It was centered more west for a while. Everybody got used to warmer winters--and said that it's westward shift was a sign of global warming. That was probably not a useful statement--sort of post-hoc reasoning there based on lack of information.
This is historically not uncommon. Texas citrus was periodically nixed by freezes--it's why citrus production moved to California and TX stopped being a citrus co-capital with Florida.
We tend to have short-term memories and tend to be very, very conservative when it comes to weather. It used to snow enough where I grew up that one year when my father could cut the grass on Xmas it shocked everybody. Usually we'd be out sledding a bunch of times every winter. Friends would routinely go skating on some ponds in the woods. Thirty years later after a rare snowstorm my mother could declare that it never used to snow there--it seldom froze more than a day or two--and everybody would nod in agreement, the implication of my battered sled hanging in the garage notwithstanding.
NOAA doesn't include AO in its projections ... yet. Mostly because it's hard to predict. The group that says it found a way of predicting it scored in the last few years with predictions far more accurate than NOAA's, but this year waffled--there were two important predictors that until this year were always in agreement but this year gave different predictions.
The AO's history was still being reconstructed based on proxies when I read up on it years ago. It wasn't recognized until a few decades ago and not considered a big deal until more recently--when it shifted east again. It's likely responsible for the cold spell we had in the winters in the 1940s into the '60s s and the warmer winter weather we had in the '70s through the '90s. Everybody wants to interpret it as part of the immediate global climate change threat, just as they want to do the same with the return of drier conditions to the American SW after a prolonged wet spell and as they wanted to do with ENSO and hurricane intensity. As soon as a regular, recurring pattern is self-servingly interpreted to be part a unidirectional change instead of something superimposed on the unidirectional change it bodes poorly for PR.
= new reply since forum marked as read