You could very well be right, but it’s just impossible to control for such variables, so we can’t say one way or the other with a degree of certainty. Especially considering that a 20% swing in numbers consists of a mere 1 or 2 people per 100,000 difference.
While I can’t necessarily disagree with your correlation≠causation argument…
Better gas mileage probably contributed a more than zero amount to that decrease.
You could very well be right, but it’s just impossible to control for such variables, so we can’t say one way or the other with a degree of certainty. Especially considering that a 20% swing in numbers consists of a mere 1 or 2 people per 100,000 difference.