delivered the opinion of the Court.
In this criminal appeal, we assess whether joinder principles were applied properly in the context of defendant’s trials on offenses that arose out of one burglary and three other burglary and sexual assault episodes, which occurred over a span of three years. If joinder of defendant’s offenses was improper, we also are asked to consider whether all of the convictions must be reversed.
The State contended that the factual underpinnings to the offenses with which defendant was charged—burglaries of women’s homes and related sexual assaults—bore indicia of “signature crimes” and sought to try defendant in a single trial. Defendant wanted the offenses tried separately and filed a motion for severance, which was granted in part. One burglary and sexual assault incident was severed from two others. The State was permitted to try the fourth incident, which involved burglary but not sexual assault, in either of the two trials. The State opted to try defendant in the first proceeding for two burglary and sexual assault episodes and for the separate burglary incident during which defendant was apprehended. A second trial before a new jury addressed the offenses involving the severed burglary and sexual assault episode, followed by a trial on a certain persons offense.
*72On appeal, the Appellate Division reversed defendant’s convictions from the first trial. The panel concluded that the threshold for a signature crime was not satisfied on the facts present for the burglary and sexual assault incidents. It further concluded that the trial court abused its discretion in not granting defendant relief from prejudicial joinder in respect of those charges and also the charges related to the burglary. The Appellate Division also reversed defendant’s convictions for the burglary and sexual assault incident that was separately tried based on the wrongful admission of extensive other-crimes evidence related to the burglary in which defendant was apprehended. The State’s petition brought this matter before the Court.
As did the appellate panel, we find error in the joinder of offenses in defendant’s first trial. We also agree with the panel that admission of the other-crimes evidence in the second trial was error. However, not all of the convictions require reversal and retrial. The quality and quantum of the evidence against defendant in two of the criminal episodes was of sufficient weight to lead us to conclude that the error, fairly viewed, did not produce an unjust result. Accordingly, we affirm in part and reverse in part the judgment of the Appellate Division.
I.
At the outset, before extensively detailing the criminal episodes at the center of this dispute, we review the basic principles governing joinder of offenses.
Rule 3:7-6 provides that
[t]wo or more offenses may be charged in the same indictment or accusation in a separate count for each offense if the offenses charged are of the same or similar character or are based on the same act or transaction or on 2 or more acts or transactions connected together or constituting parts of a common scheme or plan. Relief from prejudicial joinder shall be afforded as provided by R. 3:15-2.
Although joinder is favored, economy and efficiency interests do not override a defendant’s right to a fair trial. See State v. Chenique-Puey, 145 N.J. 334, 341, 678 A.2d 694 (1996); State v. *73Coruzzi, 189 N.J.Super. 273, 298, 460 A.2d 120 (App.Div.) (“The interests of economy and efficiency may require that similar or related offenses be joined for a single trial, so long as the defendant’s right to a fair trial remains unprejudiced.”), certif. denied, 94 N.J. 531, 468 A.2d 185 (1983). Rule 3:7-6 expressly provides for relief from prejudicial joinder, referencing Rule 3:15-2(b), which vests a court with discretion to sever charges “[i]f for any other reason it appears that a defendant or the State is prejudiced by a permissible or mandatory joinder of offenses or of defendants in an indictment or accusation.”
The relief afforded by Ride 3:15-2(b) addresses the inherent “ ‘danger[,] when several crimes are tried together, that the jury may use the evidence cumulatively; that is, that, although so much as would be admissible upon any one of the charges might not have persuaded them of the accused’s guilt, the sum of it will convince them as to all.’ ” State v. Pitts, 116 N.J. 580, 601, 562 A.2d 1320 (1989) (quoting United States v. Lotsch, 102 F.2d 35, 36 (2d Cir.), cert. denied, 307 U.S. 622, 59 S.Ct. 793, 83 L.Ed. 1500 (1939)). A court must assess whether prejudice is present, and its judgment is reviewed for an abuse of discretion. See Chenique-Puey, supra, 145 N.J. at 341, 678 A.2d 694. The test for assessing prejudice is “whether, assuming the charges were tried separately, evidence of the offenses sought to be severed would be admissible under [N.J.R.E. 404(b) ] in the trial of the remaining charges.” Ibid, (internal quotation marks omitted). N.J.R.E. 404(b) requirements must be met, State v. Cofield, 127 N.J. 328, 338, 605 A.2d 230 (1992), and the evidence of other crimes or bad acts must be “relevant to prove a fact genuinely in dispute and the evidence is necessary as proof of the disputed issue,” State v. Darby, 174 N.J. 509, 518, 809 A.2d 138 (2002) (citations omitted).
With those guiding principles in mind, we turn to the offenses that the State sought to have tried together and the decisions of the trial court and Appellate Division in this matter.
*74II.
A.
Defendant originally was charged with offenses for acts committed against five different victims on five different dates between July 2002 and May 2005. Ultimately, he was tried in two trials for offenses committed against four victims.1 The episodes are summarized, based on the facts as adduced at trial, in the order in which the events occurred.
Sexual Assault of J.L.
On July 13, 2002, twenty-oné-year-old J.L., preparing to leave for a vacation, returned to her home in New Brunswick at approximately 2:30 p.m. Intending to be brief, she left the door unlocked. As she was gathering her clothing, she heard the front screen door swing shut and turned to see a man standing in her doorway. At the time, J.L. was living with her boyfriend and three other roommates. Thinking the man was there to see one of her roommates, J.L. asked the man if she could help him.
The man, wearing reflective sunglasses and a baseball cap, told J.L., “take your clothes off or I’m going to shoot you.” J.L. saw that the man was carrying “a very large gun,” which she could see was “shiny silver.” She screamed and turned around. The man came up behind her, put his hand over her mouth, and told her to “shut up” or he would shoot her. He pushed her down, face-first, onto the bed and locked the bedroom door.
Crying, J.L. offered the man $600 in cash, but he did not respond. The man returned to the bed and removed J.L.’s shoes, pants, and underwear; he also pulled down his own pants. When J.L. lifted her head to look up at him, he hit her in the jaw with *75the gun and his fist and warned her not to look at him again. The man attempted to penetrate her vagina with his penis, but he was unable to do so. He turned J.L. onto her back and penetrated her. J.L. thought she had heard the snap of the man putting on a condom, but she was not certain.
When the sexual act concluded, the man asked J.L. where she kept the money she offered. Still in fear he would shoot her if she looked at him, J.L. responded, while lying on the bed, that the money was in her car. The man went to the window and looked out. He told her to wait five minutes and that if she came outside sooner, he would shoot her. J.L. waited five minutes and then, fearing he would return, drove to St. Peter’s Hospital. The police then took J.L. from St. Peter’s to a rape crisis center. At the rape crisis center, J.L. submitted to a gynecological exam. Her clothes were bagged, and oral and vaginal swabs were taken. The investigation revealed that no money was taken from J.L.’s car.
J.L. told the police that she did not look at the man after he told her not to or she would be shot. She did, however, vaguely describe a black male of medium build, about six feet tall, wearing a baseball hat, reflective glasses, and a white t-shirt with a design on it. She also described the gun the man carried as very large and “really, really shiny and silver.”
Subsequent to defendant’s arrest, on May 27, 2005, J.L. viewed a lineup, which included defendant. She was unable to make an identification.
At trial, a State Police forensic scientist testified about the results of the DNA testing. The expert compared the swabs taken from J.L. against her DNA profile and defendant’s DNA profile, concluding that the cervical swab taken from J.L. matched defendant’s DNA profile. She also testified that the DNA profile obtained from the swab occurs in “approximately one in 4.50 trillion of the African-American population, one in 24.7 trillion of the Caucasian population and one in 29.3 trillion of the Hispanic population.” On cross-examination, the expert admitted that she had knowledge of prior instances of cross-contamination in the lab; *76however, she was not aware of any contamination issues since she began her employment there. She further testified that the sample contained no indication of contamination, which she would have been able to recognize if it occurred.
Sexual Assault of K.G.
