←back to thread

817 points dynm | 1 comments | | HN request time: 0.279s | source
Show context
davidanekstein ◴[] No.43308307[source]
If anyone is interested in running their own self guided experiments, I made an app for this called Reflect [1]. You can run self guided experiments for anything you can model as a metric in the app. We got in the top 3 on ProductHunt not too long ago [2], and I just wrote about an experiment I did with nootropic coffee [3].

[1] https://apps.apple.com/us/app/reflect-track-anything/id64638...

[2] https://www.producthunt.com/products/reflect-c052fea3-a982-4...

[3] https://open.substack.com/pub/reflectapp/p/my-experience-wit...

replies(3): >>43308410 #>>43308545 #>>43309990 #
hombre_fatal ◴[] No.43309990[source]
Does your app make you randomize the when you do the intervention you’re tracking?

If not then almost everything you track is just a proxy for something else like how someone might only take melatonin on nights they knew they were going to sleep badly, or read in bed on nights they were already feeling good, not stressed.

Or they only take their meme supplements on days they aren’t stressed out by other things in life that make you neglect yourself, so the meme pills always correlate with good things.

It’s a major problem with Whoop’s insights feature. It needs a way to make you coin flip an intervention to be useful.

replies(1): >>43312044 #
1. davidanekstein ◴[] No.43312044[source]
Yes, you can choose a random intervention schedule and also ABAB schedules with configurable number of phases and phase length.

Reflect integrates with Whoop and so can act as an augmented or relacement Whoop diary feature.