Scott Adams was asked what it would take for him to believe that climate change is real.
He replied that if climate scientists are to persuade him they need to stop using climate models in a way that looks so much like it's all a scam (or hoax).
Adams is not a scientist, which he freely admits. He has no way of judging the science. But he can see that the climate model business resembles known scams. And he asks, if it's not a scam why does it look so much like a scam?
People are well aware of financial scams. One type is when you're offered an investment that has had a great return in the past, or, at least, that's what you are told. The scammers will tell you that they'll make you 20% a year on this investment. That is, they claim that they can predict future performance. But there is no economic model that can accurately predict future performance. Hence the warning: "Past performance does not guarantee future results." So it is, or should be, with climate models.
For example, there are a multitude of climate models. The scientists won't settle on one model. They tweek these models by adjusting them to match known data as they go along. Many of the models turn out to be wrong, and maybe one of them is right by chance, so the modelers claim their science if validated. But it's not science; it's a bunch of blind mice in a field of acorns. Some of them are going to find acorns. This doesn't mean the science is right.
Can't the scientists act in a way that would look more credible? Adams asks. For example, they could settle on one climate model, it's underlying assumptions and hypothesis would be well understood. Then they test it, and when they compare the prediction with the real climate data, they transparently conclude that the model is accurate or it is not. If they miss they don't try to shade the results, they don't try to change the data or go back and change the model
post hoc.
How Leonardo DiCaprio Can Persuade Me on Climate... | Scott Adams' Blog