Tag: Election Lawfare

A Statistical Addendum to the Ballot Date Rule Discussion

Jul 16th, 2025

As described in our most recent blog post, lawfare is ongoing in Pennsylvania that seeks to nullify a rule in state election law that undated or misdated mail-in ballots “shall not be counted.” This simple requirement is facing both federal and state court challenges that are yet to be resolved.

An initial motivating factor behind the opposition to this rule was the belief that many thousands of votes had been tossed during the 2022 general election because of it, thereby “disenfranchising” the affected voters. This belief has been reflected in an oft-repeated claim that during this election, upwards of 10,000 mail-in ballots were rejected for failure to be properly dated.

For example, a press release from the ACLU from about a year ago contained the following snippet: ¹

“In Pennsylvania, thousands of voters have their mail-in ballots rejected each year solely because of a common trivial mistake: writing an incorrect date, or no date at all, on their ballot envelope…officials do not use the date to determine whether the ballot was returned on time … and the date is not used to determine voter eligibility. Nonetheless, this essentially meaningless requirement caused the rejection of at least 10,000 ballots from voters who had submitted their mail-in ballots on time in the 2022 general election.”

Judge Baxter (U.S. District Court for the Western District of Pennsylvania) herself cited this claim in her Memorandum Opinion explaining her March 2025 decision which sided with the plaintiffs on their amended complaint, striking down the rule:²

“Under current practice, if the date on the outer envelope is incorrect or missing, the ballot is not counted. It is undisputed that in the 2022 general election, over 10,000 mail ballots were disqualified for this reason.”

As the judge makes clear, the “over 10,000” figure has gone undisputed within the contours of the legal proceedings that have taken place to date. But the origin of this statistic is quite murky, and there is good reason to believe it is a significant overstatement.

This blog post seeks to set the record straight in this regard. Additionally, the post presents a rigorous analysis of data from the 2024 general election in Pennsylvania to quantify the effect of ballots being cancelled due to a missing or invalid date. The analysis finds that 2,055 individuals ended up not voting for that reason, corresponding to just one-tenth of one percent of all returned mail ballots in that election.

Hidden in plain sight: an official count much less than 10,000

Within eight weeks after the 2022 general election, state officials announced that the total number of ballots rejected due to a missing signature; a missing or improper date; or some combination thereof was just under 8,000, per Associated Press reporting: ³

“The Department of State said this week more than 16,000 mail-in ballots were disqualified by county officials because they lacked secrecy envelopes or proper signatures or dates … The agency said 8,250 Pennsylvania mail-in ballots were rejected because they were sent in without being contained within a secrecy envelope, making it impossible for them to be tabulated without putting voter privacy at risk … The remaining 7,904 invalidated ballots were tossed out because the exterior envelopes used to send in those ballots did not have the voters’ signatures, or because those exterior envelopes were either undated or improperly dated.”

Common sense dictates that many of those nearly 8,000 rejected ballots were tossed primarily because of lacking a signature, most likely along with missing a date. The precise number of ballots rejected solely because of a missing or improper date is unknown. Undoubtedly, though, this number would have been substantially less than 8,000.

Moreover, a rejected mail ballot does not necessarily become a lost vote; for instance, the voter may proceed to cast a provisional ballot at the polls on election day. The analysis later in this post will clearly demonstrate this point.

Regardless of how the claim of “more than 10,000 rejected ballots” came about, it is devoid of credibility. Considering both its murky origin and that it is straight-out contradicted by the official figure cited in the Associated Press report, it is an unwelcome imposter and should be ejected from the proceedings.

How many lost ballots in the 2024 General Election?

In a response brief to the defendants’ filing of an appeal of Judge Baxter’s decision in the U.S. 3^rd Circuit Court of Appeals, the plaintiffs reiterated their belief that “thousands” of votes continue to be lost “due to the date requirement. Specifically, they declare that about 4,500 votes were lost in the 2024 general election due to missing or incorrect dates, seen as insufficient progress:⁴

“… before the 2024 election, the Secretary of the Commonwealth redesigned the mail ballot declaration in an effort to reduce widespread errors… in any event, the 2024 election results show that the date requirement has continued to disenfranchise voters by the thousands, even after this redesign. Despite the Secretary’s best efforts to reduce errors, approximately 4,500 eligible Pennsylvania voters had their ballots tossed out for missing or incorrect dates.”

The source cited is a January, 2025 official state press release summarizing what became of the 2.2 million approved mail-in and absentee ballot applications had been distributed to Pennsylvania voters for the November 2024 election, including the following details: ⁵

The distributed mail ballots had an 89 percent return rate
The returned ballots had a 1 percent overall rejection rate
Of the rejected ballots, 23 percent were tossed due to missing or invalid date

Carrying out the requisite calculations implies 1,958,000 returned ballots of which 19,580 were rejected (based on the first two facts).⁶ In turn, this implies (based on the third fact) that 4,500 ballots were rejected because of the date rule.

Not so fast, please!

But how accurate is the count of 4,500 lost votes due to the date requirement? Not very, as will shortly be demonstrated.

The count apparently is based on a simplistic calculation using a database maintained by the state’s Bureau of Elections known as the Statewide Mail Ballot (SwMB) file.⁷ This calculation ignores several important details:

This count pertains to rejected ballots, not rejected voters—the same voter may have more than one mail ballot recorded, which may include replacements for rejected ballots
This count combines ballots recorded as cancelled with those last recorded as pending (which, presumably, the voter had been given a chance to cure)
Ultimately, many of the cancelled or pending ballots did not result in lost votes, because the voter either cured the ballot or cast a provisional ballot on election day

Table 1 presents summary information from the file on the status of mail ballots that were distributed but have not been recorded as votes as of 11/22/2024, including a breakdown of the reasons they remain unrecorded. The total number of ballots cancelled or pending due to mistakes or lapses on the part of the voter (19,334) in Table 1 is a close match to the number calculated above. The total number of ballots cancelled or pending due to missing or incorrect dates is 4,330, which is close to the count of 4,500 calculated above. Thus, Table 1 within reasonable approximation appears to replicate the information reported in the January press release.⁸

Table 1: Unrecorded ballots by reason for current status

	Cancelled	Pending	Combined	Percent of Total
Incorrect date	1,969	440	2,409	12.5%
No date	1,556	365	1,921	9.9%
No ID	312	1,291	1,603	8.3%
No secrecy envelope	2,316	528	2,844	14.7%
No signature	2,480	1,008	3,488	18.0%
Other	882	60	942	4.9%
Returned after deadline	6,124	0	6,124	31.7%
Challenged	3	0	3	0.0%
TOTAL	15,642	3,692	19,334	100.0%
Not yet returned	0	211,353	211,353

Table 1 highlights a key distinction in the data—between cancelled and pending ballots. Cancelled ballots are indeed rejected—withdrawn for violating a requirement, plain and simple. Pending ballots are those for which consideration has been deferred, presumably awaiting remediation from voters who had been granted an opportunity to cure.

