Six Predictions for 2026 the Models Won’t Like

Six Predictions for 2026 the Models Won’t Like

Models are useful tools. They simplify complex systems, quantify trade-offs, and provide decision-makers with concrete arguments to support their decisions. But models also share a weakness: they assume cooperation from reality. In energy, electrification, and climate...
No Penalties for Attribution

No Penalties for Attribution

This week, we finish the series on attribution studies. First, let me explain, while I beat up attribution assessments, they are necessary. This post will conclude with how I think they should be used. Experimental v Quasi-Experimental Second, I want to make a couple...