Michael Connor v. Brittany Greene
Opinion
Opinion
In the United States Court of Appeals for the Seventh Circuit ____________________ No. 23-1595 MICHAEL J. CONNOR, Petitioner-Appellant, v. BRITTANY GREENE, * Respondent-Appellee. ____________________ Appeal from the United States District Court for the Central District of Illinois.
No. 20-CV-1126 — Colin S. Bruce, Judge. ____________________ ARGUED DECEMBER 8, 2023 — DECIDED MARCH 3, 2026 ____________________ Before RIPPLE, ROVNER, and SYKES, Circuit Judges.
SYKES, Circuit Judge. Michael Connor was convicted in 2013 of five counts of predatory criminal sexual assault of a child following a bench trial in Livingston County, Illinois.
The assaults occurred from 2008 to 2011 when the victims— his daughter and stepdaughter—were between five and * ȱȱȱĴ¢ȱ ǰȱȂȱȱ ǰȱȱȱ respondent, replacing Anthony Willis. See FED. R. APP. P. 43(c)(2).
2 No. 23-1595 seven years old. The penalty for this offense is 6 to 60 years in prison, and sentences for multiple convictions must be served consecutively. The penalty increases, however, to a manda- tory term of life in prison when the offender is convicted of assaulting more than one child. The trial judge sentenced Connor to the mandatory life term.
After an unsuccessful direct appeal, Connor sought post- conviction relief in state court. With the assistance of new counsel, he claimed that his trial attorneyȂȱȱȱȱ plea-bargaining stage was constitutionally ineffective. More specifically, he alleged that his counsel had failed to inform him that he faced a mandatory life sentence if convicted of assaulting both victims. That omission, he claimed, led him to reject an offer from the prosecutor of an 18-year sentence if he pleaded guilty.
At a hearing on the petition, Connor testified that he would have accepted the plea offer had his attorney advised him about the possible mandatory life sentence. The trial judge rejected his testimony as incredible because Connor had steadfastly maintained his innocence throughout the pro- ceedings—including at sentencing when he professed his in- nocence and declared unequivocally that he would not have pleaded guilty to get a lesser sentence even if he had known the outcome of the trial. The judge therefore denied postcon- viction relief, ruling that Connor failed to show a reasonable probability that he would have pleaded guilty but for coun- Ȃȱ. The Illinois Appellate Court affirmed.
Connor petitioned for federal habeas review under 28 U.S.C. § 2254. The district judge denied relief. He was right to do so. A ȱȂȱactual findings are presumed correct on federal habeas review and may not be disturbed unless the No. 23-1595 3 state prisoner rebuts the presumption by clear and convincing evidence. 28 U.S.C. § 2254(e)(1). Credibility findings are enti- tled to particularly great deference. The state court found ConnorȂ testimony unbelievable because it flatly contra- dicted unambiguous statements he had made earlier in the case. That negative credibility finding is both presumptively correct and well supported by the record. Connor has not clearly and convincingly rebutted it. We affirm the judgment.
I. Background Michael Connor and Amanda Garcia have two daughters: La.C., born in 2004, and Le.C., born in 2006. Michael, Amanda, and their daughters lived together from 2004 to 2010. Garcia has another daughter from a prior relationship, Z.T., born in 2003, who also lived with them.
The couple split up in 2010. In August 2011 Garcia took La.C. to a counselor because she had expressed thoughts of homicide and death. The counselor asked Garcia if La.C. had ever been sexually abused. Garcia did not think so at first but soon started suspecting that Connor may have sexually abused the girls. She called the police to report her suspicions, and a forensic interviewer met with all three children.
A few days later, the Livingston County StateȂȱ¢ȱ charged Connor with five counts of predatory criminal sexual assault of a child in violation of chapter 720, section 5/12-14.1 of the 2010 Illinois Compiled Statutes. 1 The counts were grouped by victim and charged in two separate charging doc- uments—three counts for assaults against Z.T. between 2008
1 ȱěȱȱȱȱęȱȱchapter 720, section 5/11-1.40 of the Illinois Compiled Statutes. We refer to the 2010 statutes.
