I'm inclined to think that the lower numbers earlier on and the higher numbers now are more the result of time passing. Griff has been on insulin since February with very little to indicate that he's getting the right dose yet. That's kind of a long time to be hanging out in high numbers. But it's also possible that the bouncing we're seeing is indicating that he's getting closer to an effective dose and so he's getting more reactive.
One big question is that we have no data in the PM cycle to tell us if he's dropping lower at night. If you look at
@Jen and Alfie's spreadsheet, we see really active PM cycles, and then high flats during the day. It's possible something like that is happening here too - we're getting closer to a good dose, so the night cycles are getting active later on, and causing the current high flats during the day, and those icky black numbers.
Most of us need more sleep than Jen seems to though

. I love the data she has, but I honestly couldn't stay up that late myself on a regular basis! At some point, those lower numbers should start popping up during the AM cycle. So I would say to keep on with the six cycle increases, and get data whenever you can since there is still the possibility that we've just had some bad luck with the current data set and somehow managed to only catch the high numbers and miss the low ones! And especially if you or anyone in the house could grab an occasional middle of the night test, that could be really helpful.