Surely it depends on what you want to ‘improve’. (:
I’ve got plenty of goals that I measure from different angles, so you might want to track both. Though I suspect that tracking ‘time in bed’ may be a better (i.e. directly controllable) measure to beemind than ‘time asleep’ (which is an outcome of many factors).
You’ve doubtless already found these relevant threads:
- Beeminding Bedtime
- [Someone posted about experimenting with several sleep-related variables]( “I couldn’t find it…”)