Iowa and New Hampshire are racially unrepresentative of America at large. Few of their residents vote in their primaries. These are just a few of the reasons why a primary system giving these states primary influence over our elections primarily sucks. (Despite the loveliness of Daily Kos users from Iowa and/or New Hampshire, of course.)
What we need is an unbiased primary system that puts all of the states in an order that will give us the best candidates for the general election. And I have The Perfect Solution. It's an algorithm. (Booooo!) But I promise, there is no math after the jump. (Yayyyy!)
Here's what you do. First, you look at the national general presidential election results. What was the point spread? Then, you look at the statewide general presidential election results. What was the point spread? The state that's closest to the national results wins!
Looking at the 2004 results, here's what the order of the 2008 primaries would be:
1. Nevada
2. Ohio
3. New Mexico
4. Iowa
5. Florida
6. Wisconsin
7. New Hampshire
8. Missouri
9. Pennsylvania
10. Virginia
Look at that, Iowa and New Hampshire, you're still in the top ten. I know that won't keep you from protesting loudly, but it will help all the rest of America to think you're being unreasonable.
After the first handful of states, clusters of states could vote in Super Tuesday-like groups.
The main defense of the IA/NH lock on early primaries is that there are fewer voters in those states, which helps non-establishment candidates be more competitive via small-scale retail politics and inexpensive TV ad rates. But the Internet has facilitated a new way of helping non-establishment candidates be more effective: getting huge numbers of individuals to play active roles in campaigns. And doesn't that sound like more fun than this "fewer voters" nonsense?