President Trump speaks to reporters on Aug, 15 in the lobby at Trump Tower in New York. (Jim Watson/AFP/Getty Images)

It’s often difficult to figure out the results of elections based on Zip codes because Zip codes, frustratingly, don’t align neatly with political boundaries. The U.S. Postal Service’s geographic needs are different than the needs of elected officials, making it tricky to figure out how the residents of Zip code 44444 (Newton Falls, Ohio) cast their ballots in the 2016 presidential election.

Happily, Ryne Rohla, working on behalf of Decision Desk HQ, spent a good deal of time after the 2016 election asking different jurisdictions for precinct-level data on how people voted. That’s a level of granularity that we don’t usually get to see, and it allows us to cobble together tallies for Zip codes across the country.

And that, in turn, allows us to do an analysis that’s otherwise tricky: Compare fundraising totals — for which Zip codes are readily available — with the results of the presidential election.

So here, for example, is how Donald Trump’s fundraising over the entire election cycle compared with the final vote results in those places.

He raised a lot of money in the 10022 Zip code — Midtown East in Manhattan, home of Trump Tower. Otherwise, the four Zip codes where he raised the most money were all ones that he won: Palm Beach, Houston, Scottsdale and Orange County. (Those are broad descriptors, of course. 77024 is no more Houston in its entirety than 10022 represents all of New York City.)

Notice how this graph is weighted. The bubbles (scaled to the number of contributions from each place) form a mound with a peak that looks like it’s just to the right of the middle line (which, in this case, represents a perfect tie between Trump and Hillary Clinton in the 2016 election).

Compare that graph with this one, which compares contributions to Clinton with how she fared in those Zip codes.

(That big, outlined bubble at the top is contributions collected through ActBlue, headquartered in Massachusetts. ActBlue reports donations that may not be large enough to trigger reporting requirements for the donor, so they’re reported as from the organization. Since those aren’t all from Somerville, we’ve de-emphasized that Zip code.)

Clinton vastly outraised Trump. Notice that the scale on that graph is millions, not thousands. Notice, in case you’d somehow missed it, how heavily weighted those contributions are toward places that she won more easily. Lots of giving in New York and D.C. The Zip code won by Trump where Clinton raised the most? Palm Beach, home of Mar-a-Lago.

If we consider the number of contributions per Zip code and the total raised in each place as a percentage of the total, we see that the weight of how that giving was distributed mirrors the graphs above.

The Zip codes where Trump was most likely to get a donation were ones that he won by about 20 percentage points. The Zip codes where Clinton was most likely to get a donation were ones she won by at least 70 points.

Overall, Trump raised about $27.5 million in Zip codes he won and $19.1 million in Zips he lost — meaning that about 59 percent of his money came from places he won. Clinton raised about 83 percent of her money in Zip codes she won: $248.6 million to $52.3 million. Yes, Clinton raised almost twice as much in Zip codes Trump won than did Trump — but this includes fundraising from the primaries, when Trump didn’t raise much.

Clinton raised so much in places that she won (that includes a lot of heavily populated Zips in big cities, of course) that it sort of breaks the scale when looking at the average amount and number of contributions, broken out by the winner of each Zip code.

Trump received contributions from about 10,700 Zip codes he won and 5,300 he lost. Clinton received contributions from more Zip codes that Trump won than ones that she won, because Trump won 75 percent of the Zip codes for which we have data. People in about 11,500 Zips Trump won gave to Clinton and about 6,700 Zips she won did.

But Clinton received way more money and way more contributions, on average, in Zip codes she won than Trump did in Zips he won. (In fact, Trump raised more, on average, in ZIP codess Clinton won than Zips he did.)

Clinton actually raised about three times as much money in Palm Beach as did Trump, and seven times as much in the Zip code that’s home to Trump Tower. Absent everything else, though, those numbers don’t tell us much about how the election turned out.