# Bayesian Time Machine

**URL:** <https://discourse.datamethods.org/t/bayesian-time-machine/28884>\
**Category:** modeling strategy\
**Tags:** bayes, rct, design\
**Created:** [October 10, 2026, 5:08pm UTC](https://discourse.datamethods.org/t/bayesian-time-machine/28884 "2026-10-10T17:08:46Z")\
**Posts on this page:** 1\
**Showing post:** 5

<div class="post-metadata">

**Author:** ![llynn](https://discourse.datamethods.org/user_avatar/discourse.datamethods.org/llynn/32/2827_2.png) [@llynn](https://discourse.datamethods.org/u/llynn)\
**Post date:** [October 12, 2026, 6:10pm UTC](https://discourse.datamethods.org/t/bayesian-time-machine/28884/5 "2026-10-12T18:10:54Z")

</div>

Thank you, Frank. I would add: combining synthetic entry gates with BTM and RAR, as applied in REMAP-CAP, creates a potentially toxic brew. We should teach that RAR should not use results from apples to allocate treatment to oranges.

---

_[View the full topic](https://discourse.datamethods.org/t/bayesian-time-machine/28884)._