FireAng85 -- those are just the spots where the model runs initialized Ivan, not other storms. It is just that not all of the models were started at the same time, and Ivan was at different spots when they were run....thus the different storm symbols.

Jason234 -- they do that to some degree in the official forecasts. And, it is not always hard and fast that the AVN/GFS will do well on timing but bad on location, the UKMET will do this, the NOGAPS will do that, and so on. To tell the truth though, taking all of that into account is what the FSU Superensemble and other multimodel ensemble forecasts (including model consensus tracks, to a lesser degree) try to do -- eliminate the known biases in model output tracks to come up with an improved forecast, both in terms of timing and location.

Note though how these multimodel ensembles are different than single-model ensembles, such as the GFS Ensemble forecast system. These single-model ensembles work by creating new model runs by tweaking the actual conditions in the atmosphere to see what the effects are on the weather pattern in the future. By combining all of these in a weighted fashion into one forecast, the hope is to eliminate a lot of biases that may arise in the main model forecast from bad data, missing data, or a bad handle on the current situation. The ultimate goal is the same, but the manner at which it is done - single model versus multiple model, actual weather/atmosphere dynamics versus pure statistics - is completely different.