Dissenting
I am compelled to dissent from the Majority’s analysis and ultimate allowance of the bad acts evidence in this case. In my view, the Majority contorts the exceptions to the prohibition against the admission of bad acts evidence, thereby stripping the Appellant, Charles Ray Hicks (Hicks), of his presumption of innocence and allowing a conviction based upon Hicks’ prior propensity to commit violent acts.
Historically, this Court has held that use of prior bad acts evidence should be strictly limited—recognizing that “[i]t is not proper to raise a presumption of guilt on the ground that, having committed one crime, the depravity, it exhibits makes it likely [the same person] would commit another,” thereby relieving the Commonwealth of its constitutional burden of proof beyond a reasonable doubt. Shaffner v. Commonwealth, 72 Pa. 60, 65 (1872). Rule 404(b)1 was adopted to codify our *493common law prohibition on the admission of propensity evidence while also providing, as at common law, that bad acts may be admissible, under special circumstances, to prove “motive, opportunity, intent, preparation, plan, knowledge, identity, absence of mistake, or lack of accident.” See Pa.R.E. 404(b). To patrol the boundaries between the prohibition and its exceptions, we have consistently required evidence of purportedly admissible bad acts to evince either signature-like similarities or a true logical connection to the charged crime.
The Majority posits that the testimony of Kim Alston, Misty Chavez and Lakisha Washington was probative of Hicks’ common scheme, intent, lack of accident, and identity. Majority Op. at 467-68, 469-70, 156 A.3d at 1127, 1128. Under the guise of our established standards, but in fact advancing a troubling departure therefrom, the Majority allows evidence of three alleged choking incidents—bearing little relationship either to this case or to each other—as proof that Hicks beheaded the victim in this case with a saw blade. The Commonwealth conceded that it needed the bad acts testimony to prove its case.2
At most, the bad acts evidence admitted in this case is probative of Hicks’ status as a drug-addicted man who, when faced with similar, random encounters, has (allegedly) exhibited violent tendencies toward women. Because the similarities *494between the circumstances surrounding the victim’s death and those of the prior bad acts are grossly insufficient to establish a true signature, no grounds exist for proving identity, common scheme, intent or lack of accident. And because the bad acts do not support an inference that Hicks conceived of a single, overarching plan encompassing each of the prior acts and the charged crime, the testimony does not demonstrate a logical connection for purposes of showing motive or common scheme. Therefore, I would hold that the trial court abused its discretion in allowing the Commonwealth to introduce its 404(b) evidence.
Two exceptions to the prohibition against propensity evidence are embedded in our decisional law: Bad act evidence is admissible if 1) a logical connection exists between the bad acts and the crime on trial, linking them together for some purpose the defendant intended to accomplish, or 2) the bad acts evince a signature crime. These principles derive from our landmark decision in Shaffner v. Commonwealth:
To make one criminal act evidence of another, [1] a connection between them must have existed in the mind of the actor, linking them together for some purpose he intended to accomplish; or [2] it must be necessary to identify the person of the actor, by a connection which shows that he who committed the one must have done the other.
Shaffner, 72 Pa. at 65 (emphasis added); see also Majority Op. at 463-64, 156 A.3d at 1125. Shaffner thus provides substantial guidance as to the type of link that will create an exception to the “general rule that a distinct crime, unconnected with that laid in the indictment, cannot be given in evidence against a prisoner.” Shaffner, 72 Pa. at 65. If such a link exists, bad act evidence may be admitted to prove motive, common scheme, intent, lack of accident, or identity, so long as the evidence is more probative than prejudicial. See Rule 404(b).3
*495In Shaffner, the defendant stood trial for murdering his wife. The Commonwealth sought to offer bad act evidence at trial that:
John Sharlock died from poison, the same kind of which Nancy, the [defendant’s wife, died; that his symptoms were the same as hers, that the [defendant] attended upon both, and that both died at the [defendant’s house; Sharlock on the 17th of February 1871, and Nancy, the wife, on the 11th of June 1871. In substance, this was an offer to show that the [defendant] poisoned Sharlock, as evidence that he also poisoned his own wife.
Shaffner, 72 Pa. at 65-66. Despite some apparent similarities between the circumstances surrounding the death of Sharlock, the defendant’s paramour’s husband, and the death of the defendant’s wife, we held the evidence of Sharlock’s death lacked the required connection to the death of Nancy.
It is obvious that to connect together the deaths of Sharlock and Nancy, and make the former bear upon the latter, they must have been both contemplated by the prisoner as parts of one plan in his mind, in which the taking of Sharlock’s life was part of his purpose of taking the life of Nancy. He must, therefore, have contemplated the death of Nancy before taking the life of Sharlock. In order to let in the poisoning of Sharlock, the judge must have had before his mind some fact or facts exhibiting this pre-existing determination to take Nancy’s life. Herein the evidence was defective.
Id. (cautioning that “if the judge does not clearly perceive the connection, the benefit of the doubt should be given to the prisoner,” and concluding that the other act evidence was inadmissible).
Thus, to show motive, plan, design or scheme (which in turn may tend to show identity, intent, absence of accident, or some other fact in issue), it must be possible to conclude that the *496bad acts and the charged crime were “both contemplated by the prisoner as parts of one plan in his mind” such that “it is obvious” that committing the prior act “was part of his purpose” in committing the charged crime. Id.4 It must appear “that the other offenses, though distinct crimes, are in fact ... part of a larger field of operation, previously conceived and in part executed.” Commonwealth v. Chalfa, 313 Pa. 175, 169 A. 564, 565 (1933) (citing Shaffner, 72 Pa. at 65); see also Commonwealth v. Weiss, 284 Pa. 105, 130 A. 403, 404 (1925) (explaining “if the act which is proven may indicate the motive or plan of action of the defendant, either preceding or following the commission of the crime, and is so closely joined thereto as to show the probability that he was guilty of the offense charged, it can be properly received”); Commonwealth v. Schwartz, 445 Pa. 515, 285 A.2d 154, 158 (1971) (excluding evidence “to show intent[,] design or motive for ... killing” a police officer where the prior shooting of another police officer did not “give sufficient ground to believe that the crime currently being considered grew out of or was in any way caused by the prior set of facts and circumstances”); see also McCormick, Evidence, § 190 (2013 7th ed.) (cautioning that to prove the existence of a common scheme, “each crime should be an integral part of an over-arching plan explicitly conceived and executed by the defendant”). Absent a true plan or motive, mere similarities between the bad acts and the crime on trial are insufficient to establish grounds for admissibility.
