My Apple Watch is fairly accurate for aerobic workouts. However, it isn’t accurate (and same hold the truth to other wrist based HR trackers) for those hard sessions, where there is a lot of HR spikes. In my case, it simply stops reading until my HR drops down a bit.
Tim.
I’m another user with a Fenix 5 who doesn’t use the hr at all due to inaccuracies. I do however use it as a sleep tracker and it works well for that. I would switch to Apple if they made something that lasted long enough to get me through a few days and nights. The Fenix lasts for more than a week. I reply on the sleep tracking, so the Apple is a deal breaker. If anyone knows of another wristwatch that does music, text notifications, and sleep tracking, and has a battery that lasts, I’m all ears.
I cant comment on wrist as I don’t find that a comfortable place to have a tight band during cycling. The chest strap bothers me inside so I use a scosche rhythm+ between elbow and bicep bulge (hah!) while indoors. I’ve never observed it giving odd readings (I’m fairly pale and don’t tan well…). It always makes sense though I haven’t compared to the Garmin strap.
N=1 of course.
I used a Mio Fuse for a while and it was very good. It had reviewed well with DC Rainmaker. Unfortunately after a few years it started having issues.
More recently I tried the Schosche Rhythm 24, which again was rated well by DC Rainmaker. However my experience has been really poor. I seem to have a ton of syncing issues with it. At times it connects, but then a few minutes later it drops completely. I’ve tried with iPad, Android tab, and Garmin head unit, but very poor results and very disappointing.
No question the Wahoo Tickr chest strap has been my most reliable heart rate monitor. Syncs well with pretty much everything and seems to be very reliable (light years better than the old Garmin straps I used to burn through). I bought a second Tickr as a back up.
I do have a Fenix 5 Plus that I got a little over a month ago. I haven’t used it to monitor HR while cycling yet, but did use it for some gym work. It seems ok, but not really a great test. Of note, when you wear it during the day it does monitor heart rate, but in a power saving mode so it reads much less frequently as compared to when you are in workout mode. I’ve left my watch on in this mode while working out and periodically looked at it and the HR it reads is way off from what the Tickr is putting out, but I think that is to be expected. I will likely play around with the Fenix 5 Plus a bit more and perhaps use it on some non-critical rides.
I should have added that we own a Peloton and use a Scosche for it. Zero issues. I just wouldn’t want to wear it outdoors on a bike.
For accurate wrist OHR, you should position the watch on the fleshy part of the wrist, well away from the wrist bone, to give it a deeper blood supply to measure. the watch band should be firm to prevent external light from entering the sensor.
Vibration from the handlebars from outside riding can negatively affect OHR accuracy, but is less of an issue with an indoor trainer.
OHR also tends to be slow to respond to changes in heart rate. This isn’t a big issue for long sweet spot intervals, but for short (20 or 40s) intervals, it can get completely out of sync trying to follow the changes in your heart rate.
I’ve got an Apple Watch series 4, still using chest strap (don’t mind them, no accuracy concerns). That said I am pleased to read DCRainmaker’s outdoor ride commentary “At which point, it honestly produces the best wrist-based optical HR sensor track I’ve ever seen during a bike ride. Seriously, it’s almost identical:”
Have given up on TickrFIT arm strap- never kept up with surges in effort. Back to Tickr chest strap which always works great. Apple Watch 3 is surprisingly accurate.
I’ve had no luck with wrist based HR’s being accurate during any exercise (garmin fenix 3, fenix 5)
What I use the wrist based HRM for is monitoring my average RHR to let me know when I am recovered from a hard workout, potentially getting sick, etc.
If I wear my TICKR for a few minutes before turning on the sensor reading apps, it’s usually fine.
If I put it on while the sensor reading apps are active or turn on the apps right after putting on the strap, readings are wonky like you described. When that happens, even after disconnecting, wetting (see next) and reconnecting it takes 10-20 seconds, sometimes longer, to stabilize. Sometimes it won’t stabilize until I disconnect and wait 20 or 30 seconds before reconnecting. I’m not sure how much of that is the HRM and how much is the sensor reader averaging displayed data. (I’ve seen some apps step down HR in two or three steps where others take five or six steps over more time, so I assume the app’s HR handling is also a factor.)
If I wet the strap with tap water about a foot on either side of the sensor snaps before I put it on, no problems. Just run that part of the strap under the tap and run your fingers along the section, stretching the strap slightly to let the water into the elastic. Takes about 10 seconds and 0.001 TSS. I’ve also tried lotion, on my chest where the strap runs, but water works better and is cheaper.
I wash the strap every two or three weeks (with 4-6 rides a week). Well, maybe that often. Maybe not.
I’ve had a TICKR for several years and have changed the battery twice–both times because I had a big ride coming up and didn’t want to risk a battery fail. I’ve seen consistent results over time. Given you’re seeing this with a several different straps, it’s probably your skin is too dry to get good contact. I suggest tap water is your friend. here.
I have this exact experience. It works most of the time but can take a minute or two to settle down and has occasional blips when I shift position etc.
Its a none issue for me as I “don’t care” about the actual data really, I just like having then extra line on the workout graph 
I’m doing some comparisons at the moment using DC rainmakers tools so thought I’d do one of my Tikr HR Strap and my Mio Velo Optical wrist strap. I have to say I was pleasantly surprised by the closeness of the 2.
Use a Garmin Vivoactive HR and get very accurate readings over the last few years running and cycling.
I would say I do pretty much the same as you. Wet straps using tap water before each wearing, rinse off in shower afterwards and stick through the washer on a cold wash every few weeks. This wasn’t the case with my old Polar straps that were rigid plastic but I wonder if it’s possible to over stretch straps like the Wahoo Tickr? My strap was loose the last workout I did before Sunday so I tightened it up and then saw the flatline. I wonder if it was too tight? Seemed reasonably ok this morning again (a couple of strange plateau’s then drop downs of nearly 20 bpm’s during rest periods) but I did slightly loosen strap from Sunday (and I’ve replaced battery).
I’m getting sucked into a 735 black friday deal, but the only real reason I can come up with for day to day use is OHR, and that’s mainly for the bike commute. If I have to use chest HRM, there’s probably not much reason over my 920xt?
I like multisport capability, but I do mainly bike these days, and if the bike is where OHR is at it’s flakiest…
I’ve also seen oddness with a too-tight strap or too-loose strap, but I’ve never had trouble adjusting the strap to get a good tension. YMMV depending on your chest measurement.
Looking at Ray Maker’s reviews of HRM devices, especially wristbands, the TICKR is his go-to reference HRM (also noted in his 2018 recommendations article), but it sometimes throws unexplainable data. Then again, some of the wristbands get lost when he stops to take pictures. Sometimes all the HRMs he’s wearing for a test are lost for 5 or 10 minutes. He often mentions the difficulties of getting a power meter right (true), but once the kinks are worked out, PMs usually just work. Maybe we need (accurate) HRMs that attach long term so they aren’t subject to the vagaries of how we put them on today vs. yesterday. 
Also, I know people complain about uncomfortable HRM straps, but I often forget I’m wearing the TICKR. For example, today I came home from a ride, took the bike out back to wash it off, spent over an hour getting organized to clean it, clearing some out-of-control plants near the back door, cleaning the bike, putting away the bike cleaning gear. Only when I went inside to change did I remember I was still wearing the TICKR. (Maybe I’m not a sensitive guy. 
)
I have a Garmin vivosport wrist based monitor. Ive tested it back to back with a garmin chest strap on the trainer and during a couple of Cx races and I would have to say I’ve found it bordering on useless.
Typically it will show about 100 to 120 bpm whilst my chest strap is showing over 160, and can often drop back to about 80 mid interval.
It’s useful for recording distance and resting heart rate, but that’s about it in my experience.
Depending on your needs, you could also consider the lack of accuracy in terms of measuring heart rate variability (HRV) with optical sensors.
The correlation between your HRV and the stress on your sympathetic/parasympathetic nervous system can be used to to estimate fatigue level and recovery status. This is used by applications, such as Garmin, EliteHRV among others.
A chest strap will supply the HRV in milliseconds. Even though that does not necessary imply they have that precision, the optical sensors do not offer the same granularity as a chest strap.
In regards of simply measuring heart rate, it will vary from device to device, but your skin type will greatly affect the result.
BR
I’m currently testing my TickrX against my new Apple Watch Series 4 - and so far the results are very good. I’ll try and get some proper tests done - I have a sub to DCR’s Analyzer tool. But just glancing at the watch on my Mills workout today they were withing a couple of beats of each other. The AW4 was a little slower to respond - but not of real concern. I do cinch the strap up one hole on a workout.
I would still wear a HR strap for cycling most of the time. But not for running that I’m just getting into (with Stryd). And I like the idea that I could use something like iSmoothrun for a casual bike ride and get power from my Stages and HR from the AW without any extra kit. Or even if my Garmin Edge 1000 ran out of charge (which has happened on a long 100+ miler).
This is worth a read (Apple Watch accuracy tested within 0.67 BPM of benchmark chest strap): Who Has the Most Accurate Heart Rate Monitor? | Tom's Guide
I would love to use my AW4 with TR. My Tickr HR strap lasted 2 years, then bought a Tickr X which has been useless - randomly outputs HR at 10-20BPM lower than it should be on 30% of rides. Switched back to an old Garmin strap for now.
