(concurring in part and dissenting in part).
/¶ 138] ISSUE ONE
DID THE TRIAL COURT ABUSE ITS DISCRETION IN ADMITTING “PRIOR BAD ACTS” EVIDENCE INVOLVING THREE SEXUAL ASSAULTS COMMITTED BY MOELLER IN 1973, 1979, AND 1990?
[¶ 139] I respectfully dissent as to the rationale and result of Issue One. I would affirm the trial court on this evidentiary ruling.
SDCL 19-12-5 provides:
Evidence of other crimes, wrongs, or acts is not admissible to prove the character of a person in order to show that he acted in conformity therewith. It may, however, be admissible for other purposes, such as proof of motive, opportunity, intent, preparation, plan, knowledge, identity or absence of mistake or accident.
[¶ 140] The trial court admitted the testimony of Moeller’s attacks upon Carolyn Beshaw in 1973, Kenneth Moore in 1979 and Tracy Warner in 1990. In so doing, the trial court correctly applied the two criteria necessary to make such a decision:
(1) Whether the intended purpose for offering the other acts evidence is relevant to some material issue in the case, and
(2) Whether the probative value of the evidence is substantially outweighed by its prejudicial effect.
State v. Werner, 482 N.W.2d 286, 288 (S.D.1992)(citing State v. Basker, 468 N.W.2d 413, 415 (S.D.1991)). The trial court found the evidence was relevant to show common method, plan or scheme, intent, identity, and motive and that the probative value was not substantially outweighed by the danger of unfair prejudice.
[¶ 141] Crucial to analysis of this issue is the standard of review. A trial court’s determination to admit other acts evidence will not be overruled absent an abuse of discretion. State v. Larson 512 N.W.2d 732, 736 (S.D.1994); State v. McDonald, 500 N.W.2d 243, 245 (S.D.1993); Werner, 482 N.W.2d at 288.
An abuse of discretion has been defined by this Court as a decision which is not justified by, and clearly against reason and evidence. We will not reverse a decision if ‘we believe a judicial mind, in view of the law and the circumstances, could reasonably have reached that conclusion.’
Dakota Cheese v. Taylor, 525 N.W.2d 713, 715 (S.D.1995) (citations omitted); State v. Erickson, 525 N.W.2d 703, 710 (S.D.1994). “Upon review ... we must be careful not to substitute our reasoning for that of the trial *496court.” Larson, 512 N.W.2d at 736. Thus, the question is not whether, had we been the trial judge, we would have admitted the prior bad acts evidence but whether the trial court sitting in this case abused its discretion by doing so. Id.; State v. Rufener, 392 N.W.2d 424, 426 (S.D.1986).
[¶ 142] Unfortunately, the majority devotes only a single sentence to this standard upon which the entire issue hinges before proceeding into a lengthy analysis which is more appropriate to a de novo review rather than applying the abuse of discretion criteria cited above. This is most clearly established by the fact that in the entire text of this discussion, the majority devotes not a single paragraph to an analysis of the similarities between the attack on Becky and the prior bad acts. Instead, the majority focuses entirely on perceived dissimilarities and justifications not to affirm the trial court rather than balancing the facts that weigh for or against admission and applying an in-depth abuse of discretion review.
[¶ 143] I. Relevance
[¶ 144] As noted above, the trial court found the prior bad acts were relevant to show common method, plan or scheme, intent, identity and motive. Contrary to the majority’s assertion that the trial court only identified five common marks between the crimes charged in the indictment and the prior bad acts, the trial court actually identified ten. They are as follows:
(a) Each involved a folding knife similar to a Buck brand knife with approximately the same or similar* length blade.
(b) At the time the Defendant was apprehended in the City of Tacoma, WA, he had two similar knives in his possession.
(c) The object of each attack was sex.
(d) The weapon of choice in each attack was the above described type of knife.
(e) The wounds and blade length of the knife used to commit the murder was [sic] consistent with the type of knife used in the assaults.
(f) In each instance the Defendant had a passing knowledge or acquaintance of the victim, and the victim had no substantial relationship with the Defendant.1
(g) In each instance the Defendant committed the act within, his own general neighborhood.
(h) In each instance the attack was opportunistic, as opposed to a planned attack with the intent to conceal the identity of the perpetrator.
(i) In each instance the object of the attack was a demand for sex coupled with the threat to kill and prodding the victim with the knife.
(j) In each instance the victim was able to identify the Defendant.
[¶ 145] Moeller does not challenge the trial court’s findings that the prior bad acts occurred. Moeller, however, attacks the trial court’s ruling of admissibility of those acts arguing that the dissimilarities between the prior bad acts and the offense charged are so substantial that they overcome the State’s claim of relevance. Moeller claims the end result is evidence which was impermissibly admitted to show merely that he is a bad man who has committed bad acts in the past similar to the charge he is now facing. State v. Steele, 510 N.W.2d 661, 668 n. 8 (S.D.1994).
[¶ 146] I agree with the majority’s analysis that in this case the issue of identity is closely related to the other exceptions contained in the rule such as motive, plan or scheme, and intent. However, I will focus on the issues of identity and modus operandi, as they are dispositive.
The exception is more frequently used where two or more crimes appear to have been plotted by the same individual because they exhibit a similar unusual pattern. Distinctive “modus operandi” summarizes the rationale on which the evidence is admitted and points out that the perpetrator’s identity is the purpose *497for which this kind of evidence is invariably used. ‘Modus operandi’ means method of working and refers to a pattern of criminal behavior that is so distinctive that separate crimes are recognized as the handiwork of the same wrongdoer. The evidence is useful to identify the accused as the perpetrator of the crime charged. The inference is from ‘modus operandi’ to the ‘identity’ of the culprit. Since the defendant acted in a distinctively similar manner on another occasion, it is more likely he (rather than someone else) did the act on the occasion of the charged crime.
State v. Champagne, 422 N.W.2d 840, 842-43 (S.D.1988) (citations omitted).
