dissenting.
{¶ 58} The majority’s decision to deny Romell Broom an evidentiary hearing on his Eighth Amendment claim is wrong on the law, wrong on the facts, and inconsistent in its reasoning. The majority adopts the trial court’s conclusion that an evidentiary hearing is unnecessary because “[t]here is no dispute as to any operative fact in connection with the events of September 14 and 15, 2009.” Majority opinion at ¶ 31. And then, with no apparent awareness of the contradiction, the majority rejects Broom’s request for a hearing precisely because it finds an unresolved dispute of fact at the heart of the case, namely, the reason for the state’s inability to establish IV access at the start of Broom’s attempted execution.
{¶ 59} The court today holds that Broom is not entitled to an evidentiary hearing to prove his Eighth Amendment claim because he failed to prove his claim at the pleading stage; that is not the correct legal standard under R.C. 2953.21(E). The court then compounds its error by seeking evidence outside the *75record to prove that the state has cured the problems in its execution procedures and seizing on an execution protocol that is no longer in effect.
{¶ 60} For these reasons, I dissent.
LEGAL FRAMEWORK
{¶ 61} A method of execution violates the Eighth Amendment if it presents a risk that is sure or very likely to cause serious illness and needless suffering and give rise to sufficiently imminent dangers. Glossip v. Gross, — U.S.—, 135 S.Ct. 2726, 2737, 192 L.Ed.2d 761 (2015). “[T]here must be a ‘substantial risk of serious harm,’ an ‘objectively intolerable risk of harm’ that prevents prison officials from pleading that they were ‘subjectively blameless for purposes of the Eighth Amendment.’ ” Baze v. Rees, 553 U.S. 35, 50, 128 S.Ct. 1520, 170 L.Ed.2d 420 (2008) (plurality opinion), quoting Farmer v. Brennan, 511 U.S. 825, 842, 846, 114 S.Ct. 1970, 128 L.Ed.2d 811 (1994), and fn. 9. The controlling opinion in Baze recognized that a series of abortive execution attempts might demonstrate an objectively intolerable risk of harm. Id.
{¶ 62} The majority opinion offers two reasons why Broom supposedly failed to satisfy the Glossip/Baze formula; both are unpersuasive.
1. Broom’s evidence of a risk of serious harm
{¶ 63} After reviewing the evidence from Broom’s postconviction petition, the court opines that.it is “unclear from the record why Broom’s execution team was unable to establish IV access.” Majority opinion at ¶ 47. The logic appears to be that if he cannot prove why the errors occurred, then he cannot prove that they will recur.
{¶ 64} The court’s reasoning rests on a misapplication of a petitioner’s burden of proof when filing a postconviction petition. Before granting a hearing on a petition, a trial court must consider whether it presents “substantive grounds for relief.” R.C. 2953.21(C). A trial court acts within its discretion when it dismisses a petition without a hearing if the evidence and affidavits submitted by the petitioner fail to set forth “ ‘sufficient operative facts to establish substantive grounds for relief.’ ” State v. Gondor, 112 Ohio St.3d 377, 2006-Ohio-6679, 860 N.E.2d 77, ¶ 51, quoting State v. Calhoun, 86 Ohio St.3d 279, 714 N.E.2d 905 (1999), paragraph two of the syllabus.
{¶ 65} The requirement that a petitioner present “sufficient operative facts” in order to secure an evidentiary hearing does not mean that the facts presented must be undisputed. The court’s application of R.C. 2953.21(C) effectively eliminates the possibility of an evidentiary hearing in every case: either the petition fails to establish substantive grounds for relief, in which case the *76petitioner is not entitled to a hearing, or the evidence and affidavits establish grounds for relief beyond dispute, in which case an evidentiary hearing would be superfluous.
{¶ 66} When considered in its totality, Broom’s petition contained sufficient operative facts to warrant a hearing. The evidence in the record, if believed, would establish that the state has repeatedly and predictably had problems establishing and maintaining access to inmates’ veins, that these problems are the result of medical incompetence on the part of the execution team members responsible for inserting IV catheters, and that the incompetence of the execution staff makes it more than likely that these problems will recur in future executions.
{¶ 67} As the majority correctly notes, the execution team unsuccessfully inserted IV catheter needles into at least 18 separate injection sites on Broom’s body, and the actual number of catheter insertions was much greater than the number of needle marks because injection sites were used multiple times. But the majority gives scant attention to the testimony of Dr. Heath concerning the magnitude of medical incompetence displayed or to evidence of the state’s long, problematic history with IV catheters in lethal-injection procedures.
{¶ 68} As Dr. Heath explained, execution drugs are administered through a peripheral IV catheter, which consists of a hollow needle surrounded by a plastic tube called the catheter. The practitioner inserts both the needle and the catheter into the vein. Thus, although blood draws and TV catheters both involve needle pricks, the latter also includes the insertion of an object with a wider diameter.
