Pardon me if I have a healthy skepticism about prognosticators who claim to foretell weather events months in advance of everyone else.
First off, understand that because the government gives away pretty much all of their forecasts and products for free, for any purpose, it makes it extremely difficult for most honest meteorologists to make a decent living (something I wish I would have realized before I graduated college and the recession hit and made it even worse). If you're not working for the feds, there are two ways to do it: via broadcasting, or in the private consulting industry. In both cases, but particularly in the consulting industry since people directly pay for your services, you have to be able to provide something the feds don't―or, at the very least, give the ILLUSION that you can provide something.
There are a number of forecasters out there who have decided to use the tactics used by the nostrum salesmen of the 19th century. It's a gambit as old as the farmer's almanacs, but computer modeling of our atmosphere has been improving rapidly over just the past decade alone, allowing us to have a general idea of what's coming our way up to two weeks in advance, and that number is still growing. These forecasters take advantage of that to give credence to the notion that somehow we can forecast months in advance with pinpoint accuracy, something that will always be balderdash for a number of reasons. These types typically take a relatively simple pattern-recognition scheme, selectively highlight the instances where it did predict something (this works especially well if they stumble upon a prominent catastrophe they can fit to their prediction) or worse, doctor their past predictions to hide their error, then use that seemingly sterling record to sell subscriptions.
Our atmosphere is too chaotic to accurately predict storm systems months in advance other than perhaps through sheer luck.
