Harvard Law Review Harvard Law Review Harvard Law Review

Supreme Court

The Statistics

Table I(A) -- Actions of Individual Justices

a A complete explanation of how the tables are compiled may be found in The Supreme Court, 2004 Term — The Statistics, 119 Harv. L. Rev. 415, 415–19 (2005).

Table I, with the exception of the dissenting-votes portion of section (A) and the memorandum tabulations in section (C), includes only full-opinion decisions. Eight per curiam decisions contained legal reasoning substantial enough to be considered full-opinion decisions during October Term 2016. These cases were Pavan v. Smith, 137 S. Ct. 2075 (2017); Hernandez v. Mesa, 137 S. Ct. 2003 (2017); Jenkins v. Hutton, 137 S. Ct. 1769 (2017); Virginia v. LeBlanc, 137 S. Ct. 1726 (2017); North Carolina v. Covington, 137 S. Ct. 1624 (2017); Rippo v. Baker, 137 S. Ct. 905 (2017); White v. Pauly, 137 S. Ct. 548 (2017); and Bosse v. Oklahoma, 137 S. Ct. 1 (2016).

This table includes every opinion designated by the Court as a 2016 Term Opinion except for one. See Opinions of the Court — 2016, Supreme Court of the United States, https://www.supremecourt.gov/opinions/slipopinion/16 (last visited Sept. 26, 2017) [https://perma.cc/8ULE-PGZ9]. The omitted opinion is Trump v. International Refugee Assistance Project, 137 S. Ct. 2080 (2017), which granted certiorari and granted the government’s stay applications in part but did not dispose of the case on the merits.

A memorandum order is a case decided by summary order and contained in the Court’s weekly order lists issued throughout the Term. This category excludes summary orders designated as opinions by the Court. The memorandum tabulations include memorandum orders disposing of cases on their merits by affirming, reversing, vacating, or remanding. They exclude orders disposing of petitions for certiorari, dismissing writs of certiorari as improvidently granted, dismissing appeals for lack of jurisdiction, disposing of miscellaneous applications, and certifying questions for review. The memorandum tabulations also exclude orders relating to payment of docketing fees and dissents therefrom.

b This portion of Table I(A) includes only opinions authored in the sixty-nine cases with full opinions this Term. Thus, dissents from denials of certiorari and concurrences or dissents from summary affirmances are not included. A concurrence or dissent is recorded as a written opinion whenever its author provided a reason, however brief, for his or her vote.

c A Justice is considered to have dissented whenever he or she voted to dispose of the case in any manner different from the manner specified by the majority of the Court.

d A plurality opinion that announced the judgment of the Court is counted as the opinion of the Court. Thus, for example, Justice Alito’s opinion in Matal v. Tam, 137 S. Ct. 1744 (2017), is considered the opinion of the Court in that case, even though Parts III-B, III-C, and IV of the opinion received only four votes.

e Opinions concurring in part, concurring in the judgment, or concurring in both are counted as concurrences. Opinions concurring in part and dissenting in part are counted as dissents.

f Dissenting votes in memorandum decisions include instances in which Justices expressed that they would not have disposed of the case by memorandum order. Therefore, Justices Thomas and Gorsuch are treated as having dissented from the Court’s affirmance in Republican Party of Louisiana v. FEC, 137 S. Ct. 2178 (2017) (mem.) because they would have noted probable jurisdiction and set the case for oral argument. This category does not include dissenting votes in orders relating to stays of execution; that information is presented in Table II(F) and its accompanying footnotes.

Table I(B1) -- Voting Alignments -- All Written Opinions

g Table I(B1) records the frequency with which each Justice voted with each of the other Justices in full-opinion decisions, including the eight per curiam decisions containing sufficient legal reasoning to be considered full opinions. See supra note a.