[V147] There are common characteristics not unique to this case which are worthy of note. Moeller had an acquaintanceship or knowledge of Becky, Moore, and Warner but not Beshaw. The trial court found that Moeller committed these acts in close proximity to his residence. The attacks against Beshaw, Warner and Becky were committed in or near Sioux Falls, South Dakota. The attack against Moore was committed in Wright, Wyoming. Three of the four attacks were against females for sexual purposes. The fourth was against, Moore, a young male, also for sexual purposes.
[¶ 148] Granted, these are generic facts which by themselves could arguably apply to many violent sexual attacks. Were this all the evidence given the trial court, I would join the majority opinion that the relevance was minimal at best. When making such factual comparisons to justify admission, the facts must be distinctive, Champagne, 422 N.W.2d at 843; People v. Haston, 69 Cal.2d 233, 70 Cal.Rptr. 419, 427, 444 P.2d 91, 99 (1968); contain a marked similarity or a close parallel, State v. Thomas, 381 N.W.2d 232, 237 (S.D.1986); or evidence common features, State v. Willis, 370 N.W.2d 193, 198 (S.D.1985) (citing 2 Wigmore, Evidence, § 357 at 334 (Chadbourn rev 1979)). Otherwise one could argue that the attacker, in the prior bad acts and in the current charge, wore pants, tennis shoes, spoke English, wanted sex, was right-handed, or any other set of generic factors. These facts, while similar, are not so unusual and distinctive as to distinguish the modis operandi of the accused from that of many other perpetrators of the same type crime.
[¶ 149] The majority, however, fails to give sufficient emphasis to the matter of the knife and the circumstances surrounding its use relied upon by the trial court. Beshaw testified Moeller used a black-handled folding-knife with a three-inch blade. Moore testified that Moeller assaulted him with a black buck knife. Warner testified that Moeller assaulted her with a folding knife, with a three and a half-inch black and a brown handle. Following questioning by law enforcement about Becky’s murder, Moeller fled to the State of Washington. When captured in Tacoma nine months later, he still had in his possession, “two pocket knives or buck knives.” Both knives were folding-type knives, one of which was similar to the knives used by Moeller in the prior bad acts and consistent with Becky’s wound.
[¶ 150] While Dr. Randall, the forensic pathologist who performed the autopsy, was unable to testify as to the exact nature of the knife used on Becky to deliver the stab wounds, he did testify that a wound to the chest was four inches deep. The claim of the majority that the “the State’s witnesses could not identify either the type of knife or the length of the blade” used in the murder of Becky takes one sentence of Dr. Randall’s testimony out of context.
[¶ 151] If one has a four-inch deep knife wound which contains a hilt mark, one must logically conclude the blade that made the wound was of no more than four inches or, as Dr. Randall testified, “[i]f the blade isn’t inserted to the hilt, then we don’t see this type of abrasion around the edge of the wound.” I know of no way the blade can be any longer than four inches and the majority does not say how it can be otherwise. Dr. Randall further testified that a blade somewhat shorter than the four-inch depth of the wound could have produced such an injury due to the compression of the chest. Clearly however, the blade did not exceed four inches in length and could not have been substan*498tially shorter than that.2 This is much more fact-specific than the majority’s assertion that “[h]aving failed to establish the type of knife or length of blade used against Becky, the State is left with the generic observation that all the offenses involved the use of a knife.”
[¶ 152] The State represented to the trial court prior to trial that it would prove the knife wounds on Becky were inflicted by the same type of knife as was used in the assaults on Beshaw, Moore, Warner and were found on Moeller’s person when he fled to Seattle after he was questioned about his involvement in Becky’s death. The State delivered on that representation as no inconsistent facts arose at trial. All known facts on the knife wounds on Becky are consistent with the State’s claim.
[¶ 153] A second factor noted by the trial court was that Moeller used this same type of knife to threaten his victims at or near the throat area. Moeller held a buck-style folding knife to Beshaw’s throat and demanded sex “or I’ll kill you.” Moeller held the buck-style knife to Moore’s throat and demanded sex “or I’ll kill you.” Moeller held a buck-style folding knife near Warner’s throat at her upper arm and breast and said if he did not get sex, Warner and her baby were “going to get cut.” The fatal wound to Becky was a knife wound to the jugular vein of her throat. See State v. Martin, 118 Idaho 334, 796 P.2d 1007, 1011 (1990) (court found assailant’s placement of knife to his victims’ throats when demanding sex to be a “significant similarity” between charged crime and earlier offenses and “highly relevant to the issue of identity in the present case”). Moeller was later arrested with two such knives in his possession.
[¶ 154] These threats were not all idle talk. In 1973 apparently Moeller did not have sufficient motivation to cany through with his threats when confronted by Beshaw. However by 1979 he was fully capable of cutting Moore with the buck-style knife and did so again in 1990 in the Warner incident.3
[¶ 155] These attacks with the knife for sex were also unusual in that Moeller made no attempt to hide his identity, hardly a universal trait for a rapist or murderer. State v. Olson, 449 N.W.2d 251 (S.D.1989); Martin, 796 P.2d at 1011. Moeller’s initial threat on Beshaw with the knife was done on a busy Sioux Falls street near’ Axtell Park in broad daylight after he had been loudly ejected from another car. After Moeller failed in his attempted attack on Beshaw, he had her drive him to the Speedy Car Wash where she worked. At this public point, Moeller exited Beshaw’s car where she and her car were obviously well known. The attack on Moore with the knife was in a mobile home occupied by Moeller which was right next to Moore’s trailer and was occupied at the time of the attack by Moore’s father. Moeller had Moore contact Moore’s father about a planned trip to Gillette, Wyoming, and went so far as to demand Moore’s father come to Moeller’s trailer with a note to give to Moel-ler approving the trip just prior to Moeller’s attack on Moore. This provided Moore’s father with the location of his son and the *499identity of his son’s would-be attacker. Moeller was known to Warner when he wandered into her apartment for a visit. His opening words were, “[h]ello, do you know me?” Warner indicated she did. He left after a brief visit only to return a short time later to assault her with intent to forcibly molest her knowing she could readily identify him. He also parked his car in Warner’s driveway prior to the attack thus allowing Warner to obtain his license plate number when he left after the attack.
