Research notes
What we tested, what the evidence said, and what did not work. Negative results are published here alongside everything else.
Every figure in a note comes from backtests: a strategy's rules run over historical prices. None comes from money that was actually traded, and past results do not predict future returns. Most strategies we test do not beat holding, and the notes say so.
Every note we publish stays here until we take it down. The methodology explains how the numbers are produced, and each ticker verdict shows how many backtests sit behind it.
Nvidia after Jensen Huang: what is happening now
Jensen Huang has resigned as Nvidia CEO. What the 17 November results, guidance, the Rubin roadmap, China and competition mean, as of 18 November 2026.
What our NVDA trading strategy backtests show
In our backtests of NVDA, 3 of 553 beat buying and holding. Figures from the published card, with no trading costs.
Dollar cost averaging: definition and example
Dollar cost averaging for investors buying a US-listed index fund: how the method works, a dollar example, and how it compares with investing a lump sum.
Euro cost averaging: definition and example
Euro cost averaging is pound and dollar cost averaging applied to euro-priced funds: how it works, a euro example, and the currency risk for UK investors.
Momentum investing explained
A plain guide to relative and absolute momentum, how ranking rules are formed, and why past strength is not a forecast.
Overfitting in backtests
How strategy settings can fit historical noise, and how an untouched held-out period helps test whether a pattern travels.
Survivorship bias in backtests
Why testing only today’s surviving companies can flatter historical results, and how missing delisted prices limit a backtest.
Trend following explained
How trend-following rules respond to persistent price moves, what signals can and cannot tell you, and how to test them.
What is a stock market correction?
A plain English guide to market corrections, how they are commonly defined, how often they occur and how they differ from bear markets.
How leveraged ETFs work, and what volatility drag is
How a leveraged ETF such as TQQQ resets each day, why it can fall behind three times the index over time, and what our TQQQ backtests show.
How to backtest a trading strategy: a plain guide
What a backtest is, the steps to run one properly, and the mistakes that make results look better than they were.
How to read a backtest equity curve
What an equity curve shows, the five things to look for, and the shapes that should make you more careful about a backtest result.
How to read a drawdown table
What each column in a drawdown table means, how to compare two of them, and the mistakes people make reading depth without duration.
ISA or general account: what active investors should know
How a stocks and shares ISA and a general investment account differ for people who trade often, explained as education, not advice.
Maximum drawdown explained
What maximum drawdown measures, how to calculate it, and why the largest fall from a peak matters as much as the return.
Overfitting and out-of-sample testing, in plain English
Why testing many settings makes some look good by chance, and how holding back data and testing elsewhere separates luck from a result.
Position sizing basics: how much to put in each trade
What position sizing is, the common methods, how sizing changes risk as much as the choice of strategy, and how it shows up in backtests.
Pound cost averaging vs lump sum: what each means
Pound cost averaging and lump-sum investing explained for UK investors: how each works, the trade-offs, and what the evidence can and cannot settle.
Survivorship bias explained, and why it flatters backtests
What survivorship bias is, how it makes past results look better than they were, and the questions to ask of any backtest or fund list.
200-day moving average, golden cross and death cross
What the 200-day moving average is, how to calculate it, what golden and death crosses are, and how to judge rules built on them.
The Sharpe ratio in plain English
What the Sharpe ratio measures, how it is calculated, what counts as high or low, and the limits worth knowing before you rely on it.
What a regime filter is, and why results change by regime
What a market regime is, how a regime filter works, and why beating buy-and-hold in a crash means less than in a rising market.
What a stop-loss does to backtest results
How a stop-loss changes a strategy's trades, win rate and drawdown in a backtest, and the trade-offs to check before trusting the result.
What a win rate tells you, and what it hides
How to calculate a trading win rate, why a high one can sit beside a strategy that trails buy-and-hold, and the figures to read next to it.
What "buy the dip" means, and how to test it
Buy the dip in plain English: what the phrase means, why the idea appeals, how to turn it into a rule you can test, and where to see it tested.
What CAGR means and how to calculate it
Compound annual growth rate in plain English: the formula, a worked example, and why it is the fairest way to compare a strategy with holding.
What "work in progress" means on a research result
Why some StrategyLens results carry a work in progress label, what held up, did not hold up and inconclusive mean, and how to read each.
Why most strategies underperform buying and holding
Why timing rules usually trail simply holding the same stock, with figures from 38 published tickers and what the exceptions tell you.
The stocks that disappeared, and why they matter
Almost four in ten US stocks I can find no longer exist. Leave them out and every backtest looks better than it should. Here's what I'm doing about it.
How we test, and what the latest tests found
A plain account of the tests since the last post, including the ones that did not go our way, and what that means for the five headline stocks.
I tested my own strategy and it didn't work
What I expected, what the evidence said, and why I'm publishing it instead of burying it.
Investing and backtesting glossary
Plain-English definitions of 15 investing and backtesting terms, from CAGR and the Sharpe ratio to survivorship bias, with links to the research notes.