My engineering degree taught me that there was a balance around systems. You could calculate what went in, determine what occurred in process, and measure what came out. This could all be mathematically solved so that algorithms could make predictions. This is pure bullshit when it comes to the human body and diabetes. See, the thing is, you never know exactly what is going on inside any given human body.
Two nights ago, William's BG was 150 mg/dL at 3:30 a.m. I chose to do half of the suggested correction because he had been somewhat active on a hot day. He woke up in the 80s. Perfect.
The next night, he was again 150 mg/dL at 3:30 a.m. and I did half a correction because of a similar day, and he woke up in the 200s. (And, yes, I should have caught it with the Dexcom CGM, but I didn't. We'd turned off the high alarms because...well because we were somewhere where it was annoying. Sue me.)
Same input, same process, different output. Was he growing? Perhaps, just maybe, it was the two hamburgers and fat that he had right before bed at midnight? (I'd already gone to bed and missed knowing this.) Perhaps he was growing. Perhaps it was a full moon. Who knows.