What is it?
The Base Rate Times is a global news site built around prediction markets, not media narratives - like The Economist meets Metaculus
How does it work?
The Base Rate Times creates:
Pretty charts, showing how different prediction markets are assessing each news story
Original headlines, re-contextualising news around forecasts from prediction markets (examples below)
Imagine a simple website with charts and news headlines.
We match news stories to existing, liquid forecasts from multiple platforms - the plan is NOT to launch new markets.
Example #1
Conventional news: "Taiwan braces for China's fury over Pelosi visit"
--> Base Rate Times: "China-Taiwan conflict risk up ~2x after Pelosi visit"
Chart similar to this: https://imgur.com/a/yStsfJ9
Example #2
Conventional news: "William Ruto declared winner of Kenya presidential election amid dispute"
--> Base Rate Times: "2:1 underdog Ruto wins Kenya presidential election"
With additional context bubble: "9 out of the last 10 election results have been disputed, a predictable feature of Kenyan politics"
Example #3
Similar to this, except easier to read: https://ourworldinprediction.net/wp-content/uploads/2022/03/Screenshot-2022-03-20-at-08.32.54-1024x697.png
(Taken from the ACX Ukraine warcasting post)
Example #4
Conventional news: "Ukraine won't retake Donetsk by winter because of new logistical problems"
--> Base Rate Times: "12% chance Ukraine retakes Donetsk by winter, down 9 p.p. since new logistical problems"
(Taken from ACX prediction markets FAQ post)
Product roadmap
The first version prioritises prediction markets, but v2 will also integrate betting markets and expert surveys (e.g. IGM Forum https://www.igmchicago.org/).
Context bubbles with key background information will be added where needed (see example above on Kenyan elections).
The draft version of the website will be launched on 31st March for community feedback, so I will have a track record before September. Otherwise, this is my first time launching a media project.
I have been following/dabbling in 'superforecasting' for ~7 years.
My background
~5 years experience at blue chip multinationals, where I owned large P&Ls, led several priority projects and gained experience across multiple business functions
Graduated from a top 50 (globally) university with an honours degree in economics, including studies abroad in Asia and Europe
I have recently quit my job to work on entrepreneurial projects
I estimate that $20,000 will deliver ~4.5 months runway
Key cost drivers include:
Labour (~25 hours per week running & promoting website, plus ~150 hours building website)
Subscriptions for back-end tools, website hosting, domain, etc. (Webflow, Flourish, etc.)
Admin (incorporation, taxes etc.)
Automation project (automatically scraping latest data from prediction markets)
--> Total ~$4,400 per month
App status across various funders