Siehl v. Grace

561 F.3d 189, 2009 U.S. App. LEXIS 6392, 2009 WL 764538

• Court: Court of Appeals for the Third Circuit
• Decided: March 25, 2009
• Docket: 07-1568
• Status: Published

Opinion

OPINION OF THE COURT

STAPLETON, Circuit Judge:

Appellant Kevin C. Siehl was convicted of first degree murder after a jury trial in a Pennsylvania state court. Following unsuccessful direct appeal and post-conviction relief proceedings, Siehl instituted this habeas corpus proceeding in the District Court alleging ineffective assistance of tidal and appellate counsel. The District Court denied relief, and this appeal followed. We will reverse and remand to the District Court for an evidentiary hearing.

I.

On July 14, 1991, Christine Siehl’s landlord received complaints that water was running out of the apartment building. He entered Ms. Siehl’s apartment and found her body in the bathtub with the shower running. She had multiple stab wounds which resulted in her death. The time of death was estimated to be between the hours of 11:00 p.m. on July 12 and 3:00 or 4:00 a.m. on July 13. Based on testimony from the mother of a neighbor who heard commotion in the apartment at approximately 1:30 a.m. on July 13, the Commonwealth argued that she was killed at approximately that time.

The bathroom contained indications of a struggle between the victim and the murderer, including blood on the walls and floor, broken mirror pieces, and scattered cat litter. There was no sign of forced entry. Potential suspects in the criminal investigation were: (1) Kevin Siehl, who was married to but living separately from the victim; (2) Frank Wills, with whom the victim had been romantically involved while married to Siehl; and (3) Robert Prebehalla, who told people of his hatred for the victim. According to Sergeant Angelo Cancelliere, Siehl became the prime suspect because a fingerprint on the show-erhead and a blood sample from the scene were determined to match Siehl’s.

Siehl was arrested and charged with first degree murder, third degree murder and involuntary manslaughter. Public defenders David Weaver and Linda Fleming were appointed by the trial court to represent Siehl. They promptly persuaded the court to appoint a forensic expert to assist them in the defense of their client. For reasons that will hereafter become clear, we will refer to their chosen expert as “John Smith.” Smith promptly provided counsel with a page and a half “PRELIMINARY ANALYSIS,” “the purpose [of which was] to give[ ] an explanation ... as to how this crime was committed and to explain certain items of physical evidence.” App. at 73. He determined, after reviewing a photograph of the latent print and the fingerprint card of Siehl, that the latent print from the showerhead matched Siehl’s. Smith’s full discussion of the print was as follows:

This print does match the .rolled inked impression on the finger print card bearing the name of Kevin Charles Siehl. See lift photograph and finger print card. It can be stated however, that the print is an exceptionally clear print and not smudged, as one would expect to find in a homicide scenario such as this one. The other thing about this print that is unusual, is that microscopic examination of the shower head, where the print was developed, shows no trace evidence of blood which one would expect to And due to the nature of the crime. It also can be stated that no time frame can be placed on this print as to when it was made. The alleged suspect, Kevin Siehl, had access to this apartment prior to the commission of the crime, therefore, the print could have been made well before the homicide occurred.

App. at 75. Smith’s “PRELIMINARY ANALYSIS” did not explain the basis for his preliminary conclusion that the print belonged to Siehl. While it commented on three of the bloodstain evidence items out of the eighty items tested, he did not test the blood evidence and made no findings with respect to the bloodstain which the Commonwealth would maintain was consistent with Siehl’s blood. Smith did not prepare any other reports, nor did he testify at trial.

During opening statements, the Com-' monwealth emphasized that the fingerprint was Siehl’s, was a direct piece of evidence that tied Siehl to the murder scene, and was in a position which would indicate that Siehl was outside the shower when the fingerprint was made. The prosecutor also told the jury to pay attention to testimony regarding blood evidence consistent with Siehl’s blood type found on the bathroom doorframe.

