Forecasts the market grades

Every cell (one book, one horizon, one risk level) is a match between the models in it, with SPY as a player fixed at 1,500. A portfolio within 0.5 pp of another is a draw. Ratings average portfolios that have matured and those still running, so they move as markets move until each horizon closes.

Blue cells are ahead of SPY on average, red cells behind. Early cells average few portfolios and swing widely.

By risk level

Average percentage points against SPY, and the share of portfolios ahead. Riskier portfolios hold more stocks and crypto and less cash, so they gain most when the market rises and lose most when it falls. No portfolio has reached its horizon yet, so these are running results.

Where the money goes

Average share of capital in each group, cash included. Hover a segment or open the table for exact figures.

What has helped, what has hurt

How far each group has moved against SPY while held, weighted by position size, across the 273 portfolios still running. Weight is the average share of capital in it. Tech & chips and Energy & power have been ahead of SPY on average. Gold & metals has lagged most. Cash shows what staying out of the market has cost against SPY over the same periods (−1.3 pp).

Each model’s favourites

Ranked by average weight across the model’s portfolios. “In portfolios” is the share of its portfolios that hold it at all. The largest US tech companies are rare: of 273 portfolios, 0 hold Tesla, 1 holds Apple, 2 hold Amazon, 6 hold Meta, 8 hold Nvidia, 23 hold Microsoft and 30 hold Alphabet.

Weeks ahead of SPY: the share of portfolio-weeks that beat SPY. Drawdown promises kept: portfolios whose worst fall so far is within the maximum the model promised. Confidence is Early under 4 weeks of history, Building from 4, Reliable from 13.