A second way “to make one criminal act evidence of another” is to demonstrate “a connection [showing] that he who committed the one must have done other,” making it impossible not to “identify the person of the actor.” Shaffner, 72 Pa. at 65. To this end, for the purpose of proving identity, we have consistently held that the bad acts and the crime on trial must share “an almost uncanny similarity in all the details.” Commonwealth v. Wable, 382 Pa. 80, 114 A.2d 334, 337 (1955). Further clarifying the high bar for admissibility based on similar acts, our Court has explained that bad act evidence is *497admissible to prove the accused committed the crime on trial, when the incidents are:
so nearly identical in method as to earmark them as the handiwork of the accused. Here much more is demanded than the mere repeated commission of crimes of the same class, such as repeated burglaries or thefts. The devise used must be so unusual and distinctive as to be like a signature.
Commonwealth v. Rush, 538 Pa. 104, 646 A.2d 557, 560-61 (1994) (citing McCormick, Evidence, § 190 (1972 2d ed.) (emphasis added); Commonwealth v. Bryant, 515 Pa. 473, 530 A.2d 83, 86 (1987) (“Bryant I”). In sum, where the circumstances surrounding each of the crimes demonstrate a distinctive signature method, the connection between the bad acts and the crime on trial is sufficient, and allowing the bad acts to be evidence of the crime charged is appropriate.
By way of example, in Wable, the defendant was on trial for the murder of Harry Pitts, a truck driver who was shot in the head on July 28, 1953 while asleep in the cab of his truck on the Pennsylvania Turnpike. Wable, 114 A.2d at 336. The trial court admitted evidence of two other crimes—a July 25, 1953 murder and a July 31, 1953 shooting—both perpetrated against truck drivers sleeping in their cabs on or near the Pennsylvania Turnpike. In holding that the trial court’s admission of evidence was proper, this Court articulated the “signature” nature of the crimes:
Apart from the fact, however, that the crimes all occurred at three day intervals and two of them in the same neighborhood, there was a striking similarity in the manner in which they were committed. Woodward and Pitts were each found lying on the seat of his cab with his head against the door and resting on a pillow. Each had apparently been attacked in the early morning hours. In each instance the murderer had poked his gun through the door window and shot his victim in the head, death being instantaneous. In each instance the bullet had entered the head at about the same angle. In each instance the motive was evidently robbery, Woodward being actually robbed but apparently a *498hurried getaway in the case of Pitts prevented accomplishment of that purpose. Shepard also was robbed and the shooting in his case was practically identical in detail with the other crimes, except that he was fortunate that his wound did not result in death. What led to defendant’s arrest was that a watch which had been stolen from Shepard was discovered in a pawnshop in Cleveland and found to have been pawned there by defendant. The weapon with which all the shootings had been performed was identified as belonging to defendant.
Id. (emphasis added).
By virtue of demonstrating a distinct signature method, identically executed as to each crime, the evidence in Wable was properly admitted. It established a connection between the bad acts and the crime on trial such that “proof of one will naturally tend to show that the accused is the person who committed the crime on trial.” Id.; see Shaffner, 72 Pa. at 65 (indicating that “a connection which shows that he who committed the one must have done the other” is admissible to show identity). Although the bad acts in Wable did not necessarily establish that the defendant had conceived an overarching plan of which the attack on each truck driver was an integral part, or that one was committed for the purpose of the other, this Court indicated that because of the striking similarities between the bad acts and the crime charged, the bad acts were probative of a common scheme. Wable, 114 A.2d at 337. This erroneous (and perhaps inadvertent) conflation of the signature requirement with the requirement for proving common scheme opened the door to a decades-long misunderstanding about what type of connection is truly required for the purpose of proving a common scheme.
Indeed, many of our cases since Wable have, like Wable, properly allowed bad acts showing signature-like “striking similarities” to the charged crime while improperly characterizing identity evidence as admissible for the purpose of proving a common scheme. See, e.g., Commonwealth v. Miller, 541 Pa. 531, 664 A.2d 1310, 1318 (1995) (positing, incorrectly, that bad acts evincing a true signature are “relevant and admissi*499ble for the purpose of establishing a common scheme, plan or design embracing the commission of two or more crimes so related to each other that proof of one naturally tends to prove the others”); Commonwealth v. May, 540 Pa. 237, 656 A.2d 1335, 1340 (1995) (equating, incorrectly, the “common scheme” exception and the “ ‘signature crime’ exception”); Commonwealth v. Hughes, 521 Pa. 423, 555 A.2d 1264, 1283 (1989) (admitting evidence of strikingly similar bad acts for the purpose of showing identity and common scheme, despite absence of an overarching plan). Contrary to the conflation of the requirements shown in these cases, invocation of the common scheme exception should be limited to circumstances from which a true plan or motive can be inferred. Striking similarities, on the other hand, may be admissible to show identity, intent or lack of accident, which may also be shown by evidence of a common scheme.
In contrast to the bad act evidence in Wable that clearly earmarked those crimes as the “handiwork” of the accused, the bad act evidence in Commonwealth v. Bryant, 531 Pa. 147, 611 A.2d 703 (1992) (“Bryant IF’), merely showed general similarities, insufficient to establish the identity of the perpetrator, or any other permissible purpose. In Bryant II, the defendant was on trial for the murder of an elderly woman named Edith Steckle, who had been found lying on the ground floor of her home, “brutally and viciously injured in the face, head, body, arms and legs.” Id. at 706. The trial court admitted bad act testimony about a subsequent attack for which the defendant had been convicted. Id. We vacated the defendant’s judgment of sentence and remanded the case for a new trial because we could not conclude that there was anything “unusual” or “distinctive” connecting the two crimes so as to prove “that both ... were perpetrated by the same individual.” Id.