4 No. 23-1595 and 2009, and two counts for assaults against La.C. between 2010 and 2011.
Each charging document correctly listed the penalty for the offense: 6 to 60 years in prison. 720 ILL. COMP. STAT. 5/12- 14.1(b)(1). Sentences for multiple convictions must run con- secutively, 730 ILL. COMP. STAT. 5/5-8-4(d)(2), and the penalty increases to a mandatory term of life in prison if the offender is convicted of assaulting two or more victims, 720 ILL. COMP.
STAT. 5/12-14.1(b)(1.2). Neither charging document specifi- cally mentioned the mandatory life term—probably because the counts involving each victim were charged in separate in- struments. Still, the penalty enhancement is contained within the statute creating the crime, and the charging documents correctly listed that statute.
Though the counts against each victim were separately charged, the cases proceeded together. At arraignment the prosecutor correctly announced the penalty for this crime as to 60 years, with the sentences to run consecutively. The prosecutor repeated this warning at the preliminary hearing, and the judge added this explanation: “That means you are facing six to 60 in one case. That would be served. And then another six to 60 in another case.” Connor indicated that he understood and had no questions about the range of penalties he faced. The judge did not, however, mention the mandatory life sentence if Connor was convicted of assaulting both vic- tims.
Connor maintained his innocence throughout the pro- ceedings. He waived his right to a jury trial, and in 2013 the charges were tried to the court. The prosecutor presented tes- timony from Z.T. and La.C. and introduced their videotaped interviews. The prosecutionȂȱ ȱ ȱ ȱ ¢ȱ No. 23-1595 5 from a nurse who examined the girls in November 2011, a few months after the charges were issued. The nurse found no vis- ible signs of abuse, but she testified that La.C., who was just six years old when the examination occurred, tested positive for chlamydia, a sexually transmitted disease.
Connor testified in his own defense, telling the court that his relationship with Garcia deteriorated in August 2011, shortly before the charges were issued. He theorized that Garcia had coached the children into making the accusations against him.
The judge found Connor guilty on all counts. Immediately after the verdict, the prosecutor asked the court to revoke Ȃ bond, noting that the convictions carried a manda- tory term of life in prison. The judge granted the motion and ȱ Ȃȱ ǯȱ Connor filed several posttrial mo- tions, including one for a new trial. Among other claims, he ȱȱȱȂȱȱȱadvise him of the possible mandatory life sentence violated his right to due process. The judge denied the motion.
At sentencing Connor continued to assert his innocence, saying this during his allocution: I told the truth, and yet I stand here facing sen- ȱȱȱȱ ȱȂȱǯȱ… The thing [is,] even if I had known the outcome of a guilty verdict in this[,] I still would have not changed my plea because simply I cannot bring myself to ȱȱȱ ȱȱȂȱȱȱȱȱȱ lesser sentence.
The judge sentenced Connor to a 30-year prison term on one count involving Z.T. and the mandatory life term on one 6 No. 23-1595 count involving La.C. The remaining counts were merged into the judgment. The Illinois Appellate Court affirmed on direct appeal.
Through new counsel, Connor petitioned for state post- conviction relief, raising a Sixth Amendment claim for viola- tion of his right to the effective assistance of trial counsel. As ȱǰȱȱȱȱ¢Ȃȱ at the plea-bargaining stage of the case. He claimed that his counsel failed to tell him about the possible mandatory life sentence.
He further alleged that the prosecutor had offered an 18-year sentence if he pleaded guilty and that he would have accepted the offer had his counsel properly advised him.
