With this post, we begin our short review of some of the academic literature regarding panel effects.

Of course, the first question one encounters when taking a case before an appellate court is how one’s panel will be chosen.  A majority of appellate courts state, either in their operating procedures or their rules, that appellate panels for argued cases are assigned randomly.  This assumption has been built into most of the analytics work done on panel effects for at least a generation.

But is it really true?  I have spoken with lawyers in several federal circuits who are dubious about just how random assignments are.

Let’s begin by making several things thoroughly clear: no one is suggesting that panels are intentionally chosen to manipulate results in particular cases (although disappointed litigants have done so once in awhile).  Further, reasonable people could dispute just how beneficial random assignments are.  There are any number of good reasons to deviate from strictly random assignments in the federal circuits: maximizing the availability of senior status judges; respecting the vacation plans or speaking or writing commitments of judges; ensuring that a particular judge doesn’t draw several panel assignments in a short period, or go lengthy periods without an assignment; ensuring that particular judges sit with a variety of their colleagues, rather than sitting repeatedly with one or two other judges; accounting for recusals; or sending a case which was previously ruled upon by a particular panel back to the same judges following remand.  The list goes on.

But even if completely random assignments aren’t necessarily a reasonable goal, the question remains: how are appellate panels chosen?

In 2015, Professors Marin K. Levy and Adam S. Chilton published Challenging the Randomness of Panel Assignment in the Federal Courts of Appeals, 101 Cornell L. Rev. 1 (2015).  The authors gathered panel information for all twelve regional circuits between September 2008 and August 2013.  Collectively, the dataset covered the activities of 775 judges and over 10,000 panels.  The professors then wrote a program to simulate the choice of over one billion entirely random panels.  They then decided to compare their dataset of randomly simulated panels to the “real world” data by counting the incidence of an objective characteristic in both datasets – how many panels had appointees of Republican Presidents on them.  Accordingly, they developed detailed data on how common panels of zero, one, two and three Republican nominees were, and then calculated whether the actual docket results fell reasonably close to that random distribution.  The professors reported that their statistical tests showed evidence that panel assignments deviated from a strictly random result in four circuits: the D.C. Circuit, the Second Circuit, the Eighth Circuit and the Ninth Circuit.  They then tested their results for robustness and calculated that the probability of all their results being solely due to chance was less than 3%.  The data from the remaining eight Circuits fell reasonably close to the fully random distribution – although given the reasons discussed above why complete randomness may not be realistic or beneficial, there is reason to wonder just how robust that result is.  Two years later, Professor Levy published a follow-up article, Panel Assignment in the Federal Courts of Appeals, 103 Cornell L. Rev. 65 (2017).  There, she discussed her interviews about panel assignment practices with thirty-five judges and senior administrators.  She reported that no two courts approached panel assignment in the same way and argued that it was far from clear that the benefits of random assignments outweighed the drawbacks.

Next time we’ll continue our discussion of the literature on panel effects over at the California Supreme Court Review.

Image courtesy of Pixabay by Piro4D (no changes).

This week, we’re taking a short break from our usual number-heavy analysis for another glance at some of the vast academic literature on the analytic-driven analysis of appellate decision making.  This is a four-part post – two here and two over at the California Supreme Court Review – sampling some of the literature on “panel effects.”

Of course, virtually all appellate decision making takes place in panels of judges – sometimes three, occasionally five, frequently seven or at the U.S. Supreme Court (or a Circuit en banc), nine or more.  It’s easy to fall into the trap in assessing an appellate panel of treating it like a political focus group.  For example, say I’m representing a defendant, and I learn that my panel consists of two Republican nominees and one Democrat, I might think that’s a reasonably favorable panel.  But the implicit assumption built into that statement is that none of the judges’ votes will be impacted by any of the other judges.

Although most political votes are cast by voters who don’t know each other or care about others’ opinions, that’s not always true.  Studies have shown that if you put a Republican, for example, in a room full of Republicans, his or her views will drift further right than they would otherwise have been.  Put the Republican in a room surrounded by Democrats, and the opposite effect is observed – the Republican drifts more moderate.  Repeat the experiment with a Democratic voter, and the effect flips – more conservative in a room full of Republicans, more liberal with other Democrats.

Appellate panels, of course, are quite different.  Often (with the exception of the Ninth Circuit), the judges know each other well, and may have been working together for years.  They presumably have a shared commitment to something they think of as the “law of the Circuit/state/court.”  Many believe that unanimity has an intrinsic value in reinforcing the moral authority of their court.  Their vote isn’t secret – they’ll have to look a colleague in the eye, at least figuratively speaking, and tell him or her why he or she is wrong.  And if they don’t succeed in convincing their colleague, they’ll have to write an opinion explaining how thoroughly wrong their colleague is – knowing that it’s going to be preserved in thick hardbound books in the library on the computer screens of any lawyer in the country who wants to see it for pretty much all eternity.

