However, I doubt that over the entire period of your graph the number of average observed new STEEM is constant at 88191.

Yeah, you're right about that. It's the daily average throughout the time period. Breaking it down by day in PowerBI is more complicated, so I haven't figured that out yet. Given that number as an average, I can't imagine any way that the individual daily numbers would close the gap overall, though.

Update:

That number is easy enough to verify with current values, not quite the same because it's a couple days later, but it's the same order of magnitude, and not big enough to get into sync with the prediction: