Forecast model tendency to move storms, cold fronts, or specific weather features towards the south on successive model runs. Commonly used to describe cold season storms. Can be used with terms such as "trending colder". The southern trend can happen to any individual model, especially those with a northern bias. A southern trend in model consensus is more common before 120HR. After 120HR southern trends in forecast model consensus is very rare. The saying "once you go north, you stay north" generally applies
southern trend in most models today following the EC's lead. At the surface most models are pretty close to the EC. The ensembles are leaning in that direction. At 850-700mb I see some things that need to be sorted out. Storm is coming on shore now. GFS CMC storm track is actually from Dubuque to Sheboygan, EC slightly south of that. 
Image on the left was the initial forecast from Saturday. Right is from this morning. The southern trend in the last 24 hours was unexpected and the trend continued right through the start of the event. Snow was clumped over the southwest and northeast and not evenly distributed across the rest of Wisconsin. Never make the same mistake twice. I learned a few things. First, broad brushing WI with 3-6" did not work. I could have lowered them in the north. The map in between these (not pictured) was actually pretty close. Second was to post snow maps too early that forces you to bias previous ideas. There will be changes before the next system.