On June 9, 2003, twenty-five-year-old K.G. was home alone in the apartment she shared with her sister in New Brunswick. At 3:30 a.m., she was asleep in her bed. The apartment doors were locked, but the kitchen window “was left open about an inch.” K.G. awoke to the sound of her bedroom door opening; she looked up and saw the silhouette of a large man carrying a knife.
The man told K.G. to keep quiet, and he threatened to cut her throat. K.G. begged the man to leave, but he slapped her across the face and told her to “shut up.” At that point, the man pressed the knife to KG.’s cheek and then “stabbed [the knife] really hard” into the bed next to her. He climbed onto the bed, removed K.G.’s pajama pants, and used the knife to cut off one side of her underwear.
The man was wearing latex gloves, and K.G. asked the man if he was going to use a condom. He replied that he would, and K.G. believed that he used one during the attack. The man asked K.G. if she “had ever been with a black man” and said, “I know you want to be with a black man, [and] things to that effect.” The man kissed her on the mouth and fondled her breasts, and K.G. felt the man penetrate her vagina with his penis.
Afterwards, the man put his pants on, got off the bed, and backed out of the room. As he did so, he told her “to stay still, not to do anything.” K.G. waited a minute or two and called the police. She had been “too terrified to look” at the man during the attack, and she could only describe him as a black male, “on the husky side,” and “big and imposing and threatening.” The man had told her he had been “watching” her, but she had not recognized him.
*77The police found that the kitchen window was wide open, and a lawn chair had been placed underneath the window on the outside. A large knife was missing from the kitchen, but the police recovered the knife from the backyard. K.G. said the knife looked like the one used by her attacker. The police also found a condom wrapper on K.G.’s bed, which she explained was different than the brand that she used with her boyfriend. The police did not locate any witnesses that saw or heard anything, and they took K.G. to a rape crisis center. A nurse examined K.G., but the examination did not reveal any nuclear DNA evidence. However, a hair was found on K.G.’s pajama pants.
On June 6, 2005, subsequent to defendant’s arrest, K.G. viewed a lineup of six men. Each man read a quote out loud, “shut up or I’ll cut your throat.” K.G. was able to easily dismiss all the men except one: defendant. K.G. noted that his voice was “familiar” and that he had a similar build to her attacker, but she was unable to identify him with certainty.
A State Police forensic expert testified at trial as an expert in the field of forensic hair analysis and explained to the jury the significance of trace analysis and her discovery of the hair on the pajama pants. She testified that the hair found on the pants exhibited characteristics normally associated with the hair of a person of African descent and that she sent the hair out for mitochondrial DNA analysis.
The State’s mitochondrial DNA expert explained that mitochondrial DNA is often only used when nuclear DNA cannot be obtained. The expert explained that, unlike nuclear DNA, mitochondrial DNA is not a unique identifier because it is only inherited from one’s mother. He further explained that a person will have the same mitochondrial DNA type as all of the relatives on the maternal side, and, thus, when doing a mitochondrial analysis, one can only conclude that a person “can’t be excluded as the contributor of that sample.”
The expert then described his analysis of the mitochondrial DNA in this case, explaining that the mitochondrial DNA sequenc*78ing from both defendant’s sample and K.G.’s pants was the same and that defendant could not be excluded as the source of the hair. The expert also performed a database search using the F.B.I. database, which indicates how common a particular type of mitochondrial DNA is in the general population. The search revealed that this type of mitochondrial DNA sequence was very rare and that no more than .06% of North Americans would be expected to have this type. In other words, according to the expert, the same DNA sequence would not be seen in more than six out of every 10,000 people.
Sexual Assault of L.R.
On January 18, 2005, thirty-nine-year-old L.R. was alone in the apartment she shared with her son in Edison. She had just finished dinner at approximately 6:45 p.m. and went into her son’s room to use the phone to pay her bills. When she exited her son’s room, the front door was open, and a man she did not know was standing in her apartment. L.R. screamed, and the man told her not to scream and that he would not hurt her. The man asked if she had any money, and L.R. replied that it was in her bag in another room. The man then told her to take off her clothes. L.R. told him to take her money and go, but he repeated that she should take off her clothes.
L.R. began screaming, and the man again told her to stop and that he was not going to hurt her. The man approached L.R. holding a knife and cut the buttons off her shirt and cut her bra. He then told her to remove her pants. L.R. continued to scream, and the man pushed her onto the bed and took off her pants. He then cut the left side of her underwear with the knife. He licked her breasts and neck, and penetrated her vagina with his penis.2 After he concluded the sexual act, the man warned L.R. not to call the police or he would come back and rape her again. He backed out of the room and continued to threaten her if she called the *79police, saying to her, “now I have white p* * * * and I’m going to f* * * you again.”
L.R. called the police after the man left, and she was taken to a rape crisis center and examined. At the rape crisis center, L.R. was physically examined, and the nurse took swabs of the left side of L.R.’s neck, her left nipple, her left foot, and her pubic area. She described her attacker as a black male with a dark complexion, between five-ten and six feet tall, with a straight nose and round big eyes. She said he appeared to have a chubby build but that may have been due to the amount of winter clothes he was wearing. L.R. also described him as having some sort of speech defect; when he spoke, it appeared as if “his tongue come [sic] out from his teeth,” making it sound like he said, “I’m not going to hurth you.” During the attack, L.R. was facing the attacker, but the room where she was attacked only had one light with a sixty-watt bulb, and the man was wearing a jacket with a hood over his head, which obscured her view of the side of his face. On February 11, 2005, L.R. viewed a lineup of six men. She identified a man, who was not defendant, saying she was “ninety percent” sure he was her attacker. On May 31, 2005, subsequent to defendant’s arrest, L.R. viewed another lineup of six men. This time, defendant was part of the lineup. As soon as defendant came out to take his spot in the lineup, L.R. covered her mouth, moving away from the door and up against the wall. She began to cry and said, “it’s him, it’s him.” Sergeant Pamela Jeffrey, who administered the lineup, testified to L.R.’s strong emotional reaction.
At trial, L.R. testified that she was “so shocked” when she saw defendant and that she was 100 percent certain at the lineup that defendant was the man “who was standing in my apartment with a knife.” Additionally, L.R. made an in-court identification of defendant as the man who attacked her.
An expert in forensic science and DNA analysis conducted a DNA analysis on the swabs taken from L.R. DNA evidence was obtained from the saliva that the attacker left on L.R.’s breasts *80and neck. The expert compared the evidence to a swab of defendant’s DNA and determined, “within a reasonable degree of scientific certainty,” that defendant was the source of the dried secretions taken from L.R.’s breasts and neck. At trial, the expert testified that the DNA profile obtained only “occurs in approximately one in 30.8 quadrillion of the African American population^ o]ne in 419 quadrillion of the Caucasian population, and one in 2.02 quintillion of the Hispanic population.” Although defense counsel asked the expert about cross-contamination in DNA evidence, she testified that she had not heard of any specific incidents at her laboratory in the past three years that she had worked there.
Burglary of S.P.
On May 27, 2005, thirty-one-year-old S.P. was living in an apartment in North Brunswick with her fiancé and three children. That night, she and her fiancé argued, and S.P. left the apartment to go for a walk at 2:40 a.m. She was outside for approximately one minute when a man approached her and said, “hey, mama, can I talk to you?” S.P. turned around, looked at him, and headed back toward her apartment. The man followed her and told her that she did not have to worry because “he was a gentleman.” S.P. told him that her fiancé was home, and she ran into the apartment and locked the door. She told her fiancé that someone had tried to talk to her and then went to her bedroom to lie down.
A few minutes later, S.P.’s flaneé noticed that the door handle was jiggling. Then he saw the blind, screen, and window of the living room window go up, and he saw a man standing outside the window. S.P.’s flaneé yelled at the man, and the man dropped the blind and fled. S.P. called the police while her flaneé chased the man. She gave the police a description of the man who had tried to talk to her. S.P. described him as a black male, about five-eight or five-nine, wearing blue jeans and a white t-shirt with a design on it. The police arrived, and S.P.’s flaneé gave a similar description.