Two Justices are considered to have agreed whenever they joined the same opinion, as indicated by either the Reporter of Decisions or the explicit statement of a Justice in his or her own opinion. This table does not treat a Justice as having joined the opinion of the Court unless that Justice authored or joined the opinion of the Court in full, or authored or joined at least part of the opinion of the Court and did not author or join any opinion concurring in the judgment, even in part, or dissenting, even in part. Therefore, Justice Kennedy is not treated as having joined the opinion of the Court in Matal v. Tam, 137 S. Ct. 1744 (2017), because he authored an opinion concurring in part and concurring in the judgment. By contrast, Justices Thomas and Gorsuch are treated as having fully joined Chief Justice Roberts’s opinion in Trinity Lutheran Church of Columbia, Inc. v. Comer, 137 S. Ct. 2012 (2017), even though they did not join footnote 3. For the purpose of counting dissents and concurrences, however, a Justice who partially joined an opinion is considered to have fully joined it.

In Tables I(B1) and I(B2), “O” represents the number of decisions in which a particular pair of Justices agreed in an opinion of the Court or an opinion announcing the judgment of the Court. “S” represents the number of decisions in which two Justices agreed in any opinion other than an opinion of the Court or an opinion announcing the judgment of the Court. Justices who together joined more than one separate opinion in a case are considered to have agreed only once. “D” represents the number of decisions in which two Justices agreed in a majority, plurality, concurring, or dissenting opinion. A decision is counted only once in the “D” category if two Justices both joined the opinion of the Court and joined a separate concurrence. Thus, in some situations the “D” value will be less than the sum of the “O” and “S” values. “N” represents the number of decisions in which both Justices participated, and thus the number of opportunities for agreement. “P” represents the percentage of decisions in which one Justice agreed with another Justice and is calculated by dividing the “D” value by the “N” value, expressed as a percentage.

Table I(B2) -- Voting Alignments -- Nonunanimous Cases

h Like Table I(B1), Table I(B2) records the frequency with which each of the Justices voted with each other Justice in full opinions, but Table I(B2) records these voting alignments only for cases that were not unanimously decided. A decision is considered unanimous for purposes of Table I whenever all the Justices joined the opinion of the Court and no Justice concurred only in the judgment, even in part, or dissented, even in part. Removing the unanimous cases produces lower rates of agreement overall, providing a more accurate picture of how the Justices voted in divisive cases.

Table I(C) -- Unanimity

i A decision is listed in this column if at least one Justice concurred in the judgment, but not in the Court’s opinion in full, and no Justice dissented, even in part. See, e.g., Life Techs. Corp. v. Promega Corp., 137 S. Ct. 734 (2017).

Table I(D) -- Voting Patterns in Nonunanimous Cases

j Table I(D) records the frequency with which each Justice joined the opinion of the Court in nonunanimous, full-opinion decisions. This table includes the eight per curiam decisions containing sufficient legal reasoning to be considered full opinions, see supra note a, if those decisions produced dissenting votes.

k This portion of the table reports the number of times that each Justice joined the opinion of the Court, according to the rule described in note g.

l This portion of the table reports the number of times that each Justice agreed with the Court’s disposition of a case. It includes all cases in which a Justice joined the opinion of the Court, but unlike the portion of the table described in note k, also includes those cases in which a Justice concurred in the judgment without concurring in the Court’s opinion in full. Cases in which the Justice dissented, even in part, are not included.

Table I(E) -- 5–4 Decisions

m This column lists the number of 5–4 full-opinion decisions in which each five-Justice group constituted the majority. A case is counted as 5–4 if four Justices voted to dispose of any issue, no matter how minor, differently than the majority of the Court. See, e.g., McWilliams v. Dunn, 137 S. Ct. 1790, 1801–11 (2017) (Alito, J., dissenting). Cases in which any Justice did not participate are not included. See, e.g., Murr v. Wisconsin, 137 S. Ct. 1933 (2017).

n McWilliams v. Dunn, 137 S. Ct. 1790 (2017) (Breyer, J.).

o Davila v. Davis, 137 S. Ct. 2058 (2017) (Thomas, J.); Cal. Pub. Emps.’ Ret. Sys. v. ANZ Sec., Inc., 137 S. Ct. 2042 (2017) (Kennedy, J.).