Both crimes occurred at approximately the same time of night and within a one and one-half block of the Appellant’s home. However, the victims’ ages varied drastically, in addition to the victims being of different races. Furthermore, the.evidence was questionable as to whether Mrs. *500Steckle was sexually assaulted although Ms. Phillips was assaulted. It was also widely known that Mrs. Steckle lived alone while Ms. Phillips lived with her three-year-old son. Furthermore, [unlike Mrs. Steckle,] Ms. Phillips was punched only in the head. [Unlike Mrs. Steckle], Ms. Phillips was dragged to the second-floor bedroom. Finally, a television, radio, and ring were taken from Mrs. Steckle’s home, while only thirty dollars in cash was taken from Ms. Phillips even though a portable television and radio were in full sight of the burglar.
Id.
We confirmed in Bryant II that where the similarities between the crimes lack “any uniqueness” and where differences between them abound, we will “be hard pressed to conclude that the same individual committed both crimes.” Id.; see also, e.g., Commonwealth v. Patterson, 484 Pa. 374, 399 A.2d 123, 127 (1979) (finding that bad act evidence did not establish “identical method” where charged crime involved attacker who wielded an ice pick, wore light-tinted sunglasses, and raped his victim in a garage while bad act involved attacker who carried a gun, wore dark sunglasses, and raped his victim in an apartment, despite both rapes occurring in same area, late at night, and commencing in alleyways); Commonwealth v. Fortune, 464 Pa. 367, 346 A.2d 783, 787 (1975) (where victims were different ages and weapons were not identical, “too many details ... are unexplained or incongruous to say that one crime naturally tends to show that the accused is the person who committed the other”).
This analysis of our case law is consistent with commentators who are generally critical of courts admitting 404(b) evidence based on mere similarities, as opposed to either a true signature or a true plan. See 1 Imwinkelried, Uncharged Misconduct Evidence, § 3:24 (2005) (collecting authorities); see also McCormick, Evidence, § 190 (2013 7th ed.). According to Imwinkelried, proof of a number of similar bad acts
may be probative of the defendant’s status as a professional criminal; and the similarities may tend to show that when faced with similar, random opportunities for committing a *501crime, the defendant repeatedly chooses to use roughly the same methodology. However, if the similarities are insufficient to establish modus [i.e., signature] and there is no inference of a true plan in the defendant’s mind, the proponent is offering ... forbidden ... character, disposition, or propensity evidence.
1 Imwinkelried, Uncharged Misconduct Evidence, § 3:24 (2005) (emphasis added). Specifically, Imwinkelried admonishes the use of “common scheme” as a justification to admit unlinked acts on the theory that a pattern or systematic course of conduct amounts to a “plan.” Id, Neither a “spurious plan” nor a plan to commit a series of similar crimes should be permitted as evidence that the accused committed the crime on trial. Id.', see also Shaffner, 72 Pa. at 65-66.
Signature-like similarities are as essential for proving absence of accident as they are for proving identity. See Commonwealth v. Boczkowski, 577 Pa. 421, 846 A.2d 75, 89 (2004). In Boczkowski, we explained that “the remarkable similarity between the manner in which both of appellant’s wives were killed” makes evidence concerning the circumstances of one death support a reasonable inference that the other’s death “was not accidental, but rather, was a result of appellant’s deliberate act.” Id. (emphasis added). There, the defendant was on trial for murdering his wife who died in a hot tub. She was in her thirties and in good health. The defendant reported to police that she had been drinking before getting in the tub and that he and his wife had argued the previous night. Her autopsy revealed she died of asphyxiation, not from drowning, as the defendant had suggested. The court admitted evidence concerning the circumstances of the defendant’s former wife’s death, and we found no abuse of discretion. His former wife, who died in her bathtub, was also in her thirties and in good health. At the time of his former’s wife death, the defendant had reported to police that she had been drinking before getting in the tub and that they had fought the previous night. His former wife’s autopsy revealed that asphyxiation was her cause of death. In each case, the defendant had fresh scratch marks on his arms, hands and torso following his wife’s death. *502Id. at 82. Given these striking similarities, the prior bad act was admissible mens rea evidence to establish the defendant’s intent and to rebut his claim of an accident.
The learned Concurrence agrees with my view that bad acts evidence must evince “striking similarities” or constitute a “signature crime” to be admitted for the purpose of proving identity, i.e., that the accused is the person who committed the crime on trial. See Concurring Op. (Saylor, C. J.) at 472-73, 156 A.3d at 1130; see also Commonwealth v. Arrington, 624 Pa. 506, 86 A.3d 831 (2014) (Saylor, C.J., dissenting). Concluding, (as I do), that the prior acts here do not present a signature and are not “strikingly similar” to the circumstances surrounding the death of the victim in this case, the Concurrence would also find the testimony inadmissible to prove the defendant’s identity.
Chief Justice Saylor, however, would admit the evidence anyway, based upon a much lower standard for admission. Concurring Op. (Saylor, C.J.) at 483-84, 156 A.3d at 1137. To show absence of accident, the Concurrence posits that “roughly similar” bad acts may be introduced on a “doctrine of chances” rationale to prove a defendant committed the actus reus5 when the defendant asserts no actus reus occurred at all, because the victim’s harm resulted instead from some innocent force. Id. at 475, 156 A.3d at 1132. I fail to see how the admission of “roughly similar” bad acts, for the stated purpose of proving the act of murder (by none other than Hicks), is any different than allowing evidence of Hicks’ violent propensity toward women to serve as proof that he committed the charged crime.
The Concurrence urges that the doctrine of chances allows for the admission of similar acts (within the same general category as the charged crime) “to establish the objective improbability of so many accidents befalling the defendant or the defendant becoming innocently enmeshed in suspicious *503circumstances so frequently.” Id. at 1133 (citing Edward J. Imwinkelried, An Evidentiary Paradox: Defending the Character Evidence Prohibition by Upholding a Non-Character Theory of Logical Relevance, The Doctrine of Chances, 40 U. Rich. L. Rev. 419, 439). I strongly disagree and I am unpersuaded by the Concurrence’s effort to characterize the doctrine of chances—in largely academic terms—as a “non-character” theory of logical relevance.
As a practical matter, the doctrine is, in my view, merely an excuse for admitting otherwise inadmissible propensity testimony. Especially as applied to the facts of the case before us, its application threatens to swallow the rule.6 I oppose the notion that we should apply a less stringent (“roughly similar”) standard for admitting evidence when the purported purpose is to prove the actus reus.