At trial, Trooper Merril Brant testified that the latent fingerprint on the shower-head matched that of Siehl, and that the position of the print led him to conclude that it was not left by someone showering, but rather was left by the murderer who was standing outside the tub and directed the shower onto Ms. Siehl’s body. Brant used the showerhead to demonstrate this theory for the jury. Brant further testified that the print had not yet started to deteriorate, and that therefore it must have been left within 24-36 hours of when the victim was discovered. Prior to this testimony, the jury was read a stipulation of the parties that the fingerprint found on the showerhead belonged to Siehl. Defense counsel did not present expert testimony to counter Brant’s testimony regarding the timing and position of the fingerprint.

At trial, Scott Ermlick, the state crime lab supervisor, testified as a serological expert. Ermlick testified that one of the twelve bloodstains recovered from the bathroom was consistent with Siehl’s blood group markers, and that none of the blood was consistent with those of Wills or Pre-behalla. The stain, which he testified was consistent with Siehl’s blood group markers, was one of two small spatters found side-by-side on the bathroom doorframe. He identified the other spatter as consistent with Ms. Siehl’s blood group markers. Because the sample was small, no DNA or follow-up testing could be performed. Defense counsel did not present expert testimony to counter Ermlick’s findings.

Siehl presented an alibi defense. His father and brother, with whom he was living at the time, and his parents’ neighbor all testified. They stated that Ms. Siehl dropped Siehl off at his parents’ home at approximately 1:30 a.m. on July 13, and then drove away while Siehl remained at his parents’ home for the remainder of the evening.

Siehl was convicted of first degree murder. The jury unanimously decided that a life sentence would be imposed. Siehl then retained new counsel, Terry Despoy, who represented him on direct appeal. The Pennsylvania Superior Court affirmed the judgment of conviction. A petition for allowance of appeal to the Pennsylvania Supreme Court was denied.

Siehl next filed a petition for post-conviction relief in the Court of Common Pleas. With new counsel in that proceeding, he raised and sought an evidentiary hearing on the issues of (1) whether trial counsel were ineffective for stipulating that the fingerprint belonged to Siehl and for failing to secure the assistance of a competent forensic expert at trial, and (2) whether appellate counsel was ineffective for not raising those issues on appeal. In response to the Commonwealth’s argument that Siehl had waived his claims of ineffective assistance of trial counsel by failing to raise them on direct appeal, Siehl insisted that this was a situation involving ineffective assistance of appellate counsel as well and, thus, “layered claims of ineffectiveness.” He argued that under Commonwealth v. Sawyer, 355 Pa.Super. 115, 512 A.2d 1238 (1986), he was entitled to an evidentiary hearing on the ineffectiveness of both trial and appellate counsel. The Court rejected this argument and concluded that Siehl’s claims of ineffectiveness of trial counsel had been waived by his failure to raise them on direct appeal. The Court agreed to an evidentiary hearing limited to only two issues: one not here relevant and the competence of Smith.

At the first evidentiary hearing, trial attorney Fleming testified only on direct examination by Siehl. She testified that she did not recall whether Smith tested either the blood or the fingerprint evidence. She also testified that Smith did not prepare a report for counsel prior to trial, and that she did not recall whether Smith was an expert in serology. Fleming’s direct testimony was cut off by the Court because it concluded that counsel was addressing the prohibited ineffective assistance of counsel issues. Accordingly, the Commonwealth was not afforded an opportunity to cross-examine her. A second evidentiary hearing was held at which the Court again cut short Siehl’s evidence. It declined to hear any testimony from Professor Herbert Leon MacDonell, a highly credentialed forensic expert. 1 Siehl was, however, permitted to make an offer of proof. The proffer summarized MacDo-nell’s affidavit in which he opined that: (1) the developed latent fingerprint from the showerhead did not come from Siehl; (2) Brant’s fingerprint deterioration theory was incorrect, and it was impossible to determine how long the print had been on the showerhead; (3) the two small blood spatter stains came from the same source and the probability of the two having come from two different individuals was “so unlikely that for all intent and purpose it is an impossibility” (App. at 52); and (4) Smith was not qualified in any field of forensic science, including fingerprint identification and bloodstain pattern interpretation, to testify as an expert. 2 The Court denied relief and Siehl filed an appeal raising the same issues of ineffective assistance of trial and appellate counsel. The Superior Court of Pennsylvania affirmed.

