← Juncture Lab

How to read a track record

Every Juncture Lab track record is a set of certified runs through real, hidden market history. This page explains how the runs work and what each metric means, so anyone reading a profile or an exported report can judge the results for themselves.

How a certified run works

  • ·Hidden regime. The trader is dropped into a real historical market period with the dates, headlines, and macro context removed. They see live price action but cannot know what year it is or how it ended, so results reflect decision-making, not hindsight or memory.
  • ·Real data. Prices are actual historical market data for the period, not synthetic or randomized.
  • ·A thesis per decision. Trades can carry a written rationale, captured at the moment of the decision. The reasoning is recorded before the outcome is known.
  • ·Server-side, tamper-proof metrics. Every metric is computed on our servers from the trade log after the run ends. The trader cannot edit results, and each run is tied to a verifiable public profile.
  • ·Certified vs sandbox. Only certified runs count toward a track record. Sandbox runs are for practice and are never certified.

What the metrics mean

Performance

Return

The total percentage change in portfolio value from the start of the run to the end.

How to read it: The headline result. A run that starts at $10M and ends at $12.4M returned +24%.

Alpha

Return above the benchmark (SPY) over the exact same period.

How to read it: This is the number that matters most. Positive alpha means you beat the market you were dropped into, not just that the market went up. Anyone can make money in a bull run; alpha isolates skill.

Benchmark (SPY)

The S&P 500 index over the same hidden dates, the default yardstick for a broad-market passive alternative.

How to read it: Your return is always shown next to what you would have made by simply holding the index.

Risk-adjusted

Sharpe ratio

Return earned per unit of total volatility, annualized.

How to read it: How efficiently you turned risk into return. Below 0.5 is weak, above 1 is good, above 2 is excellent. A high return with a low Sharpe means you took a lot of risk to get there.

Sortino ratio

Like Sharpe, but only penalizes downside volatility, not upside swings.

How to read it: Rewards portfolios that are volatile in the good direction. Usually higher than Sharpe.

Calmar ratio

Return divided by the maximum drawdown.

How to read it: Reward per unit of worst-case pain. Rewards steady compounding over jagged, high-drawdown returns.

Risk

Max drawdown

The largest peak-to-trough decline in portfolio value at any point during the run.

How to read it: The deepest hole you were in. A -10% max drawdown means that from its highest point, the portfolio fell 10% before recovering. Lower (closer to zero) is calmer. We label it Controlled, Moderate, or Significant.

Volatility

The annualized standard deviation of daily returns.

How to read it: How much the portfolio swung day to day. High volatility is not inherently bad, but it should be paid for with return (see Sharpe).

Behavior

Win rate

The share of periods (or trades) that were positive.

How to read it: How often you were right. A high win rate with a low return can signal cutting winners early; a low win rate with a high return can signal letting winners run.

Profit factor

Gross profits divided by gross losses.

How to read it: Above 1 means winners outweigh losers. 2.0 means you made twice as much on wins as you lost on losses.

Trades / Trades per day

The number of orders executed, and the average per trading day.

How to read it: Activity level. Neither high nor low is automatically better, but it provides context for the returns.

Autocorrelation

The correlation of daily returns with the prior day.

How to read it: Near zero is expected. Strongly positive or negative can hint at trend-following or mean-reverting behavior.

The market regimes

Each run is one of these historical periods, hidden from the trader until the reveal. The name and dates only appear after the run is complete.

2000 Dot-Com Bust

Tech bubble bursts, NASDAQ collapses.

Mar 2000 – Mar 2001

crisis

2003 Recovery

Post-Iraq invasion rally, early housing boom.

Mar 2003 – Mar 2004

moderate

2008 Financial Crisis

Lehman collapse, global financial meltdown.

Jan 2008 – Jan 2009

crisis

2009 Recovery

March bottom, a massive rally off the lows.

Mar 2009 – Mar 2010

moderate

2013 Bull Market

Steady gains with low volatility.

Jan 2013 – Jan 2014

easy

2015 Volatility

China growth fears, an August flash crash, Fed uncertainty.

Jun 2015 – Jun 2016

hard

2017 Bull Run

Low-volatility tech rally.

Jan 2017 – Jan 2018

easy

2018 Volatility

February VIX spike, Q4 selloff.

Jan 2018 – Jan 2019

hard

2019 Pre-COVID

Fed pivot, trade-war uncertainty, late-cycle rally.

Jan 2019 – Jan 2020

moderate

2020 COVID

The March crash and the recovery that followed.

Jan 2020 – Jan 2021

crisis

2022 Bear Market

Inflation, rate hikes, tech selloff.

Jan 2022 – Jan 2023

hard

Verifying a track record

Every certified run lives on a public profile at juncturelab.com/u/username. Exported reports link back to that profile so a reader can confirm the numbers against the source. Results cannot be edited after a run ends.