Precision signals for systematic traders.
Granular breakdowns of technical tools, mathematical overlays, and signal reliability in volatile markets.
A TradingView backtest can overstate strategy performance by 15% to 50% before the first realistic execution assumption is added. That is not a minor reporting…
Kyle Donnelly·17 min read
A backtest can look precise while quietly omitting the part of the market that matters most.
Kyle Donnelly·18 min read
$7,400. That is what one retail algorithmic trader lost when a functioning strategy met an unprepared execution layer. The loss was not evidence that the…
Kyle Donnelly·20 min read
What is algorithmic trading? At the regulatory level, it is not "using a bot" and it is not "asking ChatGPT whether to buy NVDA." It is a system in which a…
Kyle Donnelly·14 min read
A broker API is not an execution edge. It is a transport layer with authentication, rate limits, failure modes, and an inconvenient habit of exposing every…
Kyle Donnelly·13 min read
A backtest with a 100% win rate is not impressive. It is a bug report.
Kyle Donnelly·13 min read
A correlation matrix that reads 0.30 in calm markets can print above 0.80 when liquidity disappears. That is not a rounding error. It is the entire risk model…
Kyle Donnelly·13 min read
A backtest Sharpe ratio has almost no explanatory power for live performance in one study of 888 algorithmic strategies. The reported fit was below 0.025.
Kyle Donnelly·15 min read
A strategy I tested last month printed a 2.1 Sharpe in-sample, 18% annualized return, and a maximum drawdown under 7%. Nice equity curve. Smooth enough to sell…
Kyle Donnelly·16 min read
My proprietary mean-reversion strategy posted a Sharpe of 2.4 across 14 months of out-of-sample data. The equity curve was a work of art — smooth, monotonic…
Kyle Donnelly·15 min read
