I would say. So let's have a look. So here we have two confusion matrices always with the wrist first and then the biceps. So these are looking generally similarish
[06:29]
though actually on the wrist the deep sleep agreement was a little bit better. So what do we have right here? We have the reference on top and the loop on the left. And if the two would perfectly agree, all values along the diagonal
[06:43]
right here should be 100%. They're not close to 100%. So my reference also isn't perfect. It's an EEG device that actually measures my brain waves and is good enough to give us a general impression of the sleep state quality of
[06:56]
only tested on me and it's a limited test, but so far the results have been quite consistent over time. So I'm comfortable enough sharing these as a general indication. So let's actually start with the results on the wrist.
[07:10]
Deep sleep groom is actually quite okay at 69% nice. L agreement is mwah to okayish at best at 59% and the same for RAM sleep agreement. Not the worst I've
[07:23]
seen but also not great performance. And we generally have similar patterns for those on the biceps, but again the deep sleep tracking seems to be even a little bit worse. This is a limited test. I would like to do it for even longer. I
[07:36]
had a week right now. And if we take a look at some example nights, we indeed see that much of the deep sleep here by the polar loop on the wrist was the same as the deep sleep detected by the reference device which is on top right
[07:50]
here. However, then some extra deep sleep was detected as well by the polar loop. And if we take a look at the same night right here, but now for the polar night right here, but now for the polar loop on the biceps, we can see again the
[08:03]
major deep sleep agrees quite well with the reference, but then a bunch of extra deep sleep is also detected. RAM sleep agreement is mediocre I would say. So here we have the results for the wrist where we see that some of the RAM sleep
[08:18]
agrees but some is also missed like right here and also some extra RAM sleep is detected. And we see the same for this night right here. Some RAM sleep is missed. Also, some extra RAM sleep is detected. So overall, a mediocre
[08:34]
performer I would say. Now, of course, we can also compare this performance to actually cleaned up these plots right here. But before I show you those results, I quickly want to ask for your support. The easiest way of supporting
[08:47]
channel, liking this video, and then if you want to support financially, you can use my general Amazon affiliate link. Whether you buy the Polar Loop, the Amaze Fit Helio Strap, the Garmin Circa, or you want a Fibbit Air, I also have
[09:02]
strap, and also to my favorite sleep improvement device, the Aleppod. Using those will never cost you any extra and in many cases give you a great discount. memberships which gives you early access to some of my videos and that's in a
[09:17]
small monthly contribution you make. Any or all of this would be greatly appreciated, even just subscribing to the channel. Thanks so much. But let's get back to the results. Now, in all my videos so far, I tend to show both the
[09:29]
devices I tested in 2025 and those in 2026. But for this video, I want to reduce it to just the devices I tested in 2026. So that makes this overview a lot clearer. Along the horizontal axis, we have the average sensitivity over
[09:45]
those three sleep stages. And along the vertical axis, we have the worst sensitivity. And the reason I included both is because I wanted to be good in all sleep stages, which is the vertical axis right here. And I also wanted to be
[09:57]
good on average overall, which is the horizontal axis. And we have our two polar devices right here and right here. So this is the one on the wrist and this is the one on the biceps. And they're really not amongst the best performers
[10:11]
because the best performers are all the way to the top right right here. And that includes one of the main competitors, the Fitbit Air one on both the biceps, but also the Fitbit Air worn on the wrist. We also have the Amazefit
[10:27]
Helio strap worn on the biceps, which seems to do a better job. And actually the Whoop strap is also somewhere here maybe. I don't know by heart, but somewhere amongst the top devices right here. But I didn't retest that yet with
[10:40]
new data in 2026 at least not fully. So I'll add that soon in an overview. But all those are doing better than the Polar Loop which is basically in this area right here. It's not amongst the absolute worst performers which include
[10:55]
older Huawei devices, the Yume band, the Luna band, Chorus devices, but Chorus might have actually updated that. That's also a future video. But getting back to the Polar Loop, I think there are better alternatives there. I think the Polar
[11:10]
probably get it cheaper, but the Fibbit Air at 100 bucks is already doing better right here. That's probably a better recommendation. We also have Apple watches which are doing a really good
[11:22]
job. We had the Whoop strap. The older Aura Ring did quite a good job. And then my favorite sleep improvement device, the A sleep pod is also a great sleep expensive. It's actually device that can cool and heat each side of the bed
[11:36]
independently. For me, it's been a godsend both in winter and in summer because it can cool down in summer. And in winter, I still need it cooler than my partner, so she can set it to her independent temperature and I can sleep
[11:49]
at my comfortable temperature. Now, it also has great sleep tracking, all these get it just for that because it's way too expensive for that. If you still want to know more about it and try it out, use my affiliate link up here or
