A
Mathematics, dataand discipline.
Firth Quant is a quantitative research and systematic trading firm. We apply the scientific method to financial markets, and build the technology to act on what we learn.
01
The name comes from the firth — a narrow coastal inlet where broad water is channelled into a precise course. It describes what we try to do with data.
Markets generate vast amounts of information, almost all of it noise. Our work is to channel that information through careful measurement, statistical testing and robust engineering until what remains is narrow, precise and actionable.
We bring together people from mathematics, statistics, computer science and the physical sciences who share a preference for rigorous thinking. The firm is organised around a single research process that runs from the first look at a dataset to the monitoring of a live model.
02
Four convictions that shape our research.
P.01
Principle
Evidence before conviction
A view on markets is only as good as the evidence behind it. We prefer a modest claim that survives testing to an ambitious one that cannot be checked.
P.02
Principle
Simplicity until proven otherwise
Complex models are easy to fit and hard to trust. Additional complexity must earn its place with out-of-sample improvement that outlasts its costs.
P.03
Principle
Costs are part of the signal
Spread, impact and fees are not details to subtract at the end. A prediction only matters if it remains valuable after the cost of acting on it.
P.04
Principle
Models decay
Markets adapt. Every model is monitored against its expected behaviour and retired when the evidence says its edge has gone.
03
A research process designed to say no.
STAGE 01
Observe
Study market structure and data without a conclusion in mind. Measure what is actually there — distributions, regimes, frictions — before asking what it means.
STAGE 02
Hypothesize
State a specific, falsifiable claim about market behaviour and the economic or mechanical reason it should hold. No rationale, no research.
STAGE 03
Validate
Test out-of-sample, account for every variant tried, include realistic costs and try hard to break the result. Most ideas should — and do — end here.
STAGE 04
Execute
Deploy only what survives, at sizes risk controls allow, then monitor live behaviour against expectation and retire models that drift.
04
Respect the tails.
Research at Firth Quant is collaborative and written down. Hypotheses, methods and results — including negative ones — are documented and reviewed by peers before they influence any decision. We are sceptical of our own results first, and we design tests intended to break them.
That scepticism extends to assumptions. Real returns are not normally distributed; extreme moves are more common than textbook models suggest. Our research respects that, from model design to risk limits.
05
Engineering is how research reaches the market.
A good idea poorly implemented is a bad trade. We treat software as a core research instrument: reproducible pipelines, tested code, and production systems that behave exactly as they did in simulation. Researchers and engineers share ownership of the full path from data to execution.
06
What we hold ourselves to.
- V.01Research Integrity
- Every decision rests on evidence that can be reproduced and checked. We report what failed as carefully as what worked.
- V.02Engineering Excellence
- Correctness, stability and clarity come before cleverness. Systems are built to be understood by the next engineer, not just the last one.
- V.03Continuous Improvement
- Research, test, measure, refine. Small, verified improvements compound; untested ambition does not.
- V.04Risk Awareness
- Markets are uncertain and models are approximations. Protecting capital and respecting limits is part of the job, not a constraint on it.
- V.05Intellectual Curiosity
- We are drawn to hard problems at the intersection of mathematics, data and technology — and to people who enjoy them.
Get in touch
Building the future through quantitative thinking.
Whether you are exploring a research collaboration, a partnership or a career, we would like to hear from you.