*81An officer discovered defendant hiding behind a tree in a park a few blocks from S.P.’s apartment. The officer asked the man “to come out from behind the tree, and he complied, smiling.” The officer noticed that defendant was “sweating profusely, his clothes were missed [sic] up and his zipper was down.” Defendant was handcuffed, and a pat down for weapons revealed a serrated-edge folding knife and a wallet, which contained, among other items, a condom.
S.P. and her fiancé both identified the man as the person who had tried to talk to S.P. and enter their apartment. At trial, S.P. testified that she was able to get a good look at the man’s face during her encounter with him outside her apartment. She also made an in-court identification of defendant.
As a result of defendant’s arrest, the police obtained a warrant to search his home. During the search, the police seized a silver gun. An investigator also took oral swabs from defendant to obtain a sample of his DNA.
Charges Against Defendant
For acts committed against J.L. in New Brunswick on July 13, 2002, defendant was charged with second-degree burglary, N.J.S.A. 2C:18-2; first-degree aggravated sexual assault, N.J.S.A 2C:14-2a; second-degree sexual assault, N.J.S.A. 2C:14-2c; third-degree aggravated assault, N.J.S.A. 2C:12-lb(2); third-degree unlawful possession of a handgun without a permit, N.J.S.A. 2C:39-5b; second-degree possession of a handgun for an unlawful purpose, N.J.S.A. 2C:39-4a; and third-degree terroristic threats, N.J.S.A. 2C:12-3b.
For acts committed against K.G. in New Brunswick on June 9, 2003, defendant was charged with second-degree burglary, N.J.S.A. 2C:18-2; first-degree aggravated sexual assault, N.J.S.A. 2C:14-2a; second-degree sexual assault, N.J.S.A 2C:14-2c; fourth-degree unlawful possession of a knife, N.J.S.A 2C:39-5d; third-degree possession of a knife for an unlawful purpose, N.J.S.A 2C:39-4d; and third-degree terroristic threats, N.J.S.A. 2C:12-3a.
*82For acts committed against L.R. in Edison on January 18, 2005, defendant was charged with second-degree burglary, N.J.S.A. 20:18-2; first-degree aggravated sexual assault, N.J.S.A 2C:14-2a; second-degree sexual assault, N.J.S.A 2C:14-2c; fourth-degree unlawful possession of a knife, N.J.S.A. 2C:39-5d; and third-degree possession of a knife for an unlawful purpose, N.J.S.A 2C:39-4d.
And, for acts committed against S.P. in North Brunswick on May 27, 2005, defendant was charged with second-degree burglary, N.J.S.A. 2C:18-2; fourth-degree unlawful possession of a knife, N.J.S.A. 2C:39-5d; and third-degree possession of a knife for an unlawful purpose, N.J.S.A 2C:39-4d.3
B.
The State’s theory concerning these burglaries and sexual assaults was that there were sufficient similarities in the manner in which these offenses were carried out to make them signature crimes. Specifically, defendant was captured outside of the last attempted entry into a woman’s (S.P.’s) home with items similar to those used in some of the earlier crimes, which led to the search of his home and to obtaining DNA evidence used to link defendant to all of the sexual-assault incidents. The State asserted the similarities supported a single trial of these matters.
Defendant, seeking severance, argued that there was insufficient evidence to warrant joinder because the crimes were not signature crimes, part of a common scheme or plan, or probative on the issue of identity. Defendant distinguished each incident, pointing to the differences between the crimes: the times ranged from 2:30 p.m. to 3:30 a.m., two of the assaults involved a knife *83whereas one involved a gun, the assailant wore different clothing, the victims differed in age, and the crimes took place in three different towns.
In response, the State contended that, viewed as a whole, the crimes were probative on the issue of the perpetrator’s identity because all were home invasions, the perpetrator was described similarly by the victims, and each attack involved the use of a weapon and an attempt to subdue a victim. According to the State, the crimes did not need to be perfectly identical to be joined because the similarities “pointed to the identity of the perpetrator.”
The court granted defendant’s motion in part, holding that the indictments involving the burglaries and sexual assaults of K.G. and L.R. could be tried together in a single trial, but that the burglary and sexual assault of J.L. had to be tried separately. The court further held that the burglary of S.P. could be tried with either of the sexual assault trials, and, if proven at the first trial, evidence of that burglary could be admitted as other-crimes evidence in the second trial. Applying the Cofield test to assess whether there would be prejudice from the joinder, the court concluded that the assault of K.G. would have been admissible at the trial of the assault of L.R., and vice versa, if the two eases were tried separately. Accordingly, the court concluded that no prejudice would result from joining them in a single trial.
When assessing whether the evidence of the other crime was relevant to a material issue—the first prong of the Cofield test— the court found that the identity of the attacker was a disputed issue “and the evidence of L.R.’s attack [wa]s highly probative of the identity of” K.G.’s attacker. The court relied on the DNA testing that linked defendant to both crimes, acknowledging that the DNA evidence linking defendant to K.G. was less-conclusive mitochondrial DNA. In addition, the court considered the strong positive identification in L.R.’s case, although conceding the identification by K.G. was weaker. K.G. described defendant’s voice only as “familiar.”
*84Furthermore, the court found that the attacks of K.G. and L.R. were “so nearly identical as to constitute the defendant’s signature crime or handiwork.” The court pointed to the alleged use of a condom in both cases and opined that condom usage in stranger-to-stranger sexual assault “is fairly unusual or perhaps even unique.” The attacker’s comments regarding the race of his victims and his use of a knife to “distinctively cut the victim[’]s underwear off and expose[] the—body of the victim by either ripping or cutting their tops off in a very unique manner” suggest ed that the crimes were committed by the same person. The court concluded that, when considered in total, the evidence established a “nearly identical method of committing the crime.”
With respect to the remaining Cofield prongs, the court determined that each was satisfied. Although the crimes involving K.G. and L.R. occurred one-and-one-half years apart, they were reasonably close in time and similar in nature, thus satisfying the second Cofield prong. The third prong, which requires that evidence of the other crimes be clear and convincing, was satisfied because the physical evidence and proffered victim testimony led the court to conclude “that these crimes did take place.” Finally, the court found that the fourth prong was satisfied:
The fact that the probative value, of course, is not outweighed by any prejudice, I think is a conclusion that the Court must reach. The identity of KG.’s attacker clearly is a[sic] issue. The strong similarities between the two assaults suggest that the same person committed them. And the Court does not find that there’s anything particularly more heinous or necessarily prejudicial beyond the probative impact, which is great.
Although the court found sufficient similarity to join the trials relating to L.R. and K.G., it ruled that J.L.’s assault had to be tried separately from the other assaults. Given that none of the identifying features, such as the use of a knife or racial comments, were present, the court concluded that it did not present the same level of similarities.
The trial court also concluded that the burglary of S.P. could be joined with either of the trials or admitted as other-crimes evidence. The court stated that those charges could be joined, “[n]ot *85so much pursuant to the standard 404—rule 404B analysis, but because the crimes that took place against S.P. demonstrate a strong connection to the identity of the defendant.” The court pointed to the fact that defendant’s arrest for the S.P. burglary enabled the police to obtain swabs of his DNA and led the police to search his home, where they discovered the gun that J.L. claimed was “similar to the one” with which she was assaulted. Based on those facts, the court determined that, “with respect to demonstrating to the jury the identity of the defendant and how he came to be linked to these earlier crimes, clearly there is an appropriate connection.” The court also found that “the nature of the crimes against S.P. are not so prejudicial as to outweigh the probative value of having a jury understand how it came to be ... that the defendant was identified as having committed the crimes.”
In addition, the court noted that the sexual assaults would be admissible in the trial of the burglary of S.P. because “the State ... is required to prove the defendant’s intent to commit a sexual assault during the burglary.” Thus, the assaults of K.G. and L.R., which also involved the use of a knife and condom, would be admissible to prove defendant’s intent when he burglarized S.P.’s home.
C.
The State tried defendant first for the charges related to K.G., L.R., and S.P. in September 2006. The State portrayed the crimes as a series of sexual assaults, suggesting that S.P. was about to become defendant’s next victim.
