Yes I did measure the resistance and it was out of spec - primary measured 1.1 ohms (should be 0.5~0.7) and secondary measured 12k ohms (should be 5.5 to 7.5k ohms). I measured the new one when it arrived out of interest and it was bang on spec (0.7 and 6.5k). My bike is 13 years old at this point, 100,000km on it.
I didn't think about it learning, but you're right of course it would while in closed loop. I haven't touched the ECM other than reading fault codes and live data. I did change the O2 sensor to a Bosch unit from Amazon while I had the engine rotated, hopefully that's not causing issues (it was suspiciously cheap!). I'll leave it a while and just keep my eye on it, I've only ridden it about 700km since fitting the new coil. According to the seat-of-the-pants-O-meter the bike is running fine, nice clean power through the rev range, so it doesn't feel like I'm chasing a fault at this stage.unless you've disabled closed loop, the AFV should be "correcting" itself, unless you have a out of spec O2 (or other) sensor providing false readings to the ECM. These can be tricky to ferret out, especially when a part is functional enough to work, yet malfunctioning just enough to cause a problem.