The table indicates that 805 out of the total 4,330 ballots that had gone unrecorded due to the date requirement were not outright rejected—the ballots are tagged as pending, implying an opportunity to cure. In the case of these pending ballots, the voters bear at least part of the responsibility for their vote ultimately not being recorded, as they have not availed themselves of that opportunity.

As noted above, another significant complication in regard to interpreting Table 1 is the presence in the underlying data of more than one ballot per voter. In particular, some voters with a cancelled ballot are identified with another ballot that served as a replacement for the rejected ballot.

Therefore, the raw count of cancelled due to a missing or invalid date may overstate the number of votes lost due solely to the violation. If a replacement for the cancelled ballot had been supplied to the voter, then the final outcome depends on whether the voter returned the replacement ballot.

Table 2 summarizes the distribution of ballots per voter within the subpopulation of voters for whom at least one ballot had a missing or incorrect date. Most voters, totaling 3,503, have only a single observed ballot, of which 2,714 are cancelled and 789 are pending. Another 826 voters within this subpopulation have two or more ballots observed in the database.

The total number of voters within this population is 4,329, one less than the total number of ballots with missing or incorrect dates. The difference is due to the fact that one voter had a repeat violation of the date requirement, after being supplied with a replacement ballot.

Table 2: Ballot count distribution for the subpopulation of interest

Ballots per voter	Voter count
1	3,503
1 —cancelled	2,714
1—pending	789
2	770
3	53
4	3
All	4,329

The 2,714 voters whose sole mail ballot had been cancelled had no replacement ballot accorded to them. After cancellation, their sole option would be to vote by provisional ballot at a polling place.

Whether they exercised that option can be determined from their recorded voting histories in the state’s Full Voter Export (FVE) database, by matching their unique voter id numbers across the two databases. It can likewise be determined whether the 789 voters whose sole mail ballot was left pending decided to cast a provisional ballot.

Table 3 summarizes the results from carrying out this analysis. More than a quarter of the voters whose sole ballot had been cancelled due to missing or invalid date recorded in the FVE database as having voted. This leaves 2,020 not voting as a consequence of the ballot rejection. Among those whose sole ballot was pending due to missing or invalid date, nearly 30 percent are indicated as having voted, leaving 559 in pending status.⁹

Table 3: Final outcomes—voters with a single cancelled or pending mail ballot

Status Reason	Non-voted	Voted	Total	Percent voted
Cancelled – incorrect date	1,075	390	1,465	26.6%
Cancelled – no date	945	304	1,249	24.3%
Cancelled – total	2,020	694	2,714	25.6%
Pending – incorrect date	310	118	428	27.6%
Pending – no date	249	112	361	31.0%
Pending – total	559	230	789	29.2%

It remains to ascertain the final outcomes for voters with two or more observed ballots in the SwMB file. Again, this is accomplished by matching to their recorded voting histories in the FVE database.

Table 4 summarizes the results from carrying out this analysis. About 4 of 5 voters in this population (672 out of the total 826) were recorded in the FVE data as having voted. Most of these individuals also had a returned and recorded mail ballot indicated as their final status in the SwMB file.¹⁰

There are only 54 voters in this population who had an undated or misdated cancelled ballot as their final status in the SwMB file, of whom 35 had no recorded vote. Adding these 35 to the 2,020 whose sole mail ballot had been canceled due to a missing or invalid date, and who had no recorded vote, yields a total of 2,055 not voting due to their ballot being cancelled for that reason. The corresponding “rejection rate”, relative to the statewide total number of returned mail ballots, is just one-tenth of one percent.

The second largest category of voters listed in Table 4 are those whose final status is pending due to an unreturned ballot, implying that these individuals were supplied with a replacement for their undated or misdated ballot but had not returned it. More than half of these voters are recorded in the FVE data as having voted.

In total (combining unreturned with pending undated or misdated), 254 voters in this population had pending as their final status in the SwMB file, of whom 113 had no recorded vote. Adding these 113 to the 559 whose sole mail ballot remained pending and who had no recorded vote, yields a total of 662 voters left with a pending ballot and no recorded vote. It bears repeating that these voters bear at least part of the responsibility for having no recorded vote, to the extent they did not avail themselves of an opportunity to cure their flawed mail ballot.

Table 4: Final outcomes—voters with two or more observed mail ballots

Last status in SwMB File	Record in FVE File	Count	Percent
Pending — Unreturned	Vote Recorded	139	56.0%
Pending — Unreturned	No vote / unknown	109	44.0%
Pending — Undated or Misdated	Vote Recorded	2	33.3%
Pending — Undated or Misdated	No vote / unknown	4	66.7%
Cancelled — Undated or Misdated	Vote Recorded	19	35.2%
Cancelled — Undated or Misdated	No vote / unknown	35	64.8%
Cancelled — Other	Vote Recorded	0	0.0%
Cancelled — Other	No vote / unknown	5	100.0%
Ballot Returned — Vote Recorded	Vote Recorded	512	99.8%
Ballot Returned — Vote Recorded	No vote / unknown	1	0.2%
All	Vote Recorded	672	81.4%
	No vote / unknown	154	18.6%
	Total	826

Concluding remarks

Pennsylvania election law requires in no uncertain terms that undated or misdated mail-in ballots “shall not be counted.” This simple requirement is facing federal and state court challenges, motivated in part by a claim that during the 2022 general election, upwards of 10,000 mail-in ballots were rejected for failure to be properly dated.

The origin of this claim is unclear, and though the claim is often repeated, it has no credibility. It is straight-out contradicted by an announcement from the Pennsylvania Department of State in January, 2023, reported by the Associated Press, stating that the number of ballots rejected due to date and signature requirements combined was no more than 8,000.

More recently, opponents of the date requirement have asserted that 4,500 individuals were deprived of their vote during the 2024 general election on account of a missing or invalid date. However, a rigorous analysis of data from the 2024 general election in Pennsylvania indicates that fewer than half that number, 2,055 individual, ended up not voting due to their ballot being cancelled for that reason. That corresponds to just one-tenth of one percent of all returned mail ballots in that election.

An additional 662 voters who had submitted a ballot with a missing or invalid date ultimately were left with a pending ballot and no recorded vote in that election. Arguably, these voters bear at least part of the responsibility for having no recorded vote, to the extent they did not avail themselves of an opportunity to cure their flawed mail ballot.