The trial judge held an evidentiary hearing on the petition, at which Connor and his mother testified. Connor told the judge that at no time before trial did his attorney advise him that he faced a mandatory life sentence. He further testified that he and his parents met with his attorney to discuss the ȂȱŗŞ-year plea offer, and his attorney told him he faced a sentencing range of 6 to 30 years in prison, of which he would serve 85 percent. (Recall that the statutory maxi- mum for this crime is 60 years, not 30.) His attorney made no mention of the mandatory life sentence. Connor testified that he then discussed the plea offer with his family, and they agreed that because he was a first-time offender, any sentence would likely fall at the low end of the range of 6 to 30 years— at most, 12 years in prison—so the 18-year plea offer seemed too high. With their support, he rejected the plea bargain. Ȃ postconviction attorney asked him if he would have accepted the 18-year plea deal if his attorney had ad- vised him about the mandatory life sentence. Connor replied that he would have taken the deal even though he was No. 23-1595 7 innocent, saying: ”I Ȃȱȱȱwith my life on it. Even innocent, you know, you go for life you will never see anything again.” Ȃȱȱcounsel also asked him to explain his statement at the sentencing hearing when he told the judge that he would not have changed his plea to get a lesser sentence even if he knew he would be found guilty at trial.
Connor said that when he prepared that statement before sentencing, he was “under the assumption that I was doing ¡ȱ ȱ řŖǯȱ ȱ Ȃȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ sentencing.” When he found out about the mandatory life sentence at the sentencing hearing, he “just kind of stuck to” what he had written because he was nervous. On cross- examination Connor said he was confident in his innocence and believed he had a strong defense. Ȃȱmother similarly testified that ȱȂȱtrial law- yer did not mention the mandatory life sentence during the meeting to discuss the plea offer. She also testified that if she had known that her son faced a life sentence, she would have recommended that he accept the 18-year plea deal. She con- ceded on cross-examination that Connor had continuously professed his innocence, never expressing a willingness to plead guilty.
ConnorȂȱ trial attorney did not testify. Nor did Connor provide any details about the plea offer. He did not explain, for example, which assaults he would have admitted, how the proposed 18-year sentence would be structured, or whether the plea deal could have been presented to the court in a way that the judge would find acceptable.
8 No. 23-1595 The judge denied the postconviction petition, assuming for the sake of argument ȱȱȂȱȱhad been deficient and holding that Connor had failed to prove prejudice. To prevail on a Sixth Amendment claim of ineffec- tive assistance of counsel in the plea-bargaining context, the defendant must establish a reasonable probability that he ȱȱȱ¢ȱȱȱȂȱȱȬ mance. Missouri v. Frye, 566 U.S. 134, 147 (2012). The judge held that Connor had not made this showing. She expressly rejected his postconviction testimony, finding that he “was not a credible witness” and “seemed to be saying what he needed to say or more accurately [seemed to be] answering ȱ¢ȂȱȱȱȱȱǯȄȱThe judge also ex- plained that Ȃȱ¢ squarely contradicted other evidence in the record—notably his own unambiguous decla- ration at sentencing that he was innocent and would not have pleaded guilty to get a lesser sentence even if he had known that he would be convicted at trial.
The judge rejected Ȃȱ effort to explain away this statement. He testified in the postconviction proceeding that he did not mean what he said in allocution because he did not learn about the mandatory life sentence until the sentencing hearing and simply stuck to his prepared remarks. Yet in his motion for a new trial—filed more than four months before sentencing—he argued that he should receive a new trial be- cause the judge had not advised him of the possibility of a mandatory life sentence. So he knew about the mandatory life sentence well before sentencing, contrary to his claim in his postconviction testimony.
Finally, the judge noted that Connor “was advised on more than one occasion that he faced a sentencing range of No. 23-1595 9 to 60 years on each case and the sentences would run con- secutive to each other.” Assuming he received a sentence in the middle of each range, that meant he faced a de facto life sentence: For all practical purposes, [Connor] was well aware [when he rejected the plea deal] there was a very real and likely possibility that given the consecutive sentences, he could spend the rest of his life in prison. … [He] was willing to risk what would effectively be a life sentence … rather than plead guilty in exchange for an 18-year sentence.
Accordingly, the judge concluded that Connor had failed to prove prejudice because he had not shown a reasonable prob- ability that he would have pleaded guilty ȱȱȱȂȱ deficient performance.
