I think I figured out why Trilogy thought Powell's streak of times on base safely was only 12, not 14 as it should've been. After the 12th time on base, in Powell's next PA, he was batting with 2 out when Blair was CS2. So Powell led off the next inning (and singled). But Trilogy saw that Powell came to bat in the first inning, but he didn't get on base (since his PA was interrupted by the 3rd out on the bases).

That would explain the cut-off date that Trilogy gave for Powell's streak: 1st inning on 8/6, which seemed weird since Powell didn't technically bat in that inning.

Any ideas how to fix something like this? I'd really like to use DMB & Trilogy to compare simulated streak totals to what has really happened in MLB history, but I'm not sure that Trilogy's output for these particular streaks will be very helpful unless we can fix these glitches.