{¶ 69} Dr. Heath testified that in Broom’s case, the medical team inserted the catheters improperly and in a fashion likely to cause greater pain:
And what [Broom] described did not sound right to me at all in terms of proper procedure. He described the depth into his muscle which they were pushing the catheter, going all the way up to the hub of the catheter, which is well over an inch deep, and going in at a very steep angle, which does not comport with proper technique or what’s going to be a successful technique for getting a catheter into a vein.
Q: And what kind of — what kind of response would a human body have to that kind of technique?
A: It would hurt.
The fact that the medical team attempted to insert the catheter into Broom’s arm in a nearly perpendicular line “bespeaks of bizarre and certainly poor technique” *77that had “no possibility of achieving insertion.” Even the Lucasville physician inserted the catheter into Broom’s ankle at the wrong angle, which is why she struck bone.
{¶ 70} On two occasions, after getting the catheter into Broom’s veins, team members “botched up that relatively simple process of attaching the I.V. tubing into the hub of the catheter, and blood sprayed everywhere.” Compounding the problem, the team member who inserted the catheter was not wearing gloves during the procedure, thereby increasing the risk of unsuccessfully attaching the tubing.
{¶ 71} Dr. Heath testified that clinical competence demands that attempts at peripheral IV access be abandoned long before the 19th attempt. The Lucasville physician agreed with this assessment, testifying that if she had known that 18 attempts had already been made on Broom, she never would have stuck him with a needle. And although it can be proper practice to “físh” a vein — partially withdrawing a catheter, then reinserting it at a different angle — it was “completely unacceptable [and] far beyond any acceptable standard” to have made as many insertion attempts through the same needle holes as the medical team apparently did.
{¶ 72} The Broom execution attempt was not the first time the medical execution team had difficulty inserting an IV catheter. During Ohio’s very first lethal injection, that of Wilford Berry in 1999, the medical team members responsible for the IV catheter line were unable to locate a vein in Berry’s arm and had to call a third team member for assistance. When the state executed Christopher Newton in 2007, the medical team needed approximately 90 minutes to establish two IV lines. And during the 2009 execution of Marvallous Keene, the team had to switch to a backup line because they could not get fluid to flow.
{¶ 73} But the most disastrous execution prior to the attempted execution of Broom was that of Joseph Clark on May 2, 2006. Based on the end-of-the-day needle count in Clark’s case, the execution team used 17 or 18 needles trying to establish two IV lines and never did manage to install the backup line before ultimately executing him. As in Broom’s case, Dr. Heath was critical of the decision in Clark’s case to continue fruitlessly trying to establish the IV, testifying that he would not subject a patient to more than six to ten attempts, because “any normal practitioner” will recognize by that point that the effort is futile and is only inflicting needless pain on the patient.
{¶ 74} Dr. Heath offered a simple explanation for the state’s failures: incompetence. “[I]t is my opinion that the veins on Mr. Broom’s arms, other areas of his body, should be easily accessible by a competent team.” Part of the problem, he opined, arose from the fact that Ohio had lax standards regarding who may install a peripheral IV catheter during an execution. The person inserting the *78peripheral IV catheter should be someone who performs that task as part of his or her daily job duties. But Ohio’s protocol required only one year of experience, with no limitation on how remote in time that experience might be.
{¶ 75} Team Member 9, who participated in the Broom procedure, is a phlebotomist. But drawing blood, which is what phlebotomists do, is a “much easier and simpler” process that is “far less fraught with problems” than inserting a catheter in order to inject drugs into the body. Dr. Heath testified that allowing a phlebotomist such as Member 9 to handle the insertion at an execution is “deeply inappropriate and reflects a deep misunderstanding and deep inability of what is needed to assemble an appropriate [execution] team.”
{¶ 76} Nor was severe pain from inserting the IV catheter the only potential harm Dr. Heath identified. During the execution of Joseph Clark, a process known as infiltration occurred. “Infiltration” means the catheter needle is no longer in the vein, and the drug is being administered into tissue surrounding the vein. Infiltration of sodium thiopental may cause “very significant pain,” either due to a pH imbalance or from the sudden accumulation of fluid in the tissue.
{¶ 77} The protocol in effect at the time of Broom’s attempted execution called for administration of sodium thiopental as the first of three drugs. The current protocol, effective June 29, 2015, permits the use of sodium thiopental at the discretion of the warden. Ohio Department of Rehabilitation and Correction, Policy No. 01-COM-ll, at 15, http://www.drc.ohio.gov/web/drc_policies/ documents/Ol-lCOM-ll.pdf (“Current Protocol”) (accessed Feb. 29, 2016). But there is no evidence in the record to indicate whether infiltration is a sign of negligence or merely an unavoidable risk, even if performed by the most skilled practitioners.
