# Outlier ratio

Story / Test and publish / Reading the numbers. Checked 2026-11-13.

One number says whether a video did well for that account, whatever its size.

- Use when: Choosing references and judging your own posts.
- Phrase: plays divided by the median plays of the account's other recent clips
- Page: https://directorchair.ai/publish/outlier-ratio
- JSON: https://directorchair.ai/api/publish/outlier-ratio.json

## Watch out

- Leave the candidate out of its own baseline, and flag a small sample. (practitioner; Director Chair notes on AI UGC army (Jake Castillo))
- Seven of the nine reference edits in our Video library carry this ratio, from 0.6 to 71.9 times. (measured; Director Chair measurements of published shorts (measured 22 and 23 Oct 2026))

## Prompts

### Any chat model

```
Give me a step-by-step plan I can follow this week, for a short video about [your product], for [your audience]. The technique is outlier ratio: one number says whether a video did well for that account, whatever its size. How it works: plays divided by the median plays of the account's other recent clips. It is right for: choosing references and judging your own posts. Leave the candidate out of its own baseline, and flag a small sample. Seven of the nine reference edits in our Video library carry this ratio, from 0.6 to 71.9 times. Be concrete: numbers, seconds and settings, no theory. Ask me one question first if something you need is missing.
```

## Goes well with

- Start from an outlier: https://directorchair.ai/publish/start-from-an-outlier
- Read two numbers, then decide: https://directorchair.ai/publish/read-two-numbers

## Source

Own words from Director Chair notes on AI UGC army (Jake Castillo)

Text CC BY 4.0, Director Chair (directorchair.ai). Reference clips and their card texts belong to their makers.