As a final thought, consider that failure to take note of the date rule or to fill in the correct date can be viewed as a manifestation of the sorts of cognitive limitations we all are subject to. Everyone is more or less prone to forgetfulness, inattention, oversight, miscalculation, and various behavioral biases such as procrastination, making it inevitable that the rule would cause slip-ups. For the hapless 2,055 individuals that lost their vote in 2024, the rule might seem to be the brainchild of a noodge.¹¹

But could it then be the case that, also because of cognitive limitations, for another small percentage of mail-in voters the date requirement might actually be helpful? Could it perhaps be the nudge that leads them to submit their ballot on time? Recalling and writing down the calendar date when completing a mail ballot may well serve as a memory or attention booster, prodding the voter to bring the ballot to a mailbox or drop box promptly, thus precluding a missed return deadline.

This is not mere random speculation. The concept of nudges as mitigants to cognitive limitations has a strong foundation in behavioral economics and the behavioral sciences more generally.¹² Moreover, many more mail ballots were disqualified in the 2024 general election because of missing the return deadline than for having missing or invalid dates. Thus, this hypothetical tradeoff may well be worthy of serious consideration in evaluating the merits of the original complaint.

See “Advocates for voters sue Pennsylvania for rejecting thousands of mail-in ballots due to inconsequential handwritten date mistakes”, May 28, 2024 (link) ↩︎
See Memorandum Opinion, signed by Judge Susan Paradise Baxter on 3/31/2025. Cases.justia.com, pages 5-6 (link) ↩︎
See Mark Scolford, “Majority of 16k cancelled Pa. mail-in ballots were from Dems”, Associated Press, January 6, 2023 (link). ↩︎
See “Brief of Plaintiffs-Appellees Bette Eakin, DSCC, DCCC, AND AFT Pennsylvania”, United States Court of Appeals for the Third Circuit, Case 25-1644 Document 90, Filed 06/04/2025, at page 9 (link). ↩︎
See “Shapiro Administration Announces 57% Decrease in Mail Ballots Rejected in 2024 General Election”, Pennsylvania Department of State, January 24, 2025 (link). ↩︎
These counts closely align with data published by the U.S. Election Assistance Commission which indicate 1,933,102 returned ballots and 19,270 rejected ballots for the 2024 general election in Pennsylvania. See Election Administration and Voting Survey 2024 Comprehensive Report, U.S. Election Assistance Commission, June 2025, Chapter 1 Appendix A, Overview Table 1, pages 34-37 (link). ↩︎
This file contains point-in-time detailed data pertaining to the processing of absentee and mail-in ballots. It is refreshed daily and made available to the public for informational purposes during the course of an election. ↩︎
The count of pending and cancelled attributed to missing or incorrect dates corresponds to 22.4 percent of total cancelled or pending in Table 1. This is close to but falls short of the reported 23 percent in the January press release, likely reflecting updates or corrections made to the data during the interim period. ↩︎
To be more precise, 10 percent of the voters in this population who were recorded in the FVE data as having voted were indicated to have voted by mail, indicating that the SwMB file did not have the updated status for their mail ballots. ↩︎
A single voter in the SwMB file is indicated to have returned a ballot and recorded a vote, but was not found in the FVE file, presumably due to no longer being registered in Pennsylvania as of 1-20-2025. ↩︎
The colloquial term noodge derives from Yiddish and typically refers to an annoying or pestering person. ↩︎
The term nudge in such contexts refers to a subtle intervention to influence individuals to make wiser decisions, countering behavioral biases or limitations on the ability to process information. See, for example, John Beshears and Harry Kosowsky, “Nudging: Progress to Date and Future Directions”, Organizational Behavior and Human Decision Processes Volume 161, Supplement, November 2020, Pages 3-19. https://pmc.ncbi.nlm.nih.gov/articles/PMC7946162/ ↩︎

The Great Ballot Date Debate

Jul 1st, 2025
Case Ongoing after Pennsylvania Federal Judge Rules in Favor of Counting Undated Ballots

As disagreements over election rules and procedures expand and intensify nationwide, a major battle has been taking place in the key swing state of Pennsylvania over a basic prescription in state election law. The Pennsylvania Election Code holds that undated or misdated mail-in ballots “shall not be counted.” This simple requirement has faced both federal and state court challenges that are yet to be resolved.

On the federal side, on March 31, 2025, Judge Susan Baxter of the Western District of Pennsylvania issued a controversial ruling that would allow undated or misdated ballots to be counted.¹ Judge Baxter ‘s ruling stemmed from a pair of lawsuits filed after the November 2022 election–by organizations and individuals represented by the ACLU, and by the John Fetterman campaign, DCCC and DSCC.² These lawsuits claimed that voters’ constitutional rights had been violated when their ballots were annulled because of the rule.

The state case relates to ongoing disputes in three individual counties over handling undated or misdated mail-in ballots. The state’s Supreme Court, not known for judicial conservatism, nonetheless issued rulings back in October and November 2024 blocking lower court decisions that would have voided this requirement in the law. The Court has since taken up the matter itself.

The Federal Case

In November 2023, for background, Judge Baxter had ruled in favor of the plaintiffs in response to the original Fetterman campaign lawsuit, that undated or misdated mail-in ballots must be counted.³ Baxter agreed with the plaintiffs’ central argument that the law violated the “materiality clause” of the Civil Rights Act, which prohibits trivial impediments to voting, such as cancelling of votes due to minor errors by the voter.

Of course, what constitutes “minor error” is open to question. The Republicans fought back claiming the rule’s efficacy in preventing voter fraud, but Baxter was unconvinced by this claim.

Baxter’s initial decision in 2023 was overturned on appeal. Specifically, on March 26, 2024, the Court of Appeals found that the date requirement did not violate the materiality provision.⁴

Subsequently, a single voter from the original group of plaintiffs in the 2022 case, joined again by some Democratic Party affiliated groups, returned to Baxter’s court with an amended complaint. This time, they argued that the rejection of undated or misdated ballots in 2022 violated their rights to free expression under the First Amendment and Civil Rights Act, which includes voting rights.

As, noted, on March 31, 2025, Baxter again sided with the plaintiffs. She determined that the dating rule serves no functional рurpose and therefore concurred that the rule deprives people their right to vote, in this case citing said right to “free expression”. She also again dismissed the Republican claim that the dating requirement helps deter fraud, citing lack of evidentiary support. Other Republican arguments, including that the date requirement preserves the “solemnity” of the act of voting, were dismissed as too vague or “nebulous”.

On April 1, the Republican National Committee, Pennsylvania Republican Party and National Republican Congressional Committee filed an appeal with the Third Circuit Court of Appeals, which is currently hearing the case.⁵ If the Republicans lose the appeal, the Supreme Court may end up ultimately ruling on it, which could lead to a national decision on the rule.

The State Case

In May 2024, a group of Pennsylvania voters, claiming that they were disenfranchised in a local election by the ballot dating rule, initiated legal action in the State’s Commonwealth Court.⁶ The plaintiffs argued that the dating requirement violates the Free and Equal elections clause in that it does not have any meaningful application or function, being merely a technicality. Therefore, it is an undue impediment to exercising the right to vote and unenforceable.