The defense did not contest that the events occurred but rather maintained that defendant was not the perpetrator. Defense counsel attacked the DNA evidence as contaminated and emphasized to the jury that the State had not met its burden on the issue of identification. The defense cautioned the jury against using other crimes to prove identity in any of the criminal *86episodes and highlighted dissimilarities between the separate crimes.
The court instructed the jurors on the significance of the evidence actually presented and warned the jurors to consider the proof with respect to each separate charge before convicting:
Now, there are twelve crimes charged in this indictment. These are separate crimes, and in your determination as to whether the State has proven any of these crimes beyond a reasonable doubt, the defendant is entitled to have you consider each crime separately with respect to the evidence that relates to that crime based on the law as I give it to you. You just can’t take the position, well, since I found one way on this crime I am going that way across the board. You have to look at each crime based on the law that I explained to you and the evidence that was presented to prove that particular crime and reach a separate conclusion as to whether the State has proven its case beyond a reasonable doubt as to that crime.
At several points during the jury instructions, the court reiterated the need to consider each crime separately. The jury was instructed that
our law and our rules ordinarily exclude and do not permit evidence that a defendant has committed separate, numerous crimes only if it’s offered to show that a defendant has a disposition or a tendency to commit crimes and therefore must be guilty of all the crimes charged.
However, one exception to this rule is that evidence of separate crimes may be used for a specific, narrow purpose, and in this case the evidence has been offered to attempt to demonstrate that the separate sexual assaults in this case, those committed [against K.G. and L.R.] are so similar and so unique that you may, if you chose [sic] to, infer that the same person committed all of them.
You may not draw this conclusion or this inference unless you conclude that the sexual assaults with which the defendant is identified are so nearly identical in method as to earmark the defendant’s personal handiwork.
... [Y]ou can’t find the defendant guilty of one crime simply because you find him guilty of another. You still have to decide the cases individually based on the evidence that relates to that crime.
On September 15, 2006, the jury found defendant guilty on all charges involving K.G., L.R., and S.P.
The charges involving the attack of J.L. were tried in January 2007. During that trial, evidence about the burglary of S.P.’s home was presented. S.P., her fiancé, and the arresting officer testified for the State. Their testimony detailed the entire exchange between S.P. and defendant, including a lengthy deserip*87tion of his arrest and that defendant had his pants zipper down when apprehended and a knife and eondom in his pockets.
At the end of the second trial, the judge instructed the jury regarding the use of the other-crimes evidence:
Now, our Rules of Evidence ordinarily exclude evidence that a defendant may have engaged in some other type of conduct that’s not in the indictment when it’s offered to show that, well, he must be a bad guy or he must have done other things wrong or whatever, and, therefore, he must be guilty of the crimes charged in the indictment____Here the evidence of this encounter with [S.P.] and [her fiancéj was only admitted for the limited purpose of explaining to you or trying to show you how the defendant came to be charged with the offenses contained in this indictment, so you can’t use that evidence to decide that the defendant has a tendency to commit crimes or that just because he approached [S.P.j he must be guilty of the crimes in this indictment. Okay?
The jury found defendant guilty on all charges,4 and all offenses were joined for the sentencing hearing. Defendant was sentenced to an aggregate term of eighty years, with a period of parole ineligibility of sixty-three years.
D.
The Appellate Division consolidated defendant’s appeals from both trials and reversed defendant’s convictions for all charges relating to K.G., L.R., S.P., and J.L. Only defendant’s conviction and ten-year sentence for the certain persons offense were affirmed on appeal.
With respect to the first trial (K.G., L.R., and S.P.), the panel found insufficient evidence to support the finding that the assaults of K.G. and L.R. constituted a signature crime. The panel expressed skepticism that use of a condom is unique in stranger-to-stranger assaults so as to amount to a signature. It also questioned whether using a knife to cut off the victim’s clothing was *88unique, noting that defendant only resorted to using the knife in L.R.’s case after she did not comply with his request to remove her clothing. The panel pointed to other differences between the assaults—including the attacker’s use of a knife from K.G.’s kitchen instead of bringing his own—to highlight the lack of a unique signature.
Additionally, the panel found that, even if those features were unique, their distinctiveness was not “self-evident,” and therefore, an ordinary juror would not necessarily understand that they constituted a signature. Relying on State v. Fortin, 189 N.J. 579, 917 A.2d 746 (2007) [hereinafter Fortin III ], the panel stated that,
[a]t the very minimum, the State was required to introduce expert evidence to assist the court in determining at a hearing outside the jury’s presence whether the use of a condom in a sexual assault, along with the use of a knife and racial comments ... was such a unique pattern that a jury could find that the crimes were committed by the same person.
The panel also concluded that it was error to join defendant’s burglary of S.P.’s home with the offenses involving the attacks on KG. and L.R. “because they were not based on the same act or transaction, not part of a common scheme or plan, and not of the same or similar character,” as required under Rule 3:7-6. Even if the burglary could have been charged in the same indictment as the assaults, the panel found that the trial court failed to consider the prejudicial nature of the burglary evidence under Rule 3:15-2(b). No Cofield analysis was performed, and defendant’s mere possession of a condom and a knife did not uniquely link the S.P. burglary to the assaults as a signature.
The panel was not persuaded that the burglary was admissible as a basis for the jury to understand how defendant’s DNA was obtained, stating “[ujnder that reasoning, every crime that provided the basis for an order to compel DNA samples would be admissible in a defendant’s trial for any other crime.” Further, the panel noted, the reason the State obtained the DNA was not relevant to an issue before the jury. Moreover, the panel observed that there was other available admissible evidence on the issue of identification. Joinder of the offenses thus was prejudicial *89because it allowed the jury to consider extensive other-crimes evidence when there was adequate, less prejudicial evidence that could have been presented on the same issue. Conversely, the sexual assaults were not admissible to prove defendant’s intent in the burglary of S.P. because the indictment did not specify that defendant intended to commit a sexual assault when committing a burglary of her home.
With respect to J.L.’s separate trial, the panel concluded that the court should not have admitted evidence of the S.P. burglary because it was irrelevant, excessive, and prejudicial. The panel distinguished this ease from others where evidence of a later crime was admissible to prove identity, see State v. Pierro, 355 N.J.Super. 109, 117-18, 809 A.2d 804 (App.Div.2002), certif. denied, 175 N.J. 434, 815 A.2d 479 (2003); State v. Hardaway, 269 N.J.Super. 627, 630, 636 A.2d 128 (App.Div.1994), and emphasized the excessive nature and use of the evidence of the S.P. burglary during the prosecutor’s opening statement and the “more than fifty pages of trial transcript” in which S.P., her flaneé, and the arresting officer described the burglary and arrest.
Finally, the panel concluded that the errors were not harmless. With regard to J.L., the panel found that J.L.’s testimony identifying the gun was too tenuous to provide a basis for upholding the conviction. The panel stated that, “[although the DNA evidence was very strong, it was based on an analysis done many years earlier that generated only an ‘interpreted sperm donor profile’ because the specimens showed mixed male and female DNA” Therefore, the panel found that “[i]t cannot be said that the prejudice [from the burglary evidence] ... did not influence the jury’s consideration of the DNA evidence.”
The panel noted that the evidence of defendant’s identity in S.P.’s burglary was “very strong,” since he was captured near the scene and identified by both S.P. and her fiancé. For L.R., the panel found that the DNA evidence “strongly supported his guilt.” Indeed, the panel reiterated its assessment that the DNA evidence from the attacks on K.G. and L.R. “provided strong, scientific *90proof that identified defendant as the perpetrator of both sexual assaults.” Nonetheless, the panel did not find that there was sufficient, independent evidence of guilt because it did not view the evidence as overwhelming as in State v. Gillispie, 208 N.J. 59, 93, 26 A.3d 397 (2011), stating “[although there was independent evidence in all three cases [involving K.G., L.R., and S.P.], none of them involved the plethora of evidence available in Gillispie,” except “in the ease against defendant for the burglary of S.P.”5
The State petitioned for certification. Defendant opposed the State’s petition and filed a cross-petition on issues raised in a supplemental brief. We granted the State’s petition for certification and denied defendant’s cross-petition. State v. Sterling, 209 N.J. 596, 39 A.3d 200 (2012).