One can imagine that might temper your enthusiasm for constantly dissenting in a hurry.

So that’s where the literature of “panel effects” comes in.  How much difference does the composition of your panel really make?  And if you’re trying to predict the vote of Judge A, do you only want to know his or her philosophy – or are the leanings of the other judges on the panel a powerful predictor of Judge A’s vote?  We’ll begin our review next time.

Image courtesy of Flickr by A Syn (no changes).

This time, we’re concluding our three-week trip through the data for originating jurisdictions, looking at the Court’s criminal cases between 2010 and 2020.

For these years, Cook County accounted for 156 criminal cases.  Will County was next, followed by Peoria, Du Page, Kane and Lake counties.

Several small counties made our second Table, producing between two and four criminal cases apiece.  Interestingly, Madison and Sangamon counties, both of which tend to account for quite a few civil cases, produced only two criminal cases each over the past eleven years.  An additional twenty jurisdictions produced one case apiece.

Join us back here next week as we continue our ongoing analysis of the Supreme Court’s decision making.

Image courtesy of Flickr by David Wilson (no changes).

This week, we’re wrapping up our three-week series on the geographical origins of the Court’s docket by reviewing the data for the years 2010 through 2020.

Between 2010 and 2020, the Court decided 150 civil cases which began in Cook County (once again, Cook is removed from the Table for readability).  There was a three-way tie for second between Du Page, Lake and Sangamon County.  Once again, Madison County was high on the list, as were Kane and Champaign County.  Several administrative bodies were high on the list as well, including the Department of Revenue, the Workers Compensation Commission and the Commerce Commission.

In the second table, we report the next eighteen jurisdictions on our list, all accounting for either two or three cases each.  An additional 49 jurisdictions produced one case apiece.

Join us back here next time as we review the criminal data for the past eleven years.

Image courtesy of Flickr by Daniel X. O’Neil (no changes).

For the past several posts, we’ve been reviewing the jurisdictions where the Illinois Supreme Court’s civil and criminal cases originated.  This time, we’re looking at the criminal cases for the years 2000 through 2009.

Once again, we omit Cook County from the Table to make it easier to read.  Cook County produced 225 criminal cases from 2000 to 2009.  Second was Du Page County, followed by Kane and Champaign County, then Lake and Peoria County.

Twenty additional jurisdictions, reported in Table 1718 below, accounted for two or more criminal cases.  An additional twenty-three jurisdictions produced one criminal case each.

Join us back here next week as we review the years 2010-2020.

Image courtesy of Flickr by artistmac (no changes).

This week, we’re reporting the data for the Supreme Court’s civil and criminal cases during the years 2000 through 2009.

Once again, Cook County led by a wide margin, accounting for 172 civil cases.  Du Page and Lake counties – the second and third largest counties by population – were next, followed by the Industrial Commission.  St. Clair County from the Fifth District followed, with Sangamon County, Will County and Madison County next.

The remaining sixteen jurisdictions accounting for two or more civil cases each are reported in Table 1716 below.  An additional 64 jurisdictions produced one case apiece.

Join us next time as we review the criminal cases for the same years 2000 through 2009.

Image courtesy of Flickr by Ron Cogswell (no changes).

In this post, we report the originating jurisdictions for the Court’s criminal cases during the years 1990 through 1999.  Once again, we omit Cook County from the chart to make it more readable; Cook County produced 268 criminal cases during the nineties.  Lake County, the third biggest county in terms of population, is second in criminal cases.  DuPage County, the second biggest county in Illinois, was third in criminal cases during the 1990s.  Will County, the fourth biggest county by population, was next, followed by disciplinary cases from the ARDC.

The rest of the jurisdictions producing multiple cases are reported below.  Twenty-four additional jurisdictions contributed one case each to the Court’s criminal docket.

Join us back here next week, when we’ll address the civil and criminal cases from the years 2000 through 2009.

Image courtesy of Flickr by Gary Todd (no changes).

Revised to correct an mistake in the original post resulting from coding errors.

This week, we’re looking at where the Supreme Court’s cases originate.  This is important for the same reason that tracking which Districts and Divisions of the Appellate Court the Court is taking its cases from.  Just as it’s possible for the Court to conclude that a particular part of the Appellate Court is out of step with its views, it’s equally possible for the Court to be concerned about a trial court or agency.  If the Supreme Court accepts cases from the jurisdiction where you are every year, that’s a good sign for the prospects of getting review.  On the other hand, if the Court seldom takes cases from your jurisdiction, that tells you something about the odds of going up.

In Table 1711, we report the first half of the data.  To make the chart easier to follow, we omit Cook County, which produced 253 civil cases from 1990 to 1999.  Cook County has roughly five times the population of any other county in Illinois, so the disparity makes sense.  Du Page County – second in population – is second in total cases.  St. Clair County is third, despite ranking fifth in population, likely because of the very high reversal rate of the Fifth District.  Madison County, eighth in population but also in the Fifth District, is next.  Lake County, the third most populous county, is next, followed by Sangamon County, which is 11th in population, but also the home of state capitol Springfield.