The Superior Court addressed the merits of Siehl’s ineffective assistance of trial counsel claims. It recognized the validity of the concept of “layered” ineffective assistance of counsel claims, but it had no occasion to express an opinion on the waiver issue. 3 It recognized that the control ling principles were those endorsed by the Supreme Court in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). With respect to the claim that counsel was ineffective for stipulating that the fingerprint on the shower-head belonged to Siehl, the Superior Court assumed arguendo that the print was not Siehl’s but concluded that counsel had a reasonable basis for making the stipulation and that Siehl had failed to establish prejudice. Specifically, the Court found:

[T]he record reflects that counsel had a reasonable basis for making the stipulation. The record reflects that Appellant did not deny, at that time, that the fingerprint was his. Thus, trial counsel had a reasonable basis for making this stipulation as part of his strategy.

[Moreover] we observe that no prejudice has been demonstrated. The murder victim was Appellant’s ex-wife or estranged wife, in whose apartment Appellant often visited. A police officer testified that Appellant told the police that ‘he showered there [at Christine’s apartment] many times and that his fingerprints would be all over the place.’

... Thus, even before his arrest, Appellant had explained the presence of his fingerprints in the victim’s apartment. Appellant fails to establish prejudice due to the stipulation because, if trial counsel had not made the stipulation, there is no probability that the outcome of the trial would have been different.

App. at 24-25 (citation omitted).

With respect to the claim that counsel was ineffective because they did not obtain a qualified forensic expert to assist them at trial, the Court determined only that Siehl could not demonstrate prejudice:

The record is clear that Appellant was a frequent visitor to the apartment and that his fingerprints were throughout the apartment. Appellant fails to demonstrate that a challenge to the fingerprint evidence or the retention of a qualified forensic expert would have affected the outcome of the trial.

App. at 25.

Thereafter, Siehl instituted this habeas proceeding in the District Court. The Magistrate Judge’s Report and Recommendation addressed the merits of Siehl’s ineffective assistance of trial counsel claims and recommended denial of the petition and of a certificate of appealability because Siehl failed to show that the state court’s adjudication of the claims was contrary to or an unreasonable application of clearly established federal law. The District Court adopted the Report and Recommendation, and this appeal followed.

This Court granted the following certificate of appealability:

The application for a certificate of ap-pealability is granted as to these issues: whether appellant’s Sixth Amendment right to effective assistance of counsel was violated (1) where trial counsel stipulated that the fingerprint on the show-erhead belonged to the appellant without having the assistance of a qualified fingerprint expert to examine the fingerprint and render a competent scientific opinion; (2) where he did not have qualified forensic experts in fingerprint identification and bloodstain pattern analysis to assist him; and (3) where post-trial and appellate counsel failed to raise the issue of trial counsel’s ineffectiveness for stipulating to the fingerprint without having the assistance of a qualified fingerprint expert to examine the finger print and render a competent scientific opinion. The parties also should address whether the Pennsylvania Superior Court’s determination of these Sixth Amendment issues was based on an unreasonable application of Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), or on an unreasonable determination of the facts, see 28 U.S.C. § 2254(d)(1), (2), and whether appellant is entitled to a hearing in the district court on these issues, Campbell v. Vaughn, 209 F.3d 280, 287 (3d Cir.2000).

App. at 98-99.

II.

Under the Anti-Terrorism and Effective Death Penalty Act (“AEDPA”), habeas relief on behalf of a person in custody pursuant to a judgment of a state court cannot be granted with respect to any claim that was adjudicated on the merits in state court proceedings unless the decision is contrary to, or involves an unreasonable application of, clearly established federal law, or is based on an unreasonable determination of the facts in light of the evidence presented in the state court proceeding. 28 U.S.C. 2254(d). AEDPA thus limits a federal court’s authority to grant habeas relief when a state court has previously considered and rejected the federal claim on the merits.

“Under the ‘unreasonable application’ clause [of § 2254(d)], a federal ha-beas court may grant the writ if the state court identifies the correct legal principle from [the Supreme] Court’s decisions but unreasonably applies that principle to the facts of the prisoner’s case.” Jermyn v. Horn, 266 F.3d 257, 281-82 (3d Cir.2001) (quoting Williams v. Taylor, 529 U.S. 362, 407, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000)). In determining whether the state court unreasonably applied Supreme Court precedent, the question is whether the state court’s application of federal law was objectively unreasonable, not whether the application was, in the judgment of the federal habeas court, erroneous or incorrect. Id. at 282.

“Sixth Amendment claims of ineffective assistance of counsel are governed by Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), which qualifies as clearly established Federal law, as determined by the Supreme Court of the United States.” Taylor v. Horn, 504 F.3d 416, 430 (3d Cir.2007) (internal quotation marks and citation omitted). According to Strickland, a court deciding an ineffectiveness claim must “determine whether, in light of all the circumstances, the identified acts or omissions [of counsel] were outside the wide range of professionally competent assistance.” 466 U.S. at 690, 104 S.Ct. 2052. Since Strickland, the United States Supreme Court and this Court have emphasized the necessity of assessing an ineffectiveness claim in light of all the circumstances. See Taylor, 504 F.3d at 430; Jacobs v. Horn, 395 F.3d 92, 107 (3d Cir.2005) (citing Supreme Court and Third Circuit cases).

Under Strickland, a petitioner must also show that counsel’s deficient performance prejudiced the defense. See Taylor, 504 F.3d at 430. This requires a defendant to show “ ‘that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.’ ” Id. (quoting Strickland, 466 U.S. at 694, 104 S.Ct. 2052).

III.

We must now determine (1) whether the Pennsylvania Superior Court’s rejection of Siehl’s ineffective assistance of trial counsel claims involved an unreasonable application of Strickland, and (2) if so, whether Siehl’s proffer of evidence, including the MacDonell report, together with the other evidence of record, entitles Siehl to an evidentiary hearing on those claims in the District Court.

A. The Superior Court Decision

The Superior Court found as a fact that, at the time counsel stipulated regarding the fingerprint on the showerhead, Siehl “did not deny ... that the fingerprint was his.” App. at 24. We, of course, must and do accept that finding. In the Court’s view, this meant that counsel had a “reasonable basis” for making the stipulation. Id. at 24. The reasonableness of this view, and the Court’s further view that there was no prejudice from the stipulation and from the failure to secure an expert to assist at trial, depends on the soundness of a single proposition — Siehl’s failure to challenge Brant’s identification of the fingerprint cannot have prejudiced Siehl because he “was a frequent visitor to the apartment and ... his fingerprints were throughout the apartment.” App. at 25. This foundational proposition was not an objectively reasonable one, however, because it wholly ignored the core of the Commonwealth’s case which trial counsel were well aware of from the preliminary hearing, the pretrial proceedings, and the Commonwealth’s opening statement.

The forensic evidence core of the Commonwealth’s case was such that the failure to challenge it would likely lead the jury to conclude not just that Siehl had on some occasion been in the bathroom, but also that (1) he had been in the victim’s bathroom within 24 hours of the discovery of the fingerprint; (2) he had stood outside and beside the tub and directed the show-erhead toward the place where the victim’s body was found lying in the tub; (3) during his violent struggle with the victim in the bathroom, his blood and hers spattered together on the bathroom doorframe; and (4) none of the 20 items in the bathroom that tested positive for blood was consistent with the blood of the two other suspects.

While trial counsel cross-examined the Commonwealth’s forensic evidence witnesses, they did so without the advice of a forensic expert, and the defense countered with no forensic evidence of its own. Indeed, counsel failed to seek additional forensic assistance even after Smith’s “preliminary analysis” had alerted them to the fact that Brant’s crucial 24 hour print aging testimony was probably unsound.

In short, given the Commonwealth’s expected testimony regarding the age and position of the print, and the position and character of the blood samples found in the apartment, any decision to stipulate that the print was Siehl’s without an intention to counter that expected testimony was ineffective because it effectively admitted that he was the murderer. Thus, assessing the ineffective assistance claim in light of all the circumstances, we conclude that the Superior Court’s application of Strickland in this case was not objectively reasonable and that the District Court was entitled to review the record de novo.