III.
A.
The State argues that the trials for the assaults on L.R. and K.G. were properly joined as signature crimes. It points to the similarities in the manner these two crimes were committed and contends that the Appellate Division compared the facts of this case too strictly to the facts in State v. Fortin, 162 N.J. 517, 745 A.2d 509 (2000) [hereinafter Fortin /]. The State additionally asserts that the Appellate Division erred by suggesting that the State must call an expert to establish whether multiple aspects of similar crimes are so unusual as to be deemed signature crimes. Further, the State maintains that the burglary of S.P.’s home properly was admitted at both trials. Finally, the State argues that even if the joinder of the crimes in this case were error, the *91error was harmless because of the DNA evidence linking defendant to the assaults and the in-court identification of defendant by one of the victims.
B.
Defendant argues that although the sexual assaults in this case generally were similar, the crimes were not unique enough to meet the stringent requirements of signature crimes. He also contends that the S.P. burglary was improperly joined with the sexual assaults of L.R. and K.G. because the S.P. burglary bore no similarities to the sexual assaults and no assault was committed. In addition, the effect of trying the burglary with the sexual assaults was so prejudicial that no limiting instruction could overcome it. Moreover, defendant maintains that admitting evidence of the burglary at the trial for the assault of J.L. was prejudicial because it was not relevant to any contested issue and served only to suggest defendant’s propensity for committing sexual assaults. Lastly, defendant argues that these errors were harmful despite the DNA evidence linking him to the crimes.
IV.
A.
Charges need not be identical to qualify as “similar” for purposes of joinder under Rule 3:7-6. See, e.g., State v. Baker, 49 N.J. 103, 105, 228 A.2d 339 (finding offenses involving sale of heroin to different people on different dates were “of the same or similar character”), cert. denied, 389 U.S. 868, 88 S.Ct. 141, 19 L.Ed.2d 144 (1967). Rule 3:7-6 expressly permits joinder when there is some connection between separate counts rendering the evidence probative of a material issue in another charge. See State v. Long, 119 N.J. 439, 514, 575 A.2d 435 (1990) (Handler, J., concurring in part and dissenting in part); Pitts, supra, 116 N.J. at 600, 562 A.2d 1320 (charged offenses were “based on ... 2 or more acts or transactions connected together” because, viewed *92together, evidence from one was probative of motive for later crime).
That said, the inquiry only begins with an assessment of whether there is similarity or a connection between charges because one involves evidence probative of another charge. Even if that threshold standard is satisfied, the court remains obligated to assess for prejudice, which may require the granting of relief from joinder. For that, the test employed under N.J.R.E. 404(b) provides the appropriate analytical framework. We therefore turn to the N.J.R.E. 404(b) framework for approaching the joinder question presented in this matter.
B.
N.J.R.E. 404(b) addresses the admission of other-crimes evidence to prove a defendant’s identity. As exemplified in different circumstances, evidence of a later crime may be admitted on the issue of identity when defendant’s connection to the first crime was established by specific evidence discovered during the second crime. In Fierro, supra, joinder of two separate burglaries was deemed an appropriate use of other-crimes evidence for purposes of identity where the defendant was caught near the scene of the second burglary, hiding under a bush, sitting on top of a social security card and credit cards obtained from the home of the first burglary. 355 N.J.Super. at 114, 809 A.2d 804. The court reasoned that the State was entitled to produce evidence that the defendant was found with the stolen goods in the trial of the first burglary because that possession would be relevant to prove his participation in the first offense. Id. at 117, 809 A.2d 804. Similarly, in State v. Loftin, evidence of the defendant’s separate credit card fraud was held admissible in his trial on a murder charge because the defendant had used the credit card belonging to the murder victim. 146 N.J. 295, 321, 680 A.2d 677 (1996). In both of these cited examples, evidence of the defendant’s possession and use of the exact items obtained during the commission of *93one crime linked him to the other crime and, therefore, was highly probative on the issue of identity.
Other-crimes evidence may also be admitted on the issue of identity when a particular weapon or disguise used in one crime connects a defendant to another offense. See Gillispi,e, supra, 208 N.J. at 88, 26 A.3d 397 (admitting evidence of prior crime where defendant used same gun); Hardaway, supra, 269 N.J.Super. at 630, 636 A.2d 128 (allowing limited evidence of later robbery to prove defendant’s presence at killing because same gun was used in both crimes); State v. Porambo, 226 N.J.Super. 416, 423-24, 544 A.2d 870 (App.Div.1988) (permitting evidence of separate robbery where distinctive mustache disguise connected defendant to both robberies).
The standard for admitting other-crimes evidence to prove identity becomes more stringent when the State attempts to link a particular defendant to a crime on the basis of modus operandi, or a signature way of committing the crime. See Fortin I, supra, 162 N.J. at 530-31, 745 A.2d 509; see generally Biunno, Current N.J. Rules of Evidence, comment 14 on N.J.R.E. 404(b) (2013) (“When the State tries to introduce other-crime evidence to prove identity by suggesting that there is a signature to the crimes, however, the standard for admissibility is more rigorous.”). The case law reflects the practical consequences of applying a higher burden in such matters.
In State v. Reldan, 185 N.J.Super. 494, 496-98, 449 A.2d 1317 (App.Div.), cetif. denied, 91 N.J. 543, 453 A.2d 862 (1982), the defendant was indicted for the murder of two women killed by the same method of strangulation: a pantyhose ligature around their necks. The trial court determined that the charges could be joined for trial, given the unique similarities and the use of the pantyhose. Id. at 500-01, 449 A.2d 1317. To establish the defendant’s connection to the crime, the State sought to offer evidence of the defendant’s prior conviction for another rape and robbery in which the defendant also tried to choke his victims, id. at 498, 449 A.2d 1317; however, in those other crimes the defen*94dant did not use pantyhose to choke the victims, id,, at 498-99, 449 A.2d 1317. Emphasizing the stricter standard required for identity evidence in signature crimes, the Appellate Division found error in the trial court’s admission of evidence of the prior rape and robbery, determining that the use of force to the throat in all four instances was not “so unique as to indicate particular, peculiar and individual handiwork.” Id. at 503, 449 A.2d 1317.
On the other hand, in State v. Sempsey, 141 N.J.Super. 317, 324, 358 A.2d 212 (App.Div.1976), certif. denied, 74 N.J. 272, 377 A.2d 677 (1977), the heightened burden was determined to have been met, and the Appellate Division affirmed the admission of other-crimes evidence on the issue of identity. Sempsey was charged with attempting to rape two roommates. Id. at 320, 358 A.2d 212. The State sought to admit evidence of a prior attack committed by the defendant where he was confirmed as the attacker through fingerprint evidence. Id. at 322, 358 A.2d 212. In both attacks, the assailant placed tape over his victims’ eyes, wore “peculiar head gear,” had a gun, smelled of grease, and was unable to perform sexually. Id. at 320-22, 324, 358 A.2d 212. That said, there were differences emphasized by the defense. Id. at 322, 324, 358 A.2d 212. Nonetheless, the Appellate Division found that it was appropriate to admit evidence of the separate attack to establish the identity of the perpetrator, noting that, although the “[d]efendant stresses certain dissimilarities in the attacks[,] ... none is of such a nature as to eliminate him as being involved in the second crime.” Id. at 324, 358 A.2d 212. The Appellate Division explained that there must be “sufficient similarities of novel or unusual means so as to permit the introduction of evidence of the first attack as bearing on the identity of the defendant as the person involved in the second.” Id. at 323, 358 A.2d 212.
The seminal ease on signature crimes is Fortin I. There, we utilized the heightened standards described in Sempsey and Roldan when stressing the high burden that would be required when other-crimes evidence is admitted to prove identity through the *95use of a signature-crime analysis. 162 N.J. at 530-31, 745 A.2d 509. In Fortin I, Fortin was charged with the murder of a woman in New Jersey. Id. at 519, 745 A.2d 509. The victim was found naked from the waist down and was brutally beaten on her face and head. Id. at 520, 745 A.2d 509. The ultimate cause of death was manual strangulation. Ibid. Additional autopsy evidence revealed rectal tearing and bite marks on her left breast, left nipple, and left side of her chin. Ibid.