The Illinois Appellate Court affirmed, accepting the Ȃȱcredibility findings and agreeing that the record con- ȱ Ȃȱ ȱ ¢ȱ essentially the same reasons. The appellate court added some analysis of its own. Connor had claimed at his postconviction hearing that he thought he faced a prison term of 6 to 30 years based on his meeting with his attorney about the plea offer. The court noted that this claim was contradicted by the charging documents, both of which listed the correct term of 6 to years, as well as the record, which confirmed that Connor had been advised at several hearings that he faced 6 to years.
10 No. 23-1595 The appellate court noted the obvious disparity between the 18-year plea offer and a life sentence; the court also acknowledged that a discretionary de facto life sentence is not the same as a mandatory life sentence. The court reasoned, however, that the disparity could not ȱ Ȃȱ continuous insistence that he was innocent, his belief that he had a strong defense, and his firm declaration at sentencing that he would not have pleaded guilty even if he had known the outcome of trial. Applying the general framework of Strickland v. Washington, 466 U.S. 668 (1984), and the specific requirements of Frye, the appellate court held that Connor had not shown that he would have pleaded guilty but for his Ȃȱȱǯ The Illinois Supreme Court declined review.
Connor then turned to federal court with a petition for ha- beas review under § 2254. Because the state courts had cor- rectly cited and reasonably applied the legal standards in Strickland and Frye, Ȃ petition turned entirely on § 2254(d)(2), which permits federal habeas relief only if the ȱȱȱȱȱȂȱȱ ȱȱ on an unreasonable determination of the facts.
Importantly here, the habeas statute implements this al- ready deferential standard of review with yet another layer of deference. Under § 2254(e)(1), a ȱȂȱȱȱ are presumed correct, and the applicant must rebut the pre- sumption by clear and convincing evidence.
The district judge denied Ȃȱ ȗȱ ŘŘśŚȱ petition, con- cluding that he had not met his burden to rebut the presump- tion. But the judge authorized an appeal, see 28 U.S.C. § 2253(c)(2), finding that reasonable jurists could disagree about whether the disparity between the 18-year plea offer No. 23-1595 11 and a mandatory life sentence is enough to overcome the state Ȃȱnegative credibility finding.
II. Discussion Federal collateral review of state criminal judgments is highly deferential: a federal court may not grant habeas relief from a state criminal judgment ȱȱȱȂȱȬ sion (1) was “contrary to” or “an unreasonable application of” clearly established federal law, § 2254(d)(1); or (2) “was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding,” § 2254(d)(2). These steep standards of review “were designed to prevent federal habeas retrials and to ensure that state- court convictions are given effect to the extent possible under law.” Dassey v. Dittmann, 877 F.3d 297, 301 (7th Cir. 2017) (en banc) (quotation marks omitted). ȱ ȱȱȱȂ decision denying relief de novo and focus our attention on the “last reasoned state court decision to decide the merits of the case.” Pierce v. Vanihel, 93 F.4th 1036, 1044 (7th Cir. 2024). Ȃȱ habeas claim rests on § 2254(d)(2): he focuses ¢ȱ ȱ ȱ ȱ Ȃȱ ȱ ǯȱ ȱ ȱ ȱ faces the additional hurdle of § 2254(e)(1), which ȱ Ȃȱ noted applies a presumption of correctness to a state courtȂ factual findings—a presumption that can be overcome only by clear and convincing evidence. And because Connor asks us to disturb a state-court credibility finding, his burden is heavier still. On federal habeas review, we give “great deference” to ȱ ȱ Ȃȱ credibility determinations, Gambaiani v. Greene, 137 F.4th 627, 640 (7th Cir. 2025), which “are notoriously difficult to overturn,” Sanders v. Radtke, 48 F.4th 502, 511 (7th Cir. 2022) (quotation marks omitted).
