This paper presents an experimental method for predicting tool wear in drilling Inconel 718 superalloy. The method combines analysis of drilling force signals and tool wear progress. Force characteristics were studied both in time and frequency domains (power spectrum and wavelet decomposition) in order to find best correlation with tool wear progress. These analyses show that the mean value of the thrust force component, the high frequency component of the force, the frequencies that arise during drilling, and the evolution of the wavelet decomposition details are all sensitive to tool wear progress. Therefore, these characteristics can be employed as indicators for drill failure prediction. Among all those indicators, the mean value of the thrust force and the standard deviation of high frequency components of that force have shown the greatest sensitivity to drill wear.