The state Republican Party and the RNC, as intervenors, tried to broadly argue that it helps prevent vote fraud, but failed. The Commonwealth Court ruled four-to-one on August 30, 2024 in favor of the plaintiffs–that ballots not marked with the correct date, or any date, could not be rejected for that sole reason.⁷

The opinion of the four Commonwealth justices noted that the statute is unambiguous (must sign and date), but must pass a “functional” test. Accordingly, since the date serves no constitutional or administrative purpose, “disenfranchising voters” violates the Free and Fair Election Clause and therefore is invalid.

Judge Wolf on the Commonwealth Court dissented, stating removing the requirement right before the election could spread confusion and legal uncertainty. In addition, Judge Wolf contended that the constitutional issue should be addressed by the state Supreme Court, not an intermediate appellate court.

On September 13, following an appeal from the state’s Republican Party and the RNC, the Pennsylvania Supreme Court voided the Commonwealth Court decision. However, the Supreme Court ruled only that that the lower court lacked jurisdiction based on certain technicalities; the Court did not address the substance of the lower court decision.⁸

On September 23, 2024, the ACLU of Pennsylvania, Public Interest Law Center, and Arnold & Porter filed a related suit in the Philadelphia Court of Common Pleas on behalf of two voters, challenging their disqualifications in the September 17 special election Philadelphia.⁹ These voters, one of whose names, Brian Baxter (no relation to Judge Baxter), came to be associated with the case, had submitted a ballot on time but it lacked a date. In all, 69 votes had not been counted in due to a missing or improper date. On September 26, Common Pleas Judge Crumlish ruled in the voters’ favor, ordering that those 69 ballots be counted.

Soon after, on October 30, the PA Supreme Court stayed that decision, citing the Purcell principle. This means, as Justices Dougherty and Donohue emphasized in concurrences, that ongoing changes to electoral rules mid-campaign undermine electoral stability, consistent with Judge Wolf’s dissenting view.¹⁰

The Supreme Court ruled separately against Bucks, Montgomery and Рhiladelphia counties, on November 18, after they sought to circumvent their October 30 ruling. These counties had been attempting to include undated or misdated ballots in the final count for the November 5 election, including for the US Senate race which was narrowly won by Republican Dave McCormick.¹¹

On January 17, 2025, the Pennsylvania Supreme Court agreed to hear Baxter v. Philadelphia, where it will ultimately decide the fate of the state court challenges. ¹²

The Unrivaled Politics Perspective

The Supreme Court’s final ruling, expected in the coming months, will either uphold or strike down the lower court’s ruling on enforceability of the date requirement. The questions on constitutionality of the rule which the Court will explore, are:

• Does the date requirement burden voting rights without meaningful purpose?
• Can the will of the legislature in setting this requirement be nullified based on the supposition that the Free and Equal Elections Clause (or similarly, the First Amendment in the context of the federal case) invalidates it?

Plaintiffs in the state cases have argued in their briefs that the Free and Equal Elections clause, which bars any interference with “the free exercise of the right of suffrage,” prevents the enforcement of the ballot date rule. The appellees have argued that the date requirement is purposeful, and that it is the role of the legislature to decide on election rules and protocols, not the courts, so as not to destabilize government or future elections.

The state’s constitution, as well as the United States Constitution, would seem to agree with the latter assertion. The U.S. Constitution delegates the power to implement election law to the state legislature. Act 77 was passed by the Pennsylvania state legislature in September 2019, among it codifying the ballot dating requirement. The Constitution of Pennsylvania also grants to the state legislature, not to state courts, the decision on laws related to statewide elections, including рrocedures, equipment, and roles of рersonnel.

Complicating matters, though, Pennsylvania’s Supreme Court famously “overruled” the legislature shortly before the 2020 election on the key issue of the mail ballot receipt deadline for that election. The U.S. Supreme Court later refused to overturn that ruling, opining that state legislatures do not necessarily have the sole authority (Moore v. Harper). These decisions were met with an outcry from constitutional conservatives, who saw them as undermining the authority granted to state legislatures by the U.S. Constitution.¹³

Even if the U.S. Constitution no longer grants to legislatures sole authority to rule on this issue, the Constitution of Pennsylvania still grants clear authority in-state and the state’s Supreme Court should reinforce this. The “Free and Equal” clause would seem irrelevant in the case of a protocol (the date requirement) in place for many years (originally for traditional absentee ballots and later extended to the new mail-in ballots.) It reflects the lawmakers’ original intuition and will on election integrity and as such should not be subject to the clause.

This segues into a second line of argument, one that the Republican Party has thus far overlooked in its appeals, which relates to the actual purpose and merits of the rule in question. The ballot date requirement is not for naught–it is not a meaningless barrier to the voter, as the plaintiffs argue.

A ballot’s signature date provides an important data point. It is an indicator of when a ballot was filled out and mailed, and hence whether the ballot was likely submitted in time to be counted. It can help validate claims by election officials that certain ballots arrived late and are therefore invalid, wherefore the ballots aren’t counted. In essence, ballot dating serves the purpose of protecting voters.

It is important to note that with few exceptions, election officials and staff, like other government workers or law enforcement officers, are hard-working, fair-minded, and dedicated to their jobs. The mail-in voting process does not always run smoothly, however, as experience demonstrates. Ballots may be delayed in processing due to inadequate resources and resulting backlogs, or to glitches or errors. The possibility of malfeasance cannot be entirely ruled out either.

Therefore, it is imperative that the process be as transparent as possible, and transparency depends on accessible information. Ballot dates in aggregate provide useful, real-time information for assessing the efficiency and timeliness of election officials in processing and recording of mail ballots. Likewise, ballot dating provides information useful for identifying and remediating problems along these lines.

Such information may be particularly valuable in cases of delays that differentially affect one group of people (distinguished, for example, by party affiliation or by ethnicity) or one particular district or county. Ballots signed on the same date can be assumed (on average) to have been mailed around the same time. One group having a larger share of ballots rejected for lateness than others should raise concerns about fair treatment, especially if observed between voters with the same signature dates.

Thus, like the ballot secrecy envelope (which also at times has been deemed a trivial requirement despite its role in ensuring voter privacy), ballot dating can be integral to ensuring the fairness and integrity of the vote. Hence, the Unrivaled Politics view is that the date requirement should remain, not only because it has been duly legislated, but because it helps promote efficient, fair, and trustworthy elections.

Not only that, but because of its potential information value, ideally the signature dates should be digitized and included as a data item in the Statewide Mail Ballot File for each election.¹⁴

In the modern-day age of voting and elections, it seems that the political parties are at each other’s throats in never-ending court battles over election rules and procedures. The left fights fiercely for as little regulation as possible, seeking to increase voter participation even possibly at the expense of legitimacy. The right seeks to preserve traditional rules and regulations intended to ensure security and integrity.

