This paper explores a channel whereby new anomalies can appear as investors alter portfolios suggested by the dissemination of academic knowledge. In particular, I find that assets with a low realized CAPM alpha outperform those with a high one. In a multifactor world the CAPM is misspecified. I find evidence that its widespread application since the 1960s and the Smart Beta factor literature that followed generated incentives for managers (and possibly other investors) to tilt portfolios systematically towards high CAPM alpha. To do this, managers must tilt portfolios away from low CAPM alpha assets, causing such assets to be undervalued.ct.