We report the remainder of the data in Table 1712.  In addition to the jurisdictions reported here, fifty-six jurisdictions produced one case each.

Join us back here next time as we address the criminal docket in the 1990s.

Image courtesy of Flickr by Teemu008 (no changes).



Today, we’re reviewing the contents of our database, which includes every case decided by the Illinois Supreme Court since January 17, 1990.  For every case, we’ve captured the following data points:

Official Reporter Citation

  1. E. Reporter Citation

Docket Number

Case Name


Petitioner Governmental Entity (Y/N)

Respondent Governmental Entity (Y/N)

Source of Appellate Jurisdiction

Originating Jurisdiction

Trial Court

Trial Judge

Source of Case (Appellate Court)

Lower Court Dissent (Y/N)

Lower Court Published (Y/N)

Lower Court Disposition

Lower Court Disposition Direction

Petition Granted

Oral Argument


Grant to Oral Argument

Argument to Decision


Issue Area

Disposition Direction

Dissent Direction


Winning Party

Majority Opinion Writer

Special Concurrence(s) Writer(s)

Dissent(s) Writer(s)

Majority Votes

Minority Votes

Length of Majority Opinion

Length of Concurrence(s)

Length of Dissent(s)

Amici Supporting Petitioner

Amici Supporting Respondent

Amici Supporting Neither

Since 2007:

Total Questions to Petitioner

Total Questions to Respondent

For Each of Seven Justices:


Questions to Petitioner

Questions to Respondent

Questions in Rebuttal

Join Concurrence (Y/N)

Join Dissent (Y/N)

Direction of Vote

Recusal (Y/N)

Arguing Counsel for Prevailing Party

Arguing Counsel for Losing Party

Image courtesy of Flickr by Roger W (no changes).

Last time, we began our analysis by addressing the competing theories of judicial behavior.  Formalism, the oldest theory, teaches that judicial decision making can be explained and predicted based upon the facts, the applicable law and precedent and judicial deliberations – and nothing more.  But if formalism explains all of judicial decision making, then many of the factors studied by empirical analysts, such as the judges’ individual ideologies and voting records, the lower courts involved and the nature of the parties to the litigation, should have little ability to forecast voting and outcomes.  But many studies have shown that such factors do have predictive power.

One of the two primary alternative theories was set forth in The Supreme Court and the Attitudinal Model, by Professors Harold J. Spaeth and Jeffrey A. Segal.  Attitudinalism holds that judges vote based upon their individual ideologies set against the facts of a specific case.  For example, a judicial conservative will require substantially more extreme facts before being willing to condemn the conduct of a police investigator than a liberal will.  Conversely, a judicial liberal will approve of government interference in business based upon a lesser showing of need than a conservative will require.

Attitudinalists have proposed two principal methods for proxying the ideologies of judges.  First, a Federal judge is presumed to be of the same party as the President who nominated him or her.  Although simple enough to determine, this model has been criticized as a blunt instrument, not allowing for the possibility that a Democratic president might nominate a judge who is equally if not more conservative than a Republican one.  Professor Segal and Professor Albert D. Cover have proposed “Segal-Cover scores,” which are based upon analysis of newspaper editorials published prior to a Justice’s confirmation.  Segal-Cover scores have proven to be valuable predictors of judicial voting patterns.  Other analysts have attempted to derive ex post ideological measures by tracking judges’ actual votes over a substantial period.

A third theory of judicial behavior is represented most prominently by former Judge Richard Posner of the Seventh Circuit and is known as legal pragmatism or realism.  Legal realism is based upon the idea that the law evolves over time as society moves forward.  Judge Posner has written that the task of the judge “is to decide cases with reasonable dispatch, as best one can, even in what I am calling the interesting cases – the ones in which the conventional materials of judicial decision making just won’t do the trick.”  Legal realism attempts to integrate the other theories into a kind of unified theory.  To a legal realist, a not insubstantial fraction of every appellate court’s caseload can be explained using traditional formalist techniques.  Another portion of the docket can be explained by attitudinalism – more so in appellate courts of last resort than in intermediate appellate courts.  But the rest cannot be entirely explained by either theory, since formalist rules do not dictate a determinate answer to the question, and judicial concerns, such as the limitations on what courts can practically do or the value of stability in the law, constrain judges from following their ideological preferences.

In the decades since C. Herman Pritchett’s work on the Roosevelt Court – and especially in the past thirty years – data analytic researchers have provided considerable evidence to suggest that the attitudinal and realism theories have considerable power to illuminate judicial decision making.

Image courtesy of Flickr by Tom Shockey (no changes).