Deep Dives
Deep Dives
Deep Dives is for readers who want the numbers behind the claims: how many trades it takes to tell skill from luck, how position sizing and the Kelly criterion change growth and drawdown, how Monte Carlo resampling reveals the risk a single equity curve hides, why the best of many backtests is usually overfit, why martingale and grid systems look perfect until they blow up, how much transaction costs eat a busy signal, and what decades of research say about carry and technical rules in currencies. Each piece is worked through with real arithmetic and the original research named.
Start here
Statistical Significance of a Track Record: How Many Trades It Takes to Tell Skill From Luck
How to test whether a signal provider's results reflect skill or luck, using t-statistics, the Sharpe ratio standard error from Lo (2002), and worked numbers.
Read this first →More in Deep Dives
Position Sizing and the Kelly Criterion: Fixed Fractional, Full Kelly, Half Kelly and Why Overbetting Ruins
How fixed fractional sizing and the Kelly criterion work, what Kelly, Thorp and MacLean, Thorp and Ziemba proved, and simulations of full vs half Kelly growth and drawdowns.
Read the explainer →Monte Carlo Simulation for Trading Systems: Resampling Trades to Find the Real Drawdown Risk
How Monte Carlo resampling of a trade record works, shuffle vs bootstrap vs block bootstrap, and a worked 20,000-path simulation of drawdowns, losing streaks and loss odds.
Read the explainer →Overfitting and Data Snooping: White's Reality Check, Haircut Sharpe Ratios and the Deflated Sharpe Ratio
Why the best of many backtests looks skilled by chance, how White's Reality Check, Harvey and Liu's haircuts and the Deflated Sharpe Ratio correct for it, with worked simulations.
Read the explainer →Martingale and Grid Strategies Explained: Why They Look Great Until They Don't
The maths of doubling-down and grid trading in forex: why the expected value is zero before costs, how margin limits and spreads change it, and simulations of ruin.
Read the explainer →Transaction Costs in Retail FX: Spreads, Commissions, Swaps, Slippage and the Cost of a Busy Signal
How retail forex costs are built, what the research says about them, and a worked example showing why a high-frequency signal can lose more than the account to costs in a year.
Read the explainer →The Carry Trade Explained: Interest Differentials, the Evidence and Crash Risk
How the currency carry trade earns the interest gap between currencies, what Brunnermeier, Lustig, Burnside and others found, and a worked crash scenario on Fed data.
Read the explainer →Do Technical Trading Rules Work in FX? Fifty Years of Evidence, and a Test on Fed Data
What the academic record says about moving-average and filter rules in currencies, why their profits faded after 1990, and a net-of-cost test on Federal Reserve H.10 data.
Read the explainer →