In order to prove the defendant’s identity for the crime committed in New Jersey, the State sought to introduce expert testimony comparing the murder to a sexual assault committed by the defendant, later in time, against a female state trooper in Maine. Id. at 521-22, 745 A.2d 509. Fortin was arrested in connection with that event and pled guilty to sexual assault. Id. at 520-21, 745 A.2d 509. During his assault on the trooper in Maine, he strangled her, sexually assaulted her, and punched her in the face. Ibid. The victim also suffered bite marks on her left chin, left nipple, and left breast, and suffered injuries from vaginal and anal tearing. Ibid.
In our Fortin I decision, we adopted a heightened standard for signature-crime evidence and stressed the difficulty involved in such case-by-case determinations, commenting that the facts—the distinctive bite marks and patterns of injuries—might be enough to constitute a signature crime. Id. at 532-33, 745 A.2d 509. In assessing whether to allow such damaging evidence, we emphasized the need for evidence that is as identical as possible in order to overcome substantial prima facie dissimilarities: ‘“the prior criminal activity with which defendant is identified must be so nearly identical in method as to earmark the crime as defendant’s handiwork.’ ” Id. at 532, 745 A.2d 509 (quoting Reldan, supra, 185 N.J.Super. at 502, 449 A.2d 1317). “‘The conduct in question must be unusual and distinctive so as to be like a signature, and there must be proof of sufficient facts in both crimes to establish an unusual pattern.’ ” Ibid, (quoting Reldan, supra, 185 N.J.Super. at 502, 449 A.2d 1317). We observed that stating the law “is *96easier than to apply the law.” Ibid. Ultimately, we concluded that the evidence of the Maine assault, coming in the form of opinion testimony from the State’s criminal investigator, could be admissible as signature-crime evidence if the expert testifying on the uniqueness of the injuries could produce a reliable database to support his opinions. Id. at 533, 745 A2d 509.
C.
The subsequent history to Fortin I is pertinent to the instant matter. Fortin was retried after his conviction was reversed because the State failed to present a reliable database as a predicate for the expert’s testimony. Fortin III, supra, 189 N.J. at 584, 917 A.2d 746. When retrying the defendant, the State again sought to introduce other-crimes evidence as indicative of a signature crime but contended that the evidence from the Maine conviction, with the unusual aspects to the assault on the female trooper, was so unique to be admissible signature-crime evidence without the need for expert testimony. Id. at 589, 917 A.2d 746. The State contended that a series of bite marks and tearing on the victim in both crimes were so unusual that, viewed together, the jury could find that they were “akin to a signature” identifying Fortin as the murderer. Id. at 584, 917 A.2d 746. In other words, such an inference was well within the understanding of an average juror, and, therefore, no expert testimony was required. Ibid.
Based on the facts in Fortin, we held that the signature-crime evidence was beyond the general understanding of an average juror and “require[d] expert testimony to explain those features that uniquely tie the two crimes together.” Id. at 584, 917 A.2d 746. We explained that, “when the signature-like aspect of a crime would not be apparent to the trier of fact, expert testimony may be necessary to explain the significance of the evidence.” Id. at 596, 917 A.2d 746. With specific reference to the Fortin facts, we found that the “average juror cannot be expected to have the knowledge or experience to discern whether bite marks are a *97common or highly distinctive feature of violent sexual assaults.” Ibid. Therefore, “without expert testimony explaining the unique similarities between the [two] sexual assaults, the [first] assault is not admissible as signature-crime evidence.” Id. at 606-07, 917 A.2d 746.
To the extent that the Appellate Division in this matter concluded that Fortin III signaled that signature-crime evidence required, in every instance, expert testimony for its admission, the panel mistook our emphasis on an expert in that discussion. Our discussion about the need for expert testimony in Fortin was rooted in the difficulty presented by its facts. Admission of evidence as supportive of a signature crime must remain a case-by-case analysis, assessing what it is about the method of a crime’s commission that might make it a signature crime. Bite mark analysis was beyond a jury’s ken, but that does not mean that every aspect of a signature crime’s analysis is so obtuse that only an expert can be expected to translate its importance and significance to the jury.
With that understanding of our discussion in Fortin III, we turn to the State’s argument that the crimes involving K.G. and L.R. were signature crimes that could and should be tried together; or alternatively, whether the crimes were properly joined because they in combination were relevant to the material issue of identity and were not unduly prejudicial.
V.
A.
We dispense first with the signature-crime argument. Although there were some similarities between the burglaries and sexual assaults involving K.G. and L.R., we agree with the Appellate Division that there is no uniqueness to the manner in which those crimes were committed. The most significant signature elements were the use of a condom, the racial comments, and the manner in which the perpetrator cut the victims’ underwear. *98Individually, these acts certainly do not rise to the level of signature elements of a crime. Condom use was not established by the State as being unique, or even unusual, in a sexual assault.6 Nor was there anything unique, or ritualistic, in cutting off clothing in order to rape the victims. Indeed, in these two crimes, defendant did not bring a knife to the scene where the incidents occurred. Thus, there was no element that was so distinguishing that no one else could likely have known to do these “unique” acts. See Fortin III, supra, 189 N.J. at 584, 589, 917 A.2d 746. For a crime to meet the heightened standard articulated in Fortin I, much more than these facts must be present, as the Appellate Division correctly concluded.
B.
Here, the crimes neither satisfy the heightened standard for admission as signature-crime evidence, nor pass the multi-part test applicable when assessing a request for relief from prejudicial joinder.
The test is whether the evidence from one offense would have been admissible N.J.R.E. 404(b) evidence in the trial of the other offense, because “[i]f the evidence would be admissible at both trials, then ... a defendant will not suffer any more prejudice in a joint trial than he would in separate trials.” Chenique-Puey, supra, 145 N.J. at 341, 678 A.2d 694 (internal quotations omitted). Generally, other-crimes evidence is admissible to prove motive, opportunity, intent, preparation, plan, knowledge, or identity; however, in the context of a sexual assault, evidence of other sexual assaults most soundly may be used to show intent and, at times, identity. Illustrative of this latter point is State v. Oliver, 133 N.J. 141, 627 A.2d 144 (1993).
*99In Oliver, the defendant was convicted of sexually assaulting two women, each of whom was his longtime friend. Id. at 145, 627 A2d 144. The crimes were strikingly similar. In both, the victims willingly went to the defendant’s house because they had known him for years. Ibid. The defendant took each of “the victims into his third-floor room while other family members were downstairs; engaged in conversation with his victims; drank some beer; and then resorted to brute force to cut off the victims’ air supply until they relented.” Ibid.
At trial, the defendant filed a motion to sever the charges, but the trial court denied the motion, finding that the similarities of the crimes established the “defendant’s intent to commit crimes against those women, and/or established his common scheme or plan to lure female friends to his room under false pretenses.” Id. at 150, 627 A.2d 144. If tried separately, the trial court reasoned, the evidence of one assault would be admissible in the trial for the other assault, thus satisfying N.J.R.E. 404(b). Ibid.
However, our Court rejected that reasoning, noting that the acts were not part of a “larger continuing plan of which the crime on trial is a part. They are simply discrete, albeit similar, acts.” Id. at 152, 627 A.2d 144. We cautioned courts against confusing a “ ‘single purpose’ binding together several crimes with having the same purpose several times.” Ibid. As in Oliver, defendant in the instant case did not have an “ ‘integrated plan’ of which the sexual assaults ... were components.” See id. at 153, 627 A.2d 144.
In Oliver, we noted that the same evidence may have been admissible to prove other facts in issue, namely, the feasibility that the defendant could assault a woman in his room without the other family members at home knowing and to show the success of the defendant’s pretext to lure women to his room. Ibid. However, in the present case, the fact patterns in each assault do not raise any similarly compelling issue. At the core of the present case is the issue of identity. Defendant denies committing the crimes. He does not dispute that the crimes occurred; his defense is that he was not the perpetrator.