[¶ 156] While the rape and murder of Becky probably occurred in rural Lincoln county, that misses the point. Likewise, the claim that Becky was murdered to hide the identity of her attacker is also off the mark. Moeller had previously overtaken Becky at the intersection of Main and Russell, two of the busiest streets in Sioux Falls at the height of the 5 o’clock rush hour only minutes before the attack upon her. Just prior to that he had been spotted by a clerk next to Becky in a busy store in which he was a well-known customer. Moeller was known to the clerk as the store stocked an unusual brand of cigarettes for Moeller at his request. Moeller’s failure to hide his identity prior to his abduction and attack on Becky is even more bizarre when one considers he did substantial prison time for all three prior bad acts either upon a guilty plea (Moore), conviction (Warner), or pursuant to a plea bargain concerning other crimes (Beshaw). Rather than attempt to hide his identity, Moeller seems to display it.
[¶ 157] In keeping with his lack of desire to hide his identity in all four attacks, initial contact with his intended victims occurred within fifteen blocks .or less of his victims’ homes.
[¶ 158] The trial court further found it significant that Moeller did not have a “substantial” relationship with Becky, Moore, Beshaw and Warner prior to the attacks. Most rape victims know their attacker on more than a passing basis.4
[¶ 159] While the facts surrounding the knife are significant in themselves, each fact is not to be examined in isolation when applying the abuse of discretion standard of review.
It is apparent that the indicated inference does not arise ... from the mere fact that the charged and uncharged offenses share certain marks of similarity, for it may be that the marks in question are of such common occurrence that they are shared not only by the charged crime and defendant’s prior offenses, but also by numerous other crimes committed by persons other than defendant. On the other hand, the inference need not depend upon one or more unique or nearly unique features common to the charged and uncharged offenses, for features of substantial but lesser distinctiveness, although insufficient to raise the inference if considered separately, may yield a distinctive combination if considered together. Thus it may be said that the inference of identity arises when the marks common to the charged and uncharged offenses, considered singly or in combination, logically operate to set the charged and uncharged offenses apart from other crimes of the same general variety and, in so doing, tend to suggest that the perpetrator of the uncharged offenses was the perpetrator of the charged offenses.
Haston, 70 Cal.Rptr. at 427-28, 444 P.2d at 99-100 (cited with approval by McCormick’s Handbook of the Law of Evidence § 190, n. 37 (2d ed 1972))(emphasis added).
[¶ 160] This also disposes of Moeller’s claim that lack of similarity of victims (in that Becky was a young girl, Beshaw and Warner were adult women, and Moore, a young boy) automatically precludes admission. The comparison of victims is not dispositive, by itself, of the admissibility issue. State v. Houghton, 272 N.W.2d 788 (S.D.1978), reinforces the rule of law that the modus operands issue is determined based on all factors, rather than pointing to one factor in iso*500lation. Houghton is particularly instructive as it was decided by the same Court who had only five months earlier adopted SDCL 19-12-5 as a court rule.
[¶ 161] In Houghton, the defendant was charged with raping an adult woman. The State sought to introduce evidence of two prior rapes of women under SDCL 19-12-5. The opinion identifies no common characteristics of the three purported victims but goes into great detail on the defendant’s unusual methodology in carrying out his alleged attacks. Although the Court found identity was not in issue, it stated the rapes were “committed in a similar, unique manner sufficient enough to be like a ‘signature.’ ” Id. at 792.5
[¶ 162] Thus, one looks to the totality of the circumstances to determine if similar circumstances exist to justify admissibility and not to a checklist, brightline rule, or point counting. The sex and age of the victims are merely factors which weigh for or against admissibility along with all other relevant facts. See State v. Dokken, 385 N.W.2d 493, 497-98 (S.D.1986) (adopting the same analysis under a “similar circumstances” criteria concerning the crime of manslaughter); see also State v. Fender, 358 N.W.2d 248, 253 (S.D.1984)(defendant’s prior violent acts committed on his wife held admissible in defendant’s conviction for aggravated assault on a policeman).
[¶ 163] Is the use of the buck-style folding knife or similarly-bladed knife, under these circumstances, sufficient to constitute a marked similarity or evidence common features, when combined with the other relevant facts? Moeller claims he is entitled to a “dispassionate, factual comparison between the prior acts and the allegations against Defendant.” That should be carried one step further with a dispassionate, factual comparison between the prior and present acts of this case and acts of a similar nature in other cases in this jurisdiction.6 For if what he claims is true, that this type of knife and the circumstances surrounding its use are common, surely other such common cases must have arisen' since 1978. However, Moeller fails to cite us to a single South Dakota case which is even remotely similar as to the facts. For that matter, he also fails to cite to any factually similar cases in any jurisdiction which support his argument.7
[¶ 164] The following South Dakota murder and manslaughter cases involved the use of a knife, however, none revealed the use of a buck-style knife or folding knife: State v. Blue Thunder, 466 N.W.2d 613 (S.D.1991)(butcher knife); State v. Bennis, 457 N.W.2d 843 (S.D.1990)(butcher knife); *501State v. Jenner, 451 N.W.2d 710 (S.D.1990), cert. denied, — U.S. -, 114 S.Ct. 81, 126 L.Ed.2d 49 (1993), (Chicago cutlery knife); State v. Bittner, 359 N.W.2d 121 (S.D.1984)(knife of unknown description); State v. Adams 418 N.W.2d 618 (S.D.1988), cert. denied, — U.S. -, 115 S.Ct. 754, 130 L.Ed.2d 653 (1995), and State v. Boykin, 432 N.W.2d 60 (S.D.1988)(fillet knife); State v. Gregg, 405 N.W.2d 49 (S.D.1987)(knife from a sheath of unknown description); State v. Ashker, 412 N.W.2d 97 (S.D.1987) and State v. Novaock, 414 N.W.2d 299 (S.D.1987)(knife of unknown description).
