My guess is that if the model accurately represents all the realities being modeled, and if they ran the simulation enough times, they would get Otis in the model. It's very easy to get chaotic behavior out of such a model, or even simple circuitry.
At that point the question is not 'can the model do a run that produces Otis', because at that point it can and does, one run in ten thousand.
The question is, do the forecasters refuse to mention that one run of the model that produced that result. What seemed to happen this time was, not enough data or attention, and nobody ran their model enough times to produce the Otis result in the model. I don't think anyone saw it and discarded it, I think it didn't show up in anybody's runs of their models.
Odds are good that it COULD have, if the models are any good. It's just unlikely.
My guess is that if the model accurately represents all the realities being modeled, and if they ran the simulation enough times, they would get Otis in the model. It's very easy to get chaotic behavior out of such a model, or even simple circuitry.
At that point the question is not 'can the model do a run that produces Otis', because at that point it can and does, one run in ten thousand.
The question is, do the forecasters refuse to mention that one run of the model that produced that result. What seemed to happen this time was, not enough data or attention, and nobody ran their model enough times to produce the Otis result in the model. I don't think anyone saw it and discarded it, I think it didn't show up in anybody's runs of their models.
Odds are good that it COULD have, if the models are any good. It's just unlikely.