*100Concerning the offenses tried together involving K.G. and L.R., there simply is insufficient probative evidence linking these crimes on the basis of identity to justify the risk of the jury relying on the evidence for the prohibited purpose of propensity. Although the sexual assaults of K.G. and L.R. were similar in kind and occurred within one-and-one-half years of one another, and there was clear and convincing evidence that each crime did occur, thus satisfying prongs two and three of the Cofield analysis, they were not signature crimes. Because the crimes were not signature crimes, the probative value from introducing one in the trial of the other must be regarded as of minimal value on the issue of identity. There was less-prejudicial admissible evidence on the issue of identity—the DNA comparison in each and the victims’ identifications—which would have bolstered the State’s position in both trials on the issue of the attacker’s identity. See State v. Jenkins, 178 N.J. 347, 366, 840 A.2d 242 (2004) (noting preference for use of least prejudicial evidence). In sum, the slim probative value of the similarities between these crimes is outweighed by the prejudice that was generated by trying the two burglaries and sexual assaults together. By trying them together, identification of defendant as the perpetrator in each was bolstered.
We would be remiss if we did not commend the State for conceding, candidly, that the State’s ease in the matter involving K.G. needed some bolstering. KG.’s identification was less compelling than that advanced by L.R. about her attacker. However, the bolstering of the case involving K.G.’s assault should not have come from joinder of the two crimes involving K.G. and L.R.
C.
As for the criminal episode involving S.P., we note at the outset that it was not a crime of the same sort as the ones involving K.G. and L.R. It involved a burglary. There was no sexual assault, and the State did not proffer sexual assault as the intended criminal purpose of the burglary in its indictment, as the Appellate Division noted. Nevertheless, the State advanced that *101angle in arguing the joinder issue to the trial court and again in its presentation to the jury. Its argument, accepted by the trial court, was that the S.P. burglary should be tried with the burglaries and sexual assaults of K.G. and L.R. in order to admit evidence of how defendant was found by the police and for the purpose of explaining how the State obtained his DNA evidence. That argument is unpersuasive.
The State only needed to prove that it had obtained defendant’s DNA and that it had obtained a search warrant to find a gun at defendant’s home. It was not necessary for the S.P. incident to be tried with the offenses involving K.G. and L.R. It was not necessary for the State to go into detail about how defendant was caught outside of S.P.’s home in order to tie in the DNA evidence and gun used against defendant. And, it prejudiced defendant in defending against the charges involving K.G. and L.R. The evidence adduced on the burglary of S.P. posed the substantial risk that the jury would draw the conclusion that defendant was about to commit a sexual assault on S.P., so he likely committed the assaults on K.G. and L.R.
The evidence here is precisely the type of unduly prejudicial other-crimes evidence that N.J.R.E. 404(b) means to exclude. The trial court should have excluded it.
VI.
A.
Our analysis does not end with the conclusion that it was error to have joined the three crimes—involving K.G., L.R., and S.P.—in one trial. We must assess whether the error “ ‘led to an unjust result. The possibility must be real, one sufficient to raise a reasonable doubt as to whether [it] led the jury to a verdict it otherwise might not have reached.’ ” See State v. Lazo, 209 N.J. 9, 26, 34 A.3d 1233 (2012) (quoting State v. R.B., 183 N.J. 308, 330, 873 A.2d 511 (2005) (alterations in original)); State v. Cabbell, 207 N.J. 311, 337-38, 24 A.3d 758 (2011) (“The violation of defendants’ *102federal constitutional right is a fatal error, mandating a new trial, unless we are ‘able to declare a belief that it was harmless beyond a reasonable doubt.’ ” (quoting Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 828, 17 L.Ed.2d 705, 710-11 (1967))); accord State v. Castagna, 187 N.J. 293, 313, 901 A.2d 363 (2006) (applying same “harmless beyond a reasonable doubt” standard to determination under State Constitution (internal quotation marks and citations omitted)). In doing so, we conduct an independent analysis of the quality of the evidence of defendant’s guilt on a eonviction-by-conviction basis.
B.
Applying a harmless error standard, we cannot but conclude that the assault of L.R. must be affirmed based on the nuclear DNA evidence tying defendant to the crime, coupled with the victim’s strong identification of defendant. While no New Jersey case bears directly on the sufficiency of DNA evidence alone, other jurisdictions have affirmed convictions based solely on DNA evidence. Although we need not go so far in our holding, several state appellate courts have held that a DNA profile match alone, without corroborative individualized evidence, is sufficient to prove guilt. See, e.g., State v. Abdelmalik, 273 S.W.3d 61, 66 (Mo.Ct.App.2008) (“We conclude that where, as here, DNA material is found in a location, quantity, and type inconsistent with casual contact and there is a one in one quintillion likelihood that someone else was the source of the material, the evidence is legally sufficient to support a guilty verdict.”); People v. Rush, 165 Misc.2d 821, 630 N.Y.S.2d 631 (N.Y.Sup.Ct.1995) (“[T]he statistical probability that anyone else was the source of [the DNA evidence is] 1 in 500 million.”), aff'd, 242 A.D.2d 108, 672 N.Y.S.2d 362 (N.Y.App.Div.), appeal dismissed, 92 N.Y.2d 860, 677 N.Y.S.2d 91, 699 N.E.2d 451 (1998); State v. Hunter, 169 Ohio App.3d 65, 861 N.E.2d 898, 901 (2006) (“[T]he probability of the sample obtained from the rape kit belonging to anyone other than appellant was one in 756 trillion.”); State v. Toomes, 191 S.W.3d 122, *103127 (Tenn.Crim.App.2005) (“[T]he probabilities of finding the defendant’s profile within the African-American population was 1 in 5 billion, 128 million; within the Caucasian population it was 1 in 22 billion, 870 million; within the Southeastern Hispanic population it was 1 in 90 billion, 910 million, and within the Southwestern Hispanic population it was 1 in 185 billion, 700 million.”); Roberson v. State, 16 S.W.3d 156, 167 (Tex.Ct.App.2000) (“[T]he reported frequency or chance of appellant’s DNA profiling occurring in another person was 1 in 5.5 billion for the Caucasian, African-American, and Hispanic populations, and that the actual frequency was 1 in 420 billion for the African-American population.”).
A recent United States Supreme Court decision also emphasizes the “unparalleled accuracy” of DNA evidence in linking defendants to crimes at which their DNA is found. Maryland v. King, 569 U.S. -, -, 133 S.Ct. 1958, 1964, 186 L.Ed.2d 1, - (2013). Although confronting a Fourth Amendment challenge to the seizure of the defendant’s DNA in King, the Court explained that “the utility of DNA identification in the criminal justice system is already undisputed.” Id. at-, 133 S.Ct. at 1966, 186 L.Ed.2d at-. Since many patterns of DNA are shared by all individuals, forensic analysis focuses on repeated DNA sequences that are known as short tandem repeats (STRs), as well as the variations in the STRs’ size and frequencies, which are known as “alleles.” Id. at -, 133 S.Ct. at 1967, 186 L.Ed.2d at - (citing John Butler, Fundamentals of Forensic DNA Typing (2009)). Because individuals have similar alleles at a certain locus on a gene, STR alleles are analyzed at several different loci; “[tjhese loci make possible extreme accuracy in matching individual samples, with a ‘random match probability of approximately 1 in 100 trillion (assuming unrelated individuals).”’ King, supra, 569 U.S. at -, 133 S.Ct. at 1968, 186 L.Ed.2d at - (quoting Butler, supra, at 270). “Future refinements may improve present technology, but even now STR analysis makes it ‘possible to determine whether a biological tissue matches a suspect with near certainty.’ ” Id. at-, 133 S.Ct. at 1962, 186 L.Ed.2d at-(quoting District Attorney’s Office for Third Judicial Dist. v. *104Osborne, 557 U.S. 52, 62, 129 S.Ct. 2308, 2316, 174 L.Ed.2d 38, 47 (2009)).
