Figure 1 makes a dramatic visual point about the stability of technical trading rules. It opens deep
philosophical questions about data snooping, and stationarity. Has something about the dynamics of stock
prices changed over the past 10 years, or was the original trend following strategy mined out of the previous
90 years of data? Results in (Sullivan et al. 1999) suggest that it was a change in the data, since their test
attempts to adjust for data mining in the previous sample. However, no test for data mining is perfect,