[¶ 165] As this is a case involving a rape-murder, it is interesting to note that none of the other reported rape-murder cases in this jurisdiction were committed with any type of knife: State v. Davi, 504 N.W.2d 844 (S.D.1993)(death by strangulation); State v. Rough Surface, 440 N.W.2d 746 (S.D.1989)(deeedent beaten to death); State v. White, 538 N.Y.2d 237 (S.D.1995)(death by rupture of a blood vessel in the victim’s brain).
[¶ 166] Only one non-murder case could be located where a rape was committed with a knife: State v. St. Cloud 465 N.W.2d 177 (S.D.1991)(knife of unknown description).
[¶ 167] The attacks against Beshaw, Moore, Warner, and Becky were also aggravated assaults. This jurisdiction’s case law concerning aggravated assaults shows the following involved a knife: State v. Washington, 537 N.W.2d 380 (S.D.1995)(carpet knife); State v. Chase in Winter, 534 N.W.2d 350 (S.D.1995)(knife of unknown description); Two Eagle v. Leapley, 522 N.W.2d 765 (S.D.1994) and State v. Gallegos 316 N.W.2d 634 (S.D.1982)(defendant pulled a knife of unknown description upon policemen). The only two aggravated assault cases which are remotely similar concerning weapons are State v. Ganrude, 499 N.W.2d 608 (S.D.1993) and State v. Rios, 499 N.W.2d 906 (S.D.1993). However the balance of the facts are strikingly different from the case now before us. In Ganrude, the defendant pulled a switchblade of unknown size on an adult male at the State Fair. The victim was not physically injured. In Rios, the defendant committed the aggravated assault with a lock-blade knife of unknown size. Rios involved an altercation between two young males in a Rapid City mall. Apparently the victim was not injured with the knife. Neither case involved sexual assault. Further, both Gan-rude and Rios were decided more than a year after the trial court made its ruling in this case.8
[¶ 168] Moeller’s claim that the evidence of prior bad acts is generic or common fails on another note when the above South Dakota case law is reviewed. In all of the above cases dealing with weapons, the so-called common issue of prior bad acts was never even raised with the exception of Davi, 504 N.W.2d 844. In Davi, we upheld the admission of the defendant’s prior threats and harassing phone calls against the decedent.9
*502[¶ 169] It is instructive to note that in Dokken, 385 N.W.2d 493, we held it was not an abuse of discretion to admit a prior bad act committed by the defendant which occurred previous to the charged homicide where both the prior bad act and the homicide involved the use by the defendant of the same weapon, a gun. Therein we held:
[t]he term “modus operandi” is included with the “plan” exception in SDCL 19-12-5. Thus, we stated that the plan exception requires that the former acts should indicate, by common features, a plan or design which tends to show that it was carried out by doing the act charged.
Id. at 497 (citing Willis, 370 N.W.2d at 198).10
[¶ 170] The trial court relied on State v. Martin, 118 Idaho 334, 796 P.2d 1007 (1990), in its determination of admissibility as Martin closely parallels the case now before us. In Martin, two prior bad acts concerning sexual assaults were admitted as relevant to the modus operandi with the current charge also being a sexual assault. Affirming the trial court under an abuse of discretion standard of review, the Idaho Supreme Court noted that in all three cases, the perpetrator used “a kitchen knife to perpetrate the crime.” Id. at 1011. However, this was not the same kitchen knife, but similar kitchen knives obtained from the three victims’ kitchens.11 Likewise, in Commonwealth v. Keizer, 377 Mass. 264, 385 N.E.2d 1001 (1979), the Massachusetts Supreme Judicial Court stated the question of prior bad acts admissibility concerning weapons required that they be similar “not just in the generic sense, but in terms of specific characteristics.” Id. at 1004. There the court admitted a prior bad act where both crimes involved “what appeared to be a square-barrelled pistol and a sawed-off shotgun concealed by a paper bag.” Id. See State v. Breazeale, 238 Kan. 714, 714 P.2d 1356, 1358 (1986), cert. denied, 479 U.S. 846, 107 S.Ct. 164, 93 L.Ed.2d 102 (1986), (same result involving the use by the defendant of “a small handgun”); State v. Thomas, 110 Ariz. 106, 515 P.2d 851, 853 (1973)(where-in the court relied in part on the defendant’s use of “a knife” and that “the geographical location of the crimes was either in the defendant’s neighborhood or in areas where he previously resided”); State v. Braman, 191 Conn. 670, 469 A.2d 760, 764 (1983) (“a shotgun with a cut-down configuration and a small automatic pistol”); Phillips, 131 Ill.Dec. at 134, 538 N.E.2d at 509; and People v. Tate, 87 Ill.2d 134, 57 Ill.Dec. 572, 577, 429 N.E.2d 470, 475 (1981) (deeming the use of *503similar weapons a “distinctive link” between past and present offenses).
[¶ 171] The majority argues that it cannot be ascertained for certain that Becky was killed with exactly the same type of knife that was used in the prior bad acts. Although this argument is correct to a point, it does not apply the appropriate standard. The proper standard under Haston, Martin, Breazeale, Thomas, 515 P.2d 851, Braman, Keizer, Phillips, and Tate as applied to the facts of this case, establish that the weapon used in Becky’s death is consistent with the weapons used in the prior bad acts, comparing all known criteria. If major inconsistencies arise as to the type of weapon, then clearly the similarity as to the weapon used has not been established. If all points are similar, as they are here, then it is up to the sound discretion of the trial court to determine if they contain a marked similarity or establish common features. Thomas, 381 N.W.2d at 237; Willis, 370 N.W.2d at 199.
