Oh, there are certainly very big ones, but even the big ones sometimes don't have predicted impacts because at given times some things that USED to be drivers might not be just now. For example, a 2% sales tax increase, depending on the current state of people's buying, could have a minor impact or a major impact. If people are already buying only what they need, the impact to demand could be negligible. If they're splurging but wages are stagnant, maybe it has a huge impact. If they're splurging AND wages are increasing, maybe it has a negligible impact again. The basic point is, even if you understand the major drivers, without a bigger picture of the macroeconomic picture and what specific forces are driving behavior at the moment, your impact could be anything from dulled to the exact opposite of your intention.

Also, some of those factors, front running, insider trading, and market manipulation, which are evidence of a more predictable market, BECOME additional variables that impact decision making because they themselves impact other factors.

Weather forecasting might not be the best metaphor here... it's more like the human body. You might know that some protein causes some favorable condition that you want to boost, but increasing that protein production might ALSO increase production of an enzyme for breaking it down, reducing bio-availability of one of the building blocks, leading to a reduction of another protein that's critical for immune function. All of these pathways function together in ways that are extremely hard to predict, and it's natural that very often you'll be wrong.

But that's not to say I'm being defeatist... you build better models and you try things anyway - because that's what we do. I'm just saying economics is very, very, very hard, and there's not just a limit to our current ability to predict, there's a limit to how much certainty we CAN achieve.