In this paper, we identify and analyze parameters that determine the profitability of wind power operators in the German market premium model. Based on an empirical analysis of different German wind power profiles from 2007 to mid-2012, we are able to show that the profitability significantly depends on the correlation of the wind power portfolio with the overall wind power feed-in and prediction error in Germany. Significant differences between the wind forecast errors clearing cost of the analyzed portfolios can be identified. Our analysis shows that a wind power operator would profit in most cases from a reduced forecast error, which could be achieved through an improved forecast model and an increased share of the intraday cleared error. Furthermore significant locational portfolio advantages and disadvantages can be identified when comparing the different market values. In general, the empirical. analysis shows that a premium of 3.5 (sic)/MWh is suitable to cover the cost of an imperfect forecast. Taking further into account that for 2012 a premium of 12 (sic)/MWh was granted; the direct marketing option can be evaluated as highly attractive, which is furthermore indicated by the rapid increase of the directly marketed wind power and photovoltaic generation. (C) 2016 Elsevier Ltd. All rights reserved.