{¶ 78} In the face of this evidence, it is disingenuous to dismiss Broom’s petition on the grounds that it is “unclear” why the execution team was unable to establish IV access. The majority opinion cites no evidence in the record for an alternative explanation of the state’s failure, nor could it, given that the state failed to submit any evidence. And if the majority truly believes that the cause is unclear, that is all the more reason to have an evidentiary hearing to resolve the question.
{¶ 79} Nor does the majority explain its conclusory assertion that Broom failed to establish a likelihood that he will suffer severe pain in the future. There is no evidence in the record to suggest that the state has improved the training of its execution team members or increased the certification requirements to perform these tasks. In fact, Ohio’s current execution protocol still requires only one year *79of experience.3 If the state cannot explain why the Broom execution went wrong, then the state cannot guarantee that the outcome will be different next time.
2. Ohio’s current execution protocol
{¶ 80} In an attempt to bolster its conclusion, the majority reaches outside the record for evidence that the state has improved its execution procedures. Specifically, it cites the Sixth Circuit’s decision in Cooey v. Strickland, 589 F.3d 210 (6th Cir.2009), for the proposition that Ohio’s current execution protocol passes constitutional muster. Therefore, the majority asserts, “we cannot assume that the same problems with IV access will befall Broom again.” Majority opinion at ¶ 48. This argument misunderstands Cooey and improperly shifts the burden of proof.
{¶ 81} As explained in Cooey, in the aftermath of Broom’s attempted execution, the state made two changes to its execution protocol. First, it switched from a three-drug protocol to a one-drug protocol. Id. at 215. But since the one drug was still administered intravenously, id. at 219, this change did nothing to remedy the problem of incompetent insertion of IV catheters.
{¶ 82} Second, the new protocol authorized a two-drug, intramuscular injection as a backup procedure if the execution team cannot obtain IV access. Specifically, if the warden and director determine that IV injections cannot or should not be used, then the state may administer an injection of midazolam and hydromor-phone intramuscularly. Id. at 220. But in January 2015, five years after Cooey, the state removed midazolam and hydromorphone from its execution protocol.4 And that leaves the state back where it began, with a protocol that mandates intravenous injection and provides no intramuscular backup.5
{¶ 83} The majority opinion notes that the amended protocol added “a new command structure and forms that were required to be filled out as each step of the protocol was completed.” Majority opinion at ¶ 51. It cites In re Ohio Execution Protocol Litigation (Wiles), 868 F.Supp.2d 625 (S.D.Ohio 2012), for the proposition that the state “sufficiently demonstrated a commitment to follow the protocol” to persuade the federal court to allow executions to proceed. Majority opinion at ¶ 51. And it concludes by announcing that this court is “not convinced * * * that Broom has established that the state is likely to violate its execution protocol in the future.” Id. at ¶ 49. The majority’s reasoning is faulty for three reasons.
*80{¶84} First, the majority opinion does not explain how the new -command structure in the amended protocol will remedy the problem of medical incompetence identified in Broom’s petition. Additional paperwork will not improve the execution team’s ability to insert an IV catheter.
{¶ 85} Second, Wiles addressed whether the state was violating the Equal Protection Clause in its administration of its lethal-injection protocol, not the Eighth Amendment. 868 F.Supp.2d at 636-637. The district court observed “a consistent pattern of arbitrary deviations from the protocol.” Id. at 640. In the context of an Equal Protection claim, the state’s assurances that it would strictly comply with written procedures in the future was relevant. But Wiles is irrelevant to Broom’s claim under the Eighth Amendment.
{¶ 86} Third, the majority’s assertion that Broom failed to prove that the state will deviate in the future is deeply unfair. Assuming that the written protocol is even relevant (which it is not), the state’s assurance of future compliance is in essence an affirmative defense. The majority is effectively shifting the burden of proof by faulting Broom for not rebutting evidence that the state did not even introduce into the record. And when exactly was Broom supposed to introduce such evidence? All the evidence in the record comes from the postconviction petition filed September 15, 2010. The majority relies on subsequent events and documents not contained in the record. If the majority truly believes that recent events are relevant to adjudicating Broom’s petition, the answer is to remand the petition for an evidentiary hearing, not for this court to make its own assessment without a record or input from the parties.
CONCLUSION
{¶ 87} We should remand Broom’s case to the trial court for an evidentiary hearing. With this disposition, it is unnecessary and premature for the court to address any other legal questions. I dissent.
Pfeifer, J., concurs in the foregoing opinion.. See Current Protocol at 2.
. Ohio Department of Rehabilitation and Correction, Ohio Revises Lethal Injection Protocol, http:// www.drc.ohio.gov/Public/press/press436.htm (accessed Feb. 29, 2016).
.See Current Protocol at 15-16.