[¶ 172] Moeller’s claim that the facts of this case, including the distinctive knife and blade size, are generic has no basis in the record and certainly does not have any such merit when compared to other cases decided by this Court. In short, no reported case in this jurisdiction, the facts of which include a buck or folding-blade knife or, for that matter, any type of knife blade of that size used to commit a sexual assault under circumstances similar to the case now before us, involve anyone other than one person — Donald Moeller. See State v. Moeller, 511 N.W.2d 803 (S.D.1994). In that case Moeller was convicted of two counts of aggravated assault for the attack on Warner.
[¶ 173] While it is necessary that the prior bad acts evidence bear a substantial degree of similarity to the present crime in order to show a modus operandi, see Werner, 482 N.W.2d at 289-90 and State v. Christopherson, 482 N.W.2d 298, 301-02 (S.D.1992), the circumstances of the prior offenses need not be identical to those of the crime charged. Martin, 796 P.2d at 1010 and Breazeale, 714 P.2d at 1362; cf. Werner, 482 N.W.2d at 289-90. Herein Moeller’s prior bad acts demonstrate his “handiwork” (Champagne), contain a “marked similarity” or “close parallel” (Thomas), and establish “common features” (Willis).
[¶ 174] Fundamentally there is nothing novel about this issue. It is simply a circumstantial evidence question, the type of evidence we routinely instruct juries about in this state. However, there is one major difference. Unlike the jury, we do not find the facts beyond a reasonable doubt. Rather, we focus on the issue from the standard: was the trial court clearly erroneous in its findings that the prior bad acts were the defendant’s handiwork or contained common features? When applying the above facts and authority, I believe a “judicial mind, in view of the law and the circumstances, could reasonably have reached that conclusion”' and thus the trial court did not abuse its discretion when it found the evidence of sufficient probative value concerning the modus oper-andi. Dakota Cheese, 525 N.W.2d at 715.
[¶ 175] II. Prejtidice
[¶ 176] If the relevancy question has been determined in favor of the State, the trial court must then proceed to perform a balancing test to determine whether the prejudicial effect of the evidence substantially outweighs its probative value. SDCL 19-12-3; Werner, 482 N.W.2d at 289. As noted by the majority, “‘prejudice’ does not mean the damage to the opponent’s case that results from the legitimate probative force of the evidence; rather, it refers to the unfair advantage that results from the capacity of the evidence to persuade by illegitimate means.” State v. Iron Shell, 336 N.W.2d 372, 375 (S.D.1983).
[¶ 177] Moeller advances two arguments in regards to this question. He argues that he was unduly prejudiced due to the sensational nature of the case when combined with the other bad acts and that at least two of the bad acts were too remote in time to be admissible.
[¶ 178] Herein the trial court determined there was substantial need for the prior bad acts evidence by the State because this was a circumstantial evidence case and Moeller had given notice of an alibi defense. The trial court stated in its Memorandum Opinion, *504“the defendant, in his brief, admits that because the State’s case is based on purely circumstantial evidence, ‘identity of the perpetrator is the central issue that will be before the jury.’” The court found that there was no other evidence of equal probative value.
[¶ 179] Recently in White, 538 N.W.2d at 243, we adopted the same premise. We held that “the court determines whether the danger of unfair prejudice substantially outweighs the probative value of the evidence ‘in view of the availability of other means of proof and the other factors under SDCL 19-12-3 (Rule 403).” Id citing State v. Basket, 468 N.W.2d 413 (S.D.1991). In White, we affirmed the trial court’s admission of the defendant’s prior bad acts because “ ‘the probative value is high. The purported victim is dead. There are no other witnesses concerning the issue of the circumstances under which the [Defendant was admitted to the decedent’s home.’ ” Id (quoting the trial court’s balancing of the probative value against the prejudicial effect).
[¶ 180] Herein the same rationale should apply. Becky is dead. There are no eye witnesses as to the identity of her abductor, rapist and killer. There is no other evidence available to replace the prior bad acts in assisting to identify the perpetrator. Such a finding weighs in favor of admissibility. Id; Werner, 482 N.W.2d at 290. See also U.S. v. Ingraham, 832 F.2d 229, 237 (1st Cir.1987), cert. denied, 486 U.S. 1009, 108 S.Ct. 1738, 100 L.Ed.2d 202 (1988) (“Because the unknown in the case was whether X = Ingraham, the probative worth of this [prior bad acts] evidence far outstripped any unfairly prejudicial effect.”) (emphasis in original).
[¶ 181] Given the finding of need for this information, the trial court proceeded to do everything it could to avoid any unfair prejudice. The trial court determined it would instruet the jury as to the limited nature of the other acts evidence prior to the individual testimony of Beshaw, Moore and Warner as well as at the close of the trial. See State v. Means, 363 N.W.2d 565, 569 (S.D.1985). Juries are presumed to follow the instructions of the trial court. Id. (citing State v. Reddington, 80 S.D. 390, 125 N.W.2d 58 (1963)). The trial court also refused to admit into evidence at trial an alleged prior assault by Moeller on his wife with a generic knife of unknown origin because of “insufficient evidence.” Further to avoid sensationalism, the trial court refused to allow admission of evidence that, when arrested, Moeller had in his possession pornographic magazines and a “registered pervert” emblem.12
[¶ 182] I must also respectfully part company with the majority’s conclusion that “[t]he other acts testimony was a shocking and emotionally gripping contrast to State’s sterile circumstantial case.” The “State’s sterile circumstantial ease” was the brutal multiple rape and murder by repeated stabbings of a totally innocent young girl. By contrast, Beshaw was not injured at all, and Moore and Warner appear to have sustained no life-threatening or serious injuries. While the prior bad acts were hardly pleasant for the victims, claiming them to be “shocking and emotionally gripping” when compared to Becky’s murder is not supportable.
[¶ 183] In other cases, unfair prejudice has been claimed by the defendant on the basis that it requires the defendant to not only defend himself against the crime with which he is charged, but allegations of other bad act crimes which may or may not be true. Houghton, 272 N.W.2d 788. Herein no such unfair prejudice arises as Moeller does not contest the existence of the prior bad acts.