Even if I were to concede that the doctrine of chances rests upon a non-character rationale, which I do not, I would limit its application to a narrow set of circumstances, consistent with its theoretical underpinnings. Citing Imwinkelried, the Concurrence recognizes, but fails to apply, at least three important limitations. See Concurring Op. (Saylor, C.J.) at 483-84, 156 A.3d at 1136-37. First, the required degree of similarity between the bad acts and the charged crime must place them, at the very least, within “the same general category.” Id.7 In the instant case, the victim was, according to the *504Commonwealth, severely beaten on her entire body prior to being suffocated and beheaded. See infra, note 11. By contrast, the 404(b) witnesses did not allege that Hicks inflicted any blunt force trauma upon them, and they certainly were not decapitated. Id. Moreover, as to the 404(b) witnesses, Hicks did not attempt and was never charged with any class of murder. In fact, the violence he allegedly inflicted upon the 404(b) witnesses—putting his hand or hands on their necks or throats—did not lead to any charges whatsoever against Hicks, Clearly, the bad acts admitted in this case were not in “the same general category” as the one perpetrated against the victim. Accord Commonwealth v. Donahue, 519 Pa. 532, 549 A.2d 121, 127 (1988) (explaining that, for the purpose of proving absence of accident through bad acts that are not “strikingly similar,” the Commonwealth must show that “a similar result obtained in both cases”).
In Donahue, the defendant was on trial for murder in which the death of the victim was caused by alleged child abuse. Id. We affirmed the allowance of evidence of alleged prior acts of child abuse resulting in a “pattern of bruises” over the prior victim’s body because it was probative of whether the same injuries on the victim in the charged case were accidental, as the defendant claimed they were. Id. As described in detail herein, the differences between the bad acts and the charged crime in the instant matter far outnumber their similarities. Given the extreme brutality of the charged crime, it cannot be said that the same injuries or “a similar result obtained in [all] cases,” making this case distinguishable from Donahue. See id.
A second important limitation that the Concurrence acknowledges, but fails to apply, is that the doctrine of chances *505depends upon the defendant having committed the similar bad act many times, indeed many more times than “the frequency rate for the general population.”8 Concurring Op. (Saylor, C.J.) at 483-84, 156 A.3d at 1136. The large number and high frequency of the prior incidents are what give the doctrine any claim, however tenuous, to embodying a non-character rationale. Absent numerosity and frequency (not to mention “striking similarities”), the proffered 404(b) evidence cannot support the necessary determination of “objective improbability” that the defendant would face so many similar accusations, yet be innocent as to this one. Id. at 476-77, 156 A.3d at 1132-33 (emphasis added). Thus, in my view, any application of the doctrine of chances should be limited to textbook-type cases like United States v. Woods, 484 F.2d 127 (4th Cir. 1973).
In Woods, the defendant was charged with murdering her seven-month-old foster son, Paul, by means of cyanosis (oxygen deprivation). The prosecution offered and the trial court admitted evidence of twenty prior instances of cyanosis occurring in a total of nine children under the defendant’s care, including seven other deaths. Woods, 484 F.2d at 130. In all cases, the children were under the age of three. On appeal, the Fourth Circuit affirmed the trial court’s admission of the prior acts to show the defendant’s criminality, or absence of accident. The Woods court explained:
With regard to no single child was there any legally sufficient proof that defendant had done any act which the law forbids. Only when all of the evidence concerning the nine other children and Paul is considered collectively is the conclusion impelled that the probability that some or all of *506the other deaths, cyanotic seizures, and respiratory deficiencies were accidental or attributable to natural causes was so remote, the truth must be that Paul and some or all of the other children died at the hands of the defendant. We think also that when the crime is one of infanticide or child abuse, evidence of repeated incidents is especially relevant because it may be the only evidence to prove the crime.
Woods, 484 F.2d at 133.
Unlike here, the prior acts in Woods were indisputably of the same general category as the charged crime. Also unlike here, the similar incidents (cyanosis) in Woods had occurred at least twenty other times and led to seven other deaths. There was little doubt that, as a matter of objective probability (as opposed to subjective bad character), the defendant in Woods committed the criminal act of infanticide. By contrast, the three bad acts admitted in the instant case were entirely insufficient both in degree of similarity to the charged crime and in number/frequency to support anything other than an improper inference as to Hicks’ character.9
A third limitation to the application of the doctrine of chances is that there must be some disclaimer of liability both with respect to the prior instances and the charged crime. The commentators cited by the Concurrence almost universally describe the doctrine in these terms. See, e.g., Imwinkelried, An Evidentiary Paradox, 40 U. Rich. L. Rev. at 437 (explaining that the evidence is offered “to establish the objective improbability of so many accidents befalling the defendant or *507the defendant becoming innocently enmeshed in suspicious circumstances so frequently”)(emphasis added); Mark Cammack, Using the Doctrine of Chances to Prove Actus Reus in Child Abuse and Acquaintance Rape, 29 U.C. Davis. L. Rev. 355, 411 n.225 (1996) (noting that “the doctrine focuses on the improbability of repeated false accusations against the same person”); id. at 387-88 (indicating that the prior “acts” should really be called “happenings” in the context of a doctrine of chances analysis). It is difficult (if not impossible) to determine the objective improbability that one or all of the incidents in question are accidents when they are not all claimed to be such. Accordingly, this third limitation, like the first two, affords theoretical integrity to the doctrine. In the instant matter, there was no contention at the time the prior acts occurred, or otherwise, that they somehow occurred by accident. The present scenario simply does not lend itself to application of the doctrine of chances.
The Majority does not dispose of the present matter with any specific “absence of accident” analysis. Instead, with respect to the existence of a common scheme or signature, the Majority relies heavily on this Court’s recent Rule 404(b) decision in Commonwealth v. Arrington, 624 Pa. 506, 86 A.3d 831 (2014). Arrington, in my view, is the unfortunate culmination of the conflation of the requirements to establish a signature crime with those necessary to establish a common scheme or plan, which conflation had its genesis in Wable. See supra, p. 498, 156 A.3d at 1146. It falls squarely within the admonition articulated herein that where the similarities are insufficient to establish a signature crime, and there is no true plan, the evidence shows only the defendant’s propensity and must be prohibited. See supra, p. 501, 156 A.3d at 1147-48.
