Battery health is an approximate representation using algorithms.
Well, if I grep the I/O registry, I can see numbers for MaxCapacity and DesignCapacity - and those tie out to what Coconut Battery shows me. That's pretty easy math.
The entire battery health and charge levels is an algorithm as you can't have sensors inside battery cells. Numbers on the OS front can match, but it's all done via guess work. I wouldn't think too much into it.
There are several parameters that can be used to determine condition and full charge capacity. For example coulomb counting to track current flow in and out of the battery. Also temperature and voltage can be used. I have myself found out in many of my machines that battery health reading starts to drop as soon as battery does not reach the highest possible voltage just as it reaches 100% charge level. For example in my iPhone 12 Pro highest level possible is 4450mV, I had some glitch with my phone where it for some reason stopped charging past 4428mV and battery health started dropping like a rock. After a while it fixed by itself (probably new IOS did the trick) and it started allow charging past 4428mV up to 4448-4449mV and battery health (actual reading) restored back to almost 100%. So following the same parameters in MacBook with correct max mV reading (it is not the same as in iPhone 12 Pro) in mind gives similar results.
Again, those are approximations of what's in the cell given what is going in/out or the strength of the cell's EM field. It's not as exact. That said, agreed with your second part. OP is either too unlucky or something wonky is up.
Not sure what you mean, but measuring voltage, temperature or even doing coulomb counting is rather trivial thing to do accurately these days, but surely some guesswork may be needed but often outside some glitches those battery health readings go hand in hand with actual battery run times (meaning lower health = lower run time but it is not actually tied to percentage. Like 80% battery health does not mean you can get 8 hours of run times from battery that was 10 hour run time when 100% health, it is more likely half or so).
That should not be a problem with optimized battery setting but I'm not sure, I don't use it myself nor I keep mine plugged in other than when I actually charge it. But this really makes me think again my plan to sell my MBA M1 and Mini M1 and replace them with a single M1 Pro 14" and use it 99% of the time in clamshell mode connected to external display and charger all the time.
Have you actually kept any track of battery run times between how it was when new and today?
Just use the app AlDente. You can set it to limit your battery charging to any %. I have mine set to stop charging at 70% since I rarely take it off the charger.
If you are willing to take a small risk, you could try switching off "Optimized Battery Charging" and then try to charge battery as full as possible, like letting it stay plugged charger few hours after reaching full 100%. Also probably best testing that when battery is well below 50% before starting to charge. Just don't leave that setting off for a long time if you mostly keep it plugged in, but for a test. I suspect it might actually make battery health reading appear better. Of course that is just my hunch...
Some things I said in this post this post are relevant.
Just speculating here for the sake of argument but imagine situation like where "Apple's number" state battery health still 80% while ioreg command could report ~75% and your battery run times could be just like 1/4 of what it was when new. So you don't mind that because their "number is more stable"...?
HaHa...I would mind a lot....but I am just reporting what Apple will do! Don't shoot the messenger!
I did not mean to shoot the messenger but to point out few important things.
Yeah, there could be something like that too but mostly IMHO that value is also sugar coating things. I'm sure most peoples don't mind that but I'm sure most peoples think that battery health percentage value has actual relation to battery run time, like 80% health means you still get 8 hours out of the battery that used to be 10 hours when new - WRONG, it is likely much much lower, could be even hardly usable in the worse cases.
Apple would still replace the battery even if it reported above 80% in that situation. Getting only 4 hours of battery life on an M1 with 80% life remaining would still indicate a defective battery.
That would require someone I know to have had a defective battery. Since I don't know anyone with that problem, I haven't tested anything. But this is from Apple's website:
I'm familiar with that statement, but how would you prove your point in above case if battery health reports 80% and diagnostic cycle shows it is good? From what I read some threads in this forum that seem to be how it goes.
Apple will test the battery. They can do more than just look at the number from the system. A battery that degraded is surely going to fail any test.
Let's hope so. I was just referring to numerous threads about similar situations where that does not really seem so.
Looks a bit like it is rounded up about 5%. Which is actually about similar how battery meter works between MacOS reading vs terminal command when battery is full or nearly full where counter goes to 99% when terminal command returns 94.x%. Anyway, these two should not have any relation between, just my observation (and that is how it works in IOS devices too, however their health meter seems more consistent to me vs iMazing reading up to a certain point).
