OPINION DENYING APPLICATION FOR POST-CONVICTION RELIEF

SMITH, JUDGE:

¶1 James Coddington was tried by jury and convicted of First Degree Murder in the District Court of Oklahoma County, Case No. CF-1997-1500. He received the death penalty. This Court affirmed Coddington's conviction, reversed the sentence, and remanded the case for resentencing. Coddington v. State, 2006 OK CR 34, 142 P.3d 437. Coddington again received the death penalty after a resentencing trial. This Court upheld that decision. Coddington v. State, 2011 OK CR 17, 254 P.3d 684.

¶2 On March 21, 2011, Coddington filed an application for capital post-conviction relief.1 The only issues appropriately raised under Oklahoma's post-conviction statutes are those which: "(1) [w]ere not or could not have been raised in a direct appeal; and (2) [s]upport a conclusion either that the outcome of the trial would have been different but for the errors or that the defendant is factually innocent." 22 O.S.Supp.2006, §§ 1089(C)(1),(2). On review, this Court must determine: "(1) whether controverted, previously unresolved factual issues material to the legality of the applicant's confinement exist, (2) whether the applicant's grounds were or could have been previously raised, and (3) whether relief may be granted...." 22 O.S.Supp.2006, § 1089(D)(4). The post-conviction process is not a second appeal. The doctrines of res judicata and waiver will apply where a claim either was, or could have been, raised in the petitioner's direct appeal. Coddington raises two propositions of error in support of his application.

¶3 In his first proposition Coddington claims he received ineffective assistance of counsel.2 A claim of ineffective assistance of trial counsel is appropriate for post-conviction review if it has a factual basis that could not have been ascertained through the exercise of reasonable diligence on or before the time of the direct appeal. 22 O.S.Supp.2006, § 1089(D)(4)(b)(1). A claim of ineffective assistance of appellate counsel may be raised for the first time on post-conviction. 22 O.S.Supp.2006, § 1089(D)(4)(b)(2). We review post-conviction claims of ineffective assistance using United States Supreme Court precedents. 22 O.S.Supp.2006, § 1089(D)(4); Davis v. State, 2005 OK CR 21, ¶ 6, 123 P.3d 243, 245-46. Coddington must show that counsel's performance was deficient, and that he was so prejudiced by that performance that he was deprived of a fair trial with a reliable result. Strickland v. Washington, 466 U.S. 668, 687, 104 S.Ct. 2052, 2064, 80 L.Ed.2d 674 (1984). We will not find counsel was ineffective if Coddington was not prejudiced by counsel's act or omission. Harris v. State, 2007 OK CR 32, ¶ 3, 167 P.3d 438, 441. A finding of prejudice requires a showing of "a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Smith v. State, 2010 OK CR 24, ¶ 19, 245 P.3d 1233, 1239; Strickland, 466 U.S. at 694, 104 S.Ct. at 2068. "The likelihood of a different result must be substantial, not just conceivable." Harrington v. Richter, 131 S.Ct. 770, 790, 792, 178 L.Ed.2d 624 (2011). There is a strong presumption that counsel's conduct is within the wide range of reasonable professional conduct. Strickland, 466 U.S. at 689, 104 S.Ct. at 2065. Coddington must show "that counsel's representation was unreasonable under prevailing professional norms and that the challenged action could not be considered sound trial strategy." Davis, 2005 OK CR 21, ¶ 7, 123 P.3d at 246. A strong presumption exists that, where counsel focuses on some issues to the exclusion of others, this reflects a strategic decision rather than neglect. Richter, 131 S.Ct. at 790. Appellate counsel need not raise every non-frivolous issue. Jones v. Barnes, 463 U.S. 745, 753-54, 103 S.Ct. 3308, 3313-14, 77 L.Ed.2d 987 (1983).

¶4 We note that appellate counsel on Coddington's direct appeal raised a claim of error which required Coddington's death sentence to be reversed and remanded for resentencing. Given this evidence of effective representation, it will be difficult for Coddington to show that appellate counsel's performance was deficient.

¶5 Coddington claims in Subproposition A that appellate counsel from Coddington's first direct appeal (hereafter "appellate counsel") failed to raise several meritorious issues.

