Elroch

The random (or caused by unknown factors) component is larger than the known component. Hence, very imperfect information.

The book is an excellent lesson in money (mis)management

Interesting observation! This apparent discrepancy has never occurred to me before. The explanation is that, like humans, computer trading is largely reactive. i.e. some condition occurs in the markets (be it a break of a level, a spike causing an arbitrage opportunity or whatever). Almost all conditions occur more when the market is more lively, so computers are less active at lunchtime in response to the lower activity from human traders. While this is a plausible explanation of the observation of the original poster, it does show that even if humans are not responsible for the majority of the volume of trades executed in many markets, their actions are still the driver of what happens. This makes sense, as if computer programs are designed solely to take advantage of the actions of other computers, the resulting arms race could provide no net gain to the sum of all computer programs. More sensible to try to do things automated systems are ideal for, such as identify discrepancies in prices very quickly and take small profits from them - a great deal of program trading is arbitrage.

"Scalping" used to mean making money from the spread, as one poster said. I have observed the term get (mis)used in an increasingly different way, with people claiming to scalp off hour charts or even overnight. In response to the first poster, what he describes sounds great, except you need to be right a lot, and it is a lot easier to be wrong and lose a pip or two on your average trade. It is true that the trading costs are important. You need a sizeable edge just to break even if you trade for a few pips. But some can achieve results approaching perfection going for a few pips when everything is perfect, and make millions (proof by example). Good advice would be not to assume you can do this without plenty of hard evidence.