Right now, tornado (and severe thunderstorm) warnings are issued when one of two things happens:
1) A tornado or a funnel cloud is spotted
2) Doppler radar indicates strong rotation and the forecaster believes a tornado is possible or likely
It's always better to detect the tornado vortex signature on radar before a tornado actually develops. There is some lead time, should a tornado actually develop, and people have time to take cover. But the bottom line is that we still warn based on observations, not on a forecast.
A fundamental reason for conducting VORTEX2 is to investigate tornado formation to understand why some supercell thunderstorms are tornadic while others are not. One reason that has been suggested is that warmer rear flank downdrafts are more favorable than colder rear flank downdrafts for tornadogenesis. Physical reasons for this have been proposed, but at this time it still remains untested.
It is hoped that by investigating the storm environment, there will be a better understanding of why some supercell thunderstorms are tornadic while others are not. If the properties of the rear flank downdraft influence the tornado potential, those observations of wind speed, temperature, moisture, and pressure are important. The purpose of understanding why some supercells are tornadic while others are not isn't just for meteorologists to publish papers in journals and get NSF funding. The purpose is to better understand tornado formation in hopes that in the future, tornado warnings are based on forecast tornado potential instead of observing the rotation. If forecasters can predict when tornadic rotation will develop in storms, it will increase the lead time on warnings. It will give people valuable time to alert others of approaching dangerous weather and to take cover.