I think this argument is misleading because we're not guaranteed to find this "equilibrium" in a reasonable amount of time.
Think this of as a machine learning problem. The market is trying to get the global optimal point. I'm sure it exists, but the way we do this is pretty much by doing hill climbing, which may end up with a bad local maxima. And there are billions of parameters. You might sometimes want to try a different strategy even if it looks worse temporarily.
Think this of as a machine learning problem. The market is trying to get the global optimal point. I'm sure it exists, but the way we do this is pretty much by doing hill climbing, which may end up with a bad local maxima. And there are billions of parameters. You might sometimes want to try a different strategy even if it looks worse temporarily.