Lameness is a very common problem in breeding sows, which often goes undetected for long periods of time. This has severe consequences on the welfare and performance of sows. Automatic lameness detection could help pig farmers to recognize and treat the problem sooner. The Sow Stance Information System (SowSIS) consists of 4 force plates, built into an electronic sow feeder, providing non-invasive ‘stance’-output for each leg. Stance information variables can be extracted from the data. Stance data was automatically collected for 53 gestating group-housed sows for 74 days per sow. The gait of these sows was visually scored twice a week using a 150mm tagged visual analogue scale to determine their lameness status. This gait score was used as a reference and compared to the SowSIS measurements. First, significant variables were used in a multilevel linear model to predict lameness. With the model estimates, it was determined whether the model would classify a sow as lame or non-lame using gait score ≥55mm as the cut-off value. The model could detect lameness with 71.7% sensitivity, 89.7% specificity, 76.7% lame predictive value and 87.0% non-lame predictive value. The 5 severely lame sows (gait score >90mm) were all correctly classified as lame. Secondly, different linear models were tested to determine the lame leg on a sub-dataset. The support vector machine model and random forest model could predict the lame leg correctly by 100% when fitted to a validation dataset. The SowSIS shows great promise as an on-farm lameness detection system.