In an article in Tuesday’s New York Times, Adam Liptak reported that opponents of same-sex marriage are challenging the composition of Ninth Circuit panels that have heard cases involving gay rights. The opponents assert that, against what they say are 441-1 odds, certain liberal judges have repeatedly been assigned to panels in such cases, while eighteen other judges of the Ninth Circuit have never sat on a panel in a gay rights appeal. The article goes on to report that there are serious questions about the validity of the opponents’ challenge, and cites some scholarly articles about how Circuit Courts of Appeal assign judges to panels. But the Times article provoked this question: what are the odds of getting any particular Appellate Division judge on an appeal?
The Appellate Division normally consists of eight Parts of four judges. The Parts are created anew by the Chief Justice, with input from the Appellate Division’s Presiding Judge for Administration, in the General Assignment Order at the beginning of every court term. An effort is made to equalize the number of Democrats and Republicans on each Part, and to ensure that each Part has one of the most senior judges, who become the Presiding Judges of the each Part, and that the remaining judges on each Part include some more experienced judges and some judges who are newer to the Appellate Division. Thus, the assignment of judges to Part is partially random, but far from entirely so.
When cases are ready for assignment to a Part, the Calendaring Unit of the Appellate Division allocates cases to Parts in a largely random fashion. However, the Calendaring Unit does consider conflicts that a judge or judges on a Part may have, and in cases where oral argument has been requested, some effort is made to calendar the case before a Part that sits where most or all of the attorneys are based. In those limited respects, therefore, assignments of cases to Parts are not entirely random.
In most but not all weeks, only three members of each Part are assigned to appeals scheduled for oral argument or disposition without argument that week. The identity of the judge who is “off” is determined by a rotating schedule that is promulgated at the beginning of the term. Thus, from the outset, for any set of appeals before a particular Part in a given week where one judge is “off,” there is a one in four (25%) chance that a particular member of that Part will take no part in that appeal.
Appellate Division appeals are heard by two or three judges of the assigned Part. The Presiding Judge of the Part decides which cases will be heard by two judges and which by three judges.
Based on all of this, here are some rough calculations of the odds that Judge 1, a member of the Appellate Division, will sit on a specific appeal. The calculations are rough because (perhaps among other things), to simplify the issues, I will assume that one judge on a Part is “off” every week instead of just most weeks. Also, we cannot know the likelihood that any particular case will be considered by a two-judge panel as opposed to a three-judge panel.
To begin with, there is a one in eight (12.5%) chance that a case will be calendared before Judge 1’s Part. If no judges on a Part were “off” in any week, there would be four possible combinations of three-judge panels (Judges 1, 2, and 3; Judges 1, 2, and 4; Judges 1, 3, and 4; and Judges 2, 3, and 4) and six possible combinations of two-judge panels (Judges 1 and 2; Judges 1 and 3; Judges 1 and 4; Judges 2 and 3; Judges 2 and 4; and Judges 3 and 4). As a result, the chances would be six in ten (60%) that Judge 1 would wind up considering a particular appeal. However, since one of the four judges will be “off” that week, the viable panel combinations are reduced to four (one three-judge panel possibility, consisting of all three judges who are “on,” and three two-judge panel possibilities). Of those four possible panels, Judge 1 has a 75% chance of being included on a panel in any week in which he or she is “on” (in such a week, Judge 1 is on two out of the three potential two-judge panels and on the only potential three-judge panel), and Judge 1 is “on” 75% of the time. Multiplying the 75% chance that Judge 1 is “on” by the 75% chance that Judge 1 is included in a panel when he or she is “on” yields a 56.25% chance that Judge 1 will be considering a particular appeal before a particular Part. To reach a final result, we multiply that 56.25% chance by the 12.5% chance that the appeal would come to Judge 1’s Part in the first place. That yields a final likelihood of 7.03%, when an appeal is filed, that Judge 1 will sit on that appeal. Whew!
In short, the assignment of Appellate Division judges is almost entirely random, and the odds that a given judge will hear any particular appeal are small. That does not mean, of course, that an attorney or party may not encounter the same judge repeatedly. Simple statistics says that there is always some chance of that.