← back

Hindcast: Replaying Prediction Markets to Evaluate LLM Forecasters

Forecasters are evaluated by backtesting, which replays resolved questions and grades the probability the system would have assigned before the outcome was known. For LLMs, two channels leak the answer into this test. A model that retrieves

https://arxiv.org/abs/2607.14051v1 ↗
Thesis fit
Good fit

Within your typical scope; diligence still required.

Edit thesis
In your usual scope
Idea match Light

How close the company’s idea is to your thesis statement

Sector llm

Overlap with sectors you care about

Geography Unknown

Location unknown — scores 0

Your thesis: “We back exceptional technical founders building AI-first products and infrastructure, deploying $100K checks within 24 hours.”

Founder → stable Traction → stable Idea vs market ↓ declining
Generate memo
Add / edit details

Correct facts used on the next screening or memo.

Similar baseline plays (YC · idea space)

Baserun · Active
Observability and evaluation platform for LLM apps.
founders not scraped yet
Datacurve · Active
Frontier coding data for training and evaluating LLMs
founders not scraped yet
River Markets · Active
Prime brokerage for prediction markets
founders not scraped yet
LLM Stats · Active
Independent AI evaluations lab
founders not scraped yet
Oddpool · Active
Institutional infrastructure for prediction markets
founders not scraped yet