Table I(F) -- Average Opinion Length

p The data in this table reflect the length of opinions as published in the Court’s slip opinions, estimated to the nearest tenth of a page. Though the slip opinions are eventually superseded by official case publication in the United States Reports, the total opinion length, in pages, is generally preserved in the final publication. Average opinion length is obtained by summing the number of pages written by each Justice within each category of opinion and then dividing by the number of opinions of that type written by that Justice. For the number of opinions written by each Justice, see supra Table I(A).

q Opinions concurring in part, concurring in the judgment, or concurring in both are categorized here under Concurring in Judgment. Opinions concurring in part and dissenting in part, or concurring in the judgment in part and dissenting in part, are categorized here under Dissenting Opinion.


Table II(A) -- Final Disposition of Cases

a All numbers in Tables II(A), II(B), and II(C) are derived from data provided by the Supreme Court.

b The appellate docket consists of all paid cases.

c The number of cases remaining on the appellate and miscellaneous dockets is calculated by adding the number of cases not acted upon in the 2016 Term to the number of cases granted review in the 2016 Term but carried over to the 2017 Term.

d The miscellaneous docket consists of all cases filed in forma pauperis.

Table II(B) -- Cases Granted Review

e Table II(B) reports data that versions of Table II prior to 1998 reported under the label “Review Granted.” For a full explanation, see The Supreme Court, 1997 Term — The Statistics, 112 Harv. L. Rev. 366, 372 n.d (1998). Table II(B) does not include cases within the Court’s original jurisdiction.

f The number of cases granted review includes only those cases granted plenary review in the 2016 Term. It includes neither cases summarily decided nor those granted review in a previous Term and carried over to the 2016 Term. It does include cases granted review in the 2016 Term but carried over to a subsequent Term.

g The number of petitions considered is calculated by adding the number of cases docketed in the 2016 Term to the number of cases carried over from prior Terms and subtracting the number of cases not acted upon in the 2016 Term.

Table II(C) -- Method of Disposition

h Table II(C) does not include cases within the Court’s original jurisdiction.

Table II(D) -- Disposition of Cases Reviewed on Writ of Certiorari

i Table II(D) reports the disposition of cases reviewed on writ of certiorari and decided on the merits. It does not include cases reviewed under other bases of jurisdiction, such as Cooper v. Harris, 137 S. Ct. 1455 (2017) (reviewed under 28 U.S.C. § 1253 (2012)).

j This category includes cases reversed in part and affirmed in part, as well as cases reversed in part and vacated in part.

k This category includes cases vacated in part and affirmed in part.

l In Jaffe v. Roberts, 137 S. Ct. 2191 (2017) (mem.), eight Justices disqualified themselves. In Arunga v. Obama, 137 S. Ct. 2194 (2017) (mem.), four Justices disqualified themselves. Because a quorum was therefore unobtainable in both cases, the Court affirmed the judgments below as if by an equally divided court. These opinions are not included in Table I(C).

Table II(E) -- Origins of Cases and Their Dispositions

m Table II(E) counts consolidated cases disposed of by the same lower court opinion as a single case. Thus, for example, Ziglar v. Abbasi, 137 S. Ct. 1843 (2017), is counted only once. Table II(E) does not include original jurisdiction cases.

n This section reports only full opinions decided on the merits. It thus includes eight per curiam decisions containing sufficient legal reasoning to be counted as full opinions. See supra Table I, note a.

o This category includes cases reversed in part and affirmed in part, as well as cases reversed in part and vacated in part.

p This category includes cases vacated in part and affirmed in part.

q Consolidated cases originating from multiple circuits are counted in each circuit, but once each toward the sum. Thus, Advocate Health Care Network v. Stapleton, 137 S. Ct. 1652 (2017), is counted in the Third, Seventh, and Ninth Circuits, but as a single Supreme Court case originating from the circuit courts of appeal.

r This category includes statutorily authorized direct appeals from district courts.