¶6 Appellate counsel interviewed several jurors after Coddington's first trial, but did not ask about their backgrounds. Coddington's first claim is that appellate counsel should have investigated the background of members of Coddington's first trial jury, and interviewed jurors regarding that specific investigation. Coddington claims that, had appellate counsel done so, counsel would have discovered that six members of that jury gave inaccurate answers to voir dire questions; with this information, Coddington argues, appellate counsel could have claimed that Coddington was denied his right to an impartial jury. It is important to remember what claim of error, precisely, this Court is reviewing. While Coddington explains at length the substantive basis for this claim - the responses jurors gave to voir dire questions, and information apparently contradicting those responses which was discovered in post-conviction investigation - this Court is not reviewing the substantive claim of error. The issue raised in this subproposition is not whether any or all of these six jurors were less than candid in their responses; nor is it whether trial counsel, had she known of these misrepresentations, would have challenged any of these jurors for cause or through a peremptory challenge. The issue before the Court is whether appellate counsel's failure to investigate these jurors and raise any issues resulting from that investigation on direct appeal, raises a substantial likelihood that the result of the proceeding would have been different. In conducting this review, this Court is focusing on Coddington's conviction as the outcome which would have been affected by different actions on appellate counsel's part.

¶7 Coddington apparently fails to understand these limitations. In addition to claiming ineffective assistance of appellate counsel in this proposition, he claims that he was denied his right to fully conduct voir dire, and his right to an impartial jury, by what he characterizes as material omissions in the responses of three (and particularly two) of the six jurors. This is not the question before the Court. These substantive issues have been waived.

¶8 Coddington fails to show a reasonable probability that the outcome of his trial would have been different, had appellate counsel discovered the information about the jurors he presents to this Court on post-conviction. Coddington claims that this Court has held appellate counsel has the duty to investigate juror misconduct. This claim is not supported by his cited cases; while this Court has ruled on claims of juror misconduct developed through appellate investigation, we have not imposed upon appellate counsel a duty to conduct such investigation as part of effective representation. Harris, 2007 OK CR 32, ¶ 10, 167 P.3d at 443. Coddington argues that any misrepresentation by a juror goes to the juror's impartiality, and is structural error. This is not the case. Structural error is reserved for a very limited category of errors, and the strong presumption is that trial error is subject to harmless error analysis. Robinson v. State, 2011 OK CR 15, ¶ 4, 255 P.3d 425, 428. We have held that an error in voir dire as basic as the outright denial of a peremptory challenge is reviewed for harmless error. Robinson, 2011 OK CR 15, ¶ 12, 255 P.3d at 430. Our cases involving juror misconduct consistently subject that issue to harmless error analysis.

¶9 The Constitution guarantees defendants a disinterested jury; even where jurors' voir dire responses are deliberately misleading, relief is required only where the record "casts sufficient doubt on the juror's ability to be impartial." Underwood v. State, 2011 OK CR 12, ¶ 95, 252 P.3d 221, 255; 22 O.S.2001, § 660 (juror bias involves relations between prospective juror and defendant, a witness, or the case itself); 22 O.S.2001, § 659 (juror bias where, in reference to the case or either party, juror's state of mind is not impartial). Relief is not necessary unless a juror's misrepresentations or lack of candor have "any relationship to, connection with, or bias toward or against any party or witness in this case," involve a situation similar to the charged crime, or suggest a personal interest in serving on the jury so strong the juror would commit perjury to do so. Underwood, 2011 OK CR 12, ¶¶ 98-99, 252 P.3d 221, 256-57. A juror's omission must be material or affect the outcome of the trial. Harris, 2007 OK CR 32, ¶ 15, 167 P.3d at 444. Coddington must show actual injury from a juror's non-disclosure of even material information. Grissom v. State, 2011 OK CR 3, ¶ 25, 253 P.3d 969, 979.

¶10 Coddington argues that, because appellate counsel did not follow the ABA guidelines for capital defense counsel, which recommend that appellate counsel aggressively investigate all aspects of the case, counsel must be ineffective. The United States Supreme Court has rejected the suggestion that compliance with published ABA standards or guidelines is required to meet the standards of effective assistance set forth in Strickland. Bobby v. Van Hook, 130 S.Ct. 13, 17, 175 L.Ed.2d 255 (2009). This Court has also rejected this argument, stating, "we will not find that capital counsel was per se ineffective simply because counsel's representation differed from current capital practice customs, even where the differences are significant." Torres v. State, 2005 OK CR 17, ¶ 12, 120 P.3d 1184, 1189.

