Not sure it's all garbage. Case number is rapidly rising, near exponential. If we start approaching 2000 deaths/day, it's off to the races I think.
We've never seen a weather model waver either.
I mean these disease models get new inputs all the time (cases, state-wide COVID precautions, mortality rates, etc). So it's not surprising that numbers and the prediction horizons change. I assume they are referencing the IMHE model, which is updating every 2-4 weeks, which is why the horizon keeps moving about a month each time.