CastIron

I favor the idea of chip timing. If they can make it work for a marathon with 5k people on a mass start, then surely it can work for 75 cyclists.

Administering the thing is clearly another matter. I'd like to see USAC transition to it with video finish being the final word. For $100 per racer you make officials' lives easier (again, if done correctly) and get a wealth of comparative data. Some potential for new prime methods.