In Arrington, the defendant stood trial for first-degree murder in the shooting death of his ex-girlfriend, Tondra Dennis, who had him arrested following years of physical and emotional abuse, including beatings, death threats and break-ins. Ten days after Arrington’s release from custody, an unidentified man shot Dennis on the street near a home Arrington had once shared with his mother. Arrington, 86 *508A.3d at 839. The trial court admitted evidence that Arrington had physically assaulted and harassed three other girlfriends, characterizing the abuse in each relationship as triggered by the woman attempting to break up with him or interacting with other men. Id. at 842.
One former girlfriend testified that Arrington had followed and telephoned her incessantly, threatened to harm her brother, set fire to her apartment, slashed furniture, threw her belongings off a terrace, and inflicted two beatings upon her and two attacks upon her brother. Id. at 843. A second ex-girlfriend testified that Arrington had physically assaulted her three times over the course of their two-year relationship, including a punch to the face, and two separate pistol-whippings to her jaw and head, respectively. Id. The jury heard additional testimony that Arrington had struck this woman’s male friend on the shoulder with an axe and fired a gun at her friend’s brother. Id. The third ex-girlfriend testified that Arrington had punched her in the face, continually harassed and followed her, and threatened to kill her and her family when she ended their relationship. Id. at 843-44.
A Majority of this Court held that the “shared characteristics of each relationship”—monitoring his girlfriend’s daily activities; resorting to violence when his partner wanted to end a relationship or interacted with other men; inflicting head or neck injuries with his fist, a handgun or an edged weapon; and harming or threatening to harm members of his girlfriend’s family—rendered the evidence admissible under Rule 404(b)(2), Id. at 844. Notably, the Arrington Majority did not purport to find a “signature method” or “striking similarities,” departing even from our Wable line of cases. See supra, pp. 498-99, 156 A.3d at 1146-47. Indeed, these crucial terms appear nowhere in the opinion, and for good reason; there was clearly nothing signature-like to connect Arrington’s disparate violent acts. They were not carried out in an identical fashion, as the Arrington Majority ostensibly conceded by listing the five distinct methods Arrington employed to inflict *509injury on his partners or their male friends. See Arrington, 86 A.3d at 844.
Instead, we explained that “in order for evidence of other criminal activity to be admissible to establish a common scheme ... ‘a comparison of the crimes must establish a logical connection between them.’” Id. at 842 (quoting Commonwealth v. Miller, 541 Pa. 531, 664 A.2d 1310, 1318 (1995)). The Arrington Majority then concluded that Arrington’s “repeated efforts to preserve intimate relationships through harassment, intimidation, and physical violence culminating in the use of a deadly weapon” amounted to a “common plan or scheme.” Id. at 844. Thus, in Arrington, we ultimately held that similar, repeated, but entirely unconnected acts, constituting, at most, a pattern of abusive behavior toward women, establish the kind of logical connection necessary for the purpose of showing a common scheme to establish identity.10 See id. As my historical overview of our jurisprudence makes clear, they do not. The other bad acts in Arrington were clearly not part of the defendant’s purpose in committing the crime charged. Thus, the holding in Arrington lacks prece-dential support and, under a very thin veil, permits evidence of the violent propensity of a defendant to prove the crime charged. For the same reasons I dissent in the present matter, I would overrule our decision in Arrington as it is a departure from our cornerstone jurisprudential requirements for establishing a common scheme or plan and evinces nothing close to the required signature for purposes of showing identity. See Arrington, 86 A.3d at 860-61 (Saylor, C.J., dissenting) (cautioning that “as the decisional law gravitates further and further away from the centering ground of signature crimes, the identity/propensity distinction devolves to a matter of semantics”).
Turning to the case at bar, if there is anything “unusual and distinctive” about the circumstances surrounding the victim’s death, it is the fact that she was brutally beaten, strangled, *510decapitated and dismembered with a saw blade; her body parts were then strewn in garbage bags along two highways; and her hands were found buried in the wall of Hicks’ bedroom closet, wrapped in athletic socks, newspaper, layers of detergent and plastic bags. See N.T., 11/5/2015, at 91-93; N.T., 11/6/2014, at 28-90; N.T., 11/7/2014, at 162-68. Had the Commonwealth sought to introduce evidence that Hicks had previously attacked a number of women in a manner demonstrating an “uncanny similarity in all the details” surrounding the victim’s death, there would be little question as to its admissibility. See Wable, 114 A.2d at 337. Instead, the bad acts admitted here merely share some generic characteristics with each other and, to a lesser degree, with the violent acts perpetrated against the victim. There are simply no “striking” or “remarkable” similarities. See id. at 336; see also Boczkowski, 846 A.2d at 89. Moreover, there is nothing “distinctive” about the bad acts that would set them apart from other “crimes of the same class.” See Bryant I, 530 A.2d at 86 (citing McCormick, Evidence, § 190 (1972 2d ed.)). Indeed, the bad acts relied on here are not even “of the same class” as the charged crime.11 See id.
*511The absence of a signature is particularly apparent from the extraordinarily broad categories the Majority creates in its strained effort to elucidate the required “striking similarities.” See Majority Op. at 453-54, 156 A.3d at 1127. The Majority indicates that, in each case, Hicks “(1) was introduced to drug-dependent women of similar body types for purposes of using drugs” and “(2) showed a sexual interest in the women, sometimes involving prostitution.” Id. The fact that Hicks allegedly attacked women is hardly a distinctive feature. It merely narrows his victims to approximately half the population. That these women were drug dependent and encountered Hicks in the context of drug-use and, sometimes, prostitution, is also unremarkable, especially given that Hicks was himself drug addicted. Moreover, I find no support in the record for the Majority’s statement that all of the women had similar body types.12
*512The Majority next posits that, in each case, Hicks “(3) resorted to violence when the women behaved in a way he found disagreeable; (4) inflicted injuries on each woman by targeting her neck area with his hands, a sharp edged object, or both; and (5) verbally threatened to kill each woman.” Id Regrettably, these are descriptions that could easily be applied to the vast majority of violent crimes, maybe even more so to crimes of violence that arise between drug-addicted sexual partners. Even putting drugs and sex aside, one might safely assume that a substantial majority of violent interactions in this world occur as the result of one person finding another person’s behavior “disagreeable.” Moreover, as a factual matter, the record offers no evidence that the victim behaved in a way Hicks found disagreeable prior to her death or that he verbally threatened to kill her.
