About Quotient

Turns market noise into
actionable trading strategies.

We combine broad source coverage, structured AI reasoning, and human contributor signal to surface where consensus breaks down.

The system

Built to find what others miss.

Q groups related markets into narratives organized around a shared thesis. You get context, not just a probability.

1,000+ global sources

News, data, and event signals tracked across markets and geographies.

AI superforecasting

A team of agents weighs evidence, identifies key factors, and translates uncertainty into a reasoned view.

900+ human contributors

Contributors review forecasts, surface context, and help sharpen Q's reasoning.

Performance

Q's performance, verified.

Benchmarked against frontier models and tested against real market outcomes. Q performs at a level worth paying attention to.

Q vs. frontier AI models

How often each model's forecast matched the eventual market outcome.

Q
0.0%
Gemini 3.1 Pro
50.9%
Claude Opus 4.6
42.6%
GLM 5
25%

Based on 120+ resolved forecasts. Frontier model data via PredictionArena. As of March 23, 2026.

Q vs. top forecasting benchmarks

Lower is better. Brier score measures forecast accuracy and Q scores alongside elite human superforecasters and ahead of leading AI model benchmarks.

← BetterWorse →
Perfect (0)Coin flip (0.25)
Q
0.076
Superforecasters
0.086
Best LLMs
0.103
Public median
0.123
Building on

Where Quotient plugs in.

Polymarket
Bankr
World
Base
The track record

Where Q called it and where it learned.

Every resolved market is reviewed. Wins and misses are published so you can evaluate conviction.

Where Q called it right

Recent markets where Q diverged from consensus, and the outcome proved Q right.

US or Israel strike Iran by Mar 31?
Q Called
Yes
Outcome
Yes ✓
Market Odds
63%
Return
+60.0%
Houthi strike on Israel by Mar 15?
Q Called
No
Outcome
No ✓
Market Odds
36%
Return
+55.0%

Where Q missed and what we learned

Q is always learning. When a forecast misses, we trace the reasoning, find the breakdown, and come back sharper.

U.S. strike on Iran by [date]?
Q Called
No
Outcome
Yes
Market Odds
40%
Miss
Escalation risk
Israeli strike triggers U.S. action?
Q Called
Low
Outcome
Yes
Market Odds
[TBD]%
Miss
Linked markets

What we learned from it

Markets don't move in isolation. An Israeli strike on Iran raises the odds of a U.S. strike too. Q was missing that dependency.

We updated the pipeline to track correlated outcomes across linked markets. Forecasts on escalation events are now better calibrated, and Q's spreads reflect it.

Read the full postmortem on Substack
The Improvement Loop

How Quotient Gets Better

Quotient improves through a loop of forecasting, review, and feedback. Each cycle tightens the signal.

1
Forecast
Q analyzes markets, sources context, and assigns probabilities with structured confidence.
2
Review
Contributors surface context, flag errors, and add signal that Q may have missed.
3
Learn
Every resolved market feeds back into the system. Q gets sharper with each outcome.
Contributor Leaderboard

The People Sharpening Q

Contributors review forecasts, surface context, and compete for monthly rewards. Here's who's leading.

Top contributors
Rank
Contributor
Score
#1
R
ruminations
77.8
#2
K
kan1902
66.8
#3
Y
yiky
63.0
#4
M
monteluna
63.0
#5
S
shira
56.1
View full leaderboard

Scores reflect contributor accuracy across resolved forecasts. $250 in monthly rewards for the top 25. Leaderboard updates live via the Signal app.

Why This Team

Built by People Who Understand Intelligence, Infrastructure, and Markets

Meet the team