12 No. 23-1595 Under these standards of review, “federal habeas relief from state convictions is rare.” Dassey, 877 F.3d at 302. “It is reserved for those relatively uncommon cases in which state courts veer well outside the channels of reasonable decision- making about federal constitutional claims.” Id. The merits legal standard under Strickland is also highly deferential. A defendant raising a Sixth Amendment claim of ineffective assistance of counsel must prove that his attor- neyȂȱ ȱ ȃȱ ȱ ȱ ȱ ȱ ȱ Ȭ sonableness” under all the circumstances, Strickland, 466 U.S. at 687–88, and that the deficient performance prejudiced his defense, id. at 692. To prove prejudice, the defendant must es- ȱȃȱȱ¢ȱǰȱȱȱȂȱȬ fessional errors, the result of the proceeding would have been different.” Id. at 694. As applied in the context of plea bargain- ing, the Strickland prejudice inquiry requires the defendant to prove a reasonable probability that (1) he would have ac- ȱȱȱȱȱȱ¢ȱȱȱȂȱȬ ficient performance; (2) his guilty plea “would have been entered without the prosecution canceling it or the trial court refusing to accept it”; and (3) the end result “would have been more favorable by reason of a plea to a lesser charge or a sen- tence of less prison time.” Frye, 566 U.S. at 147.
ConnorȂȱȱȱȱȱȱȱȱȱFrye prejudice framework under ȱ ȱ Ȃȱ highly deferential standard of review for factual findings in general and credi- bility determinations in particular. Connor claimed in his tes- timony at the postconviction hearing that he would have accepted the 18-year plea offer and pleaded guilty had his counsel advised him of the mandatory life sentence he faced if convicted of assaulting both victims. But that testimony No. 23-1595 13 squarely contradicted what he said at the sentencing hearing when he professed his innocence and declared unequivocally that he would not have pleaded guilty just to get a lesser sen- tence even if he had known the outcome of trial. Based on that glaring contradiction, and for additional reasons relating to his demeanor and other conflicts between his testimony and the record, the state trial judge found him to be “not a credible witness.”
The Illinois Appellate Court affirmed, largely on the same reasoning. That ruling is well supported by the record. It is also presumptively correct under § 2254(e)(1), and Connor bears the burden of rebutting the presumption by clear and convincing evidence. His argument for overturning the state Ȃȱcredibility finding rests solely on the significant dis- parity between the 18-year plea offer and the mandatory life sentence he faced if convicted at trial (and is now serving). He contends that the significance of this disparity is enough, without more, to rebut the adverse credibility determination. Ȃȱ ȱ ȱ a sizable disparity between the sentence called for in a putative plea bargain and the sentence the de- fendant faced if convicted at trial is objective evidence that can inform the Strickland prejudice question in this context— i.e., whether the defendant would have pleaded guilty but for ȱ¢Ȃ errant advice. See, e.g., Julian v. Bartley, 495 F.3d 487, 498–99 (7th Cir. 2007); Moore v. Bryant, 348 F.3d 238, 242– (7th Cir. 2003). ȱȂȱȱenough on its own. Ȃȱ¡Ȭ plained that a defendant must do more than “show that most or many defendants in his position would accept the plea.”
Quintana v. Chandler, 723 F.3d 849, 857 (7th Cir. 2013). Rather, ȱȱȱȱ ȱȱ¢ȱȂȱȱȱ the plea-bargaining stage, the defendant “must offer 14 No. 23-1595 something more than the large gap in sentence lengths to show that he would have accepted the [plea deal].” Id. Applying this principle here, the large gap in sentence lengths is not enough to ȱ ȱ ȱ Ȃȱ ȱ credibility finding. It shows, at best, that many defendants in Ȃȱ ȱ ȱ ȱ ȱ ȱ ; it does not clearly and convincingly show that he would have done so. In more concrete terms, it does not show that he would have pleaded guilty to sexually assaulting his daughter or step- daughter—especially (though not exclusively) because he said precisely the opposite at sentencing when he was well aware that he was about to be sentenced to life in prison.