In general, the sensible solution should be to follow the text of the U.S. Constitution, by which the framers intended to delineate and preserve the roles of our government institutions. This requires settling matters on voting via legislation. It is primarily the role of lawmakers, not the judiciary, to decide on the rules that protect voters and guard election integrity while ensuring access to those legally eligible to vote. –but rather a measure that helps secure the integrity of the vote.
1. See Peter Hall, “Federal judge rules Pa. ballot dating rule violates constitutional right to political expression.” Pennsylvania Capital-Star, March 31, 2025 (link). ↩︎
2. The filing of these lawsuits had been preceded by several skirmishes in state courts over the date requirement, culminating in a November 1 order from the Supreme Court of Pennsylvania that undated and misdated ballots should be segregated and not counted. See Ellis Champion, “Voters, Fetterman Campaign, DSCC and DCCC Sue Over Mail-in Ballots in Pennsylvania.” Democracy Docket, November 7, 2022 (link). ↩︎
3. See Sean Kitchen, “Ballots that arrive in time should be counted, Pennsylvania judge rules.” The Keystone, November 22, 2023 (link). ↩︎
4. See Nate Raymond, “Pennsylvania can discard undated mail-in ballots, US appeals court rules.” Reuters, March 27, 2024 (link). ↩︎
5. See Justin Sweitzer, “GOP appeals federal judge’s mail-in ballot decision.” City & State Pennsylvania, April 3, 2025 (link). ↩︎
6. See Carter Walker, “Pennsylvania’s rule on dating mail ballots faces new challenge in state court.” Votebeat, May 29, 2024 (link). ↩︎
7. See Carter Walker, “Mail ballots can’t be rejected over improper date on envelope, Pennsylvania court rules.” Votebeat, August 30, 2024 (link) ↩︎
8. See Carter Walker, “Pennsylvania Supreme Court restores date requirement for mail ballots, voiding lower-court ruling.” Votebeat, September 13, 2024 (link). ↩︎
9. This special election was held to determine representation in state house districts 195 and 201. See “Pennsylvania Philadelphia Undated and Wrongly Dated Mail-in Ballots Challenge”, Democracy Docket, September 23, 2024 (link). ↩︎
10. See Carter Walker, “Philadelphia must count undated and misdated mail ballots from September election, Pa. court rules.” Votebeat, October 30, 2024 (link). ↩︎
11. See Carter Walker, and Sarah Anne Hughes, “Pennsylvania Supreme Court stops 3 counties from counting undated, incorrectly dated ballots.” Votebeat, November 18, 2024 (link). ↩︎
12. See Carter Walker, “Pennsylvania Supreme Court to hear latest case over undated mail ballots.” Votebeat, January 17, 2025 (link) ↩︎
13. Conservative talk show host Mark Levin, for example. ↩︎
14. This file contains point-in-time detailed data pertaining to the processing of absentee and mail-in ballots. It is refreshed daily and made available to the public by the state’s Bureau of Elections for informational purposes during the course of an election. ↩︎
Bonus Ballots for Biden: A Case Study of Election Lawfare Excess

Aug 27th, 2024
As the 2024 Presidential campaigns ramp up, election lawfare is underway as well, with the political parties seeking to exploit ambiguities in voting rules or bend them in their favor, and we can only hope that its role will remain limited and not affect trust in the election. This article highlights the problematic nature of election lawfare, focusing on a controversial case from the 2020 Presidential election in Pennsylvania.

In 2020, in states with newly expanded mail-in voting processes where the parties battled in court over what would be permissible under the new rules and what would not be, with Pennsylvania a particularly active lawfare battleground. The case discussed and analyzed here resulted in extension of Pennsylvania’s originally mandated deadline for receipt of mail-in ballots, and likely had a material effect on the final vote tally.

The analysis, which is in part data-driven, calls into question the primary, stated rationale for extending the deadline, which was to ensure sufficient time for return of mail-in ballots for those who had been relatively late in requesting a mail-in ballot. It also offers a rough quantification of the net vote gain for Joe Biden arising from the extension.

While this number may have been material–possibly as many as 9,000 votes, it was far from enough to swing the election. Nonetheless, in combination with other perceived shortcomings of the process such as lax security around ballot drop boxes, the case contributed to a broader erosion of trust in the election outcome.

The Democrats’ legal challenge to the ballot return deadline

No-excuse mail-in voting was introduced into Pennsylvania by PA Act 77, which was legislated in 2019 with bipartisan support in the state legislature and the blessing of the Democrat governor (Tom Wolf). Although the Democrats in the state’s General Assembly in 2019 mostly opposed Act 77, the party had a change-of-heart by the summer of 2020. Conversely, at the national level during 2020, President Trump and other Republican leaders increasingly were expressing fears that no-excuse mail-in voting might facilitate election fraud.

Observations on voter turnout during the 2020 Primary in Pennsylvania might have helped motivate the Democrats’ embrace of mail-in voting. No-excuse mail-in ballots accounted for almost 40 percent of total votes in the 2020 Primary Election, underscoring the preference of many voters for the mail-in option. Although data on mail-in ballot shares by party are not publicly available, it is a reasonable presumption that Democrats in the Primary were more likely to vote by mail than Republicans, as turned out to be the case in the general election, and that this fact was known to strategists in each party. ¹

Moreover, it seems that, because of the perception that liberalized vote-by-mail favored their party, Democrats were not content with a literal or narrow reading of Act 77. They adopted a strategy of lawfare to try and stretch the limits of mail-in voting as far as possible in order to maximize their perceived vote potential.

On July 10, 2020, the state Democratic Party, joined by some individual Democratic officials and candidates, filed a petition for review in the Commonwealth Court against Secretary of State Kathy Boockvar and all 67 county boards of elections. The petition requested declaratory and injunctive relief “to protect the franchise of absentee and mail-in voters.”² Specifically, the plaintiffs asked the Court to rule in favor of a number of requested flexibilities, among which was to grant an exception to the deadline in Act 77 for receipt of mail-in ballots, such that any ballot postmarked by 8:00 p.m. on election day (November 3) and received by 5:00 p.m. on Tuesday, November 10 can be counted.

Technically, Boockvar was an opponent of the petitioners in the case. However, she herself was a Democrat, and on August 13, she filed a response that objected only to the requested seven-day extension of the received-by deadline, primarily on logistical grounds, proposing instead a three-day extension. Her response strongly concurred with the petitioners’ other requests.

On August 16, Boockvar filed an application requesting that the state’s Supreme Court exercise extraordinary jurisdiction over the petition for review; that is, take over for the intermediate appellate Commonwealth Court, in order that the resolution of these matters be expedited to preclude any disruption of the 2020 General Election. On August 19, the petitioners filed an answer to the Secretary’s application, indicating no objection to the Supreme Court exercising extraordinary jurisdiction. The Pennsylvania Supreme Court, where 5 of the 7 justices are Democrats, then took over the case.

