1. As your trade frictions increase which is typical with trading larger amounts of money or OPM/etc, the advantage tends to swing toward quantitative or systematic trading. The reason that this is the case is that it placed a greater demand on the discretionary trader. If your costs are high then you need to be able to evaluate the costs in relation to your edge, and that's going to be difficult to do if you are trading on intuition. However, by the same token, if you are able to reduce your costs/fees then you may be able to take even greater advantage of the strengths of discretionary trading.
2. One of the major problems with traditional backtesting environments is that it is very difficult to consider factors other then price when backtesting. I'm not saying it is impossible but it is certainly not trivial or easy. For example, Tradestation doesn't really support testing of order flow concepts. Tape reading is a big part of my trading. If I develop a system in TS then I have to spend a ton of time trying to figure out "correlates" to order flow which are just rough approximates. I like to think markets are cognitive. Well, let's think if someone knows the factors that influence their markets then they can know better how they are likely to trade. And not all of these factors are price based.
3. I feel like the answer regardless is some importance on super structures which can help control bet size and risk.