I'm also less inclined to believe the RescueTime productivity number when they take their own solid figures on user involvement and then mix them in with a few SWAGs to come up with what people now believe is a solid figure. Once you throw BS into the recipe you've spoiled the final product.
From the DOE report--- Changes in national traffic volume and motor gasoline consumption for passenger vehicles in 2007 were determined to be statistically insignificant and therefore, could not be attributed to Extended Daylight Saving Time.
http://www1.eere.energy.gov/ba/pba/pdfs/epact_sec_110_edst_r...
In the spring the sunrise is getting earlier as time goes by. Then the clock jumps forward an hour, which makes sunrise an hour later, causing the repeat the op talked about.
In the fall, sunrise is getting later. The clock jumps backward an hour, which makes sunrise an hour earlier. This causes the same repeat, in reverse (aka, trending darker rather than lighter).
edit: Also, I have the experience to back this up. I've been one of the living west of my workplace people for too long.
Twice a year, once in the spring and once in the fall, commuters experience a period during which the sun is right in their eyes. For those of us driving east in the morning, and west in the evening, the sun is right in front of us all the time, and when it's at the right (wrong) elevation, vision is painfully obscured.
Earlier in the year the sun is too low to bother; later, the sun has risen high enough that your visor can block it. But for that time period, traffic is terrible because drivers can't see where they're going.
The thing about DST is that it resets this period. After having made it through once, DST drags the clock around so that we must endure the same sun glare again.
Once you start to think about this effect, it's clear that it must be costing fuel and lost productivity from the traffic, and likely even injuries and lives from accidents.