Research Hub
Topic Hub · Scored in public.

Predictive Modeling.

Forecasts with a scoring rule attached — calibration, Brier scores, market baselines.

Personalized feedUpdated · 2026-07-08

Articles

0

Quick Takes

0

Videos

0Soon

Podcasts

0Soon

Clarity Index

Soon

Followers

Soon

The Standard on Predictive Modeling

A forecast without a scoring rule is an opinion with decimals. Prediction is one of the few areas where honesty can be measured directly — so we measure it.

What this field actually measures

Brier score

Mean squared error of probabilistic forecasts.

ReferenceCompared against a baseline such as the historical base rate or the market.

Calibration curve

Predicted probability plotted against observed frequency.

ReferenceA well-calibrated 70% call should land about 70% of the time.

Market baseline

Prices from liquid betting markets as a comparison forecast.

ReferenceA hard benchmark to beat, and the honest yardstick for any public model.

Evidence & metrics

The published work behind the numbers above, and the public datasets you can open to check us. If a claim can’t be traced here, we don’t print it.

Citations

  1. Verification of forecasts expressed in terms of probability

    Brier — Monthly Weather Review · 1950

    Defined the Brier score, the standard measure of probabilistic forecast accuracy.

  2. Superforecasting / the Good Judgment Project results

    Tetlock & Mellers et al. — Science and subsequent reports · 2014–2015

    Calibration training and aggregation measurably improved forecast accuracy over expert intuition.

Datasets

Still open

  • How should public predictions be archived so they cannot be quietly revised?
  • What horizon makes sports forecasting meaningful rather than decorative?
  • Where do human analysts still beat models, and why?

What we won’t say

We will not publish a prediction we are not willing to score in public afterwards.

  • VO-1Evidence before wit
  • VO-2Punch at claims, never at people
  • VO-3Say the uncertainty out loud
Read the Brand Voice Standard
Latest Articles

Everything published on this topic.

No articles have been published for Predictive Modeling yet. New long-form pieces will land here.

The Archive

Everything published on Predictive Modeling.

The permanent record of every piece we publish under this topic — grouped by year, oldest kept as history.

Quick Takes

Short reactions, polls, and open questions.

Faster than the long form. Same standard.

No quick takes match this filter yet.

2CS Analysis Framework

Evidence, counterpoints, patterns, and the standard.

Every entry links back to its source article. Nothing duplicated.

Framework entries appear here as articles publish.

Future Watch

Open questions and measurable predictions.

We log the question, the confidence, and the review date. When the evidence moves, so does our position.

No active predictions are currently being tracked for this topic.

Video

Coverage on screen.

Long-form, shorts, and playlists connected back to the source articles.

Video coverage is coming soon.

Coming Soon

We haven't published video on Predictive Modeling yet. Playlists, chapters, and shorts will surface here as they roll out.

Podcast

Long-form conversations.

No podcast episodes yet.

Coming Soon

When we publish podcast episodes on Predictive Modeling, each will link back to its source article.

Topic Discussion

The conversation, on the record.

Community discussion is coming soon.

Coming Soon

A dedicated space for readers to discuss this topic will live here — moderated to the same standard as the editorial.

The 2ND Take

Follow Predictive Modeling.

One weekly dispatch. Signal over noise. No spam.

No spam. Unsubscribe anytime.