Over 1,200 Mail-In Ballots Were Rejected So Far, and There Are Some Weird Patterns in the Data
Of the 1,248 rejected ballots, there are identifiable patterns in whose mail-in ballots are being rejected, by age, ZIP code, and reason for rejection.
If you put a ballot in the mail in Philadelphia this year, your name might be on a list the city started publishing on September 24. As of September 30 it had 1,248 names on it, one per registered voter, each with a birth year, a ward, a ZIP code, and the reason it was flagged.
Voters have until November 9 to fix whatever the problem is and re-submit their votes, most don't because it requires checking this list online.
There are four reasons for mail in ballot rejections. Undeliverable is 969 of the 1,248. The definition of undeliverable includes a missing signature on the declaration envelope or a missing secrecy envelope. One label, possibly two different problems.
What “undeliverable” actually means
The city’s own name for the category is “Undeliverable or Potentially Flawed,” and it covers two different situations. Some of these ballots came back because USPS could not deliver them to the address on the envelope — the post office codes the reason and mails a notice back. A ballot goes to the address on the voter’s registration record, not one they typed themselves, so the city typed it, printed it, and mailed it, which means the address was often already wrong before the voter was ever involved. The rest were flagged by a clerk as potentially flawed, usually for lacking a signature on the declaration envelope or lacking a secrecy envelope. The published file gives one label for both, so I can’t split the 969.
The rejections split into two groups 33 years apart
The median “undeliverable” ballot belongs to a 41-year-old. The median age of every other voter whose ballot was rejected is in the 70s: 73 for no signature, 75 for ID not verified, and 74 for no secrecy envelope.
There’s a clear monotonic correlation too, the younger you are, the more likely your ballot came back undeliverable. For voters under 25, 97% of rejected ballots were undeliverable. For voters age 25-34, it was 96%. For voters 85 and over, it was 47%.
These 279 failed somewhere else entirely. The post office delivered every one. The ballot arrived and the city couldn’t confirm the voter was the person who cast it.
ID not verified is 79 voters, which means the ID on the mail application didn’t match the state’s registration record — someone married and took a different surname, someone has a hyphen now, someone registered under an address from 1998 and the system has two spellings of a middle name. Those 279 run from birth year 1924 to 2003. Two of them were born in 1924, which puts them at 102, and both are flagged for a missing signature, one in Ward 50 and one in Ward 60. A third centenarian in Ward 36 is on the list for an ID that would not verify. Voters 85 and over are 19% of this group and 4.9% of the undeliverables.
Several areas are disproportionately negatively affected by mail-in ballot rejections. Interestingly, they tend to be the wealthier areas of Philly
One of Philadelphia's wealthiest neighborhooods, Rittenhouse, has more flagged voters than any other ZIP, 102 of them, or about 8% of the total. 90% of those are marked as undeliverable. Graduate Hospital, another well-to-do area, is the next most affected at 90. There are 46 zip codes in Philadelphia, and yet the top 10 account for 49% of all the rejected ballets.
By ward it is more concentrated still. Ward 8 has 140 flagged voters, more than any other in the city, Ward 5 has 81 and Ward 21 has 67, and then it drops off a cliff. Twenty-five of the city’s 66 wards have ten or fewer, and the median ward has 13.
A ward number does not tell you where in the city it is, which is the problem with reading the chart below. A ZIP code is one neighborhood, so 19103 is Rittenhouse. A ward is an electoral district cut across neighborhood lines, and Ward 15 runs through Fairmount, Francisville, Brewerytown and part of Spring Garden. The ward counts say where the concentration is. They do not say what it is.
As I was doing this analysis, I started with the assumption that rejected ballets were mostly going to be from poorer areas, as thats the way media tends to portray the disparate impacts of tighter voter and ballot verifcation rules. So I was surprised to find that, no, in fact, in Philadelphia mail-in ballots at least, there is a STRONG correlation with higher income and greater likelihood of mail in ballot rejection. The below two graphs analyze this.
The one on the left takes a simplistic look - does zip wealth correlate with ballot rejection, and the answer is "no." But that's misleading, because zips have different populate densities and voter rates. The graph on the right redoes the analysis but adjusts for zip populations - and the result is a stark pattern. There is a strong positive correlation between higher income and higher mail in ballot rejections.
One thing to disclose here: higher income zips have higher voter turn out rates, both in person and mail in by a factor of about 2x (17% vs 34%). While I was not able to accurately adjust for this, my rough analysis shows that this does not compensate for this unusual ballot rejection pattern. Moreover - and critically - wealthier voters are generally better informed, hav emore housing stability and generally follow the mail in ballot and USPS rules better, this is well established by national voting patterns research.
So what I see in the data is very, very odd.
Rejected ballots
Rejections per 1,000 residents
Suggested next steps for the Voting Commission and/or election experts to go deeper
Ask USPS for the reason codes. Undeliverable is too vague. For example, it could be the ballot had a 5 digit zip code instead of a 9 digit, or used hard-to-see pink ink. Who knows?
Notify people the day the city knows via phone. Many voters register their phone numbers as part of their voter registration (I did as part of a recent update), so call them, don't just mail the invalidated ballot back. This is particularly important for lower-income, less housing-stable voters.
Match names before throwing a ballot out. National research shows ID-not-verified ballots and 151 missing signatures are usually clerical errors, either by the voter or by the government in the system. In the latter case, the onus can not be on a typical resident to fix government systems' errors they cannot know about.
Investigate why the age gap is so sharp. Younger voters' mail in ballots are rejected because they are undeliverable, whereas older voters' ballots are rejected for missing a signature or invalid ID. Divergences in rejection reasons mean distinct root causes, which require distinct curative actions.
Notes, Sources, and Methodology
The data is the City Commissioners’ own deficiency list, published on September 24, 2026 and current as of September 30. vote.phila.gov has the spreadsheet. 1,248 rows, one per registered voter. Ages are 2026 minus the birth year on the file, so they are accurate to the year. The voter turnout rates cited above come from the city’s own Voter Turnout Details dashboard.
What this cannot tell you. The city does not publish how many mail ballots it sent, so every share here is a share of the 1,248 and not of all mail voters. The file also carries no street address, so nothing here can say why a particular voter’s ballot came back, and the list shrinks as people fix their ballots. Ages are the year, and the per-1,000 rates divide by residents rather than registered voters, which is the wrong denominator and reads high in ZIP codes with a lot of students. November 9 is the identification deadline specifically. The file names private individuals, so every figure here is aggregated and no voter is identified.