Research · notes from the desk

We publish the hard parts.

Not the strategies — the discipline. These are working notes on the problems that quietly kill quantitative research: signal decay, survivorship, overfitting, costs, and the reproducibility you can't fake.

§ methodology§ data§ backtesting§ risk§ execution
2026-08-18
· 6 min read

On the half-life of a signal

Every signal decays; the only question is the slope. We walk through how we measure factor decay, why it should set your rebalance frequency rather than your gut, and what happens when you ignore the half-life entirely.

methodologysignal decayturnover
2026-07-02
· 8 min read

Survivorship is not a data bug

It's a research bias. When a history quietly drops the companies that delisted or failed, every backtest looks better than it was. We explain why point-in-time data is the difference between a strategy and a survivorship artifact.

datasurvivorshippoint-in-time
2026-05-14
· 7 min read

Backtests that flatter you

Multiple testing, parameter search, and the quiet pressure to show a line that goes up. On how we keep our own backtests honest — and the specific traps we built the platform to refuse.

backtestingoverfittingmethodology
2026-04-09
· 6 min read

Transaction costs are a feature, not a footnote

The strategy that looks great before costs and mediocre after isn't a strategy — it's a cost transfer. Why execution-aware backtesting has to be the default, and what "net of costs" actually has to include.

executioncostsbacktesting
2026-02-19
· 5 min read

Why we version research, not just code

Code without the data it ran on is a claim you can't check. On pinning datasets to version hashes, and why "it worked last month" is not a reproducible result.

methodologyreproducibilitydata
2026-01-14
· 6 min read

Point-in-time means two timestamps

A number has both a date it describes and a date you could have known it. Most datasets only carry the first. Why the second is where look-ahead bias actually hides, and how we store both so a backtest can't quietly use the future.

datapoint-in-timelookahead
2025-12-04
· 8 min read

Volatility is not risk, but it's a start

A risk model that only sees realized volatility misses the drawdown that's still coming. On the limits of variance as a proxy, and what ex-ante risk actually needs to capture.

riskvolatilitydrawdown
2025-10-22
· 6 min read

The quiet cost of corporate actions

A split applied a day late, a dividend that never landed. These aren't edge cases — they're the difference between a clean backtest and a silent error that compounds for years. What we check, and why.

datacorporate actionscleaning
2025-09-08
· 7 min read

Regime detection without the hindsight

Every regime-change model looks brilliant in hindsight and mediocre live. On the lookahead bias hidden in "adaptive" strategies, and how to build regime awareness that only uses information available at the time.

methodologylookaheadregime
2025-08-12
· 4 min read

What we mean by "production-grade" research

The founding note. Why we started VertiMax, and the definition of rigor we hold ourselves to: no black boxes, no survivorship, and every number traceable to its inputs.

methodologyabout
2025-06-30
· 7 min read

The deflated Sharpe, and why we report it

Search enough parameters and you'll find a great backtest by accident. On multiple-testing correction, how the deflated Sharpe ratio prices the search itself, and why we log every parameter we tried instead of only the one that worked.

backtestingoverfittingmultiple testing
2025-03-20
· 5 min read

Drawdown is a duration, not a number

A −20% drawdown that recovers in a month and one that takes three years are different risks wearing the same label. On time-to-recovery, the underwater curve, and why the peak-to-trough number alone hides the part that hurts.

riskdrawdownrecovery
Method · how we keep ourselves honest

Three rules we don't break.

Every result is tied to a versioned dataset and a cost model — not a screenshot and a shrug.
If it can't be reproduced by someone who wasn't in the room, it didn't happen.
We assume our own backtest is lying to us, and make it prove otherwise before we trust it.
Release notes · what shipped, and when

Changelog.

We ship in public. Each release carries a version number and a record of what changed, so you can pin your research to the platform you actually used.

v2.4.0
2026-08-11
Execution analytics (beta). Implementation shortfall and slippage measured against arrival price; TCA reconciliation of simulated fills to actual executions; borrow and financing carried through the cost model.
v2.3.2
2026-06-24
Point-in-time restatement for fundamentals across all equity universes, with as-of and known-at timestamps on every field.
v2.3.0
2026-04-02
Walk-forward and anchored out-of-sample splits as native constructs; deflated Sharpe and parameter-search reporting.
v2.2.0
2026-01-19
Square-root market impact model with per-venue and liquidity-band calibration.
v2.1.0
2025-11-06
Survivorship-free index membership and corporate action engine v2 — splits, dividends, mergers, and spin-offs applied on effective date.
v2.0.0
2025-08-14
Unified data layer GA. Deterministic replay and version-hash pinning for every dataset.
v1.4.0
2025-05-02
Point-in-time price history and the first cost-aware backtester in early access.