[¶ 184] Regarding the issue of remoteness, it is clear the Beshaw incident occurred in 1973 and the Moore incident occurred in *5051979. Thus, seventeen and eleven years have passed between these acts and Becky’s rape and murder. In Werner, this Court held:
Werner claims the other acts testimony was too remote from the acts charged. When standing on its own, the testimony from these women may appear remote in time. In fact, the other acts evidence spanned several years. However, in determining probative value, remoteness must be considered with other factors, such as reliability and necessity. State v. Titus, 426 N.W.2d 578 (S.D.1988).
Furthermore, ‘[t]he trial court must consider the nature of the offenses, the similarity of occasions and locations as well as the time elapsed between incidents.’ Id. at 580. ‘[W]hether prior acts are too remote must realistically depend on their nature.’ State v. Wedemann, 339 N.W.2d 112 (S.D.1983). Moreover, ‘each case depends upon its own particular facts as to a limitation, regarding vintage, on the remoteness. Admission of prior acts must realistically depend upon their nature.’ Titus, 426 N.W.2d at 582 (Henderson, J. concurring specially).
Werner, 482 N.W.2d at 289.13
[¶ 185] The trial court followed Werner and concluded on the remoteness issue:
Attached to the State’s brief is a copy of the rap sheet for the defendant. It is clear that the defendant has been confined in prison the greater portion of his adult life. It further appears that the criminal activity of defendant abated only during periods of incarceration. Defendant plea-bargained a sexual assault charge to a theft plea in 1973 and was incarcerated more or less continuously from 1973 to 1979. Defendant was again in the penitentiary from 1979 to 1988. It would seem that defendant’s incarceration is really the only obstacle to a continuous life of crime. Further, this would explain the lack of other similar criminal activity of more recent vintage.
It is the opinion of this court that the State should not be deprived of the use of such evidence simply because defendant was denied his freedom by reason of his own criminal activity which deprived him of the opportunity to commit such crimes. When defendant’s period of incarceration is taken into account, the period of opportunity between the prior crimes and the principal offense is approximately two years.
See Breazeale, 714 P.2d at 1356 (ten years’ time not found to be remote in light of fact defendant spent intervening years in prison), and Martin, 796 P.2d at 1014 (period of ten and twelve years not found to be remote where defendant was incarcerated nearly the entire period).
[¶ 186] Based on the extensive pre-trial hearings and briefs on the prior bad acts issue, the trial court clearly had sufficient information to properly perform the balancing test as required by SDCL 19-12-5. State v. Chapin, 460 N.W.2d 420, 422 (S.D.1990). The trial court’s determination whether the probative value of the bad acts evidence is substantially outweighed by its prejudicial effect is an issue left to the sound discretion of the trial court, and we will not overturn it on appeal absent a “clear abuse of discretion.” Werner, 482 N.W.2d at 290-91, Champagne, 422 N.W.2d at 842. I conclude that based on the record, no such abuse of discretion occurred as “a judicial mind, in view of the law and circumstances, could reasonably have reached that conclusion” concerning the prejudice issue. Dakota Cheese, 525 N.W.2d at 715.
[¶ 187] In conclusion, I respectfully dissent on Issue One. I am unable to find any evidence in the record or relevant case law from this jurisdiction to support Moeller’s attempt to show that this type of knife and the totality of the circumstances of its use are not unusual in the commission of rape-murders in this state. Thus, the theory is reduced to *506arguably perusing the aisles of local stores to conclude such knives are generic or commonly offered for sale to the public. The so-called fact that these knives are commonly offered for sale proves nothing by itself. So are many other articles of merchandise, some harmless and some possessing potential for harm. For such an argument to have merit, however, one must make the leap from items commonly offered for sale to items used to commit forced rapes, aggravated assaults or murders. Herein Moeller and the majority opinion are unable to cite any facts or cases from this jurisdiction to support this thesis.
[¶ 188] I end this issue as I began it. Our standard of review is abuse of discretion. Did the trial court lack “a judicial mind” in view of. the law and the facts of this case in arriving at this decision of admissibility? Another jurist once defined abuse of discretion as shooting at a target. You did not need to hit the bulls-eye but did need to hit the target. I would submit herein the trial court hit the target.
[•¡¡189] ISSUES TWO, THREE, FIVE, SIX, AND SEVEN
[¶ 190] I concur.
A191] ISSUE FOUR
DOES A TRIAL COURT’S FAILURE TO INSTRUCT THE JURY AS TO THE DEFINITION OF “REASONABLE DOUBT” AT THE SENTENCING HEARING REQUIRE REVERSAL AND VACATION OF THE JURY’S VERDICT OF DEATH?
[¶ 192] As the majority would reverse on Issue One, it did not reach this issue. With my view that Issue One should be affirmed, I am compelled to address this issue on the merits.
[¶ 193] In South Dakota a person cannot be sentenced to death unless the jury determines the existence of an aggravating factor beyond a reasonable doubt. SDCL 23A-27A-5. Herein the jury was correctly instructed at the guilt phase of the trial as to the definition of reasonable doubt. However, due to inadvertence at the sentencing phase, no similar instruction was given.
[¶ 194] At the outset, the State argues that this issue was waived since Moeller failed to offer a proposed instruction on the subject. State v. Holloway, 482 N.W.2d 306, 309 (S.D.1992). The State then proceeds to argue that the “plain error” doctrine can not apply since this error “is neither obvious nor substantial.” Holloway, 482 N.W.2d at 309.
[¶ 195] I would apply the plain error rule as I disagree with the State on both points. SDCL- 23A-27A-3 specifically requires the trial court to properly instruct the jury at the penalty phase of a proceeding. “Upon the conclusion of the evidence and arguments of counsel, the judge shall give the jury appropriate instructions _” (emphasis added). Therefore the State’s argument that any error is corrected by the fact the jury was properly instructed at the guilt phase conflicts with the explicit terms of SDCL 23A-27A-3.