The state Supreme Court ultimately agreed to an extension of the ballot receipt deadline by three days so that any ballot cast by 8 p.m. on Election Day, (Nov. 3, 2020) and received by 5 p.m. on Nov. 6 would be counted. The Court’s decision, which was based on majority opinion, was announced on September 17.

Review and critique of the petitioners’ arguments

The petitioners had argued that special circumstances arising from the Covid-19 pandemic justified a seven-day extension to the received-by deadline for mail-in ballots. Essentially, they contended that, due to the preference of many voters to stay away from polling places on account of the ongoing pandemic, the volume of mail-in ballot applications would likely overwhelm county resources, such that bottlenecks and delays in the processing of ballot applications and distribution of ballots would likely arise. Secondarily, they noted the potential for delays in the United States Postal Service (USPS) delivery system as an aggravating factor.

They argued that under the state constitution’s Free and Equal Elections Clause, which prohibits interference in or prevention of “the free exercise of the right to suffrage”, the state was obligated to take action to remedy the anticipated disruption to mail-in voting. As described in the Court’s written opinion:

Petitioner interprets this provision as forbidding the Boards from interfering with the right to vote by failing to act in a timely manner so as to allow electors to participate in the election through mail-in voting.

A key component of the petitioners’ argument was a claim that the kinds of obstacles to the timely processing of ballot applications that were encountered prior to the June 2^nd Primary Election would recur and be even more severe. As summarized in the Court’s written opinion:

Petitioner avers that the difficulties encountered by Boards processing the ballot applications prior to the June Primary will only be exacerbated in the November General Election. It emphasizes the continued grip of the pandemic and a potential second wave of infections…. It highlights that the Secretary estimates that 3 million electors will seek mail-in or absentee ballots for the General Election in contrast to the 1.5 million votes cast by mail at the Primary.

However, the petitioners offered no substantive analysis to support this claim—only gripes about difficulties during the Primary and a poorly supported, inflated estimate of increased processing burden for the General Election.

During the Primary, many counties had indeed been inadequately prepared for the unexpectedly large volume of ballot applications they received, as this was the state’s first encounter with no-excuse mail-in voting. Thus, in some counties, most notably in Bucks and Delaware Counties, a significant number of ballots were mailed out too late for voters to return them by the deadline.

But clearly, county boards would have learned from their experiences in the Primary and taken action to shore up their capabilities in anticipation of needs for the General Election. The Court did not adequately consider what remedies might already have been implemented or what preparations were currently underway.

At least anecdotally, such ongoing efforts were substantive. A press report in late August contained the following assessment:³

Democratic leaders also contend they’ve remedied the issues that plagued them in the June primary when election officials were overwhelmed by a deluge of mail-in ballot requests during the coronavirus pandemic. It was the first time the state had conducted a mail-in election since a new law authorized the procedure last year.

For example, according to this report, Montgomery County had “used federal funds from the Cares Act to add eight staffers and more technology to help process the vote faster” since the Primary.

Another press report in early September included the following snippet:⁴

We were learning on the fly from February to June,’ says Jeff Greenburg, the director of elections in Mercer County until August, when he stepped down to work for The National Vote at Home Institute. ‘I really think Pennsylvania is in a better position now. … To me, we have a much better chance of succeeding because we now know what to do.’

Not only did the petitioners fail to back their claim that county boards remained inadequately prepared, but they overstated the anticipated increase in application processing volume. As evidence of increased burden, the petitioners had alluded to the gap between 1.5 million “votes cast by mail” in the Primary and 3 million anticipated applications for mail-in or absentee ballots for the General Election. This comparison, referenced in the Court’s written opinion, is misleading, as it is not apples-to-apples. The 3 million applications anticipated for the General Election is more appropriately measured against the total number of applications for mail-in or absentee ballots received for the Primary, which was 2 million.

Even that corrected numerical comparison, however, materially overstated the potential increase in processing burden, because much pre-processing of applications had already been completed. In fact, about 1.1 million individuals who had requested and been approved for a mail-in ballot in the Primary had signed up to be automatically sent a mail-in ballot for the upcoming general election, a highly pertinent fact missing from the discussion.

No further processing was required for the latter requests aside from the eventual automated printing and mailing of the ballots. Netting these out from the projected 3 million implies 1.9 million new applications for the General Election, not surpassing the 2 million processed for the Primary.

Review and critique of Boockvar’s arguments

Secretary Boockvar did not concur with the petitioners’ claim regarding the lack of preparedness of county boards, nor their request for a one-week extension. Instead, she advocated for a three-day extension based on the contention that those applying for mail-in ballots on or close to the statutory deadline for doing so, October 27, would face a high likelihood of missing the receive-by deadline for returned ballots if they chose to return them by mail.

The USPS delivery standards in effect at the time specified 2-5 day delivery for domestic First-Class Mail and 3-10 day delivery for domestic Marketing Mail. As described in the Court’s written opinion, “the USPS had advised that voters should mail their ballots no later than October 27 in order to meet the received-by deadline”. Boockvar argued that under these standards, a voter applying for a mail-in ballot on Oct. 27 would face a substantial risk of not meeting the November 3^rd deadline:

…even if a county board were to process and mail a ballot the next day by First Class Mail on Wednesday, October 28th, according to the delivery standards of the USPS, the voter might not receive the ballot until five days later on Monday, November 2nd, resulting in the impossibility of returning the ballot by mail before Election Day, Tuesday, November 3rd…This mismatch [between the USPS’s delivery standards and the Election Code deadlines] creates a risk that ballots requested near the deadline under state law will not be returned by mail in time to be counted…

Further, Boockvar argued that mail delays associated with the pandemic would make the problem worse, selectively disadvantaging those who are applying for mail-in ballots shortly before the October 27 deadline and therefore necessitating a one-time extension of the received-by deadline:

The Secretary emphasizes that the remedy sought here is not the invalidation of the Election Code’s received-by deadline, but rather the grant of equitable relief to extend temporarily the deadline to address ‘mail-delivery delays during an on-going public health disaster’.

This argument, however, is specious. To begin with, it rests on the illusory notion that the application and received-by deadlines in Act 77 were originally flawed because of inconsistency with standard USPS delivery times.

The legislature could have opted for an earlier application deadline, whereby voters returning their ballot by postal mail would be more assured of meeting the received-by deadline. Instead, the legislature opted to provide more flexibility on application timing with the later application deadline.

Act 77 grants voters the choice to wait up to a week before Election Day to apply for a mail-in ballot, with the implicit, common sense understanding that voters will bear an increased risk of missing the received-by deadline the longer they wait to apply. To mitigate this risk, Act 77 also provides voters the option of hand delivering their mail-in ballots, including the option of applying for and submitting ballots in-person at an election office on the same day. To claim the deadlines were flawed is disingenuous.

