prior/odds

№ 00 · PRIOR/ODDS · 2026

Prior odds, posterior work.

I'm Elvis Liu — recovering internal auditor → risk & forecasting. I build small self-hosted systems and treat prediction markets as the lens for everything. The point is calibration, not certainty.

  1. 01 Start with a number Not a feeling. A probability you'd actually put money behind.
  2. 02 Let the world push back New data, a surprise, a miss — evidence has weight.
  3. 03 Move the odds Update, and write down by exactly how much.
A coin with a hidden bias
The chance of heads runs left to right, from 0 to 1. A dashed curve is the prior: a wide guess centred on 0.5. A solid curve is the current belief; it narrows as coin flips arrive. belief 0.50 prior
Flips
0
Heads
0
Belief
0.50
90% range
0.14–0.86

Before any evidence: a wide prior, centred on a fair coin.

Not louder. Better calibrated.

№ 01 · OPERATING PRINCIPLES

How I think

  1. 01

    Calibration over confidence

    A loud take is worthless if your 70%s only land half the time.

  2. 02

    Small bets, all logged

    Every call gets a number, a date, and nowhere to hide.

  3. 03

    Self-host the loop

    Own the tools and the data. No SaaS between me and the work.

  4. 04

    Evidence over vibes

    Update when the world pushes back, not when it feels good.

  5. 05

    Wrong in public

    A scoreboard only counts if you can't quietly edit it later.

What I build

№ 02 · THE WORK

Cover of Steeped in Data: two teacups beneath a teal and an orange bell curve BOOK · EN / 中文 Steeped in Data A plain-English book on data analysis, pipelines and A/B testing — one case to solve, with live SQL in your browser. Find the missing 12%, and learn the craft on the way. Read it online Cover of Inside the Guardrails: a closed book in a circle of lamplight, at night BOOK · EN / 中文 Inside the Guardrails A plain-English book on how today's AI models are made safe — one fictional lab, a small real model, every number measured. Keep one secret for six weeks, and learn how AI safety works. Read it online BOOK · 中文 · PARTS I–II One Ledger China's six CPA exam subjects and financial analysis as one subject, taught through one fictional company — how the numbers are made inside, and how outsiders read them. The prologue and Parts I–II (to Ch 57) are online, updated for the 2027 exams; the rest follows as it is written. Read it online
  1. 03 SELF-HOSTED Small systems, run myself A weekend rack: a private notebook, a forecasting brain, a chat gateway — no SaaS bill, no tracking. soon
  2. 04 MARKETS Prediction markets as a lens Reading the world as live probabilities. Risk control, but for everything. soon
  3. 05 AI AGENTS Agents that ship Claude Code as a teammate — wiring tools, loops, and harnesses that do real work. soon
  4. 06 CALIBRATION An honest scoreboard Every call logged, scored, and left to age in public. Calibration, not bravado. soon
  5. 07 PRIORS When a prior is too sharp On overconfidence, and the cost of refusing to update. soon
  6. 08 NOTEBOOK Read the whole thing Working notes on risk, forecasting, and building in the open.

What I run

№ 03 · THE STACK · SELF-HOSTED

predicted 0.05 → observed 0.05 predicted 0.15 → observed 0.16 predicted 0.25 → observed 0.22 predicted 0.35 → observed 0.37 predicted 0.45 → observed 0.46 predicted 0.55 → observed 0.52 predicted 0.65 → observed 0.66 predicted 0.75 → observed 0.71 predicted 0.85 → observed 0.83 predicted 0.95 → observed 0.93 predicted → observed →

№ 04 · THE SCOREBOARD

Predicted, then observed.

On the diagonal, your 70%s come true 70% of the time. The gap is the lesson — and it's left in public to age.

Featured priors

prior → posterior · 7d
NO TRACKERSNO COOKIESSELF-HOSTEDOPEN SCOREBOARD

Recent from the notebook

all entries →

№ 05 · POSTERIOR ODDS

Bet with me, not against the evidence.

Working on forecasting, risk, or self-hosted AI? I like a good prior and a sharper posterior.

hi@priorodds.com chat ↗ github ↗