For a forecasting model to be as accurate as possible, it needs to be based on accurate historical data as this would influence the model in such as way as to produce more accurate results.
Even when sufficient quantitative data is available, qualitative date is also needed because not everything can be explained by numbers.
A good forecast model brings out results such that half of them are above the actual results and half are below. That way the average will be more aligned with the actual results.
