Trend sleeve
11.4% CAGR · 1.11 Sharpe · WF 0.73. Solid validation and acceptable benchmark-relative behavior.
Benchmark-Aware Compare Example
Compare is most useful when several candidates all look plausible in isolation. This example shows how Alphrex uses validation and benchmark-relative framing to choose the cleaner strategy instead of the loudest headline.
Example fit
Use this page when the real job is selection under uncertainty: one momentum sleeve, one crossover sleeve, one regime-conditioned variant, and a need to decide which one deserves further time.
Example candidate set
Fast crossover sleeve
13.2% CAGR · 1.05 Sharpe · WF 0.22. Better headline return, but the holdout support is thin and turnover costs are doing more damage than the first chart suggests.
Regime-conditioned sleeve
10.6% CAGR · 1.18 Sharpe · WF 0.81. Slightly lower headline return, but stronger fold stability and a cleaner story about when the strategy should be allowed to act.
Why the leader wins here
1. Validation survives the sample split
The regime-conditioned sleeve is not the prettiest on CAGR, but its walk-forward average and worst-fold behavior are materially stronger than the fast crossover variant.
2. Benchmark-relative context stays coherent
Its active return remains positive without requiring a suspiciously soft benchmark. That matters more than raw outperformance against one easy proxy.
3. Decision contract stays cleaner
The strategy that deserves the next unit of attention is the one with the fewest unresolved objections. Compare is useful because it makes those objections sit in the same table.
4. The outcome is a workflow action
The leader should move into paper or a stricter saved-result review. The weaker candidates should be dropped or demoted, not carried forward out of optimism.
What compare still does not settle
Forward drift
Compare only tells you which historical case is cleaner right now. It does not replace the paper layer that tests whether the winner still behaves coherently after promotion.
Economic benchmark judgment
Benchmark-aware compare helps, but it does not remove the need to choose a defensible opportunity set. A technically tighter basket can still be economically wrong.
Execution feasibility
Turnover, cost, and benchmark checks make the workflow harder to game, but they still do not replace broker-side implementation design.