Abstract In this paper, we formulate the problem of early classifification of time series data, which is important in some time-sensitive applications such as healthinformatics. We introduce a novel concept of MPL (Minimum Prediction Length) and develop ECTS (Early Classifification on Time Series), an effective 1-nearest neighbor classifification method. ECTS makes early predictions and at the same time retains the accuracy comparable to that of a 1NN classififier using the full-length time series. Our empirical study using benchmark time series data sets shows that ECTS works well on the real data sets where 1NN classifification is effective