The differences between the attacks in this case are far more pronounced than the similarities, making this case much more like Bryant II, Patterson and Fortune than Wable or Boczkowski. Hicks allegedly choked Ms. Alston, causing her to lose consciousness. He threatened her with a gun (his) and a pocket knife (hers), and raped her during the assault. He also threw a glass table at her. She was not a prostitute and was not trading sex for drugs. To the contrary, she was providing drugs to Hicks, who was apparently upset that she was not *513more freely sharing her supply. The incident occurred in Hicks’ Hampton, Virginia home on the first and only occasion the two met, in 2006. See N.T., 11/6/2014, at 184-200.
Unlike Ms. Alston, Ms. Washington was a prostitute. She allegedly did drugs with Hicks on the occasion of the alleged attack, which occurred during their first and only encounter, in Fort Worth, Texas in 2003. Unlike the incident with Ms. Alston, Hicks did not threaten Ms. Washington with a gun or a knife, but with his car. Rather than grabbing her throat while in his home, he grabbed the back of her neck while in a car. She never lost consciousness. In fact, following the alleged assault, she was able to drive the car, which Hicks allowed her to do. He told her that he wanted someone to love him, but he did not attempt to penetrate her during the assault. He did not throw a glass table (or anything else) when she tried to leave. Although he allegedly threatened to run her over with his car if she got out, he did not follow through on this threat. See N.T., 11/7/2014, at 60-86.
Ms. Chavez was a prostitute at the time Hicks allegedly attacked her. She had been seeing him a couple times a week (or every couple of weeks) for three to five months. She testified that their encounters always involved her using drugs but did not always involve sex. Hicks was reportedly clean and sober at the time he attacked her, unlike his state when he allegedly attacked the other two witnesses. He allegedly grabbed Ms. Chavez’s neck with both hands while they were driving in Hicks’ truck together. She testified that she felt like she was blacking out but did not say that she became unconscious. Hicks threatened to kill Ms. Chavez when she opened the car door but he ultimately pulled off the road and calmed down. He did not throw anything at her or use an edged weapon. He apologized for hurting her and he brought her to his house to clean blood from her neck caused by his fingernails. Their encounter occurred in Arlington, Texas in 2002-2003. See N.T., 11/7/2014, at 86-106.
To the extent there are similarities between the bad acts and the charged crime, they are entirely insufficient to show a true signature for purposes of proving identity, intent or lack *514of accident. The infrequency of the prior acts, which do not fall in the same general category as the charged crime, also weighs heavily against their admission for the latter purpose, even if we applied some diluted version of the doctrine of chances. Moreover, the required connection from which a true common scheme or motive might be inferred is absent. The admitted testimony is indicative of nothing more than a pattern of violent behavior. There is no evidence from which to draw an inference that Hicks conceived of an overarching plan encompassing any of the bad acts and the charged crime, or that he committed any of the bad acts as “part of his purpose” to murder the victim. See Shaffner, 72 Pa. at 65-66. Neither the Commonwealth nor the Majority claim otherwise. By the Commonwealth’s own admission, there was no plan at all, as Hicks’ violent behavior was merely a spontaneous reaction to a speculated disagreement. Rather, it advances a definition of “common scheme” that requires no scheme at all but instead justifies the admission of random, unlinked acts so long as it is possible to discern some similarity with the charged crime, no matter how attenuated or unintentional.
If Rule 404(b) is to accomplish its intended purpose, it must be strictly construed in light of the common law principles underlying it, as articulated by this Court more than a century ago. See Shaffner, 72 Pa. at 65-66. Instead, the bad acts testimony allowed here was used to establish precisely what Rule 404(b) prohibits, i.e., that because Hicks had acted violently toward three women in the past, he was therefore capable of, and would have a propensity to commit, the far more extensive, brutal and ultimately fatal acts of violence perpetrated on the victim in this case, several years later. By permitting such an inference to be drawn, the Majority all but eliminates Hicks’ presumption of innocence and the Commonwealth’s burden to prove, in this case, that Hicks is guilty of first-degree murder. See Commonwealth v. Stanley, 484 Pa. 2, 398 A.2d 631, 633 (1979).
Finally, the admission of bad act evidence completely lacking in the necessary connection to the charged crime is not harmless error. See Fortune, 346 A.2d at 787 (remanding for a *515new trial because “whenever a jury improperly receives evidence of other crimes, the impact is significant”). Our standard for determining harmless error is clear. If an appellate court “concludes beyond a reasonable doubt that the error could not have contributed to the verdict,” the error is harmless. Commonwealth v. Mitchell, 576 Pa. 258, 839 A.2d 202, 214 (2003) (citing Commonwealth v. Story, 476 Pa. 391, 383 A.2d 155, 164 (1978)). However, “if there is a reasonable possibility that the error may have contributed to the verdict, it is not harmless.” Id. The Commonwealth bears the burden of establishing harmlessness beyond a reasonable doubt. Id. Specifically, we will find harmless error where:
(1) the error did not prejudice the defendant or the prejudice was de minimis;
(2) the erroneously admitted evidence was merely cumulative of other untainted evidence which was substantially similar to the erroneously admitted evidence; or
(3) the properly admitted and uncontradicted evidence of guilt was so overwhelming and the prejudicial effect of the error was so insignificant by comparison that the error could not have contributed to the verdict.
Commonwealth v. Young, 561 Pa. 34, 748 A.2d 166, 193 (1999).
In the instant matter, the Commonwealth itself has repeatedly and unambiguously argued that its case against Hicks was largely circumstantial and that it viewed the bad acts testimony as essential to proving the elements of first-degree murder beyond a reasonable doubt. See Commonwealth’s Brief at 12-14 (arguing the “fact that the case was circumstantial in its evidence coupled with the defense of accident being the suggested manner in which the victim died, all warrant the conclusion that the Commonwealth’s need for the [404(b) ] evidence was, in fact, great”); see also Offer of Proof, 4/27/2011, ¶ 15 (arguing that “the need for other crimes evidence is high in that the case against [Hicks] is primarily based on circumstantial evidence and [Hicks] has already placed an [sic] issue the precise cause and manner of the victim’s death”); see also Statement Pursuant to Pa.R.A.P. 1925(b), 8/8/2011 (explaining that “the proof of murder in this *516case is entirely circumstantial” and “[t]he [404(b) ] evidence would allow the jury to conclude that the victim’s death was the result of defendant’s assault”).
