Oh, good, further evidence that TrackBot is conceptually flawed and ineffective for distances 1000m and under.
Optimized time for 2:01.3 800.0m based on even pacing:
I am a bot. Info: habs.sdf.org/trackbot
OP splits 59/62, great splits for a 2:01, and TrackBot says if he went out in 60 he'd even-split for a 2:00.0? ðŸ˜‚ðŸ˜‚ðŸ˜‚
TrackBot relies on an incorrect assumption, that "even pacing" is best for all distances. It simply isn't, obviously, for 600-800-1000.