¶11 To support this claim Coddington provides the Court with an Appendix of Exhibits containing various official records which, he argues, show involvement in or knowledge of court cases not admitted by the six jurors in voir dire.3 For purposes of this review, we assume without deciding that the documents apply to the jurors in question. Taken together, these documents do not support a conclusion that any of the jurors' answers in voir dire reflect any bias or partiality for or against this case or this defendant. We repeat, whether the responses were inaccurate or whether trial counsel might have used peremptory challenges based on this information is not the issue before the Court. Appellate counsel had no duty to investigate jurors' backgrounds. Coddington has not shown he was prejudiced by counsel's failure to conduct such an investigation, or to raise any issues resulting from that investigation on appeal. Appellate counsel was not ineffective for failing to investigate jurors' backgrounds.

¶12 In the second part of subproposition A, Coddington claims appellate counsel was ineffective because appellate counsel did not raise trial counsel's failure to question Juror M. in voir dire about her health, illness, and ability to stay awake during testimony. Again, our scope of review does not encompass the substantive claim underlying the ineffective assistance argument. That is, we do not decide whether trial counsel should have entered into a dialogue with Juror M. regarding her health after the juror stated that she retired for health reasons. We decide only whether appellate counsel was ineffective for failing to claim that trial counsel was ineffective for failing to ask such questions.

¶13 As Coddington cannot show any prejudice from this failure, the record does not support such a finding. Appellate counsel raised the substantive issue of Juror M.'s ability to discharge her duties as a proposition of error on direct appeal, and claimed that trial counsel was ineffective for failing to bring the issue to the trial court's attention. We held that the trial court's failure to remove Juror M. was not plain error. Coddington, 2006 OK CR 34, ¶ 27, 142 P.3d at 446. We further found that trial counsel was not ineffective for failing to ask that Juror M. be removed or replaced; we refused to evaluate this trial strategy in hindsight, and found the record showed neither that Juror M. missed any testimony, nor that Coddington was prejudiced by trial counsel's failure to ask that she be removed. Coddington, 2006 OK CR 34, ¶ 28, 142 P.3d at 446. On post-conviction, Coddington fails to show how any additional colloquy about Juror M.'s health would have had any effect on the outcome of the proceedings. The most Coddington argues is that, had such questioning occurred, Juror M. might not have served. This is wholly speculative and cannot serve as a basis for a finding that Coddington was prejudiced by appellate counsel's failure to raise this particular issue.

¶14 In subproposition B, Coddington claims appellate counsel for Coddington's resentencing (hereafter "resentencing appellate counsel") was ineffective for failing to raise meritorious issues.

¶15 Coddington first claims resentencing appellate counsel should have raised as error particular claims of prosecutorial misconduct in the resentencing trial. Coddington admits that resentencing appellate counsel raised several issues of prosecutorial misconduct, which this Court found were unsupported by the record. Coddington, 2011 OK CR 17, ¶¶ 72-77, 254 P.3d at 712-13. Insofar as Coddington's claims were raised and decided on direct appeal, they are barred. Smith, 2010 OK CR 24, ¶ 38, 245 P.3d at 1243. Insofar as Coddington appears to suggest that resentencing trial counsel was ineffective for failing to object to questions allegedly constituting misconduct by the trial court or prosecutor during voir dire, the claims could have been raised in the resentencing appeal and are waived. Harris, 2007 OK CR 32, ¶ 12, 167 P.3d at 443. Coddington argues that resentencing appellate counsel should have included more and different claims of prosecutorial misconduct. Post-conviction review is not intended to serve as a second appeal. "We will not allow a defendant to subdivide claims in order to relitigate an issue in an application for post-conviction." Harris, 2007 OK CR 32, ¶ 13, 167 P.3d at 443, quoting Browning v. State, 2006 OK CR 37, ¶ 4, 144 P.3d 155, 157; Davis, 2005 OK CR 21, ¶ 17, 123 P.3d at 248. The additional prosecutorial misconduct claims Coddington wishes to discuss have been waived. Harris, 2007 OK CR 32, ¶ 13, 167 P.3d at 443.