B. The Case for an Evidentiary Hearing

An adequate record upon which to evaluate trial counsels’ performance in this case has not yet been developed. Accordingly, a final judgment about their performance cannot be rendered at this time. We can and must make a judgment, however, about whether Siehl has shown enough to entitle him to an opportunity to create the necessary record. This involves two issues: (1) whether this is a situation in which AEDPA bars an evidentiary hearing in the federal habeas proceeding; and (2), if not, whether Siehl has proffered sufficient evidence to demonstrate that “a new hearing would have the potential to advance the petitioner’s claim.” Campbell v. Vaughn, 209 F.3d 280, 286-87 (3d Cir.2000).

“AEDPA and uniform case law interpreting it provide that if the habeas petition ‘has diligently sought to develop the factual basis of a claim for habeas relief, but has been denied the opportunity to do so by the state court, [AEDPA] will not preclude an evidentiary hearing in federal court.’ ” Id. at 287 (quoting Cardwell v. Greene, 152 F.3d 331, 337 (4th Cir.1998)). Here, Siehl has shown that he diligently sought and was denied an evidentiary hearing on the relevant issues before the Court of Common Pleas in his post-conviction relief proceeding. While this would not be a sufficient showing for purposes of AEDPA if the waiver issue had been finally resolved against him, we conclude that this was a sufficient showing given that Siehl may be able to show that there was no waiver because appellate counsel was ineffective in failing to raise the ineffectiveness of trial counsel on direct appeal. 4

Turning to the second issue, we also conclude that Siehl has shown enough to demonstrate that an evidentiary hearing would have the potential to advance his claim to habeas relief. The record evidence and the evidence tendered by Siehl, including the MacDonell report, if credited, would suggest that he received ineffective assistance of counsel and that “ ‘there is a reasonable probability that, but for counsel’s unprofessional errors, the result of [his trial] would have been different,’ ” i.e., a probability “ ‘sufficient to undermine confidence in the outcome.’ ” Hull v. Kyler, 190 F.3d 88, 110 (3d Cir.1999) (quoting Strickland, 466 U.S. at 694, 104 S.Ct. 2052). This prima facie showing is sufficient to entitle him to an evidentiary hearing in the District Court if through no fault of his own he was unable to establish the necessary record in the state courts.

Counsel’s duty to investigate does “not force defense lawyers to scour the globe,” and limited investigation is reasonable where counsel has good reason to think further investigation would be wasteful. Rompilla v. Beard, 545 U.S. 374, 383, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005). Here, even if counsel may have had good reason to think that further investigation as to the identity of the print would be wasteful, 5 they apparently did not have good reason to think that further expert opinion on the timing of the print would not be helpful. Moreover, because Smith’s preliminary report did not address the bloodstain evidence that was said to impli cate Siehl, it would appear unreasonable that counsel had no expert opinion regarding that bloodstain evidence, the only other physical evidence linking Siehl to the murder. Although counsel’s strategic choices made after full investigation are “virtually unchallengeable,” Strickland, 466 U.S. at 690, 104 S.Ct. 2052, in the absence of some explanation not found in the current record, the strategic choices of counsel here would appear to have been made without a full investigation. If so, they were not reasonable in light of the circumstances and facts known to counsel at the time.

Based on the current record, we find the situation before us much like that presented to the Sixth Circuit Court of Appeals in Richey v. Bradshaw, 498 F.3d 344 (6th Cir.2007). There, the state maintained that Richey had deliberately set fire to a house, using accelerants, thereby occasioning the death of a child. There were no eyewitnesses, and there was some evidence suggesting that the fire was caused by careless smoking. Accordingly, the state’s scientific evidence of arson was fundamental to its case. Counsel retained an expert to evaluate that scientific evidence, and the expert advised that he agreed with the opinions of the state’s experts. Counsel accordingly did not challenge the state’s scientific evidence of arson. The Court found his performance deficient:

The scientific evidence of arson was thus fundamental to the State’s case. Yet Richey’s counsel did next to nothing to determine if the State’s arson conclusion was impervious to attack. True, Richey’s counsel retained [an expert] to review the State’s arson evidence, so this case does not exemplify that most egregious type, wherein lawyers altogether fail to hire an expert. But the mere hiring of an expert is meaningless if counsel does not consult with that expert to make an informed decision about whether a particular defense is viable.

Having been simply served up with [the expert’s] flat agreement with the State, and not having known either what [the expert] did to arrive at his conclusion or why he came out where he did, [counsel] was in no position to make this determination.

Id. at 362, 363.

