Rather than go point by point, I'll refer you to someone more eloquent and credentialed. Check out some of Mark Buchanan's comments on http://physicsoffinance.b

Rather than go point by point, I'll refer you to someone more eloquent and credentialed. Check out some of Mark Buchanan's comments on http://physicsoffinance.blogspot.com http://physicsoffinance.blogspot.com He talks about many things, but if you're curious you'll find several lit review posts about this topic. The heart of good science is solid theory followed by experimental evidence. Simulations such as the El Farol Bar problem demonstrate how systems of prediction agents go through a phase shift from stable to chaotic as the ratio of memory to number of participants decreases. Historical phenomena match the prediction -- as more low-latency algorithms participate, the market exhibits more dramatic movements.