Supporters of Republican presidential candidate Donald Trump get autographs after a rally in Rochester, New York. Photo / AP
Five big states are left on the presidential calendar, four for the Republicans and four for the Democrats.
The significance of those states differs for each party, of course - Donald Trump will be scrambling to cobble together enough delegates to win while Hillary Clinton will be hoping to close the door on Bernie Sanders once and for all. But the states are largely the same: New York, Pennsylvania and California - plus New Jersey for the Democrats and Indiana for the Republicans.
Today, Fox News released new polls in the New York and Pennsylvania contests. After a slew of brutal defeats for Trump (both procedural and electoral) and a string of losses for Clinton, some by wide margins, how do these two states look? How have the contests shifted?
Not much.
NEW YORK (April 20)
Trump and Clinton have long been favoured in New York, the home state (adopted in Clinton's case) of each. In the new Fox survey, each still has a big lead.
Trump gets 54 per cent of the vote to 22 per cent for John Kasich. Ted "lol New York values" Cruz is in third, with 15 per cent.
That Trump tops 50 per cent is key. If Trump gets 50 per cent of the vote in the primary, he gets all of the state's at-large delegates, and three delegates in each congressional district that he wins with 50 per cent of the vote. If he's below 50 per cent, statewide or in the congressional districts, he splits the delegates, too.
And for a guy scrambling to hit the 1237-delegate mark to clinch the nomination, that counts. It's why he cancelled plans to campaign in California so that he could hold down the fort at home.
Clinton's up 53 per cent to Sanders's 37 per cent in New York. It's a narrower lead than in past polls (particularly one from Emerson College at the end of last month that showed her getting more than 70 per cent of the vote), but it's still a wide margin.
If that result holds, Clinton will get about 145 delegates to Sanders's 102 - increasing her delegate lead by 43, which is just shy of what Sanders earned by winning Washington by a wide margin. That was Sanders's biggest win in his recent string of victories - and his biggest net delegate haul overall.
That string of victories was largely a function of being in states that had smaller black populations and held caucuses.
In New York, according to the new Fox poll, Clinton gets 61 per cent of support from black voters to Sanders's 29 per cent. But Clinton also wins among whites in her adopted home state, by a 13-point margin. Sanders wins among men (by four points) and people younger than 45 (by 11). Clinton wins everywhere else.\
PENNSYLVANIA (April 27)
Trump's lead in Pennsylvania is not quite as wide, but it is substantial. The problem Trump has here, though, is that most of the state's delegates will head to the convention unbound, meaning that they can vote for whoever they want. He can win the statewide delegates (and this poll suggests that he will) - but that's only going to add 17 delegates to his total.
Clinton's lead in Pennsylvania - again thanks to a small lead among whites and a big one among blacks - means that she is likely to seize somewhere around 72 of the state's 127 delegates, further increasing her delegate lead.
What does the trend look like?
RealClearPolitics's polling averages in both New York and Pennsylvania show how Trump and Clinton's leads in each state have evolved over time. This takes recent polling and averages it, essentially, so things move around as new polls are added to the mix.
At no point has either been leading by less than 10 points in either state.
Sanders supporters may point to the big drop in Clinton's New York lead as evidence of a shift toward their candidate after the Washington primary, which was in late March. It's more of a reflection of that wacky Emerson poll being gradually taken out of the average.
Siena College had Clinton up by 21 in January and February; since March 29, Clinton's been up 12, 10, 18 and 16 points - all within a fairly narrow band.
So what of this talk of momentum, that New York Republicans and Democrats would look at how their peers in Wyoming and Wisconsin and Washington voted and shift their votes accordingly? Almost certainly just rhetoric.
The nature of these primary contests is necessarily deceptive, prompting us to find patterns where none exist. (Blame evolution.) Sanders won a string of contests that favoured him demographically and that happened to be bunched together on the primary calendar - just as Clinton won a string of contests that favoured her demographically and that happened to be bunched together on the calendar. (And, no, the South didn't vote first intentionally to help Clinton.) It's just that Clinton's demographically favourable states had a lot more delegates at stake.
So what's going to happen?
If Trump doesn't hit that 50 per cent mark in New York, a state where he should do well, he's in deep trouble.
He's in trouble anyway, still having to scramble through the remaining states to hit 1237, but if he misses the mark in the friendly turf of his home state, he'll have to do much better in California on June 8 than he does already.
His team has been confidently predicting that it will head into the convention with the majority it needs. If he doesn't hit 50 per cent in New York, we'll see how that language changes.
Sanders doesn't only need to win New York, he also needs to win it by a wide enough margin to take a big bite out of Clinton's 200-plus delegate lead. Any loss for him from here on out means he watches that gap widen, and it's already almost certainly too wide for him to be able to close.
Sanders has been arguing that his campaign has momentum and that this will persuade Democratic super-delegates to vote for him at the convention. (Setting aside Clinton's near-certain pledged delegate and vote-total leads.)
There are a lot of flaws in that argument, not the least of which is that his "momentum" is a function of winning states that he should have won anyway. But it's also flawed in that Clinton will probably have some significant victories on the tail end of the calendar that will give the lie to his overall argument.
In 2008, she won a number of late contests against the opponent she then trailed; it didn't make much difference.
Are we closer to the end?
New York and Pennsylvania most probably won't be the end of either contest. For that, we'll have to wait for California, two months away.
But each state will help push each party's nomination contest further down the path to an eventual outcome - more so in the case of the Democrats.
The new Fox polls, in other words, show that nothing much has changed on either side of the equation, which is useful information in and of itself.
Starship, at 123m tall, is key to the billionaire's Mars colonisation plans.