Table II(F) -- Dispositions of Applications for Stays of Execution

s This table treats multiple applications from the same death row inmate as a single application. Although the Court entertained thirty-five applications for stays of execution last Term, these applications pertained to only seventeen individuals.

This table includes only those dispositions that appear in the Supreme Court Reporter and excludes applications to vacate stays of execution.

For useful background information on how the Court handles stays of execution, see generally Eugene Gressman et al., Supreme Court Practice §§ 18.1–.8, at 897–911 (9th ed. 2007); A Reporter’s Guide to Applications Pending Before the Supreme Court of the United States (2017), https://www.supremecourt.gov/publicinfo/reportersguide.pdf [https://perma.cc/NS43-67UL]; and The Supreme Court, 2006 Term — The Statistics, 121 Harv. L. Rev. 436, 446 n.t (2007).

t This Term, the Court granted one stay application pending its decision whether to grant certiorari in the underlying case. See Arthur v. Dunn, 137 S. Ct. 14 (2017) (mem.). The stay was to terminate automatically upon the Court’s denial of the associated petition, or if certiorari was granted, upon issuance of the judgment of the Court. Id. at 14–15. Certiorari was denied. See Arthur v. Dunn, 137 S. Ct. 1521 (2017) (mem.).

u Twenty-five denials, counting multiple applications from the same inmate, were unanimous.

In McGehee v. Hutchinson, 137 S. Ct. 1275 (2017) (mem.), the Court denied the inmate’s application for a stay of execution, in addition to denying certiorari. Id. at 1276. Justices Ginsburg and Sotomayor would have granted both the stay of execution and certiorari. Id. Justices Breyer and Kagan would have granted the stay of execution. Id.

Eight further denials were nonunanimous. Justices Ginsburg, Breyer, Sotomayor, and Kagan would have granted the application in Smith v. Alabama, 137 S. Ct. 588 (2017) (mem.). Justices Breyer, Sotomayor and Kagan would have granted the application in Johnson v. Kelley, 137 S. Ct. 1622 (2017) (mem.). Justice Ginsburg would have granted the application in Christeson v. Griffith, 137 S. Ct. 910 (2017) (mem.). Justice Breyer dissented in Ruiz v. Texas, 137 S. Ct. 1246 (2017) (mem.), and McGehee v. Hutchinson, 137 S. Ct. 1622 (2017) (mem.). Justice Sotomayor dissented in Jones v. Kelley, 137 S. Ct. 1284 (2017) (mem.), Williams v. Kelley, 137 S. Ct. 1284 (2017) (mem.), and Arthur v. Dunn, 137 S. Ct. 1521 (2017) (mem.).

Chief Justice Roberts and Justices Kennedy, Thomas, Alito, and Gorsuch did not dissent from any denial of an application for a stay of execution.


Table III -- Subject Matter of Dispositions with Full Opinions

a Table III records the subject matter of dispositions by full opinion, including the eight cases with per curiam opinions on the merits containing sufficient legal reasoning to be considered full opinions. See supra Table I, note a.

b Each case is categorized as primarily constitutional or not. Cases invoking a mixture of statutory interpretation and constitutional adjudication are particularly difficult to classify.

c “Government” refers to federal, state, or local government, or an agency thereof, or to an individual participating in the suit in an official capacity. A decision is counted as “for” the government if the government prevailed on all contested issues. When the federal government opposed a state or local government, a decision is counted as “for” the government if the federal government prevailed on all contested issues. When two states, two units of local government, or two federal agencies opposed each other, the decision is counted as neither “for” the government nor “against” the government. When the government prevailed on at least one but not all of the issues before the Court, a decision is counted as neither “for” nor “against” the government.

dTurner v. United States, 137 S. Ct. 1885 (2017), which reached the Court on writ of certiorari to the District of Columbia Court of Appeals, is included in this category.