Moreover, Boockvar’s argument presumes that voters have little or no agency — that those who apply close to the application deadline and consequently miss the receive-by deadline do so by fate — suggesting an obligation to provide such voters with flexibility on the receipt deadline when mail delays occur. However, voters can choose when to submit a ballot application. The possibility of pandemic- related mail delays should just change their calculation, motivating them to apply sooner rather than later.

Additionally, the argument fails to consider factors that would mitigate the effect of any pandemic-related mail delays. Drop boxes are one obvious mitigant. Surely, the availability of drop boxes for returning a mail-in ballot limits the risk that late applicants will miss the receive-by deadline. Another is to inform voters about the potential for mail delays and to encourage them to apply early, which was, in fact, the sensible advice being disseminated by the USPS at the time.

The Court decides

The Court’s written opinion indicates no awareness of any of these weaknesses in the arguments presented by the petitioners and Boockvar. The judges apparently were swayed by those arguments:

In light of these unprecedented numbers and the near-certain delays that will occur in Boards processing the mail-in applications, we conclude that the timeline built into the Election Code cannot be met by the USPS’s current delivery standards…The Legislature enacted an extremely condensed timeline, providing only seven days between the last date to request a mail-in ballot and the last day to return a completed ballot. While it may be feasible under normal conditions, it will unquestionably fail under the strain of COVID-19 and the 2020 Presidential Election, resulting in the disenfranchisement of voters…this Court can and should act to extend the received-by deadline for mail-in ballots to prevent the disenfranchisement of voters.

In short, the Court undertook to alter the deadline ensconced in plain print in the legislation, potentially skewing the vote outcome in favor of one party. The decision lacked factual support, and instead relied on an inaccurate and misleading estimate of processing volume and on nonsensical griping about a “condensed timeline.”

Quantifying the effect on the vote tally

The Court’s decision to grant a three-day extension to the ballot return deadline favored Biden due to the overwhelming ratio of Democrats relative to Republicans anticipated to vote by mail. It can also be seen as implicitly disfavoring those who were planning to vote at the polls, insofar as they were not extended any comparable flexibility, such as expanded hours at polling places or more than one day of in-person voting.

How many mail-in and non-overseas civilian absentee ballots did the Democrats manage to gain during the extended timeframe, net of ballots returned by Republicans? To address this question, we turn to a dataset from the Pennsylvania Department of State which has detailed information on each approved application for an absentee or mail-in ballot for the 2020 General Election. A link to the dataset, which is housed within the Department’s online Open Data Portal, was shared with the author in response to a “Right-to-Know Law” (RTKL) request.⁵

Individual data items include the date the application was received by the county in which the voter is registered. In addition, the data distinguish between traditional absentee ballots available to voters and the newly available, no-excuse mail-in ballots (henceforth to be referred to simply as absentee and mail-in, respectively) and further differentiate among types of traditional absentee ballots, such as military, overseas civilian, regular civilian, and so on.

The database also provides the date the ballot was received back by the county election office (the ballot return date). Non-return of a mail-in ballot that was sent to a requester is indicated by missing information in the return date field: if the ballot was not recorded as returned, then no return date is present.

Chart 1 shows the counts of returned mail-in and non-overseas civilian absentee ballots received in the two weeks prior to and including November 3, the statutory deadline for returning these ballots, and the return counts for the three-day extension period granted by the Court. The counts are separated by party affiliation of the voters. During these three days, 9,432 more ballots were returned by Democrats compared to the number returned by Republicans.

While some of these ballots from Democrats might have been votes for Trump and some from Republicans might have been for Biden, there is no way to determine how many of each, and a best guess is to assume that such cross-party votes are few and mutually offsetting. Under the further assumption that Independent and other voters split equally between Trump and Biden, the net additional vote count for Biden is simply this difference between the returned ballot count of Democrat versus Republican-affiliated voters.

Does the extension seem justified in retrospect?

Chart 1 also indicates that the number of ballots returned from October 21 through November 3 swamps the number returned during the added three days. The number returned during the added days sums to 45,761, just 2.3 percent of the total number (1,987,793) returned on or after October 21.⁶ This small share seems inconsistent with the notion that pandemic-related mail delays would present a major obstacle to the timely return of mail-in ballots—the primary rationale put forth to justify the extension.

The three-day extension was motivated by a perception that those applying close to the application deadline would require this flexibility due to the potential for “systematic mail delays”. Chart 2 shows how ballots with November 4-6 return dates were distributed by application period. As seen in the chart, over half of the mail-in and non-overseas civilian absentee ballots with extension period (November 4-6) return dates were associated with applications filed prior to October 15; that is, applications filed well in advance of the deadline. Thus, most of those benefiting from extension of the deadline had not been late filers.

Another question worth asking is, how many of the ballots with application dates close to the October 27 application deadline would not have arrived in time had the extension not been granted? If there were many such ballots, then this could provide some ex-post justification for the Court’s decision.

Chart 3 shows the percent returned during the November 4-6 extension period by date of application for each application date in the late October (October 15-27) application period. Not quite 10 percent of individuals applying within three days of the deadline had return dates on November 4-6, only a slightly higher percentage compared to individuals applying four to seven days before the deadline.

These outcomes seem inconsistent with the anticipated “systematic mail delays” put forth to justify the extension, and quite plausibly can be attributed instead to procrastination on the part of some individuals or to idiosyncratic factors affecting ballot return times. In retrospect, there seems to have been little justification for extending the deadline, especially as some individuals intending to vote in person may have missed their opportunity to vote due to idiosyncratic factors, and it seems unfair to have provided extra accommodation only for the mail-in voters.

Not all ballots with recorded return dates in November 4-6 are votes that would have been lost

If the Court had left the statutory deadline unchanged, then it seems quite plausible that some of the individuals whose votes came in during the added three days would have behaved differently, returning their ballots sooner.

In other words, voters’ behavior cannot be presumed independent of the stated deadline—voters would have adapted their behavior in response to changing the deadline. It is not possible, however, to determine how many voters with return dates on November 4-6 would have met the original deadline had it not been extended.

Likewise, the actions of those whose duty it is to collect and record ballots may have adapted to the extension of the deadline. In some cases, those responsible for collection may have left some ballots uncollected as of the close of business on November 3, knowing that there would be additional time for completing the task. Even if all ballots had been collected on the 3^rd, some officials might have delayed opening and recording them until the next day, at which point they might be assigned a November 4 return date.⁷ The actual net gain to Biden due to the extension would depend on the total number of ballots incorrectly attributed to the extension period and how they divide up between Biden and Trump, which are also unknown quantities.

Enabling underperformance?

