Testing methodology: data quality, walk-forward analysis, Monte Carlo, overfitting detection, forward testing.
Backtests are easy to manipulate. Any decent developer can curve-fit a strategy to historical data until it looks like a gold mine. When you rely on backtest-only rankings, you…
Most backtest reports give you a single max drawdown number. That is a dangerous way to look at risk. It assumes your trade history is a fixed sequence, but in reality, the order…
Most beginners obsess over win rate, thinking a 70% win rate guarantees success. It doesn't. If you don't account for the size of your wins versus your losses, you are flying…
Most EA vendors focus entirely on the equity curve, but they conveniently ignore the math behind the risk of ruin. If you want to survive, you need to understand the probability…
That '90% modelling quality' bar you see in MT4 backtest reports is effectively a vanity metric. It usually just means the tester successfully interpolated price movements between…
Walk-forward analysis is your best defense against curve-fitting. Instead of testing over the whole dataset at once, you break it into chunks. You train on an in-sample (IS)…
Quick map of **quant backtesting** stacks that show up here (pros/cons, not ads): 1. MT5 Strategy Tester — free, broker-book dependent 2. Python notebooks — flexible, you own…
Publishing the walk-forward checklist I refuse to skip: - IS/OOS window lengths written down before seeing results - Anchored vs rolling chosen on purpose - Parameter stability…
Trying to map the [QuantMogul](https://quantmogul.ai) surface honestly for a retail workflow: - Web: research / agentic / marketplace-ish views - Desktop: walk-forward, Monte…
After reshuffling a backtest trade list 10k times, median max DD is often ugly. Question: which percentile do you size to — 90th, 95th, 99th — and why? I use…
Choosing a backtesting stack usually comes down to how much you value coding flexibility versus raw speed of implementation. Most retail quants start with TradingView. Its…
Does the difference between 90% and 99% modelling quality actually change results, or is it strategy tester vanity? I assume the answer is "it depends", so the real question is:…
Ran 5 years of backtests, 99% modelling quality, profit factor 2.1, realistic spread, slippage modelled. Live result: -40% in 11 weeks. What did I miss? Genuinely asking — I want…
After 8 years of building EAs, this is my personal checklist. Score 3 or more and the strategy is probably noise, in my experience: 1. More than 4 optimized parameters 2.…
I'm building walk-forward optimization into my process. The literature says everything from 4 to 50 windows. Practically: how many in-sample / out-of-sample windows do you run…
If you reshuffle the trade order of a backtest 10,000 times, the drawdown distribution is brutal — the median max drawdown is often 1.5-2x the single historical sequence you…
I ran two EAs live (small size) and re-backtested the identical period every month, logging the divergence. Six months of data: **EA-A (EURUSD H1 trend):** backtest PF 1.61, live…
Sharing a full walk-forward run because most WFO posts show only the winners. Strategy: deliberately boring — Donchian 55 breakout entry, ATR(20) trailing exit, EURUSD H1, 0.5%…
Retail options I see: - MT5 Strategy Tester - Python (backtrader / vectorbt / custom) - [TradingView](https://www.tradingview.com) Strategy Tester + Pine - Desktop quant apps such…