There were no witnesses to the crime and Hicks did not confess to the murder. See Commonwealth Brief at 12-13. As described hereinabove, whether strangulation and decapitation played a role in the victim’s death was very much in question. See supra, note 11. The 404(b) evidence permitted the jury to draw the improper inference that because Hicks had allegedly inflicted injury to the throat or neck of the 404(b) witnesses, he must also have strangled the victim and beheaded her. In fact, the trial court stated in its opinion ruling on post-sentence motions that the 404(b) evidence helped “elucidate that it was [Hicks] who killed the victim,” as opposed to another. See Trial Court Opinion, 8/18/2015, at 16. As we said in Young, “[i]t is difficult to imagine any evidence more prejudicial to a defendant than that which identifies [him] as a perpetrator of a capital crime.” Young, 748 A.2d at 193. Accordingly, there is no question that the testimony of Ms. Alston, Ms. Washington and Ms. Chavez significantly prejudiced Hicks.
For these same reasons, it is impossible to conclude that the 404(b) evidence “was merely cumulative of other untainted evidence.” See id. By the Commonwealth’s own admission, absent the 404(b) testimony, it lacked sufficient evidence to prove that Hicks caused the victim’s death beyond a reasonable doubt. Moreover, the “uncontradicted evidence” did not overwhelmingly point to Hicks committing the murder, as opposed to merely committing the crime of abuse of corpse following the victim’s death by drug overdose. See supra, note 11. Even the most damning of the uncontradicted evidence— that the victim’s hands were found buried in Hicks’ wall—is consistent with Hicks’ guilt of abuse of corpse and innocence as to the murder itself. Because the bad acts evidence was wrongly admitted and because its admission was far from harmless, I would reverse the judgment of sentence and remand this case for a new trial.
. Pennsylvania Rule of Evidence 404(b) was enacted in 1998 and its basic principle was intended to be consistent with both Federal Rule of Evidence 404 and prior Pennsylvania case law. See Pa.R.E. 404, Order of Feb, 28, 2006, Cmt. Rule 404(b) provides:
(1) Evidence of a crime, wrong, or other act is not admissible to prove a person’s character in order to show that on a particular occasion the person acted in accordance with the character.
*493(2) This evidence may be admissible for another purpose, such as proving motive, opportunity, intent, preparation, plan, knowledge, identity, absence of mistake, or lack of accident,
(3) Evidence of other crimes, wrongs, or acts proffered under subsection (b)(2) of this rule may be admitted in a criminal case only upon a showing that the probative value of the evidence outweighs its potential for prejudice.
Pa.R.E. 404(b).
. The Commonwealth argues "the bad acts evidence was necessary in this circumstantial evidence case. The evidence rebutted any defense of accident. The evidence established a motive for the crime. The evidence helped to identify the defendant as the perpetrator of the murder. The evidence also helped the fact finder in determining the defendant's intent at the time of the murder.” Commonwealth’s Brief at 1; see also infra, pp, 515-16, 156 A.3d at 1156-57. That bad acts evidence is helpful, or even necessary, does not render it admissible. Indeed, such evidence is typically exceptionally helpful to the Commonwealth but if it does not evince a logical connection to the crime at issue or a true signature, it should not be allowed.
. Although there is significant overlap between the various 404(b)(2) purposes, it is my view that a "signature” does not itself establish a “common scheme,” even though a "signature” and a “common scheme” may, for example, both prove identity. See infra, pp. 498-99, 156 A.3d at 1146-47; cf. McCormick, Evidence, § 190 (2013 7th ed.) (citing Stone, Exclusion of Similar Fact Evidence: America, 51 Harv. *495L. Rev. 988, 1026 n.190 (1938) ("Motive, intent, absence of mistake, plan and identity are not really all on the same plane. Intent, absence of mistake, and identity are facts in issue—facta probanda. Motive, plan, or scheme are facta probantia, and may tend to show any facta probanda.”)
. Or vice versa, in the case of subsequent act evidence.
. Black’s Law Dictionary defines actus reus as "the wrongful deed that comprises the physical result of human conduct and that generally must be coupled with mens rea to establish criminal liability; a forbidden act.” As to first-degree murder, the actus reus is the act of killing. Black's Law Dictionary 39 (8th ed. 2004).
. Notably, "several commentators have argued that prosecutors are now smuggling inadmissible bad character evidence into the record under the guise of invoking the doctrine of chances.” Imwinkelried, The Use of Evidence of an Accused's Uncharged Misconduct to Prove Mens Rea: The Doctrines Which Threaten to Engulf the Character Evidence Prohibition, 51 Ohio St. L.J. 575, 586 (1990). The doctrine’s potential to eviscerate the character evidence prohibition is stronger still given the reality that limiting instructions do little to curb the jury’s commonsense or perhaps unconscious mis-application of 404(b) evidence. See Concurring Op. (Saylor, C.J.) at 485 n.16, 156 A.3d at 1138 n.16.
. In discussing the degree of similarity necessary to achieve a non-character application of the doctrine of chances, another commentator (cited by Imwinkelried but not by the Concurrence), explains:
The closer the physical similarities in the acts, the less likely the acts were a coincidence. For example, a pair of accused car thieves may claim to have had a good faith belief that they had permission to take an automobile. Evidence showing that they had previously stolen a *504stereo would have very little inferential value in proving or disproving coincidence. However, evidence showing that defendants had made a similar 'mistake' with other cars would strongly indicate a lack of coincidence.
Eric D. Lansverk, Admission of Evidence of Other Misconduct in Washington to Prove Intent or Absence of Mistake or Accident: The Logical Inconsistencies of Evidence Rule 404(b), 61 Wash. L. Rev. 1213, 1230-31 (1986) (internal footnote omitted).
