March 22, 2008

The "Butler list"

Here is the deal. In a press release dated 3/06/08, WIsconsin Supreme Court Justice Louis Butler's campaign communications director Erin Celello wrote:
Justice Butler has voted to uphold criminal convictions in 97 percent of cases brought before the court. Even if you only select those criminal cases in which the Court has issued a written opinion, Justice Butler has ruled in favor of a criminal appeal only about one-third of the time, a rate not much different than any other member of the court.
Emphases added. Notice: "upholding criminal convictions" vs. "ruling in favor of a criminal appeal." Those are obviously quite different issues. Everybody knows that, right?

Recall, Celello's statement was in rebuttal to the Coalition for America's Families claim that Butler "has sided with criminals nearly 60% of the time," whatever that means. ("Right-wing guy" investigative reporting into that figure has been hard to come by.)

The "Butler list" of 70 cases is all about convictions. That's why it contains numbers in parentheses, thus: (3:0). In such a case, there were three separate convictions on three separate charges. But "ruling in favor of a criminal appeal" does not necessarily mean that the conviction was reversed. In several of the 70 cases, the defendant won an appeal, but the conviction remained in place.

Now why somebody would expect the Butler campaign to score a conviction as reversed where there was no conviction reversed is anyone's guess. Maybe some "right-wing guy" has some insight into that as well. Then again, it's easier to just call a sitting justice of the Wisconsin Supreme Court a liar, I suppose. After all, that's what any self-respecting journalist would do.

So convictions are what this list is all about. If a defendant won an appeal, and on remand the circuit court reversed the conviction, that reversal cannot be laid at the feet of Justice Butler. For one thing, the Supreme Court is often addressing questions of constitutional infirmity that are only tangentially related to the conviction itself. And, for another thing, in many of these cases Butler sides with the most conservative members of the court.

Therefore, in those numerous instances, if Wisconsin Manufacturers & Commerce wants another David Prosser, they've already got one in Louis Butler.

As several other observers have pointed out, these "percentages" are a bit silly, because there are so many other factors involved. But the bottom line here is that the Butler campaign was responding — however tactically wise or not — with percentages of its own.

As the list stands, the percentage of convictions where Butler (whatever other members of the court he aligned himself with) did something other than reverse that conviction, stands at 70.89%. In every single specific case but one that Jessica McBride objects to on these grounds, the Supreme Court did not reverse a conviction.

Pursuant to that one case, State v. Raye, McBride does appear to have located a typo. Woooo. So if we move that conviction to the reversed column, the percentage changes to 69.23%.

As for her other objections, I'm not going to address any more of them at this point. If she wants to quibble over what some of the notations beside the docket numbers mean, then, fine. Or if she wants to complain about the accuracy of the wording on either side of the backslash in the header, then, fine.

But elsewhere she vacillates between demanding only cases with the -CR suffix and insisting that cases without the -CR suffix be added to the list, and I simply have no interest in untangling such incoherence.

69.23% is what the list shows. Of all the cases on that list, once the Supreme Court had finished its decision making process, 69.23% of the convictions remained intact. Then they went to other courts.

That's what the list means.


AutismNewsBeat said...

Your blog is now the number one hit when you Google "69" and "Jessica McBride".

William Tyroler said...

Excellent catch (deconstructing the Celello release, not googling 69 and JMc). My refrigerator doesn't hold enough Guinness Stout to have gotten me through the effort required to drill down that far into the data. Hat's off to iT.

The onus is now on McBride and McIlheran to apologize and withdraw their overheated claim or show just how iT's wrong. (And no, the one apparent lowly mistake -- Raye -- won't cut it; that's what we like to call an outlier.) I'm sorry to say that Professor Esenberg, for whom I have great respect, must count himself among that number and defend or retract his claim that Butler's campaign cooked data. As we also like to say, lie down with dogs, get up with fleas.

illusory tenant said...

Thanks once again, Counsellor. I, too, was really hoping that Prof. Esenberg wouldn't jump to their defense. I even warned him.

Rick Esenberg said...

This makes no sense. The list of 70 cases that McBride put up and which I have no reason to believe did not come from the Butler campaign breaks down cases in which the campaign claims that he ruled in favor of the state versus those that he ruled in favor of the defendants. Some of them have nothing to do with upholding (or vacating) a criminal conviction. Some are not even, strictly speaking, criminal cases although I don't criticize them for that.

The list is supposed to show that he ruled in favor of the state and against the "defendant/inmate" 70% of the time. That's just not true.

There is a new spin on these numbers everytime I visit here. Maybe the easiest thing to do is just admit they aren't right.

illusory tenant said...

There is a new spin on these numbers everytime I visit here.

"Spin" is a Wittgensteinian imperative, and therefore something that you yourself engage in on a fairly regular basis, so I can't take that remark too seriously.

The only thing new is that I reverse-engineered the 70% figure to determine that the figures count convictions. Where a conviction wasn't reversed, you expect Butler to count it as reversing a conviction?

I didn't notice either McBride or McIlheran attempting to determine what the numbers actually represent before leaping madly to their preordained conclusions, did you?

Maybe it was about time somebody did.

It's not the greatest document in the world, and I certainly wouldn't have prepared it or released it as it is, but the figures are based on convictions, which is as good an indicator as any.

It's certainly a hell of a lot more illuminating than accusing Butler of lying wouldn't you say?

William Tyroler said...

I don't want this to be a game of gotcha. But the fact is, the higher the expectation you have of someone, the greater the disappointment when they fall short. I expect a lot from Professor Esenberg, whose insights are typically sharp and challenging. But in this instance he has not fulfilled his considerable potential.

Professor Esenberg initially hedged his bets, though he clearly thought that McBride's was a worthy effort. After a bit of back-and-forth, he eventually revealed his hand: "It seems clear that the Butler campaign has cooked their numbers." It is on that claim of intentional manipulation of the data that Professor Esenberg comes to grief. Not only did iT's exemplary data-drilling show that McBride is the one who got it wrong, it was never clear that the Butler campaign was massaging the data. This was a mess to begin with: raw data, unclear methodology, no control of variables, no known definitions. A number of individuals (not least iT) attempted to warn the professor away from McBride's pitchfork brigade, but to no avail.

All that said, Professor Esenberg's most recent post is, IMHO, exemplary in its analysis (putting aside a revisit of that nettlesome numbers thing). I do wish that had been his initial post, and the major thrust of subsequent ones, on the subject. Maybe next time ...