I realized conventional wisdom was underestimating Trump way back in August. But for a long time, I was dismissive of Bernie Sanders. If you’d asked me, say, in December, why I was ignoring Sanders while being bullish on Trump, I would have said two things: first, the Democratic base just isn’t as angry at the Democratic establishment as the GOP base is at the GOP establishment. Second, I’d have told you, don’t compare Trump and Sanders in isolation: on the GOP side, insurgent candidates were doing well, while establishment candidates were doing poorly. But on the Democratic side, Clinton started averaging around 55% in polling once Biden made clear he wasn’t running.
However, in mid-January, I changed my mind. What did it for me was seeing Sanders dominating polls in New Hampshire, and neck-and-neck with Clinton in Iowa. Initially, I had bought into the story that Sanders just wasn’t going to be able to make inroads with non-white voters. But what if that wasn’t why Sanders was doing so well in early states, even as he trailed Clinton nationally? What if the reason for the discrepancy was just that that was where Sanders had been campaigning, and many voters in other states weren’t really paying attention yet?
Way more important than any blather you could get from me is this chart:
This is a static version, because a month from now I want to be able to link to this post and have people see what the chart looked like at the time that I wrote this. You can see an up-to-date version here. It’s a little hard to tell from this version, but at the beginning of the year Clinton was ahead of Sanders by 20% in the polls. Now her lead is down to 7.5%. Given another month, Sanders has a good chance of passing Clinton in national polls–at the very least, I expect that one month from now, the national race will look extremely tight.
After Sanders’s loss in Nevada, many pundits are already pronouncing his campaign dead. If there ends up being any truth to this, it will only be as a matter of self-fulfilling prophecy: the media uses Nevada as an excuse to go back to ignoring Sanders, or give his campaign much more negative coverage. Just two weeks ago, many were saying Nevada would show Sanders couldn’t compete among non-whites. Yet exit polls showed Sanders winning the Latino vote, casting serious doubt on that narrative.
Some pundits are arguing that the actual results of the Nevada caucuses suggests the exit polls are wrong about Sanders winning the Latino vote. However, the same exit polls that showed Sanders winning Latinos showed Clinton winning African-Americans by a truly enormous margin, and Clark County (the area around Las Vegas, where Clinton’s support was strongest) has both a high concentration of Latinos and a high concentration of African-Americans, making precinct results difficult to interpret in demographic terms (compare this and this).
The enormous support that Clinton enjoys among African-Americans is a real problem with Sanders. Sanders will likely have to do more to win over African-Americans to win the nomination. But the Nevada results are evidence of his success winning over Latinos. Whether Sanders can win over African-Americans remains, I think, an open question.
(An aside: though Sanders’s record on immigration is deeply flawed, I was impressed when, during the debate on February 5th, Sanders spoke out against the Obama administration’s deportation of child refugees. In the same debate, Clinton defended the policy as “sending a message”, the same justification given by Obama himself.)
Attaching great importance to the results of early states makes a lot less sense for the current Democratic race than it does for the GOP race. The battle royale for the GOP nomination means strategic voting plays a huge role: any candidate who has a disappointing showing in an early state risks having his supporters switch to their second choice. That’s not an issue in the two-candidate Democratic race.
Second, the GOP primaries are governed by arcane delegate rules that, for example, allowed Trump to get all 50 South Carolina delegates even though he got less than a third of the vote. These rules differ from state to state, but their ultimate effect is to create the possibility that a string of early victories could give a candidate an insurmountable lead in convention delegates. The Democratic primary, on the other hand, allocates delegates proportionately in every state. That means there’s a real possibility the Democratic primary might not be settled until the last set of states vote in June.
(Or even until the convention itself–but my guess is that if Sanders gets a majority of pledged delegates, a majority of the superdelegates will switch to supporting him.)
Sanders is still the underdog here. He will almost certainly lose South Carolina, and he’s unlikely to win more than a few states on Super Tuesday (March 1st). Not because of his national polling, but because Clinton remains much stronger in southeastern states (a majority of Super Tuesday states) than she is nationally. But because the Democratic Party allocates delegates proportionally, Clinton’s margin of victory will matter a lot. So if you live in any state that isn’t Iowa, New Hampshire, or Nevada, I’d urge to take this primary very seriously–it’s nowhere near being over yet.