The betting markets currently believe Trump is the overwhelming favorite to become the GOP nominee. This is despite Trump winning less than 40% of the popular vote, less than 50% of pledged delegates, and actually being below his delegate targets according to Aaron Bycoffe and David Wasserman. Are the markets overrating Trump?

At this point, there are only three plausible outcomes of the Republican race:

1. Trump earns a majority of pledged delegates (1237+) and becomes the nominee.

2. Trump falls short of a majority, but convinces enough unpledged delegates to get on the Trump Train, allowing him to emerge victorious in Cleveland.

3. Trump falls short of a majority, can’t get enough unpledged delegates to cross over, and someone else emerges.

Here’s why the betting markets think the first two outcomes are much more likely than Trump losing.

MARCH

Trump currently needs 542 more pledges delegates to reach a majority (1,237) and become the presumptive nominee. Even though he’s won less than half of delegates now, the terrain ahead is favorable for his chances.

This Tuesday, Trump is likely to add 58 delegates to his count. PredictWise thinks Trump has an 87% at winning Arizona, which is a winner-take-all state with 58 delegates. Such a victory should weaken the sting of his near-certain blowout loss in Utah, which the markets think Cruz is the 94% favorite. Recent polling also suggests Cruz will clear 50% of the vote in Utah, giving him all 40 delegates.

Taken together, Arizona and Utah will give Trump 58 new delegates, expand his lead over Cruz by 18 delegates, and leave him 484 delegates short of a majority.

APRIL

The first half of April is the barren part of the election season. There are only three contests, and two of them don’t even have a public preference vote (North Dakota and Colorado). That leaves Wisconsin as the contest of note.

Wisconsin is hard to predict. There are no betting markets, no recent polls, and even its geography and demographics don’t give many hints. It borders states won by Cruz, Rubio, and Trump. It’s a midwestern, very white state, which sound good for Cruz, but it also has a low share of very conservative, frequent church-attending evangelicals–which Cruz needs. The polls we do have point to a Trump lead–but it’s a small one, and maybe now nonexistent.

Wisconsin is also a “winner-take-most” primary, so the winner of the state gets a disproportionately high amount of the state’s delegates. As we saw in Missouri, a tiny shift in the the vote outcome has huge ramifications for the allocation of delegates. For now, let’s follow the evidence we have and say Trump narrowly edges out Cruz in Wisconsin, and takes home most of its 42 delegates.

The final two weeks of April will likely be Trump’s best stretch of the election. It will have to be for him to reach 1237. On April 19th is Trump’s home state of New York and its 95 delegates, most of which are allocated at the congressional district level–three delegates at stake per CD. The winner of each CD gets 2 delegates, unless they reach 50% of the vote, which gives them all 3 delegates. Recent polling, and what we know about Trump’s supporters, suggest New York may be the first state he wins with over 50% of the vote. If he does so, he’s likely to capture nearly all 95 delegates.

One week after New York comes Connecticut, Delaware, Maryland, Pennsylvania, and Rhode Island. These five states combine for 172 delegates, although 54 of Pennsylvania’s will remain unbound. That means 118 delegates are at stake, and are allocated on a mix of winner-take-all, winner-take-most, and proportional rules. The little polling we do have, along with geographic and demographic indicators, suggest this is favorable Trump territory and he’ll take home most of those 118 delegates.

Assuming he wins Wisconsin, New York, Connecticut, Delaware, Maryland, Pennsylvania, and Rhode Island, Trump will gain between 220 and 240 delegates, on top of the 58 he’d get from Arizona. For the sake of argument, let’s assume he wins 230, which would leave him 254 delegates short of 1237.

MAY
May brings us five elections:
Indiana on May 3
Nebraska on May 10
West Virginia on May 10
Oregon on May 17
Washington on May 24

There is virtually no polling on these five states, but we do have this detailed national polling map, from Nate Cohn. From this, along with what we know about Trump’s base, Trump is a big favorite in West Virginia, a favorite in Indiana, and competitive in Nebraska, Oregon, and Washington. If you assume he wins WV, IN, narrowly wins or loses in OR and WA, and loses NE, Trump should end up with around 100 delegates, bringing him within 154 delegates of 1237.

JUNE

The last day of the primaries is June 7th, when Montana, New Jersey, New Mexico, South Dakota, and the biggest prize of them all–California–vote. At stake are 303 delegates–of which Trump will need to win around half of them.

New Jersey is easy to predict for Trump. Numerous polls and indicators suggest that NJ is one of Trump’s best states. New Jersey allocates its 51 delegates on a WTA basis, so if you subtract that from the 154 he needed, he’s now only 103 delegates short of a majority.

New Mexico only has 24 delegates to give out, and it does so on a strict proportional basis, so it doesn’t really matter if Trump wins or not. The little polling we do have and what he know about the demographics point to Trump winning at least a 1/3 of the vote–so let’s assume he wins 8 of the 24 delegate. Trump is now only 95 delegates short.

South Dakota and Montana have a combined 56 delegates, all of which are awarded to the winner. While Trump seems to fare better in these two states than the rest of Big Sky and the midwest, let’s assume Cruz hangs on and takes all 56 delegates. Trump still needs 95 more, but he has 169 to play with in California.

California only awards 10 delegates to the statewide winner. 3 delegates are given to the winner of each of the 53 congressional districts–159 delegates total. Polling suggests Trump has a state-wide lead over Cruz, but Cruz’s support typically tends to be concentrated in pockets of deeply conservatives areas, whereas Trump enjoys broader and more evenly-distributed appeal. This positions Trump to win most of CA’s districts, even if he’s running even with Cruz at the state level. If he does so, Trump will get most of those 159 delegates, pushing him over 1237 and becoming the nominee.

Trump could easily stumble and fall short. If he loses Wisconsin and Indiana, he’d have sweep nearly all of the delegates in the Northeast and California, or make up the difference with surprise wins in Nebraska, South Dakota, or Montana. So while Trump has an obvious path to a majority, he also is walking on an obviously narrow path.

The markets nonetheless remain confident in Trump’s position, because if he doesn’t get to 1237 by June 7th, he’s almost certain to be within striking distance and close enough where he could convince a handful or two of unpledged delegates to join the dark side and vote Trump.

While it’s easy to imagine elite Republicans conspiring to deny Trump the nomination if given an opportunity, such a conspiracy will be much harder to pull off in reality. One of the defining traits of this election has been the total lack of competence from the Republican establishment. Another defining trait has been Trump’s large, frenzied, and violent base of support–a base that’s currently the largest faction of the GOP rank-and-file. While it would make for a great script in House of Cards, it’s hard to imagine the bumbling Republican elite boldly standing up to the powerful, enraged mob of Trumpism, and giving the nomination to Paul Ryan, Mitt Romney, or anyone else they think could and should beat Hillary. Instead they’ll do what they’ve been doing all along:

Lose to Trump.

Nick Warino is a political scientist and does legislative and policy research in Oakland, California