Reduced battery life

Other than general tips like proper charging (to about 80% and not let it drop below ~20%) and using a stock watchface, maybe PulseOx off - I don't know, I have the same problem and am thinking of replacing the battery, but stalling on that. Battery replacement is hard for a Forerunner. It's a great watch, and so far I don't mind charging it every day. It charges really quickly.

The latest firmware is causing it.  I called Garmin a few weeks ago, and it's a known issue.  They had me install logging onto my watch, that sends logs directly to their engineers; they're still hunting it down.  Hopefully there's a fix soon.  

Thanks! I did read that awhile back.. but I'm always a bit hesitant to blame firmware for battery issues, and then I forgot about it. However this time it really must be the cause. 

Yeah, if I were to guess, it has something to do with the new sleep tracking on the watch.  Before, it would just track stats and Garmin's servers would do the heavy lifting to figure out what was REM, etc., now all that's done on the watch itself.  So who knows.  If my watch wasn't logging for Garmin to help troubleshoot, I'd reset my watch to factory and give that a shot.  The other issue I've noticed is the sleep tracking itself has been pretty jacked up the last couple weeks; it's saying I only slept 5-5:30 hours, when it was really closer to 7-8.  Just bizarre.  

I can't compare the new sleep tracking with before (I never uploaded sleep data to Connect). It seems to produce plausible data in my case, and I'm quite happy with the processing being done on the watch. It was already in beta 6.85, maybe I'll go back to that, although I can't remember what battery life was like on 6.85.

Do you know if it's possible to turn sleep tracking off to help with battery consumption?

I'm going to try:

- Removing the Sleep widget for 1 night. I suppose that won't affect sleep metrics as calculated by Connect, but I'm not sure. 

- Disabling PulseOx during 1 night.

- Disabling 'activity tracking' entirely during 1 night. That will obviously result in no sleep data. 

These won't be anywhere near accurate tests or conclusive - just battery level before and after the night, to get a rough idea. 

I did a very unscientific test: changed 3 variables at the same time. Removed Sleep widget, turned PulseOx off completely (before it was on during the night), and set the phone connection to off. Battery was 62%. 12 hours later, which included ~8 hours of sleeping: 61 %. After adding the sleep widget again, it had the sleep data immediately, which makes me think that the processing is done either very quickly by the widget, or done independently of the widget, some time after the watch detects you are awake for a while.

This overnight battery usage is perfectly OK. I am most curious if it was PulseOx or the phone connection responsible for the excessive battery drain. Next up: testing with only PulseOx off and only phone connection off. 

I had a look at the sleep-related FIT files (everything undocumented) and have a rough idea of the data the sleep widget gets (and maybe also computes) visualizes. Wish I had more time for diving into this kind of thing! Years ago I made a simple 'best-guess total sleep deep and light sleep duration' (based on Monitor FIT files), great fun, buy it took way too much time. 

Disable bluetooth (if its enabled) and recheck.