What Do You Do With Terrible Precedents Shielding Lying Prosecutors?

Bacall v. Stoddard is about a prosecutor who lied.

Bacall was accused of first-degree murder. He claimed self-defense. The prosecutor told the jury that Bacall never once raised the issue of self-defense before trial -- that it was an opportunistic argument he only now was trying to swing. This was the lie. Bacall had been emphatic in claiming self-defense since being booked for the crime, and the prosecutor was well aware of this. But following that lie, the jury (which made it clear it was agonizing over the case in deliberations), voted to convict.

The Sixth Circuit rejected Bacall's Habeas petition. They were clearly disturbed by the conduct. There was no question in their mind that the prosecutor lied, and did so intentionally (quoth the court regarding the prosecutor's statements to the jury: "This was false, and the prosecutor knew it."). The case was not one where the evidence against the defendant was overwhelming; the prosecutor's lie very well could have tipped the margin. The issue was preserved at trial (via an objection made -- on instructions of the trial court -- out of earshot of the jury).

The problem was that Supreme Court precedents and the AEDPA have made prevailing on a Habeas petition almost ludicrously difficult to manage. Even in a case like this, where there was a manifest abuse by the prosecution, the question was whether the Michigan state court's decision not to overturn the jury verdict was "was so lacking in justification that [it committed] an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement." Yeah, that's a tough standard to meet.

The panel clearly believed this case well-illustrated how the current law and precedents have gone badly off the rails. It seemed to me that they weren't saying that the only reasonable way of reading the law was to require that outcome, however. Rather, their analysis suggests that they believed this outcome was the most reasonable interpretation of the governing law.

For me, this raises an interesting hypothetical. Suppose you're the appellate judge hearing this case, and you think the following things are true:

(a) the prosecutor here committed a gross miscarriage of justice, such that, in a just and functioning legal system, Bacall's conviction should clearly be reversed; 

(b) the most accurate read of the governing statute and precedents -- entirely bloodless and indifferent to the consequences or questions of justice -- would suggest that his Habeas petition must fail; 

(c) notwithstanding the above, there is a plausible and reasonable (though not the best) interpretation of the statute and precedents which would justify granting the Habeas petition; and

(d) you suspect that, if your panel does successfully grant the petition, that ruling will not be disturbed by any further appeals (the case won't go en banc or to the Supreme Court).

What do you do?