Skip to main content
Recurring Trade
Sign in

Build a portfolio that doesn't move as one.

Pick a benchmark, like the S&P 500, or your own portfolio, and find the assets that move independently of it, or against it. Then build a basket step by step, to help reduce risk.

Join the waitlist See what's inside

Free during Alpha. No credit card required.

Recurring Trade Correlation Screener showing correlation results, a selected basket and the simulated equity curve panel

Diversify on purpose.

Set your target a benchmark, your portfolio, or a ticker

Every scan needs an anchor. Choose a standard benchmark, your own portfolio, or any ticker, then pick how far back to look. Correlation is not fixed, so the right window depends on what you're building.

  • Compare against a benchmark, your own portfolio, or a single ticker
  • Look back from 30 days to 3 years, depending on what you're building
  • Short and long windows can disagree, so it's worth checking both

What you can target

Pick the anchor everything else is measured against.

Benchmarks
Standard indices and commodities: S&P 500 (SPY), Nasdaq 100 (QQQ), Treasuries (TLT), gold (GLD), crude oil (USO).
My Portfolio
Correlates everything against your own portfolio's daily value changes, so results reflect your specific mix, not just the wider market.
Custom ticker
Enter any stock or ETF to find what moves with it, or against it.
Lookback window
30 to 90 days for tactical, short-term hedges. About a year for the standard structural view. 2 to 3 years for long-term, all-weather portfolios.

Uncorrelated, hedges or lookalikes three ways to sort the whole market

Once you have a target, sort every asset by how it relates to it. Filter by sector, correlation range or volatility, and check your own holdings for risk you're already doubling up on.

  • Uncorrelated: closest to 0.00, for pure diversification
  • Hedges: closest to -1.00, assets that have tended to move the opposite way
  • Positive: closest to +1.00, for cheaper or alternative stand-ins for an asset you already hold

Example: Turn on "Holdings only" to check your existing portfolio for redundant risk instead of scanning the wider market.

How to read the colors

Every match is shaded by its correlation to your target.

Red
Highly positive correlation: redundant risk.
Soft green
Near zero: an independent return stream.
Strong green
Highly negative: a potential hedge.

Check the overlap before you commit to a basket

Select assets from the results table and they're added to a correlation matrix, every one checked against every other one you've picked.

  • Mostly green cells mean the basket is well diversified
  • A bright red cell means two assets have been moving together
  • Drop one of a highly correlated pair without giving up much of the return

Example: CORN and SOYB, both agricultural commodities, show the highest pairwise correlation here at 0.71, a duplicate bet worth trimming. Every other pair sits under 0.40, which is why most of the grid is shaded green.

Recurring Trade correlation heatmap matrix comparing 10 selected assets pairwise, mostly shaded green

Test it risk parity, not equal dollars

Instead of putting equal dollars into each asset, risk parity puts equal risk into each one, so a volatile stock and a calm bond both pull their weight in the result.

  • Equal dollars lets your most volatile asset dominate the portfolio
  • Risk parity weights each asset by its historical volatility instead
  • Compare the simulated curve against your target, side by side

Example: On Mar 30, 2026, the example above shows the equal-weighted portfolio up 9.52% and the risk-parity portfolio up 9.09%, while the benchmark (SPY) was down 3.76% over the same stretch. Both diversified baskets held their ground while the single benchmark didn't. Hypothetical $1,000, using daily closes with no rebalancing costs, to show the structural effect of diversification, not a return forecast.

Recurring Trade simulated equity curve comparing an equal-weighted portfolio, a risk-parity portfolio and the benchmark target, with a tooltip showing returns on a selected date

Verify it holds up not just a backtest on data it already knows

Every basket also gets tested out-of-sample: built on part of its history, then checked against the rest, the part it never saw while being built. A basket that only looks good on the data it was built from is a warning sign, not an edge.

  • Compare in-sample and out-of-sample Sharpe, return, volatility and drawdown side by side
  • Choose a monthly or quarterly rebalance, and a transaction cost of 0, 10 or 25 bps
  • A Monte Carlo range and a bootstrapped Sharpe distribution stress-test the result further

Example: In the example above, both diversified baskets held up out-of-sample, with Sharpe ratios close to their in-sample values, and both cut the benchmark's -33.7% drawdown roughly in half. This run didn't show risk parity beating an equal-dollar mix on raw return, the two methods just traded return for volatility differently. Hypothetical performance, assuming monthly rebalancing and a 10 bps turnover cost.

Recurring Trade Backtest and Verify panel comparing in-sample and out-of-sample Sharpe, return, volatility and drawdown for a risk-parity and an equal-weighted basket against the benchmark

Correlation is measured on historical returns and can change. The simulated equity curve and out-of-sample backtest are hypothetical: the equity curve uses daily closes with no rebalancing costs, and the out-of-sample test assumes monthly rebalancing and a 10 bps turnover cost. Neither is a return forecast. Diversification does not guarantee a profit or protect against loss. Past performance and historical results don't guarantee future results.

About the Correlation Screener

Is the Correlation Screener the same thing as Asset Comparison?
No. The Correlation Screener searches the whole tracked universe to find candidates. Asset Comparison takes tickers you've already picked and overlays their performance, seasonality and correlation side by side. Most people use the screener to find candidates, then compare the shortlist.
What's the difference between an uncorrelated asset and a hedge?
An uncorrelated asset, correlation near 0.00, moves independently of your target, so it adds a genuinely different return stream. A hedge, negative correlation, has tended to move the opposite way, which can offset losses elsewhere. Both reduce how much of your risk rides on one thing; a hedge is the more direct form.
How does targeting "My Portfolio" work?
It calculates correlation against your own portfolio's daily value changes instead of a benchmark, so the results reflect your specific mix rather than the wider market.
Does a low correlation guarantee diversification?
No. Correlations are measured on past returns and can change, especially in stressed markets. Use the screener to research candidates, not as a promise of lower risk.
What does "Holds up" mean?
It means the basket's Sharpe ratio on data it wasn't built from, the out-of-sample period, stayed close to its in-sample Sharpe. A result that only looks good in-sample and falls apart out-of-sample is a sign of overfitting; holding up out-of-sample means it wasn't.

Find your diversifiers.

Join the waitlist and we'll email you as soon as your access is ready.

100% Free during Alpha No credit card required