The biggest issue I've seen on the means is a tendency to over spread anomalies as you get further out in time. I'm a big believer in using the plumes if you are looking for a point forecast vs the mean since you can find any clustering or see if a group is skewing the mean. When it goes to the continental or hemispheric scale its hard for a few outliers to skew the mean as a whole but you start to run into spreading of anomalies or washing things out. Remember back earlier this month the means were pretty toasty in that day 7-10 range but we averaged near average mid monthReally enjoy your podcasts Mitch. Quick question.
Like the precip ens means can these larger feature ens means be skewed by just a few members as well?