[¶ 196] Further, there is nothing in the record to reflect that the non-law-trained jury, in this most pressure-packed of situations, could and did, accurately recall the reasonable doubt instruction it had previously been given during the guilt phase of the trial. It did not have that instruction in the jury room with it at the penalty hearing. A review of the penalty phase jury instructions shows no reference to the guilt phase instructions, nor do the penalty phase instructions direct the jury to recall or refer back to the guilt phase instructions, based on its memory. How many members of the bench and bar can accurately recite the reasonable doubt instruction from memory? This clearly was an error of an obvious and substantial nature as the other penalty phase instructions made at least ten references to “reasonable doubt” in those instructions without defining it. Given the fact that the above statutes specifically required the trial court to instruct on the definition of reasonable doubt and the fact it failed to do so, I would hold this is plain error and therefore Moeller did not waive the issue for review on appeal to this Court.
[¶ 197] The State then proceeds to argue that if this is a technical violation of a statute, it is not a structural defect affecting the *507fairness of the trial or, in other words, it is harmless error.
[¶ 198] When the jury is the final sentencer, it is essential that the jury be properly instructed “regarding all facets of the sentencing process.” Walton v. Arizona, 497 U.S. 639, 110 S.Ct. 3047, 111 L.Ed.2d 511 (1990), Roberts v. Louisiana, 428 U.S. 325, 96 S.Ct. 3001, 49 L.Ed.2d 974 (1976). In Sullivan v. Louisiana, 508 U.S. 275, 113 S.Ct. 2078, 124 L.Ed.2d 182 (1993), the Supreme Court held that an improper definition of reasonable doubt in a jury instruction is a structural constitutional error in the trial which cannot be harmless. The reason is obvious:
But the essential connection to a ‘beyond a reasonable doubt’ factual finding cannot be made where the instructional error consists of a misdescription of the burden of proof, which vitiates all the jury’s findings. A reviewing court can only engage in pure speculation — its view of what a reasonable jury would have done. And when it does that, ‘the wrong entity judgefe] the defendant guilty.’
508 U.S. at 281, 113 S.Ct. at 2082, 124 L.Ed.2d at 190. (emphasis original).
[¶ 199] The State relies upon State v. Holmes, 464 N.W.2d 612, 613 (S.D.1990), wherein we held that it was harmless error when the trial court failed to instruct on the “presumption of innocence.” We arrived at this conclusion based on our view that the record established the defendant’s guilt on an overwhelming basis. The State cannot claim such is the state of the record here. It argued, and the trial court accepted the premise, as do I, that the prior bad acts were essential to be admitted based on the circumstantial nature of the facts of this case and lack of direct proof of identity of Moeller as the perpetrator of the crime.
[¶ 200] To me, allowing the jury to speculate as to the standards for reasonable doubt is as major an error as a jury that received guidance from the court, albeit, not accurate guidance. In the end, the result is the same — -a conviction not shown to be based upon the Constitution and the correct rule of law. This was a prejudicial error of both statutory and Constitutional magnitude.
CONCLUSION
[¶ 201] I would affirm the conviction on the murder charge. I would reverse and remand on the penalty verdict of death to allow the State to impanel a new jury for a retrial of the penalty phase. If the State elects not to do so, the conviction for murder in the first degree under SDCL 22-6-1 would be life imprisonment in the state penitentiary with no possibility for parole.
. For the most part, this is correct. Moeller had a passing knowledge of Becky, Moore and Warner. However, the trial court erroneously found Moeller had prior knowledge of Beshaw when he did not. The balance of the finding is correct, however. Moeller did not have a substantial relationship with any of these victims prior to his attacks upon them.
. Dr. Randall testified:
This particular wound extended approximately four inches primarily front to back and towards the right.... And the abrasion around here suggests that the blade was inserted to the hilt. Because that’s usually what we see causing this type of abrasion is a hilt mark. If the blade isn't inserted to the hilt, then we don’t see this type of abrasion around the edge of the wound.
Later Dr. Randall was asked:
Q: Were you able to estimate roughly the length of the blade that would have caused the wound?
A: No. As I mentioned, the depth of the wound was four inches but a shorter blade could have produced that due to the compression of the chest.
. In People v. Phillips, 127 Ill.2d 499, 131 Ill.Dec. 125, 130-34, 538 N.E.2d 500, 505-09 (1989), cert. denied, 497 U.S. 1031, 110 S.Ct. 3290, 111 L.Ed.2d 798 (1990), the Illinois Supreme Court held that where the defendant is charged with a crime evidencing acts of physical violence, such as forced rape and murder, admission of a prior bad act rape did not have as a prerequisite establishing violent acts causing physical injury. Rather, the test is the overall similarities of the two crimes.
Clearly if the prior bad acts crimes also contained similar violence, as do two of the three herein, that would be a point in favor of admissibility. In its conclusion of admissibility, the Phillips trial court, like the Moeller trial court, identified ten factors of similarity.
. Recent statistics indicate rape victims know their assailants well in eighty-four percent of the cases. Marcia G. Pfeiffer, Date Rape: The Reality, 17 SULRev 283, 284 (1990). One commentator noted this figure demonstrates "acquaintance rape is more common than left-handedness,' heart attacks, and alcoholism.” Linda Robayo, The Glen Ridge Trial: New Jersey's Cue to Amend its Rape Shield Statute, 19 Seton Hall LegisJ 272 at n. 87 (1994).
. The Houghton Court upheld the trial court's refusal to admit the prior bad acts on the basis of undue prejudice to the defendant combined with a minimal probative value. In Houghton, the defendant convinced both the trial court and this Court that identity was not an issue, thus negating any use of the prior bad acts to prove his identity. The issue became one of consent to sex by the purported victim. Herein, Moeller has admitted to the trial court that identity of Becky's killer is a major issue thus putting into question the matter of modus operandi. Second, in Houghton, the defendant vigorously contested he committed the purported prior bad acts. Moeller, however, did not contest his commission of the prior acts. Finally, in Houghton, we applied the proper abuse of discretion standard in affirming the trial court's finding of denial of admissibility. Here, we analyze the opposite result — the trial court’s finding of admissibility under the same abuse of discretion standard.