¶16 The record does not support Coddington's claims that resentencing counsel was ineffective for failing to raise additional, now waived, instances of misconduct. Coddington fails to show that he was prejudiced by any of the statements during voir dire, testimony and argument which Coddington alleges as misconduct. The record shows that much of the alleged misconduct was not erroneous, and that the trial court ruled in favor of Coddington on some of the issues. Where there is no prejudice, we will not find counsel ineffective.

¶17 In the second part of subproposition B, Coddington claims that resentencing appellate counsel failed to raise issues of ineffective assistance of resentencing trial counsel. Resentencing appellate counsel raised several instances of ineffective assistance of resentencing trial counsel; this Court found those claims were not supported by the record. Coddington, 2011 OK CR 17, ¶¶ 78-85, 254 P.3d at 713-16. Again, Coddington claims that resentencing appellate counsel should have raised more and different claims. These claims have been waived. Harris, 2007 OK CR 32, ¶ 13, 167 P.3d at 443.

¶18 Coddington first claims that resentencing appellate counsel failed to raise as ineffective assistance resentencing trial counsel's failure to object when the autopsy results were presented by a medical examiner who did not perform the autopsy. To the extent that Coddington may be attempting to claim that his Confrontation Clause rights were violated, or that resentencing trial counsel was ineffective for failure to object, these claims have been waived. We consider only the claim of ineffective assistance of resentencing appellate counsel. Our scope of review is limited. We do not consider the substantive issue of whether Coddington's right to confrontation was violated by this testimony. We ask only whether resentencing appellate counsel's failure to raise this issue was ineffective. Coddington admits that this issue is subject to harmless error analysis. Cuesta-Rodriguez v. State, 2010 OK CR 23, ¶ 40, 241 P.3d 214, 230. Coddington further admits that the facts of this case are similar to those in Cuesta-Rodriguez, in which we found the error was harmless, but argues that the error would not be harmless in his case. Coddington confessed to the murder, and witnesses described the crime scene and the victim's wounds in detail. The record does not support a conclusion that resentencing appellate counsel's failure to raise this issue would have affected the outcome of the resentencing trial.

¶19 Coddington argues that resentencing appellate counsel should have claimed trial counsel was ineffective for failing to show jurors the biological and social factors connected with Coddington's addiction and abuse problems. Coddington admits that resentencing appellate counsel specifically claimed trial counsel was ineffective for failing to present psychological testimony regarding Coddington's addictions which was presented in his first trial; this Court found trial counsel's strategic decision to present this evidence through a neuropsychologist was not ineffective. Coddington, 2011 OK CR 17, ¶¶ 79-82, 254 P.3d at 714-15. In reaching that conclusion this Court compared and discussed the similarities and differences between the psychologist's testimony in his first trial and the neuropsychologist's testimony in the resentencing trial. We found that resentencing trial counsel's mitigating evidence focus was more expansive than in the first trial, including the effects of substance abuse on Coddington's brain, thinking and judgment, as well as the effects of his whole history and upbringing. Coddington, 2011 OK CR 17, ¶ 82, 254 P.3d at 715. On post-conviction, Coddington argues that resentencing trial counsel should also have presented testimony from a neuropharmacologist, and that resentencing appellate counsel should have raised that omission as ineffective assistance. The neuropharmacologist's evaluation was finished by September 5, 2005, when it was included in Coddington's original post-conviction application, and was available to resentencing trial and appellate counsel. This evaluation is included as Exhibit 12 in the Appendix, and contains material similar to that included in resentencing mitigation testimony. Defense counsel's decision not to present particular evidence in mitigation may be sound trial strategy. Coddington, 2011 OK CR 17, ¶ 82, 254 P.3d at 715; Jones v. State, 2009 OK CR 1, ¶ 91, 201 P.3d 869, 891, cert. denied, 130 S.Ct. 237, 175 L.Ed.2d 163. Coddington has not shown that the outcome of the proceeding would have been different if resentencing appellate counsel had alleged that this decision constituted ineffective assistance of resentencing trial counsel.

