The original ENSO-CLIPER model proposed by Knaff and Landsea (1997) is arguably the leading statistical model for forecasting the Nino 3.4 ENSO index. In this paper, we assess the model's true independent forecast skill as a function of lead out to 10-months for the period 1987-2001. We extend the model to other ENSO indices and assess its 15-year hindcast skill for the seasonal boreal summer prediction of Nino 3, Nino 4 and Nino 3 and 4 combined. We discuss the sensitivities of the model to the assumptions made. Subject to data availability this skill will be compared to that from the leading ECMWF and Met Office coupled dynamical model ENSO predictions 1987-2001.