Indiana might now be Trump’s last hope

The Briefing, Vol. IV, Issue 15-

This week:

  • Trump’s actual worst week
  • His path to nomination isn’t closed yet
  • Could all come down to Indiana

Trump’s worst week: Last week, Donald Trump‘s campaign staff had consoled themselves with the fact that their “worst week yet” — the week before — maybe wasn’t so bad. In fact, compared to last week, it wasn’t.

The bad week began weekend before last, when Trump was trounced in a pair of Colorado district conventions. Then on Tuesday, in Wisconsin, Trump lost to Ted Cruz by 13 points amid ultra-high turnout in the Badger State. His unfavorable ratings hit new highs in the polls. He then got swept on Thursday, Friday and Saturday in Colorado’s remaining district and statewide conventions, letting slip another 34 delegates (really 37 if you include the three party officials, who will not be for Trump).

And finally, to cap it all off, Trump began bleeding his own actual pledged delegates. In North and South Carolina, several Cruz backers were installed to vote for Trump on the first ballot. It’s the same story as in Indiana, where state party leaders are expected to install an anti-Trump slate this Saturday, ahead of the May 3 primary. And this has already happened in Tennessee, Louisiana, and Georgia.

To be clear, Trump is not actually losing pledged delegates when this happens. He will get every vote he has earned in elections and caucuses in the first, second and (in the case of Florida) third ballot at the convention, as the rules require. Every delegate pledged to Trump must either vote for him or resign — the rules provide no leeway on this point.

But if he hasn’t got a majority nailed down on that first ballot, it’s increasingly clear that Trump will not be nominated. So the critical question is whether he can get to 1,237 delegates or not.

The winnowingSince March 15, the day Marco Rubio dropped out of the presidential race, the following number of delegates have been won by each candidate:

Cruz 114
Trump 65
Uncommitted 35
Kasich 0

At most, it appears that one of those 35 uncommitted is a Trump supporter. So the more important total is

Trump 66
Not Trump 148

Trump’s victory in Arizona did not depend completely on the fact that Marco Rubio was still on the ballot and got 72,000 votes. That wouldn’t have been enough for Cruz to overtake him. Still, it’s worth pointing out that with the outcomes in Utah, North Dakota, Colorado and Wisconsin might have all been more complicated had Rubio still been in the race. One can only wonder how much better the prospects for stopping Trump would look if Rubio had been out in time for the contests in Missouri and Illinois.

This is one sign that Cruz is successfully consolidating most of the Rubio vote — something he must do to stop Trump. In the three actual primary contests with voters that have taken place since March 15 (Arizona, Utah, Wisconsin), Cruz has received 43.4 percent overall, to Trump’s 36.7 percent. Trump has a few good contests to look forward to now — New York especially — but in much of America, the shine is off.

Delegate Math: At this point, the math is still there to make Donald Trump the Republican nominee. But his path is now very narrow, and any significant deviation will derail him for good.

There is more than one delegate estimate out there. Let’s look at the one most favorable to Trump — the Green Papers estimate of 758 delegates. (Note that if another count is more accurate, he might have 15 delegates fewer than that). We’ll also assume, arguendo, that Trump’s one “leaning” unpledged North Dakota delegate is a solid Trump vote — so 759 it is.

For purposes of this illustration, let’s ignore the chronological calendar for a moment, and think of the contests instead based on how the state parties award their delegates. We will divide all of the remaining contests into three groups. The first group is of states that award their delegates in a winner-take-all fashion — these have 159 delegates in all:

Nebraska (36)
Delaware (16)
South Dakota (29)
Montana (27)
New Jersey (51)

The second group is of states that award delegates proportionally (or more or less proportionally). These have 115 delegates to award in all:

Rhode Island (19)
Oregon (28)
Washington (44)
New Mexico (24)

The third group is the most complicated, because they award delegates based on both statewide and Congressional District votes. These have 495 delegates to award in all:

New York (95)
Pennsylvania (71)
West Virginia (34)
Maryland (38)
Connecticut (28)
Indiana (57)
California (172)

Group One is the easiest to allocate. Trump should win New Jersey and Delaware, for 67 delegates. Cruz should win all the others, for 92 delegates. That brings Trump’s delegate count to 826. That means Trump needs to win 411 of the remaining 610 delegates, or just under two-thirds.

Now let’s make a reasonable assumption about Group Two. Trump has received about 37 percent of the primary vote so far, so let’s just assume he does exactly that well in these states, on aggregate. We will assume that he wins 37 percent of their delegates. (This is actually a generous assumption, for two reasons: Trump’s lousy performance so far in Western states, and the fact that some of these states will likely award district delegates to each of the three candidates.) This would mean Trump gets 43 delegates out of these states, and not-Trump gets the other 72. This brings Trump’s delegate count to 869.

Group Three is the hardest to guess. The one thing we can say with certainty is that, given our (fairly reasonable) assumptions so far, Trump would need to win 368 out of the 495 delegates, or just under 75 percent.

This is no small task in an election where Trump has been getting amost exactly 37 percent or the vote. It means he cannot afford a bad day in any of the Group Three states. If we look at each one separately, here is where Trump’s target needs to be in each state — and of course, whatever he fails to get in one state he can always make up in another:

New York: 72 delegates out of 95
Pennsylvania: 54 out of 71
West Virginia: 26 out of 34
Maryland: 29 out of 38.
Connecticut: 21 out of 28.
Indiana: 43 out of 57
California: 129 out of 172

Given our other assumptions here, Trump would be mathematically eliminated from a delegate majority if he were to lose 128 of the delegates in this group. The places where delegate leakage is most likely to occur for him are California (where there’s a close race and Cruz leads in many congressional districts), Indiana (which Trump might lose outright), and Pennsylvania, where 54 of the 71 delegates are chosen by name at the bottom of the ballot. Because the Pennsylvania delegate candidates’ preferences are not actually printed on the ballot (as they were in Illinois), a well-organized campaign is likely to do much better than a disorganized one. Trump’s delegate disasters over the weekend make clear which kind he has.

There’s one more thing: Although there is no polling yet ahead of Indiana’s May 3 contest, leaders in the state seem to believe that Cruz has a decent shot at winning in the Hoosier State. It stands to reason — its economy and demographics (among Republicans, anyway) is not too dissimilar from Wisconsin. Even so, Trump is virtually assured 6 delegates out of Indiana (the reward for victories in two Congressional Districts), and to be safe we should assume he will get 9 if he loses there.

So in our Cruz-Indiana victory scenario, the bar gets higher for Trump in all of the other states. Suddenly he needs to win 359 out of 438 delegates, or 82 percent. That implies targets of:

New York: 78 delegates out of 95
Pennsylvania: 59 out of 71
West Virginia: 28 out of 34
Maryland: 32 out of 38.
Connecticut: 23 out of 28.
California: 142 out of 172

This is probably a bridge too far. Which means that Trump really needs to win Indiana. The Hoosier State, once described on its license plates as the “Crossroads of America,” could well determine whether the GOP goes the route of Wisconsin or the route of Illinois and Michigan.

To explain this more simply, I put together a video, which originally appeared here. You’ll notice it makes slightly different assumptions (a bit more favorable to Trump), but the general idea is about the same: