I have a 5S2P battery project and for this project I did tests to select the ones in good condition from the Samsung 30q batteries I have. Accordingly, I first measured the capacity of the batteries I selected. I measured the internal resistance of those between 2800-3000mah. I finally fully charged 10 batteries with an internal resistance of around 15mOhm (between 4.18-4.20V) and left them for 1 week. Seven of the batteries showed a value of 4.18, while three of them showed a value of around 4.02V when I measured them after 1 week. In this case, I wonder why these three batteries with low internal resistance and the same internal resistance self-discharged so much after 1 week of waiting? If anyone has any information and can share it, I would appreciate it.
Short answer: sometimes it just happens.
Long answer: You did a good job 1) checking for internal resistance, 2) measuring actual capacity, and 3) doing a self-discharge test. Because as you now see, just because a cell has good capacity left and good internal resistance, doesn't mean it has other hidden issues.
Some questions that can give you more things to check:
What tester did you use for individual cell capacity? Sometimes lower-quality testers may give you an inaccurate result, leading you to believe a cell has more capacity than it actually does.
What tester did you use to measure IR? There are some good ones, but there are a lot more bad/inaccurate ones.
Where did you get the cells, are they used? If they are used/reclaimed from something else, that would be more likely that you would find 3 self-dischargers.
How did you measure voltage, and was it consistent? I ask because if you used one of your testers to charge a cell up to 4.18-4.20v, and you assumed the tester was accurate when it told you it was 4.18-4.20v, but it was a poor quality tester that actually only charged up to 4.12-4.15v, then maybe it didn't discharge as much as you think it did.
I am currently testing 832 cells from the same make and manufacturer, that are supposed to be new. All are testing within 95% of their capacity, and all are testing below their IR specifications. But I am still finding 3-4 self-discharging cells so far. It's not unheard of.
Last thing. You said you're saving cells that are above 2800mah. That's 93% of their spec capacity. Which certainly isn't bad. But it is a definitive indicator of some level of aging. So it's not terribly surprising that you'd also find other signs of aging.