¶20 Finally, in this subproposition, Coddington claims that resentencing appellate counsel should have argued the State of Oklahoma induced resentencing trial counsel's ineffectiveness. He notes that when resentencing trial counsel originally asked for a continuance of the resentencing trial, she stated she was handling several capital cases at the time. He also notes that the second chair attorney for the resentencing trial changed from the time the motion for continuance was filed in August, 2007, and the resentencing trial in 2008. Coddington argues that these factors - the change in second chair attorneys, and lead counsel's heavy capital caseload - violated the ABA Guidelines for capital counsel. He claims the State bears the responsibility for this because the State provides financial resources for indigent capital defendants. Coddington argues that the resources provided by the State were not sufficient to allow competent capital representation in his case. He infers that the State deliberately failed to provide sufficient resources, in order to profit from the resulting poor capital defense representation by obtaining a death sentence.

¶21 This claim is wholly unsupported by the record. As we discuss above, we will not hold capital defense counsel ineffective if capital representation does not comply with ABA standards or guidelines. Torres, 2005 OK CR 17, ¶ 12, 120 P.3d at 1189; Van Hook, 130 S.Ct. at 17. Coddington must show, not merely that his representation may not have met ABA standards, but that he was prejudiced by the representation he received. Torres, 2005 OK CR 17, ¶ 12, 120 P.3d at 1189. This Court found that resentencing trial counsel were not ineffective. Coddington, 2011 OK CR 17, ¶¶ 78-85, 254 P.3d at 713-16. We have found in previous subpropositions of this post-conviction application that resentencing appellate counsel was not ineffective for failing to raise other claims of ineffective assistance of resentencing trial counsel. The record of Coddington's resentencing does not reflect that resentencing trial counsel was unprepared, rushed, unable to engage in the adversarial process, or less than zealous in representing Coddington. The record does not show any way in which Coddington's defense may have been prejudiced by lack of resources. Coddington has not shown that the results of the proceeding would have been different had resentencing appellate counsel raised this issue of ineffective assistance of resentencing trial counsel.

¶22 In his second proposition, Coddington claims that the cumulative effect of errors identified in both his direct appeals and post-conviction proceedings rendered his trial and appellate proceedings unconstitutional and require relief. As we have previously noted, this Court does not consider on post-conviction any errors in previous proceedings which were not raised in the post-conviction application itself. Harris, 2007 OK CR 32, ¶ 20, 167 P.3d at 445. We found no merit in Coddington's post-conviction propositions. There is no error to accumulate. Smith, 2010 OK CR 24, ¶ 39, 245 P.3d at 1243.

¶23 In connection with his Application, Coddington filed a motion for evidentiary hearing and discovery. Discovery may be had on post-conviction where parties did not comply with discovery orders at trial, and where the material sought would have resulted in a different outcome. 22 O.S.Supp.2006, § 1089(D)(3); Rule 9.7(D)(3), Rules of the Oklahoma Court of Criminal Appeals, Title 22, Ch.18, App. (2011). This Court may remand for an evidentiary hearing on the issue of discovery if failure to comply with a discovery order is the grounds for an issue raised in the application. Rule 9.7(D)(4), Rules of the Oklahoma Court of Criminal Appeals, Title 22, Ch.18, App. (2011). A request for an evidentiary hearing must present information which shows "by clear and convincing evidence the materials sought to be introduced have or are likely to have support in law and fact to be relevant to an allegation raised in the application for post-conviction relief." Rule 9.7(D)(5), Rules of the Oklahoma Court of Criminal Appeals, Title 22, Ch.18, App. (2011). In support of his motion for evidentiary hearing, Coddington incorporates all the material included in his post-conviction application, Appendix of Exhibits, and any other filings in his case. He makes no separate argument regarding the necessity for discovery. We found no merit to the propositions of error which were raised in his Application and supported by this material. Coddington fails to meet the standard for an evidentiary hearing above, and has not shown why discovery is warranted. Harris, 2007 OK CR 32, ¶ 21, 167 P.3d at 445; Davis, 2005 OK CR 21, ¶ 19, 123 P.3d at 249. This motion is denied.

¶24 Coddington also filed a document purporting to be a motion to reserve the right to supplement this application for post-conviction relief. There is no such right. Under our Rules any attempt to supplement an existing post-conviction application is considered a second or subsequent application. Harris, 2007 OK CR 32, ¶ 22, 167 P.3d at 445; Rule 9.7(B)(3), Rules of the Oklahoma Court of Criminal Appeals, Title 22, Ch.18, App. (2011). Although Coddington made no attempt to supplement this application, because the motion is improper under our Rules, it is denied.