[12:02]
wallet and the channel at the same time. But for just sleep stage tracking again I would recommend something like a Fibbit Air on the biceps which is based on my testing than the polar loop either on the wrist or on the biceps. We
[12:17]
can also use a different visualization right here which uses a slightly different average sensitivity and Coen's Kappa a very typical sleep metric used in sleep science and we still get very similar results. So the top devices are
[12:32]
somewhere up here I would say [music] and the polar loop on the wrist and the biceps are right there. So again the Fibbit Air also according to these metrics is doing a much better job than the Polar Loop. I'm not saying the Polar
[12:45]
Loop is useless because you're still getting probably pretty decent heart rate at night, pretty decent heart rate variability at night, but not the variability at night, but not the greatest sleep stages. But I suspect if
[12:57]
you have the Polar Loop, you also want to track your exercises and your heart rate during those exercises. So let's take a look at those results. But before to these results. And here I actually included both the new testing I did. So
[13:11]
these are these 2026 testings and also my old testing on the wrist and the biceps. And that sleep score we get is sort of the average sensitivity. And an automatic rating. So I don't decide this. This is just the output of my
[13:26]
data. And then the sleep stage tracking gets a score of okay, which is what it is. Okay, not great, not really bad, but I would probably look into another device if this is what you want to do. But as I said, let's now take a look at
[13:42]
the heart rate tracking performance. And those results are right here. So this is heart rate tracking during exercise. Again, we're comparing against an ECG chest trap, which is one of the most reliable ways to measure heart rate,
[13:54]
which is along the horizontal axis right here. And then my heart rate according to the polar loop on the biceps and according to the wrist, is here along the vertical axis. So, we have the wrist on the left again and biceps on the
[14:07]
right. And we're starting with one of the easiest exercises for a device to track spinning. And we can actually see that the polar loop on the biceps did quite a decent job. So most points are on or close to the blue line, which is
[14:21]
exactly what we want to see. And the correlation is quite okay at 0.99. Now the correlation cannot be higher than one. 0.99 is quite good. However, the one on the wrist has quite a low correlation. Sorry, at 0.57.
[14:38]
So not great is not very good honestly. If we take a look at some example So, here have one spinning session and we can indeed see that right here. The polar loop in red detected a way to low heart rate for a while and most devices
[14:54]
don't really struggle with indoor cycling. For the same session, the polar loop on the biceps did an amazing job. So, almost perfect. Then again, the polar loop on the wrist right here in the beginning detected a way too low
[15:07]
session, the one on my biceps didn't have any issues at all. So on the biceps for spinning or indoor cycling, I can recommend the polar loop. But what about a harder exercise? In this case, running. Well, here we again see a big
[15:25]
difference between the two. And I would say here the results for the one on the say here the results for the one on the wrist are just really bad. So many points are away from the blue line. The correlation is super low now at 0.42.
[15:39]
This is really bad. Whereas on the biceps, it's still okay, I would say, with a correlation of 0.91. And here, I really want to take a look at some of those individual sessions. But for the one on the wrist right here, we can
[15:51]
indeed see that my heart rate, according to the Polar Loop, deviated a lot. Here, they detected a heart rate up to 180, which is super high for me. I almost never get there. Whereas the reference detected much lower heart rate. The one
[16:03]
on the biceps did a much better job. It struggled a little bit in the beginning, maybe because I was still warming up. Otherwise, good job. This run, again, the one on the wrist attacks extremely high heart rates for me, whereas the one
[16:16]
on the biceps was basically perfect. So again, for running, the one on the biceps is probably good enough, but at least based on my testing, I'm relatively skinny. I would say I would not recommend the Polar Loop on the
[16:29]
wrist. Now, it's a limited test because it's my body, but I do think it's take a look at an exercise which tends to be very hard, which is outdoor to be very hard, which is outdoor cycling or biking. And now both the one
[16:42]
on my wrist and the one on my biceps are struggling. So again, left the wrist, right biceps, and both are just not doing a good job. On the wrist, the correlation is 0.54 and on the biceps 0.57. Now, for biking, I should remark
[16:57]
that for the first two bike rides that I did, I started it manually, but then for all the others, I let it auto detect. And there I also noticed with the standard sensitivity, it missed many of the bike rides or only detected part of
[17:11]
them. So now I've set it much more sensitive and it seems to be a bit more reliable, but it still misses parts of my bike rides. These first two right here, I actually started manually. So here we have the one on the wrist where
[17:23]
you can see it detects a way to low heart rate. And this is another bike ride for on the wrist. Again, a way to low heart rate. [music] Then for this exercise on the wrist, it only detected the last one/3 of my bike ride. Most was
[17:37]