A far disproportionate share — nearly half — of the mail-in and non-overseas civilian absentee ballots with return dates on November 4-6 were from voters in just two counties, Bucks and Delaware. Chart 4 shows the distribution of ballots with November 4-6 return dates across the top 10 counties ranked by number of such ballots. Delaware County alone accounted for a third of all ballots returned during the three-day extension period. Bucks had about the same share of these ballots as much larger Philadelphia.⁸

These same counties, Bucks and Delaware, were notorious for having had especially severe delays in distributing mail-in ballots for the Primary in June. Uniquely among all Pennsylvania counties, these two had gone to their county courts to seek an extension to the ballot return deadline for the Primary. In fact, the experience of these two counties in the Primary was highlighted by proponents of the November extension as portending more systematic delays in the General Election. As described in the state Supreme Court’s written opinion regarding the November extension:⁹

…in Delaware County, thousands of ballots were ‘not mailed out until the night’ of the Primary, making timely return impossible… Bucks County apparently experienced similar delays. To remedy this situation, the Election Boards of Bucks and Delaware Counties sought relief in their county courts… the Courts of Common Pleas of Bucks and Delaware Counties extended the deadline for the return of mail-in ballots for seven days, so long as the ballot was postmarked by the date of the Primary.

Hence, one might reasonably conclude that the primary accomplishment of the Supreme Court’s granting the three-day extension was to relieve these two counties from responsibility for making needed improvements to their ballot distribution and collection processes. Surely, in retrospect, this is insufficient justification for the Court’s decision.

Concluding remarks

The troubling nature of election lawfare is exemplified by a particular case from the 2020 Presidential election in Pennsylvania in which the Democrats convinced the state’s Supreme Court to extend the deadline for receipt of mail-in ballots. In short, the Court undertook to alter the deadline ensconced in plain print in the legislation, potentially skewing the vote outcome in favor of one party.

The decision lacked factual support. Rather, the judges (the majority of whom were Democrats) were swayed by inaccurate and misleading information that was presented to them, and by a nonsensical, emotionally driven argument about a “condensed timeline” for mail-in voters arriving late to the game.

The Court’s decision was wrongheaded, because of the weak reasoning brought to bear. Overriding the explicit requirement of the legislation without adequate justification amounted to judicial overreach.

A bit of irony is evident here, considering the lawfare conducted in 2020 by the Democrats against the Green Party in Pennsylvania. A Green Party official chose to fax (as opposed to hand deliver or mail) a candidate affidavit, which was technically a violation of candidacy requirements. This minimal infraction, during the state’s Covid lockdown when office hours were limited and direct contact between persons discouraged, could have merited forbearance, but the Democrats succeeded in expelling the Green Party candidate from the ballot. Whereas the Democrats would grant no leeway to the Green Party, they self-servingly pursued a far greater degree of forbearance regarding the statutory ballot receipt deadline, based on flimsy justifications using the pandemic as a pretext.¹⁰

Analysis of mail-in voting data from this election casts further doubt on the notion that extending the deadline was necessary to ensure sufficient time for return of mail-in ballots for those who had been relatively late in requesting a mail-in ballot. In particular, the data show that most of the ballots that arrived during the three-day extension period were not from late applicants.

The extension of the ballot receipt deadline yielded a material net vote gain for Joe Biden resulting from the extension–possibly as many as 9,000 votes–and alongside the Green Party’s expulsion and other factors contributed to a broader lack of trust in the election outcome. Let us hope that in 2024, experiences such as this, where lawfare is used as a means of acquiring an unfair advantage, are not repeated.

Footnotes
1. In the 2020 Presidential Election, 57.7 percent of votes cast for Biden, but only 17.6 percent of votes cast for Trump, were by mail-in or absentee ballot, according to the state’s official election results. ↩︎
2. The case history discussion is drawn from the Opinion recorded by Justice Baer for the Pennsylvania Supreme Court’s September 17, 2020, decision on this case. ↩︎
3. Roarty, Alex, and David Catanese. August 27, 2020. “Despite uncertainty, Democrats bet big on mail voting in Pennsylvania.” McClatchyDC.com. Link. ↩︎
4. Bryant, Christa Case. September 4, 2020. “Why Pennsylvania is ground zero for mail-in voting debate.” The Christian Science Monitor. Link ↩︎
5. The link was provided in a RTKL response letter dated March 25, 2021; however, the dataset itself might already have been publicly available. The version currently accessible through this link may incorporate some updates relative to the dataset used here. ↩︎
6. Broken out by party affiliation, the shares received during November 4-6 were 1.9, 3.5, and 2.7 percent for Democrats, Independents or other, and Republicans, respectively. ↩︎
7. In fact, it is almost certain that some ballots recorded as returned on the 4^th actually were returned on the 3^rd. For five outlier counties, the data show zero or near zero ballots returned on the 3^rd but a far larger number on the 4^th, which is not credible. These five outlier counties and their ballot counts for November 3^rd compared with the 4^th are: Clinton (6 versus 169); Luzerne (5 vs. 2,076); Sullivan (1 vs. 33); Lebanon (0 vs. 729); and Mifflin (0 vs. 513). In the rest of the state, more than three times as many ballots had a return date of the 3^rd relative to the 4^th. ↩︎
8. Philadelphia had about 2⅓ times as many registered voters as Bucks County. Montgomery had less than a third the number of ballots with November 4-6 return dates relative to Bucks, despite having about 40 percent more registered voters. Also note that the shares for Luzerne and Lebanon shown in the chart are inflated for the reason described in the previous footnote – zero or near-zero ballots with November 3 reported return date. ↩︎
9. See pages 21-22 the Pennsylvania Supreme Court’s September 17, 2020, decision. ↩︎
10. For a wider ranging discussion of lawfare cases initiated by Democratic Party activists during the 2020 election around the nation, see chapter two of Fund, John, and Hans von Spakovsky. 2021. Our Broken Elections: How the Left Changed the Way You Vote. New York: Encounter Books. ↩︎

Unrivaled Politics

Facts and Data First. Politics Second.

Tag: Election Lawfare

A Statistical Addendum to the Ballot Date Rule Discussion

Hidden in plain sight: an official count much less than 10,000

How many lost ballots in the 2024 General Election?

Not so fast, please!

Table 1: Unrecorded ballots by reason for current status

Table 2: Ballot count distribution for the subpopulation of interest

Table 3: Final outcomes—voters with a single cancelled or pending mail ballot

Table 4: Final outcomes—voters with two or more observed mail ballots

Concluding remarks

The Great Ballot Date Debate

Case Ongoing after Pennsylvania Federal Judge Rules in Favor of Counting Undated Ballots

The Federal Case

The State Case

The Unrivaled Politics Perspective

Bonus Ballots for Biden: A Case Study of Election Lawfare Excess

The Democrats’ legal challenge to the ballot return deadline

Review and critique of the petitioners’ arguments

Review and critique of Boockvar’s arguments

The Court decides

Quantifying the effect on the vote tally

Does the extension seem justified in retrospect?

Not all ballots with recorded return dates in November 4-6 are votes that would have been lost

Enabling underperformance?

Concluding remarks