Siehl’s prima facie case is also sufficient to suggest that an evidentiary hearing would enable him to demonstrate a reasonable probability that, but for the ineffectiveness of trial counsel, the result of his trial would have been different. As Siehl stresses, the only direct evidence placing him at the scene of the murder at the relevant time was the fingerprint and bloodstain evidence, and the Commonwealth’s evidence with respect to those matters, in the absence of contradictory evidence, was strongly supportive of Siehl’s guilt. MacDonell’s evidence would establish that competent counsel, through Smith or another expert retained to supplement his limited assistance, would have been able to show that the showerhead fingerprint evidence simply did not place Siehl at the scene at the relevant time. MacDonell’s testimony would also indicate that competent counsel would have been able to undermine the Commonwealth’s bloodstain evidence by showing that the bloodstain Ermlick identified as consistent with Siehl’s blood came from the same source as the bloodstain identified by him as coming from the victim. Undermining the Commonwealth’s only direct evidence would also have fortified Siehl’s alibi defense and underlined the fact that the Commonwealth offered very little in the way of motive.

IV.

We will reverse the judgment of the District Court and remand for an eviden-tiary hearing on Siehl’s ineffective assistance of counsel claims identified in our certificate of appealability.

1 . Professor MacDonell has taught college level criminalistics — i.e., "application of science to the investigation of crime which is primarily concerned with the examination of physical evidence and how it can be used to reconstruct prior events” — since 1960. App. at 41. He is a Fellow in the American Academy of Forensic Science and a past chairman of the Criminalistics Section of that Society. He has testified as an expert in forensic disciplines in thirty-five states and five foreign countries. He was a member of a select committee of the International Association for Identification which produced a report promulgating minimum requirements for "friction ridge identification,” a report that has been accepted by every major identification bureau in the world. App. at 43. Professor MacDonell is certified by the International Association of Identification as a Senior Crime Scene Analyst.

2 . Professor MacDonell’s affidavit indicates that Smith had been a student of his and that Professor MacDonell was familiar with his work in this and other cases. His critique of Smith's work went far beyond disagreement with Smith’s analysis and conclusions. He pointed out, for example, that Smith's curriculum vitae described his "current occupation” as a "Forensic Reconstruction Consultant” and indicated that this title is “not recognized as a forensic discipline by any forensic organization” of which Professor MacDonell was aware. App. at 46. Professor MacDonell's affidavit also expressed his opinion that Smith had "repeatedly overstated his qualifications to such a degree that his errors are far beyond simple carelessness.” App. at 50.

3 .Given its conclusion that there was no ineffective assistance of trial counsel, the Superi- or Court impliedly determined that there was no ineffective assistance of appellate counsel for failing to contend the contrary. The respondent has not argued before us that Siehl's ineffective assistance of trial counsel issues have been conclusively determined to have been waived. There appears to be no dispute that these ineffective assistance of trial counsel claims were not raised on Siehl's behalf on direct appeal. However, if it is ultimately determined that Siehl’s ineffective assistance of trial counsel claims have merit, it is likely that a similar determination will be made with respect to the claims regarding appellate counsel, in which event there would be no waiver under Pennsylvania’s "layering of ineffective assistance” line of cases, see Commonwealth v. Duffey, 579 Pa. 186, 855 A.2d 764, 768 (2004), and no procedural default by virtue of the doctrine of cause and prejudice. See Murray v. Carrier, 477 U.S. 478, 488-89, 106 S.Ct. 2639, 91 L.Ed.2d 397 (1986) (ineffective assistance of counsel is a cause for procedural default).

4 . See supra note 3.

5 . We say “may have had” because Siehl insists that what little assistance trial counsel received from Smith contained numerous red flags which should have led them to realize that he was not competent to express an opinion on the origin of the fingerprint. With the hindsight benefit of MacDonell’s report, the red flags were indeed numerous. Even without that report, however, some of those red flags were sufficiently noticeable that further inquiry into counsels’ use and consideration of Smith’s “preliminary analysis” is clearly appropriate. As we have noted, that analysis contained no comparative analysis of the fingerprints, set forth conclusory findings with no scientific explanation or discussion, and failed to discuss the allegedly incriminating blood evidence.