To hear Bernie Sanders tell it, the delegate deficit he faces is in part a function of the order in which states voted in the Democratic primaries. He's alluded to this before, but made the point plainly on "The Nightly Show With Larry Wilmore" on Wednesday.
"People say, 'Why does Iowa go first, why does New Hampshire go first,'" he said, "but I think that having so many Southern states go first kind of distorts reality as well." After all, by Mar. 1, when Georgia and Alabama and Texas and South Carolina and Tennessee had all voted, Hillary Clinton had already run up a pledged-delegate lead that Sanders hasn't been able to touch. And that, he implies, is why people think he's losing.
We've addressed this before, noting, among other things, that the fact that Southern states vote earlier in the process was not a function of skewing the vote for Clinton (or against Jesse Jackson in 1988). It's also worth noting that Clinton's doing well in Southern states didn't keep Sanders from doing well in other red states like Oklahoma, Nebraska, Wyoming and Utah -- so if there was some deleterious effect on Sanders's candidacy, it's not easy to spot.
But let's set all of that aside, and instead focus on the what-if. What if the states were all re-arranged, and there were the same number of contests on each day of the Democratic calendar but the states that voted those days were mixed up? If the results in each state were the same (which there's not really any good reason to think they wouldn't be, Sanders's complaints aside), how often would Sanders have come out of March 1 with the lead?
I ran this experiment a million times in a row, over and over again. (Literally. Computers!) And the results were always about the same. About 10 percent of the time, Sanders would lead after March 1, with states he won big earlier in the calendar. But the other 90 percent of the time, it was Clinton who led after March 1 anyway (although about one out of 10 of those times, she then lost the lead again temporarily).
(We used Daniel Nichanian's delegate estimates for each state, because 1) he's done a good job tracking the numbers and 2) he allocates Washington's delegates to Sanders proportionally. For the trend over time, we used the same number of contests per day as have been held this year.)
Why? Because Clinton has won more states, and more of her states have yielded more net delegates because they're bigger. If the record sounds like it's skipping, it isn't. We're just saying this over and over because it somehow isn't sinking in.
As the kids on "Reading Rainbow" used to say: But don't take my word for it. We built a little tool that lets you simulate how a random order of states would affect the primary. Negative numbers are a delegate lead for Clinton; positive, for Sanders. Each time you hit the "random" button, the tool will tally how often Clinton's been leading after Mar. 1. Click the button a million times, and it should end up somewhere around 81 percent.
So why is Sanders suggesting that the order mattered, when this suggests that it really didn't? Because Sanders's consistent argument is that he's doing better than the numbers suggest -- an argument he makes so that he can convince superdelegates that they should offer him their support. The superdelegates can do what they want. But in 90 percent of all of the universes where Iowa and New Hampshire were replaced with Wyoming and Arkansas (or whatever), Clinton is still the front-runner on Mar. 2, 2016.