On this record, the sentence disparity is not alone enough to overcome the presumption of correctness that applies to state-court factual findings. More specifically, it does not overcome the great deference owed to state-court credibility determinations on federal habeas review. The state appellate court ȱȱȱȂ rejection of ȂȱȬ viction testimony as not worthy of belief, a finding that is am- ply supported by the record. Connor had a heavy burden to clearly and convincingly rebut this adverse credibility find- ing. He has not carried it.
AFFIRMED No. 23-1595 15 ROVNER, Circuit Judge, concurring. Michael J. Connor re- jected a plea offer of eighteen years, thinking it was possible that he might be sentenced to as little as twelve years. At the time he was considering the plea offer, neither his counsel nor the court informed him that, should he be convicted, the court would have no choice but to sentence him to 120 years. This is an astoundingly troubling fact scenario. Nevertheless, the majority’s careful analysis correctly concludes that our hands are tied. The state post-conviction relief court made factual and credibility findings to which we must give great defer- ence, and those determinations make federal habeas relief from this state conviction impossible.
I write only to highlight two issues I believe to be worthy of consideration in future cases. First, the defense bar may wish to consider whether lawyers have a duty to advise their clients about the potential consequences of making allocution statements like the one Connor made here—that he would have never accepted a plea deal under any circumstances. Af- ter all, it is not an uncommon scenario for a defendant to al- lege in post-conviction proceedings that he was misinformed about his maximum potential sentence. A defendant can con- tinue to profess vociferously his innocence without stating that there are no circumstances under which he would have accepted a plea. A person may think he would never plead guilty to a crime he did not commit, but many of us would feel differently when confronted with the risk of spending the rest of our lives in prison. “Never say ‘never’” would have been sage advice from counsel.
Second, allocution is an important right that allows a de- fendant to present mitigating circumstances and represent himself to a judge in the most favorable light. United States v. 16 No. 23-1595 Cunningham, 883 F.3d 690, 699 (7th Cir. 2018); see also Kim- berly A. Thomas, Beyond Mitigation: Towards A Theory of Allocution, 75 Fordham L. Rev. 2641, 2644 (2007). One might expect defendants to do all they possibly can to paint them- selves in the best light. Because the purpose of these unsworn statements is to persuade, present mitigating circumstances, and seek mercy, courts ought to think skeptically about the use of these statements as substantive evidence in subsequent proceedings. But see U.S. v. Ward, 732 F.3d 175, 183 (3d Cir. 2013) (noting that the unsworn nature of allocution does not prevent its use in future proceedings); United States v. Whitten, 610 F.3d 168, 199 (2d Cir. 2010) (same). The use of Connor’s allocution statement in this case is particularly troubling be- cause it was not a confession to a particular crime or fact, but a statement meant to convey how strongly he believed in his innocence.
Moreover, contrary to the state post-conviction court’s conclusion, I am not so certain that, at the time he delivered his allocution statement, Connor understood that the court had absolutely no choice but to impose a mandatory life sen- tence. Just before Connor made his statement, his counsel ad- dressed the court and offered a theory for imposing a mere twelve-year sentence. If, in fact, his counsel thought that a twelve-year term of imprisonment might still be on the table, it seems likely that Connor could have as well. In order to demonstrate the sincerity of his claim of innocence, Connor may have been willing to risk twelve years, but as we have now noted many times, the calculus must surely change when a court has only one choice of sentence—life imprisonment.
At the moment in time when Connor had to make his de- cision to accept or reject the eighteen-year plea offer, he was No. 23-1595 17 devastatingly ill-informed of the potential ramifications of re- jecting the offer. I question the fairness of locking Connor into his statement at allocution that he would not have changed his plea had he known that he would be found guilty, given the disparity between the 18-year plea offer and the eleventh- hour discovery that he was subject to a mandatory life sen- tence. Nevertheless, the state post-conviction relief court did not believe Connor’s later contention that he would indeed have accepted the 18-year offer. As the majority correctly and fairly concludes, we have no ability to undo the state court’s credibility and factual determinations. This court is left with no choice but to deny the requested relief.
Case-law data current through December 31, 2025. Source: CourtListener bulk data.