missed. It detected a little bit over half and the quality wasn't too bad. So these were the results for the wrist so far. Now moving to the biceps. This bike in the beginning it detected a way to low heart rate. And we have these sudden
[17:52]
jumps in heart rate. Not sure what's going on there. Also right here detecting a way to low heart rate. Here detecting only one/3. Very similar to the wrist actually. And also here only detecting about half. The quality here
[18:07]
isn't too bad. But I'm just not very happy with this. And compared to something like the Vivid Air, Fivet Air does seem to be more complete in auto detection. And the same is largely true for the Whoop strap, but I still want to
[18:19]
retest that as well. So for biking outside, based on my testing, I cannot doesn't seem to be doing well. Even on the biceps, it's struggling. And it's not just the auto detection. Also, when I am recording it, it's detecting way
[18:33]
too low heart rates on me. And for weightlifting, it's a largely similar story right here. So this is weightlifting or the gym. And we can see that in the higher heart rate range, it detects a way to low heart rate. And
[18:46]
regular views will know that this means that at the peaks of your exercises, so when you're putting a lot of force on your weights, it's not able to detect there's either a lot of tension on your wrist or on your biceps. I've not seen
[19:00]
it so severely for the biceps before, but this does seem to be what's going on. So here we have a wrist result. Basically, none of the peaks according to my reference right here were detected by the polar loop. Also, right here,
[19:13]
none of the peaks were detected on the biceps. Again, a wrist exercise. None of the peaks were detected. And the same for the biceps right here. I'm also not sure what these sudden dips in heart rate are. This might be an export issue.
[19:26]
Actually got the data directly from Polar if I'm correct. So, I'm not sure what's going on here, but overall doesn't change any of the results. this Let's actually compare these results to the previous testing that I did with the
[19:40]
Polaro Loop, either on the biceps or on the wrist. And then we have the 2026 results, which are the new results. So, I'll mark them in yellow. So, the yellow results are the new results from today. And the score we calculate is the
[19:55]
average sensitivity over the cardio exercises which would mean that on the biceps the heart rate rating of the new testing is okay and on the wrist it's [music] bad and it was actually before also bad on the wrist. So the
[20:09]
correlation was a little bit higher [music] and before the heart rate rating was actually good on the biceps. Now that was a bigger test so I'm not sure why there's such a big difference. These are just the results as I have them and
[20:21]
here. If we compare the two average correlations, it's 92 for my old testing correlations, it's 92 for my old testing and 86 for the new testing. So, not a world of difference, but overall, I'm just not super impressed with the
[20:36]
performance of the Polo Loop. It's not a horrible device by any means. And for something like just basic sleep tracking, and if you wear it on the also be good enough. It's just that
[20:48]
there's so much more competition right now. Something like an amazefit device will give you much much better heart rate. Also, the Fitbit Air will have better heart rate and probably even the Garmin Circa is outperforming it by
[21:00]
quite a bit. Now, those devices I'll also be reviewing soon with a bunch of new data, but it just seems to me that there's not a lot of reasons to get the Polar Loop over these other devices. So far, the upgrades I've seen to Polar
[21:14]
Flow, so the app that you use to pair the device with, are also not that great. I still think it's not the best app out there. I think there, especially Fitbit and Whoop just have better apps. My favorite app is still the Whoop app,
[21:28]
than the Fitbit app that you need to have specific reasons to go with Whoop imagine getting the Polar Loop is because they're still a more or less European company where your data might be more secure and hopefully then fully
[21:43]
stored on European servers. I know that one of my colleagues specifically chose the Polo Loop because it's a European company and he wanted to try at least to keep his data away from the major US companies. Whether that actually is
[21:55]
happening then or not, I'm not 100% sure. But other than that, maybe one more reason to go for the Polaro Loop over one of these other devices is because you're already in the Polar ecosystem and you want to combine it
[22:07]
with one of your other Polaro watches, then it would also make a lot of sense. For me personally, if I were starting out brand new without a preference for a brand, I would try one of these other devices first because they are more
[22:20]
likely to give you good results. At least that's on my body. Also, of course, check out some other reviewers. Now, if you do decide to get an amaze with Helio strap or a Polaro loop, a Garmin Circa, a Fitbit Air, a Whoop
[22:33]
Strap, my favorite sleep improvement device, the Hleep, or anything at all on as small as toilet paper or coffee beans, you want to support the channel and in many cases get the best discount possible, use one of my affiliate links
[22:47]
down here or up here, and you'll be supporting your wallet and the channel considering. Don't forget to like and subscribe. Now, check out this video on the Fibbit Air, which you will probably also find very interesting if you're
[23:01]
interested in a Polo Loop, or this video right here on my favorite sleep right here on my favorite sleep improvement device, the H Pod.