. This type of caselaw analysis is a mandatory obligation of this Court at the sentencing phase of a death penalty case placed upon us by the Legislature pursuant to the enactment of SDCL 23A-27A-12(3). This statute requires, in part, that this Court engage in a comparison of death penalty cases with "similar cases, considering both the crime and the defendant." See generally State v. Rhines, 1996 SD 55, ¶ 85, 548 N.W.2d 415.
.Moeller relies on the following cases to support his argument. In People v. Alcala, 36 Cal.3d 604, 205 Cal.Rptr. 775, 685 P.2d 1126 (1984), cert. denied, - U.S. -, 114 S.Ct. 215, 126 L.Ed.2d 171 (1993), no weapon was used by the defendant to commit his prior sexual acts. In White v. Commonwealth, 9 Va.App. 366, 388 S.E.2d 645 (1990), overruled on other grounds by Lavinder v. Commonwealth, 12 Va.App. 1003, 407 S.E.2d 910 (1991), a knife of unknown description was used to commit the charged rape and a prior rape. In Foster v. Commonwealth, 5 Va.App. 316, 362 S.E.2d 745 (1987), a "small handgun" was used to commit the charged rape and a prior sexual assault. In State v. Hansen, 187 Mont. 91, 608 P.2d 1083 (1980), no weapon was used in a series of sexual assaults after the victims were picked up in bars.
. This is a listing of all relevant South Dakota cases that could be located decided after the adoption of SDCL 19-12-5 in 1978. The trial court made its ruling on admissibility of the prior bad acts on June 1, 1992. It clearly could not anticipate the future as to the facts and our decisions in cases that arose thereafter. Nevertheless, the subsequent cases are listed for the sake of completeness and to demonstrate that no case law has arisen since the trial court’s decision which would support Moeller's argument that the facts of this case and the prior bad acts are generic.
. Moeller and the majority argue that the use of the knife and the circumstances surrounding its use are common. Black's Law Dictionary 275 (6th ed.1990) defines “common" in part as:
Belonging or pertaining to ■ many or to the majority. Generally prevalent, of frequent or ordinary occurrence or appearance; familiar by reason of frequency. Also, usual, customary, and habitual .... (citations omitted).
Either fact situations are common or uncommon. They cannot be both. Cited herein is every rape, murder and assault case reported in this jurisdiction by written opinion or pursuant to SDCL 23A-27A-13 since 1978 when SDCL 19-12-5 was adopted. These cases are cited to show that if the events surrounding the case now before us are "common,” by definition there should be other "common” cases reported. According to the above definition such similar cases should be "generally prevalent or frequent or ordinary ... usual, customary and habitual.” Instead there are none, not one, save Donald Moeller. Given this, I am not surprised that the majority's conclusion, "[f]ocusing on only reported cases is an unrealistic and unscientific means of deciding whether the use of a folding or buck-style knife is a unique characteristic” is made without citing supporting authority.
. This Court has sought to compile information on all potential death penalty cases of first degree murder for possible proportionality review. SDCL 23A-27A-13. Many of these cases have already been cited herein. However, in numerous murder cases the defendant entered a plea of guilty and there was no appeal to this Court or the cases were not cited above because of factual differences. It appears that the following methods of killing were used in those cases not reported above: handgun, ten convictions; rifle or shotgun, four convictions; gun of unknown description, four convictions; strangulation or smothering, seven convictions; beating, three convictions; knife of unknown description, three convictions, beating and knife of unknown description, one conviction; auto jack, one conviction; and no indication, three convictions. There is not a single case involving a homicide with a buck-style folding knife with a three- to four-inch blade.
. As noted above, the facts and analysis in Martin are important because it is the framework upon which the trial court herein based its analysis and ruling of admissibility. The majority lists fourteen criteria in Martin which leads it to conclude, "[t]he numerous and distinctive similarities detailed in Martin are noticeably absent in the case.” This is curious in light of the fact that factor one in Martin is "all were rape-type cases," factor three "all victims knew Martin," factor five "all victims were surprised by their assailant," factor eight, "the assailant always used a kitchen knife to perpetrate the crime," factor nine, "the knife always came from the victim's kitchen” (the same source), factor eleven, "apparently, the assailant always left without taking objects from the residence," factor twelve, “the assailant always placed the knife by the victim’s throat to perpetrate the crime,” factor thirteen, "the assailant always threatened to kill the victim if she did not comply with his desires” and factor fourteen, "the victims were all injured by the knife.” Compare with ten factors recognized by the trial court listed at p. 496 of this dissent.
The majority concludes that Martin "identified a long list of similarities between the other acts and the charged offense.” That is fourteen versus the ten identified by the trial court herein. I know of no legal doctrine that decides such issues by rote number. Rather it is the "totality of circumstances” that controls. (See pp. 498-499 of this dissent).
. The majority concludes:
Unable to define concrete similarities among the other acts and charged offenses, the State failed to give the jury any clear guidance concerning the relevance of the other acts to legitimate issues in the case. Without more compelling direction from the State, the jury’s verdict almost certainly rested on the forbidden inference that because Moeller was involved in other sexual assaults, he likely committed the offenses against Becky.
I respectfully submit that the majority errs when it places this obligation upon the prosecution rather than the trial court and fails to take into account the effect of the trial court’s instructions to the jury on the prior bad acts.
. This Court has previously affirmed the admission of prior bad acts with a span of 17 years and more. In Werner, 19 years ran between the first act of sexual misconduct and the time of trial. In Christopherson, 482 N.W.2d 298, 17 years elapsed between the first molestation and trial. In State v. Ondricek, 535 N.W.2d 872 (S.D.1995), 25 years expired between the first sexual misdeeds and trial.