What the forecast expects
Likely to beat the market
The model's strongest chances of beating the typical tracked card over the next 60 days, with better-grounded forecasts. Compare the chance and outcome range on each card.
190 cards answer this today
Why ask this
A second model, independent of everything above it, tested on data it was not trained on and graded on how often it was wrong. It disagrees with the interest score often enough to be worth having both: only 36 of the top 100 by score appear in the top 100 by expected return.
How to read the figure
The figure is the estimated chance of beating the typical tracked card over 60 days. 50% is average by construction, so 58% is a modest tilt and not a passing grade.
Shown on every row: Chance to beat market
Where it can be wrong
Every card here also passes a check on how sure the forecast is, because without it the list fills with cards whose listings we don't trust — 20 of the top 25 would be. The range on each of these is wide; open one and read it.