This has been puzzling me for a long time.

I have successfully developed what looks to me to be some very good performing trading systems using multiple indicators. However, when I backtest them in the Enhanced System Tester, even though the results of the system are impressive, Buy & Hold performance nearly always outperforms the trading system, and does so by very extensive margins. Yet the system looks great on the chart. It's difficult for me to believe that buy & hold actually outperformed the trading system when looking at the chart.

So, I'm wondering, what am I doing wrong?

There can be a number of reasons for this that it would be hard to hazard a guess without knowing the rules and what you were trading it on.

Without knowing the specifics, there are some general things you could look at:

-Commissions

-If you are buying on margin you could be losing more than it would seem just looking at the arrows

-Are you looking at Expert Advisor arrows or are you "Plotting on Chart" from the actual System Tester?

-On the summary of the system test, how much money were you losing on average on each trade?  Highest losing trade?

Have a look at a relatively new (2006) software called StrataSearch that does everthig needed to find multisystem strategies with/without optimization and including Montecarlo simulation and screening. All this is available only for USD 650 as you can se in http://www.stratasearch.com/home.html.