Exit polls have had a hard time predicting election results in Sweden over the last decade. The introduction of the Sweden Democrats in the political scene has generally caused problems for polling organizations. Opinion polls have a big influence on the economic market and therefore any wrongdoing might cause a big disturbance in the economy. This paper develops a model based on systematic sampling differences and sociodemographic background variables in order to predict the results of the Swedish general election 2018. The study uses data from SVT´s exit poll and SCB´s voter turnout survey. The results of the study show that the model in general predicts the 2018 Swedish general election better than the prognosis of SVT.