This is a pretty fuzzy analysis. Even with a comparative cumulative relative return of -13.6% for the 51% market-weight portfolio, the investor might realize a superior absolute or risk-adjusted return, given the opportunity to invest the other 49% of her capital in an alternative portfolio. Why not rerun the analysis with the assumption that the other 49% of capital went into Treasuries (an investment that almost certainly would have lower risk as measured by sigma than the SPY portfolio)?