Tag Archives: Market Neutral

Top 50 Cointegrated Stock Pairs for September 2026: 34,500 Screened, 50 Selected, Five Years Backtested

September 25, 2026 2:02 pm |

 

Every quarter we re-screen the entire U.S. mid-to-mega-cap universe for cointegrated stock pairs and keep the fifty that earned their place across a full five-year window. The September 2026 vintage of our Top 50 USA Stock Pair Stars is live in PairTrade Finder® Ultimate Alpha 3 today. This post explains how 34,500 candidate pairs became 50, what the fifty did in the backtest, and — because a backtest is only half a story — what the backtest cannot tell you.

Hypothetical backtest results — not live trading, not out-of-sample Every figure in this post comes from backtested signals over the same period in which these pairs were selected. … Read More →

Pairs Trading Example: The Vegas Pair That Kept Paying for 5 Years

September 13, 2026 1:35 pm |

Want a real pairs trading example instead of theory? Here’s one straight from our platform: Hilton Grand Vacations (HGV) and MGM Resorts (MGM). Two Las Vegas-rooted leisure stocks. One quietly reliable trading relationship.

The numbers: 12 trades, 12 winners

We ran a 5-year end-of-day backtest at $5,000 per leg. The results speak plainly:

  • 12 completed trades — every one profitable
  • Net profit: $3,122 after $258 in commissions
  • Average profit per trade: $260
  • Maximum drawdown: just 4.02%

The equity curve is a staircase, not a rollercoaster.

… Read More →

Kalman Filters vs. PairTrade Finder®

March 26, 2026 3:38 pm |
Why We Keep It Simple (On Purpose) Kalman filters

From time to time, experienced traders reach out and ask:

“Why doesn’t PairTrade Finder® use a Kalman filter to calculate dynamic hedge ratios?”

It’s a fair question. Kalman filters are widely used in quantitative finance and are often presented as a more advanced way to trade pairs.

So this post explains our thinking clearly.

PairTrade Finder® is designed to capture most of the statistical edge—without the complexity that often reduces real-world performance.

In practice, as a rule we seek to target a platform that delivers 80% of the benefit of institutional grade models with around 20% of the complexity.

… Read More →