concurring in result.
I agree with affirming Kopsho’s convictions and sentence of death. However, I disagree with the majority’s reliance on McWatters for the proposition that the prior act of kidnapping the victim was admissible and with its reliance on Owen in upholding the cold, calculated, and premeditated aggravator (CCP). I also write to express my concerns regarding the heinous, atrocious, or cruel aggravator (HAC), on which the jury was instructed, but which the trial court did not find.
In upholding the admission of evidence concerning the prior kidnapping of the victim, the majority relies on McWatters v. State, 36 So.3d 613 (Fla.2010), stating that it is “factually similar.” Majority op. at 213. However, McWatters certainly is not “factually similar” to this case. In McWatters, a third, similar homicide was *222admissible as similar fact evidence. Id. at 628. In that case, the defendant was convicted of three counts of first-degree murder and sexual battery for the strangulation murders of three women. Id. at 619. The defendant admitted to killing all three victims, but his defense was that “the killings ‘just happened’ when he ‘lost it’ during consensual sex.” Id. at 628. This Court concluded that the third, similar homicide was relevant to the issues of lack of consent, premeditation, intent, and absence of mistake, which are the traditional purposes for similar fact evidence. Id.
In this case, the prior kidnapping was clearly not similar fact evidence. Therefore, the focus of the inquiry is on whether it is relevant to the issue of premeditation; that is, whether the prior kidnapping is evidence that the murder in this case was not a heat-of-passion crime. And even if this evidence is probative, an important question still must be asked: whether the prejudicial effect of admitting the prior crime substantially outweighs its probative value.
Although McWatters does not support the view that the prior kidnapping was relevant, I recognize that the prior kidnapping may be relevant under the case law relied upon by the trial court below in allowing the evidence. See Spencer v. State, 645 So.2d 377 (Fla.1994) (holding that evidence of two prior incidents of violence by the defendant toward his wife in the months before the murder were relevant to premeditation where defendant claimed that it was a “heat of passion” killing); King v. State, 436 So.2d 50 (Fla.1983) (holding that evidence that twenty-three days before the murder, defendant had beaten the victim, with whom he was living, to the point that she became unconscious was “proper as evidence of premeditation”); see also Dennis v. State, 817 So.2d 741, 761 (Fla.2002) (holding that evidence that the defendant had “stalked, threatened, and assaulted” the victim, his ex-girlfriend, was evidence as to the nature of the defendant’s relationship with the victim and that it was relevant to establish motive).
However, trial courts should be cautious and ever-vigilant when admitting evidence of prior violent acts toward the victim under the rubric of premeditation. Admitting such evidence runs the risk that
the jury will convict the defendant based on prior crimes because these unrelated crimes would “go far to convince [individuals] of ordinary intelligence that the defendant was probably guilty of the crime charged. But, the criminal law departs from the standard of the ordinary in that it requires proof of a particular crime.”
Victorino v. State, 23 So.3d 87, 108 (Fla.2009) (Pariente, J., specially concurring) (quoting Smith v. State, 866 So.2d 51, 71 (Fla.2004) (Pariente, J., concurring in part and dissenting in part)). Further, in some cases, the evidence may only be relevant to demonstrate the violent propensity and bad character of the defendant. Of course, collateral crime evidence is inadmissible if it is “relevant solely to prove bad character or propensity.” Wright v. State, 19 So.3d 277, 292 (Fla.2009) (emphasis omitted).
Regardless of whatever relevance a pri- or crime may have to prove premeditation or another material fact in dispute, an important inquiry remains: whether the probative value is substantially outweighed by the prejudice. See § 90.403, Fla. Stat.; see also Hodges v. State, 885 So.2d 338, 358 (Fla.2004) (“Regardless of relevancy of collateral crime evidence ... admissibility is improper where the probative value of the evidence is substantially outweighed by undue prejudice.”). Trial courts should always carefully conduct this balancing in*223quiry when admitting collateral crime evidence, no matter how relevant to a material issue the crimes may be. Here, the majority fails to conduct this balancing test.
