I present a detailed study of portfolio optimization based on cointegration, a statistical tool that exploits a long-run equilibrium relationship between stock prices and an index price. I compare the theoretical and empirical properties of cointegration optimal equity portfolios with those of portfolios optimized on the tracking error variance. From a nine year out of sample performance analysis I found that cointegration optimal portfolios clearly dominate the TEV equivalents for all of the strategies based on enhanced indexation,. Moreover, I provide some information regarding the performance of financial markets and the equity Italian funds of the last ten years. Finally, I deepen my research with a comparison between an index fund managed by Soprano SGR and the cointegration based portfolio. From a two year out of sample analysis, my tracking portfolios dominate both benchmark, the DJ Stoxx 50, and the Soprano fund.