A support vector machine classifier was trained to detect stances on vaccination in online forums, using posts from Mumsnet annotated as 'against,' 'for,' or 'undecided.' The classifier achieved a macro F-score of 0.44 for multi-class stance detection and 0.62 for distinguishing 'against' versus 'for' vaccination. These results indicate that stance detection is challenging with the current model and dataset, suggesting future improvements through expanding the training data and exploring different classifiers and features.