Notes on evidence
What we learn from measuring trading ideas on real data — including the ideas that turn out not to work.
The single-window backtest trap: how 4-fold Walk-Forward and Regime Matrices expose curve-fitting
A strategy can pass a single 70/30 out-of-sample split if the last 30% happened to be a favourable bull run. Combining 4-fold Walk-Forward and Bull/Bear/Chop regime decomposition stops that.
Read the article →- October 18, 2026 · 1 min read
Why is it moving? Multi-market catalyst attribution and event-study tracking across Crypto, US Stocks and BIST
How Evident combines 13 news feeds, empirical event-study impact matrices, derivative microstructure and a live macro/options calendar into one Market Tracking Terminal.
- October 16, 2026 · 1 min read
Order flow inside the candle: VPIN toxicity, Cumulative Volume Delta (CVD) and Short Squeeze mechanics
How intrabar volume delta, Easley-Lopez de Prado VPIN flow toxicity and perpetual Volume-to-Open-Interest (Vol/OI) velocity reveal institutional absorption and liquidation squeezes.
- October 15, 2026 · 1 min read
Why the Sharpe ratio is not enough: measuring tail risk with Omega, Ulcer Index, SQN and Half-Kelly
The Sharpe ratio assumes Gaussian returns and penalises upside breakouts as risk. Here are the institutional metrics that actually measure drawdown stress and trade quality.
- October 14, 2026 · 1 min read
Is the market trending or chopping? Regime detection with the Hurst Exponent, Shannon Entropy and FDI
How Rescaled Range (R/S) Hurst analysis, Shannon information entropy and Sevcik Fractal Dimension separate persistent trends from mean-reverting chop.
- October 12, 2026 · 2 min read
Cutting moving-average lag without overfitting: Kalman and Ehlers DSP filters on Bitcoin
Why classical moving averages lag price turns by half their window length, and what happened when we tested state-space Kalman and Ehlers SuperSmoother filters on 1,000 daily Bitcoin candles.
- September 27, 2026 · 2 min read
Optimising a strategy without fooling yourself
Parameter optimisation finds the best settings for the past, which is exactly the problem. How walk-forward folds and a held-out tail show whether the best settings mean anything.
- September 27, 2026 · 2 min read
A trading forum in every language, with moderation you can see
The Evident forum lets you write in your own language and read everyone else in yours. Here is how translation, limits, reports, mutes and bans work.
- September 27, 2026 · 3 min read
How AI credits work on Evident, and why most of the product needs none
What a credit is, what each AI job costs, when credits refill, why failed jobs are refunded, and how your own model key takes credits out of the picture.
- September 27, 2026 · 2 min read
Scoring the news is easy. Measuring what it did is the point.
Most news tools stop at "bullish" or "bearish". Evident scores each headline, then measures what the price actually did 1, 4 and 24 hours later against the market — and only shows an expectation when enough history exists.
- September 27, 2026 · 2 min read
An AI trading agent that cannot run away with your money
How Evident's AI agent decides, why every proposal passes a deterministic risk engine first, and why it trades only on decentralised venues with keys that cannot withdraw.
- September 27, 2026 · 3 min read
Does your indicator know anything? The information coefficient, explained with real numbers
An indicator can look convincing on a chart and still say nothing about the next candle. The information coefficient measures how well it ranks future returns. What it is, how to read it, and what ten popular indicators scored.
- September 27, 2026 · 2 min read
From Pine Script to evidence: test a TradingView indicator in two minutes
Paste a TradingView Pine Script indicator into Evident's lab, see which lines convert and which do not, measure whether it carries information on real data, use it in a strategy and export it back to Pine.
- September 27, 2026 · 2 min read
Why most backtests lie — and six checks that catch it
A rising equity curve is the easiest thing in trading to produce and the least informative. Six checks that separate a real edge from a lucky fit, and how Evident runs them on every strategy.
- September 27, 2026 · 3 min read
We measured 22 candlestick patterns on four markets. None beat an ordinary candle.
Hammers, engulfing candles, dojis and morning stars on the daily charts of Bitcoin, Ethereum, Solana and the S&P 500: what happened in the five candles after each one, compared with an ordinary candle.