Company
About Cade Market
We built Cade Market because risk analysts deserve a probability instrument, not a noise feed. Our founding premise: a sourced, calibrated number is more useful than a summary of what happened.
Why We Built This
The corporate risk function has a documentation problem. Analysts are expected to attach a probability to a country risk decision, a board briefing, or an insurance review. But their tools give them news feeds and country ratings. Those tools are not probability instruments.
Cade Market was founded in 2024 by a team that came out of quantitative research and geopolitical forecasting. We had spent years watching analysts manually synthesize probability estimates from sources that were never designed to produce one. We thought the aggregation and calibration work belonged in software, not in analyst hours.
The result is a scoring engine that ingests intelligence signals continuously, applies calibration weights based on historical accuracy, and produces a probability with a confidence interval. Analysts receive the number, see the sources behind it, and can document it at decision time.
4
Event category types
3
Access tiers
4
Source categories
2024
Founded
Team
The People Behind the Engine
Cade Market is built by a small, focused team with backgrounds in quantitative research, software engineering, and geopolitical intelligence.
Harrison Leggio
CEO & Co-Founder
Harrison focuses on product strategy and client relationships, with a background in risk intelligence tooling and event probability research.
Sofia Mendes
CTO & Co-Founder
Sofia architected the Bayesian scoring engine and calibration infrastructure. Her background covers NLP-driven information retrieval and probabilistic modeling across structured and unstructured data sources.
Dmitri Volkov
Head of Intelligence
Dmitri oversees the source network and calibration methodology. His background is in open-source intelligence collection and geopolitical event analysis, with particular focus on source calibration for risk-focused research.
Principles
How We Think About This Work
Calibration over confidence
A probability score is only useful if its uncertainty is quantified. We track Brier scores and resolution rates because calibration is the standard we hold ourselves to, not precision alone.
Attribution over opacity
Analysts need to see which sources moved a score. A black-box output cannot be attached to a decision record. Source attribution is a core output, not an afterthought.
Instrument over narrative
We built a research instrument, not a news product. The output is a probability with a confidence interval and a source trace, timestamped at the moment of production.
Work With the Team
If you have questions about the methodology, coverage, or how Cade Market fits your risk function, reach out directly.