battery life shorter than expected

What other fields do you have? What other CIQ fields? Do you use any wireless communication during the activity? ANT+ or BT?

My experience is actually pretty positive respectively battery life corrosponds to the manuals claims:

Recently I did an ultra run with a duration of 13 hours. GPS mode automatic, heart rate chest strap, phone connection off, AOD activated and navigation on. Battery dropped from 98% to 48%. So in may case i would get around 24h, at least 22h

Just have one Trail running data field from the CIQ shop. Phone connection is on. And I have a HRM-Dual chest trap.

That sounds really good.

I can try to switch the phone connection off, maybe that is the big consumer?

Unless you do live tracking or receive tons of notifications, the phone connection during an activity is not going to be a big battery consumer.

But do you use map how often? Drawing the map takes some power since it's normally in white background, it also increases display battery consumption when shown.

During the mentioned runs, I had typically the map shown for 95% of the time. That is the whole point of using GPS/navigation, isn't it?

Well, displaying the map whole time is not included in the Garmin's "up to" battery figures. And I would argue that the "point" of GPS for many people is to be able to get running pace, running power, distance, etc.

I have no exact knowledge on how much battery the actual drawing of the map takes (965 has four times the pixels of 955 and other MIP watches, and the fact that it has 65535 colours instead of 64 means it uses 8 times as much memory for display, but I still think that when you are running, you move slowly enough so that the map doesn' t have to be redrawn every second). But displaying lots of white under bright conditions will definitely take lots of power, especially if you have always-on-display on.

You could make some tests and try switching the map to "night mode", that makes the map background dark (just like some people keep their amoled phone in dark mode to save battery). And turn off always-on-display, if you have it on, that would mean that even if you keep the map page selected, it's only drawn when you look at the map.

I have had issues with the battery life in my 965 for awhile, there is another thread here devoted to that.

But I had a bad experience over the weekend with in-activity battery depletion on a run.  I started the run with 17% battery life (I tried to charge en route to my running location but the @#$%^ cable wouldn't actually register when connected - that's a different problem), and usually 17% is enough for at least 2.5-3 hours of running with no music.  But at 1:30 into the run I got a surprise low-battery warning with the battery around down to 2%, that's a rate above 10% per hour.  I quickly turned off the screen altogether but it was too late, only a few minutes later it shut down altogether.

Something has changed in the last few months with my watch, wish I could figure it out.

Sounds like what some people (including me) reported that when the watch was new and we charged to 100% it stayed on 100% for a long time. To me it sounds the root cause for both is that the "table" Garmin uses to translate battery voltage to percentage isn't correct, aka it's not linear to the time (what user cares when we see X%)
It also looks like the initial problems around the full battery are fixed with time (probably adaptive algorithm, so it "learns" over time) and since probably none of us lets the battery to completely discharge usually the same algorithm didn't have the chance yet to adapt. But all this is my un-educated guess. It would be way better for Garmin to fix these issues!