This paper studies the forecasting abilities of a battery of univariate models on hourly electricity spot prices, using data from the Leipzig Power Exchange. The speci cations studied include autoregressive models, autoregressive-moving average models and unobserved components models. The results show that speci cations where each hour of the day is modelled separately present uniformly better forecasting properties than speci cations for the whole time series, and that the inclusion of simple probabilistic processes for the arrival ofextreme price events can lead to improvements in the forecasting abilities of univariate models for electricity spot prices.