All of those other factors count, too, but speed's an easy one to measure and reduce with speed limits. Higher speed also increases the likelihood that an accident due to any of the factors you mentioned will be fatal. Obviously "
fatal accident could occur at 30mph," but the likelihood that a given accident will be fatal increases with the speed of the cars involved.
The effect of raising the speed limit from 55 to 65 has been heavily studied. There are a ton of links if you Google - here's one, from the Insurance Institute for Highway Safety:
How has abolishing the national speed limit affected fatalities? Institute studies show that deaths on rural interstates increased 25-30 percent when states began increasing speed limits from 55 to 65 mph in 1987. In 1989, about two-thirds of this increase — 19 percent, or 400 deaths — was attributed to increased speed, the rest to increased travel.
A 1999 Institute study of the effects of the 1995 repeal of the national maximum speed limit indicates this trend continues. Researchers compared the numbers of motor vehicle occupant deaths in 24 states that raised speed limits during late 1995 and 1996 with corresponding fatality counts in the 6 years before the speed limits were changed, as well as fatality counts from 7 states that did not change speed limits. The Institute estimated a 15 percent increase in fatalities on interstates and freeways.
A separate study was conducted by researchers at the Land Transport Safety Authority of New Zealand to evaluate effects of increasing speed limits from 65 mph to either 70 or 75 mph. Based on deaths in states that did not change their speed limits, states that increased speed limits to 75 mph experienced 38 percent more deaths per million vehicle miles traveled than expected — an estimated 780 more deaths. States that increased speed limits to 70 mph experienced a 35 percent increase, resulting in approximately 1,100 more deaths.
As for the other usual argument - that everyone speeds anyway:
Does the speed limit matter? Don't drivers speed anyway? Many drivers tend to drive somewhat faster than posted speed limits, no matter what the limits are. However, drivers do not completely ignore posted speed limits but choose speeds they perceive as unlikely to result in a ticket. The more important speed-related safety issue on freeways is the proportion of vehicles traveling at very high speeds, not the proportion violating the speed limit. The Institute's frequent monitoring of free-flowing travel speeds on interstate highways shows that, in general, higher speed limits lead to greater proportions of cars traveling at very high speeds.
For example, New Mexico raised its speed limits to 65 mph on rural interstates in 1987, and the proportion of motorists exceeding 70 mph grew from 5 percent shortly after speed limits were raised to 36 percent in 1993. In 1996, when speed limits were further increased to 75 mph, more than 29 percent of motorists exceeded 75 mph; by 2003, 55 percent of motorists exceeded 75 mph.10 In Maryland, which retained 55 mph limits on rural interstates until 1995, the proportion traveling faster than 70 mph remained virtually unchanged at 7 percent during 1988-93. By 1994, 12-15 percent of cars were exceeding 70. In neighboring Virginia, which switched to 65 mph limits on rural interstates in 1988, the percentage exceeding 70 mph went from 8 percent in 1988 to 29 percent by 1992 and 39 percent by 1994.
Sometimes I'd like to drive faster too. That doesn't make it good public policy. And arguing that there are other factors doesn't mean that we shouldn't take an easy step (or in this case, NOT take an easy step) and reduce something we can easily control.