. As Imwinkelried has explained, it must be clear that "the accused has suffered the loss more frequently than the typical person endures such losses accidentally.” Imwinkelried, The Use of Evidence of an Accused's Uncharged Misconduct to Prove Mens Rea: The Doctrines Which Threaten to Engulf the Character Evidence Prohibition, 51 Ohio St. L.J. at 591. Accordingly, "it is highly relevant to consider the number of losses the accused has suffered" and "if the judge has no basis for determining the frequency of such accidental occurrences among the general populace” or "if the judge ... genuinely doubt[s] whether the frequency of the accused’s losses exceeds the incidence for the general population^]” the judge must exclude the evidence. Id. at 590.
. For similar reasons, I question whether Donahue properly applies the doctrine of chances. In that case, a single prior incident bearing no signature-like similarities to the charged crime was nonetheless admitted to show lack of accident. The frequency requirement underpinning the doctrine was clearly absent in Donahue but, at least in that case (unlike here), the injuries in the charged crime and the prior incident were the same. Id. at 127. Moreover, Donahue (like Woods) was a child abuse case, for which other evidence is commonly unavailable. This fact may well have driven the result. See Imwinkelried, The Use of Evidence of an Accused's Uncharged Misconduct to Prove Mens Rea: The Doctrines Which Threaten to Engulf the Character Evidence Prohibition, 51 Ohio St. L.J. at 586 (discussing the particular relevance of the doctrine of chances in child abuse cases where the accused often contends the victim’s injuries were sustained accidentally).
. As noted, the Arrington Court did not rely on the bad acts evidence to support a “lack of accident" exception. Even if it had, Arrington’s conclusion that the shared characteristics of the prior abusive relationships sufficed to establish a logical connection between them and the crime charged is a monumental leap from the standard articulated in Donahue (a child abuse case) that required a similarity in the bad act and outcomes to establish lack of accident.
. Although the Commonwealth's forensic expert, Dr. Ross, testified that the victim was alive when she was beaten, strangled, and decapitated, and that her cause of death was a combination of strangulation and sharp force injury to the neck, N.T., 11/6/2014, at 59-90, Hicks’ expert, Dr. Shane, testified that the victim's bruising and lacerations occurred after her death, and that her cause of death was most likely related to substance abuse. N.T., 11/12/2014, at 15-38. Notably, Dr. Shane testified that "there was nothing to indicate that [the victim] was strangled” or that there was "any intervention while she was still alive.” Id. at 27. He further testified that there were no compression injuries, no fingernail injuries, and no other signs of “forceful strangulation” or "manual strangulation” on the victim's body. Id. at 28. He also testified that the bite marks on the victim's tongue were consistent with a drug overdose, countering Dr. Ross’ conclusion that they resulted from strangulation. Id. at 38. Another of Hicks’ experts, Dr. Mihalak-is, concurred with Dr. Shane that there was no evidence of strangulation or choking on the victim's body. See id. at 78-79. He testified, "I did not see any fingernail markings in the neck. I didn’t see any ligature marks, and I didn’t see any scratching in the neck.” Id. at 79. Moreover, both Dr. Shane and Dr. Mihalakis testified that immunohisto-chemistry, the method of forensic analysis used by Dr. Ross, is typically used on live tissue, and considered unnecessary or unreliable when used on decomposed tissue like that of the victim. See id. at 27, 67. *511In his Concurrence, Justice Baer draws an ex post facto conclusion that there was no controversy as to the cause of death. See Concurring Op. (Baer, J.) at 491-92, 156 A.3d at 1142. While I recognize that Dr. Shane’s testimony may be viewed as ambiguous, it nonetheless raised the question whether the victim could have died from an overdose. The defendant had no burden to prove, unequivocally, that the victim's death resulted from an accident. He only had to raise a reasonable doubt as to the Commonwealth’s cause of death theory. Contrary to Justice Baer's assertion, ambiguous testimony is not tantamount to uncontradicted testimony. See id. Moreover, in charging the jury, the trial court correctly presented the jury’s role in considering contradictory expert evidence: "In this case, the testimony and opinions of Dr. Ross, Dr. Shane, and Dr. Mihalakis may seem to conflict with each other in certain aspects. If you find that the conflict is more than superficial, that the conflict is real and irreconcilable, you may decide what parts, if any, of the contradictory testimony and opinions you choose to believe.” N.T., 11/14/2014, at 109.
Even setting aside any violence inflicted to the victim’s neck, none of the bad acts come close to resembling the savage treatment of the victim’s body at issue in this case. Unlike the victim, none of the 404(b) witnesses suffered blunt force trauma, let alone an estimated seventeen impacts to the head and face inflicted by a hand, fist or foot said to have been suffered by the victim. See N.T., 11/6/2014, at 27-45. Unlike the victim, none of the 404(b) witnesses suffered injuries to their torsos, let alone multiple fractures to each of their ribs by stomping or kicking. Id. at 46-54. And, unlike the victim, none of the 404(b) witnesses suffered a laceration to the back of her head, likely inflicted by a crowbar, pipe or claw end of a hammer. Id. at 35-42.
. The victim was a white woman, approximately 5’ 7” tall, 150-160 pounds with long black hair. Commonwealth’s Exhibit 35 (Autopsy *512Report). At the time she was murdered, she was approximately 36 years old. The record reveals that Ms. Alston is a black woman and was approximately 41 years old when the alleged incident with Hicks occurred. Commonwealth’s Offers of Proof, 4/27/2011, at Exhibit C (Alston Police Report). Ms, Chavez is a white woman and was approximately 24 years old at the time Hicks allegedly attacked her. Id, at Exhibit B (Chavez Police Report). Ms. Washington is a black woman who was approximately 31 years old at the time of her encounter with Hicks. Id. at Exhibit E (Washington Police Report). I have discovered no evidence in the record indicating the respective heights, weights or other descriptors of the body types of the 404(b) witnesses. The Majority cites to the trial court's 1925(a) opinion which, notably, does not cite to anything when it finds that "the victims [sic] and the three witnesses were all of a similar body type.” See Majority Op. at 460, 467-68, 156 A.3d at 1123, 1127 (citing Trial Court Opinion, 8/18/2015, at 31). To the extent the trial court was relying on its own observations as to the witnesses’ body types, I would observe that many years had elapsed between the prior bad act incidents and the trial, and there would be no *513way for the trial court to know that the witnesses’ body types had not changed during that time.