Susan Potter

A programmatic premarket overview dashboard that combines market profile, volume profile, cash index context, expected moves, gamma exposure, and economic calendar into a tabbed single-page visualization, generated automatically before the opening bell.

You can’t quantify what you haven’t observed.

Backtests have a testing problem: you often do not know the correct output.

Shuffle your signal, re-run the backtest 10,000 times, see if your strategy beats random.

You have one history of market data.

Every backtest is biased.

Walk-forward validation is the backbone of out-of-sample testing for trading strategies, but the choice of window type, window length, and step size introduces meta-parameters that are themselves easy to overfit.

Autocorrelated returns inflate Sharpe ratios, invalidate standard significance tests, and make backtests look better than reality.

A strategy built on a non-stationary signal is a strategy built on sand.

Property-based testing generates random inputs and checks invariants.

The quant research funnel maps almost directly onto the software testing pyramid.

Productivity in quantitative work isn’t about doing things faster.

Traditional databases store current state.

Market data comes in many formats: FIX messages, exchange feeds, CSV exports, JSON APIs.

What Susan Potter is focused on right now: building backtesting infrastructure and strategy validation pipelines at Referential Labs, writing about quantitative methods and property-based thinking, and reading through options market microstructure researc…

Content All articles, code examples, and analysis on this site are for educational and informational purposes only.

What This Site Collects This site uses Google Analytics (GA4) to understand which articles are read and how visitors find the site.

Get in Touch Email: me@susanpotter.net LinkedIn: /in/susanpotter BlueSky: @susanpotter.net Mastodon: @SusanPotter@mastodon.social Work With Me I build quantitative infrastructure for trading firms, family offices, funds, and commercial traders: validation…

I know, I know.

Why chasing the perfect backtest, optimal parameters, or flawless infrastructure can stall your trading research.

Explore how to apply property-based thinking to the practice of software optimization.

Explore useful mental models for software engineers to apply to the practice of software optimization

Digital twins can be valuable, but they’re not a silver bullet.

A ISO 8601 guide to date, time, duration, interval formats and more for working developers.

Smart constructors validate data at construction time, preventing invalid states from existing.

This article explores Haskell’s techniques for isolating side effects, starting with the IO monad, which encapsulates impure code to enable reasoning about pure portions.

Dive into an educational journey exploring the power of strong static typing, enforcing domain-specific constraints, and ensuring the validity of values.

In this in-depth exploration we discover the power of mapped types in TypeScript through numerous examples.

Are you a TypeScript developer who loves Deno’s safe and sandboxed environment and its out-of-the-box TypeScript, LSP, linting, formatting, stdlib testing and benchmarking support?

A comprehensive guide that explores the power of DuckDB for efficient and effective data exploration.

Dive into TypeScript's conditional types and discover how they empower developers to create flexible type mappings and enhance type safety.

Join us as we dive into the world of serverless hosting and uncover the hidden challenges that come with this revolutionary approach.

Discover how to extract and analyze your Twitter archive using DuckDB, a lightweight and fast embeddable SQL database.

This article delves into the world of functional programming with the Flix programming language with higher-order functions and user-defined infix operators.

Are you ready to embark on a coding adventure with Flix?

A comprehensive guide for CTOs and engineering leaders evaluating unikernels as a strategic technology investment, covering business impact, implementation considerations, and decision frameworks.

The reMarkable 2 e-ink tablet has become essential to my quantitative research workflow, from reading statistics papers to annotating proofs inline and teaching mathematical concepts via screen sharing.

Discover the power of scoping change differentials in Git by focusing on specific files, line ranges, functions, methods, or classes.

Discover the incredible potential of the keyof operator in TypeScript and revolutionize your code for a better development experience.

Originally published as a Twitter thread starting here.

Part 3 examines the Profunctor typeclass in detail: the dimap operation, its laws, and the Strong and Choice extensions that enable profunctors to work with products and sums.

Part 1 showed paired operations like encode/decode and serialize/deserialize.

Learn how to leverage TypeScript's enums, const enums, ambient enums and how they impact generated JavaScript code.

It recently came to my attention that not everyone overrides the default merge.conflictStyle git-config setting.

Software development involves many paired operations: encoding and decoding, serializing and deserializing, encrypting and decrypting.

Dhall's merge function provides exhaustive pattern matching for union types.

Reader is a pattern for threading read-only configuration through a computation without passing it as an argument at every call site.

PureScript has no subtyping.

Unlock the mathematical elegance of algebraic data types (ADTs) and revolutionize your understanding of data modeling!

Note: the code for the TypeScript (to help you follow along) is available here: Encoding Algebraic Data Types in TypeScript code Background Recently I've been reviewing TypeScript code at work.