To answer Mike N's question at the end of the old thread: models usually do a pretty good job in forecasting troughs out to 4 and 5 days, at least in general, but the specifics -- how strong, how fast, how far -- they sometimes have trouble with. Some models do better than others with different systems & regimes, while others do better solely in the tropics. The forecasting aspects must be incorporated into any track forecast, starting from the outset: if the model doesn't get the initial conditions of the overall regime right, it's likely not going to get the rest of it right. We saw this last year with the subtropical ridge.