Since introduction of the German refined DRG system, the, in relation to other countries, relatively long hospital stay has been reduced significantly. The analyses presented here examine the hypothesis that only few impact factors allow for good hospital management and that length of stay is a performance indicator. The analyses were based on remuneration data of 20 German hospitals for 2007-2009, comprising details of 963 923 patient cases, and cost data for 2007-2010. Using simple and multiple linear regressions, the impact of hospital length of stay and case-based factors on case mix index (CMI) and cost categories were explained. The variance of the case mix index was best explained by the length of hospital stay, showing an adjusted regression coefficient β=0.488 (p<0.001), followed by the number of coded procedures (β=0.35, p<0.001). Other predictors played a minor role. The regression coefficient was higher for conservative (non-operative) than for surgery (operative) departments. In the regression on costs, the length of hospital stay explained the variance of total costs by β=0.391 (p<0.001), among these personnel costs of the nursing staff correlated most highly (β=0.595, p<0.001). The analyses confirm the working hypothesis with length of stay impacting CMI and costs most significantly. Treatment effort and complexity level, mortality rate, patient age or gender are well discussed factors but their joint impact on the CMI is minor. Consequently, cockpits for operational length of stay management comprising key data and ratios should be introduced on the department level.