He said it an a weird way so I get where you’re coming from but his point is valid.
Models are tools, simple as that. They help meteorologists make forecasts. It is still incumbent on the forecasters to account for model biases, weigh the chances based on time from event, and decide the best way to alert the public.
as other have stated, there was ONE suite where all models showed a storm at the same time. One. Sure each model showed the storm at one point, but they were clearly picking up on a wrong piece of energy or variable that impacted the output. For us to know since Saturday evening or Sunday morning that there wasn’t going to be a storm on Wednesday or Thursday for us, 4-5 days out, is pretty solid.