Agree (not off-topic ). I would say there are two schools of validating approach: Back-testing (including Forward-testing) which is not real-time, and Realtime-papertrading which is of course real-time. Agree again, sometimes a trust in some back-testing results could cost someone something.
The more I think about it, I think I am getting your point Funky. Almost every indicator is parsing price into some form of crossover, overbought/oversold, slope, and/or divergence information. I think what you're saying is that there is an art to interpreting this information, and every indicator period you use is giving you the same information, just with a different level of sensitivity. So, the art would be feeling out the indicator periods that best fit your desired trade timeframe and style, most of the time. Like in your analogy, a fiat and a porche both drive, just at different speeds, and with different levels of sensitivity to the road. Either will get you there as long as you know which one you're driving.
In a trend use trend stuff. i.e. MA's. In a choppy market use divergence or OB/OS stuff, i.e. RSI, Stohs. That simple. Figuring out the trading environment is the hard part.