DECISION

¶25 The Application for Post-Conviction Relief is DENIED. Coddington's Motion for Evidentiary Hearing and Discovery, and his Motion Reserving the Right to Supplement Original Application for Post-Conviction, are DENIED. Coddington's Motion to File Over-Sized Application is GRANTED. Pursuant to Rule 3.15, Rules of the Oklahoma Court of Criminal Appeals, Title 22, Ch.18, App. (2011), the MANDATE is ORDERED issued upon the delivery and filing of this decision.

APPEARANCES ON APPEAL

LAURA M. ARLEDGE
CAPITAL POST-CONVICTION DIVISION, OIDS
P.O. BOX 926
NORMAN, OKLAHOMA 73070
ATTORNEY FOR PETITIONER

E. SCOTT PRUITT
ATTORNEY GENERAL OF OKLAHOMA
SETH BRANHAM
ASSISTANT ATTORNEY GENERAL
313 N.E. 21ST STREET
OKLAHOMA CITY, OKLAHOMA 73105
ATTORNEYS FOR APPELLEE

OPINION BY: SMITH, J.
A. JOHNSON, P.J.: CONCUR
LEWIS, V.P.J.: CONCUR
LUMPKIN, J.: CONCUR IN RESULTS
C. JOHNSON, J.: CONCUR

FOOTNOTES

1 At this Court's direction, the State filed a Response to Coddington's application on June 2, 2011.

2 The title of this proposition of error states Coddington received ineffective assistance of appellate and trial counsel in every previous proceeding. Coddington apparently recognizes that issues of ineffective assistance of trial counsel, at both the trial and resentencing hearing, have been waived. Browning v. State, 2006 OK CR 37, ¶ 4, 144 P.3d 155, 157. His proposition primarily argues that appellate counsel for both the first direct appeal and Coddington's resentencing were ineffective for failing to raise various claims of ineffective assistance of trial counsel.

3 These records contain identifying information for both the jurors and family members. At the State's request, this Court ordered the Appendix of Exhibits containing those records sealed. Coddington v. State, Case No. PCD-2008-920 (Okl.Cr. June 15, 2011). This Court cautions parties filing similar documents in future to either redact personal identifying information, or request that the documents be filed under seal.


LUMPKIN, JUDGE: CONCUR IN RESULTS

¶1 I concur in the results reached in this case but write separately to address several issues.

¶2 Petitioner violates Rule 3.5(A)(5), Rules of the Oklahoma Court of Criminal Appeals, Title 22, Ch. 18, App. (2011) by setting out sub-propositions. Petitioner's failure to separately set out each proposition waives review of these claims. See Cuesta-Rodriguez v. State, 2011 OK CR 4, ¶ 12, 247 P.3d 1192, 1197.

¶3 The opinion does not correctly apply res judicata and waiver to Petitioner's allegations concerning his first trial. "[C]laims that could have been raised in previous appeals but were not are generally waived; claims raised on direct appeal are res judicata." Murphy v. State, 2005 OK CR 25, ¶ 3, 124 P.3d 1198, 1199.

¶4 Finally, the opinion fails to distinguish the circumstances in Melendez-Diaz v. Massachusetts, ___ U.S. ___, 129 S.Ct. 2527, 174 L.Ed.2d 314 (2009), from the instance where an expert testifies to his or her own opinion based on the testing of other professionals if reasonably relied upon by experts in the field. Cuesta-Rodriquez v. State, 2010 OK CR 23, ¶¶ 4-10, 241 P.3d 214, 248-49 (Lumpkin, J., concurring in result). It is expected that the United States Supreme Court would make this distinction. See Bullcoming v. New Mexico, --- U.S. ---, 131 S.Ct. 2705, 2722, --- L.Ed.2d --- (2011), 131 S.Ct. at 27 (5-4 decision) (Sotomayer, J., concurring in part) ("Third, this is not a case in which an expert witness was asked for his independent opinion about underlying testimonial reports that were not themselves admitted into evidence."). In the present case, the admission of Dr. Choi's own opinions was not error. Id. The admission of Dr. Parker's sketches of the injuries and photograph were harmless beyond a reasonable doubt and thus Petitioner has failed to show that the outcome of the proceeding would have been different absent counsel's omission.