Oh ok, that's what you were doing.

Sorry, I got confused when you were accusing me of agreeing with the author.

And now you've clarified that a single study has shown a correlation:

A 2009 study showed a 14.5% decrease in bicyclist injuries after the passage of the original Idaho Stop law (though did not otherwise tie the decrease to the law).

I actually downloaded the PDF to read what was going:

Repeated stopping increases risk of injury from repetitive stress. The act of a hard stop, and the 28 subsequent resuming of speed from a hard stop, involves considerable strain on joints, >particularly the 29 wrists and knees but also the shoulders, elbows, neck and low back. (cite) The wrist suffers >strain under 30 vibration, flexion and torsion during hard stops.

To get that 14.5% decrease, they're counting the "physical strain" of having to start from a stop. But if they're supposed to be coming to a full stop anyways, they have the same amount of starts...

Still doesn't make any logical sense to me how it's safer, but it lead me time more info in the end, so thanks!