Although I disagree with the majority’s reliance on McWatters and question whether the prior kidnapping in this case was relevant, I concur in result because even if the evidence was inadmissible (either as irrelevant or because the probative value was substantially outweighed by prejudice), the admission of this evidence would be harmless beyond a reasonable doubt. Although the erroneous admission of irrelevant collateral crimes evidence “is presumed harmful error,” Robertson v. State, 829 So.2d 901, 918-14 (Fla.2002), under the unique facts of this case, there is no reasonable possibility that the error, if any, affected the verdict. State v. DiGuilio, 491 So.2d 1129, 1139 (Fla.1986). The testimony regarding the prior kidnapping was brief. Further, witnesses to the murder testified and described the murder as well as Kopsho’s actions and state of mind afterwards. Moreover, Kopsho confessed to the murder to the 911 operator and law enforcement. Although the issue of premeditation was disputed at trial, Kopsho stated in his confession to law enforcement that the murder was premeditated, admitted that he had begun planning the murder days beforehand, and described his actions in preparing for the murder. Additionally, the trial judge in this case proceeded cautiously, excluding the most inflammatory and irrelevant portion of testimony concerning the prior kidnapping and giving the jury a limiting instruction.
Next, I disagree with the majority’s holding that heightened premeditation — a requirement in order to find CCP — is proven in this case because Kopsho had the opportunity to choose whether to complete the intended murder after his original plan was interrupted, relying on Owen v. State, 862 So.2d 687, 701 (Fla.2003). In that case, we held that heightened premeditation was proven because:
When Owen first entered the home and saw the fourteen-year-old babysitter styling the hair of one of her charges, he had the opportunity to leave the home and not commit the murder. While he did exit the home at that time, he did not decide against killing Slattery. Instead, he returned a short time later, armed himself, confronted the young girl, and stabbed her eighteen times. Owen clearly entered the home the second time having already planned to commit murder.
Id. In short, the defendant left the scene and then returned to kill the victim. This clearly demonstrates a situation in which the defendant had ample opportunity to leave, but instead made the conscious decision to commit the murder, exhibiting heightened premeditation. See also Alston v. State, 723 So.2d 148, 162 (Fla.1998) (finding heightened premeditation where the defendant had ample opportunity to release the victim after the completion of a robbery). Care must be taken not to expand this prong of the CCP aggravator to apply in every case. The United States Supreme Court has held that in order for an aggravating circumstance not to be constitutionally infirm, it “may not apply to every defendant convicted of a murder; it must apply only to a subclass of defendants convicted of murder.” Tuilaepa v. California, 512 U.S. 967, 972, 114 S.Ct. 2630, 129 L.Ed.2d 750 (1994). As recently reiterated by that Court, “States must give narrow and precise definition to the aggravating factors that can result in a capital sentence.... Th[is] rule[ ] vindicated] the underlying principle that the death penalty is reserved for a narrow category of crimes and offenders.” Roper v. Sim*224mons, 543 U.S. 551, 568-69, 125 S.Ct. 1183, 161 L.Ed.2d 1 (2005). Under the reasoning of the majority, heightened premeditation is demonstrated whenever the defendant could have left the scene instead of murdering the victim. This is virtually every first-degree murder case.
Reliance on Owen is unnecessary. There is competent, substantial evidence in this case to support the trial court’s finding of CCP. Here, Kopsho even admitted that he clearly planned the murder in advance. Although the murder did not occur as planned, as recognized by the májority, “[t]he fact that Lynne’s escape from the truck forced Kopsho to modify his otherwise carefully prearranged plan does not negate the premeditated and calculated elements of CCP.” Majority op. at 215.
Finally, I write to express my concerns about the HAC aggravator, on which the jury was instructed, but which the trial court did not find. As I have previously explained, this causes difficulties because the trial court and this Court do not know whether the jury found HAC and how this may have affected the jury’s vote for death in this case. See Aguirre-Jarquin v. State, 9 So.3d 593, 612 (Fla.2009) (Pariente, J., specially concurring). I continue to believe that this Court has the authority to require special interrogatories, which would assist the trial court in giving “great weight” to a jury’s recommendation and assist this Court in reviewing death sentences.