Tools
Five free, in-browser calculators and simulators that turn the core ideas of quantitative signal research — measuring skill, alpha decay, breadth, cointegration, and overfitting — into something you can actually play with. No sign-up, nothing to install, your numbers never leave the page.
Each tool isolates one idea from the signal-research workflow and lets you change the inputs to see how the result moves — the fastest way to build intuition. Everything is illustrative and for education, not investment advice.
How predictive is your signal?
Paste your predictions and the returns that followed to get the Pearson IC, the rank (Spearman) IC, a significance t-stat, and a scatter plot — the first number to know about any signal.
How fast does the edge fade?
Enter a signal's IC at several horizons to estimate its half-life, fit an exponential decay curve, and read off the rebalancing frequency that keeps the most alpha on the table.
How many signals do you actually need?
Stack N weak signals and watch correlation cap the payoff. An interactive Fundamental Law of Active Management — effective breadth, the IC ceiling, and why ten correlated signals act like three.
Tradeable pair, or coincidence?
High correlation isn't cointegration. Build two co-moving assets, run an Engle–Granger test on the spread, and see why only a cointegrated pair holds up — and how a structural break wrecks the backtest.
Is your backtest fooling you?
Generate hundreds of zero-edge strategies, keep the best, and watch a spurious Sharpe ratio appear from pure noise — with the expected-maximum-Sharpe benchmark that tells you how much to discount it.
Does diversification survive a crisis?
Drag a market-stress slider across a twelve-asset book and watch the correlation matrix turn red as the effective number of independent bets collapses toward one — why diversification fails exactly when you need it most.
Want the theory behind the tools? Start with the Information Coefficient and Information Ratio explainers, browse the glossary, or work through the curated reading list.