While we do not suggest that DNA evidence alone is enough, based on a qualitative analysis of the evidence in this matter, the nuclear DNA evidence in L.R.’s case, viewed in combination with L.R.’s immediate and strong identification of defendant during the May 31, 2005 lineup and her in-court identification, rendered harmless the joinder error. See Castagna, supra, 187 N.J. at 312, 901 A.2d 363. We reach the fact-sensitive determination that the prejudice to defendant from joinder in respect of L.R.’s ease was overwhelmed by the strong, independent evidence of his guilt. L.R. made an immediate and unequivocal identification of defendant when she first saw him in the lineup and a second time at trial. Further, the State’s forensic expert confirmed that there was “a reasonable degree of scientific certainty”—only one in 30.8 quadrillion individuals in the African-American population, one in 419 quadrillion individuals in the Caucasian population, and one in 2.02 quintillion individuals in the Hispanic population—that defendant’s nuclear DNA matched the DNA swab taken from L.R.’s body only hours after defendant’s assault of L.R.
That evidence separates this case from the ordinary one when prejudice from joinder would require a new trial. Though we seek to not serve as the thirteenth juror when performing harmless error analyses, we conclude that on these facts the joinder error is harmless beyond a reasonable doubt because of the strong, independent proof of defendant’s guilt. Thus, all convictions relating to defendant’s attack of L.R. must stand.7
*105c.
However, we cannot similarly conclude that the joinder error was harmless in respect of defendant’s convictions for the offenses involving K.G. The independent evidence of guilt in the case of K.G. stands in marked contrast to that of L.R. and does not provide a basis for concluding that the improper joinder was harmless beyond reasonable doubt.
With respect to KG.’s assault, the DNA evidence was the less-precise mitochondrial DNA. We cannot disregard the distinct possibility that the stronger, nuclear DNA evidence obtained for the offenses involving L.R. influenced the jury in its assessment of the offenses involving K.G. In the K.G. case, the DNA did not point with “unparalleled accuracy” to defendant as the perpetrator; rather, the DNA evidence simply did not exclude defendant as the perpetrator. Moreover, there was no strong identification evidence presented. In contrast, the DNA evidence in the L.R. case powerfully demonstrated that defendant was the perpetrator in that sexual assault, and that evidence was buttressed by L.R.’s strong identification. Thus, the joinder of the L.R. offenses in the trial on the K.G. offenses could well have had considerable impact such that, on this record, we cannot conclude that the joinder error was harmless beyond a reasonable doubt in the case of K.G. See Cabbell, supra, 207 N.J. at 337-38, 24 A.3d 758; R. 3:15-2. Thus, defendant’s convictions relating to the burglary and sexual assault of K.G. correctly were reversed and remanded for a new trial by the Appellate Division.
D.
We analyze the prejudice of joining the offenses involving the burglary of S.P.’s home differently than the other two criminal episodes with which it was tried. It was, as noted earlier, *106different from the crimes involving K.G. and L.R. It was a burglary, not a burglary and a sexual assault.
Although the trial court made an error in judgment in joining the offenses in the first trial, such an error does not necessarily bring about an unjust result in defendant’s convictions for the offenses involving S.P. We find the record replete with overwhelming evidence of defendant’s guilt in the case involving S.P. Although defendant contests the identifications associated with his involvement in the S.P. burglary, the identification evidence was powerful. It can be regarded as so overwhelming that identity was not seriously at issue.
Both S.P. and her fiancé saw defendant during his attempted burglary of their home. S.P. got a good look at him when he approached her early that morning outside her apartment. When startled by the burglary that began almost as soon as S.P. locked herself in the safety of her apartment, her fiancé chased defendant from their home while S.P. called the police. Defendant was found alone, hiding behind a tree near their apartment—at nearly 3:00 a.m.—at which point S.P. and her fiancé identified defendant at the show-up that followed immediately after these events transpired. In sum, defendant was caught close to the scene of the crime and was identified at a show-up immediately by the victim and her flaneé. Moreover, S.P. identified defendant in court as the perpetrator of the burglary. On this record, we do not view the joinder as having produced an unjust result. To the contrary, we view the independent evidence of guilt practically and conclude that the joinder error was harmless beyond a reasonable doubt. Thus, we reverse the Appellate Division’s judgment in respect of defendant’s convictions concerning S.P. Defendant’s convictions relating to the S.P. burglary do not merit reversal.
VII.
As for the other-crimes evidence from S.P.’s burglary that was admitted in the trial on the offenses related to the burglary and *107sexual assault of J.L., we agree with the Appellate Division that it should not have been admitted.
The State did not need to detail how defendant was caught outside of S.P.’s home in order to tie in the DNA evidence obtained from defendant. It was embellishment, pure and simple. And, it was prejudicial in J.L.’s trial for that amount of information to have been presented to the jury when all that was necessary was for the jury to know that the State had obtained defendant’s DNA. Admitting that extensive evidence in the trial for J.L.’s assault allowed the jury to draw the inference that defendant was about to commit a sexual assault on S.P., so he likely committed the assault on J.L. That inference was impermissible.
As noted by the Appellate Division, though witnesses were instructed to avoid discussion of the S.P. burglary during the second trial relating to the sexual assault and burglary of J.L., several witnesses testified extensively about the burglary for which defendant was not on trial. Moreover, the trial court did not sanitize the evidence in an attempt to minimize the prejudice to defendant. See State v. Barden, 195 N.J. 375, 390, 949 A.2d 820 (2008) (requiring sanitization of other-crimes evidence to reduce risk of prejudice to defendant). Thus, the trial court’s mistaken admission of this evidence was not harmless. It presented a classic misuse of N.J.R.E. 404(b) evidence. For this reason, defendant’s convictions for offenses involving the burglary and sexual assault of J.L. must be reversed and remanded for a new trial. We affirm the Appellate Division judgment that so held on these convictions.
VIII.
The judgment of the Appellate Division is affirmed in part and reversed in part.
We reverse the Appellate Division judgment reversing defendant’s convictions concerning L.R. for second-degree burglary, *108N.J.S.A. 2C:18-2; first-degree aggravated sexual assault, N.J.S.A. 2C:14-2a; second-degree sexual assault, N.J.S.A. 2C:14-2c; fourth-degree unlawful possession of a knife, N.J.S.A. 2C:39-5d; and third-degree possession of a knife for an unlawful purpose, N.J.S.A. 2C:39-4d.
We also reverse the Appellate Division judgment reversing defendant’s convictions concerning S.P. for second-degree burglary, N.J.S.A. 2C:18—2; fourth-degree unlawful possession of a knife, N.J.S.A. 2C:39-5d; and third-degree possession of a knife for an unlawful purpose, N.J.S.A. 2C:39-4d.
The judgment of the Appellate Division is otherwise affirmed. The matter is remanded to the Law Division for proceedings consistent with this opinion.
Justice ALBIN, dissenting and concurring.
I agree with the majority “that it was error to have joined the three crimes—involving K.G., L.R., and S.P.—in one trial.” Ante at 101, 71 A.3d at 807. I agree that “there simply is insufficient probative evidence linking [the K.G. and L.R.] crimes on the basis of identity to justify the risk of the jury relying on the evidence for the prohibited purpose of propensity.” Ante at 100, 71 A.3d at 806. I agree that “the slim probative value of the similarities between these crimes is outweighed by the prejudice that was generated by trying the two burglaries and sexual assaults together” and that “[b]y trying them together, identification of defendant as the perpetrator in each was bolstered.” Ante at 100, 71 A.3d at 806. I agree that “[i]t was not necessary for the S.P. incident to be tried with the offenses involving K.G. and L.R.,” ante at 101, 71 A.3d at 806, and that “it prejudiced defendant in defending against the charges involving K.G. and L.R.,” ante at 101, 71 A.3d at 806. Finally, I agree that the evidence of the S.P. burglary presented in the K.G. and L.R. sexual-assault cases “is precisely the type of unduly prejudicial other-crimes evidence that N.J.R.E. 404(b) means to exclude.” Ante at 101, 71 A.3d at 806-07.
*109The majority’s chronicling of the qualitative and quantitative prejudice engendered by improperly joining three separate crimes in one trial would seem to lead to one ineluctable conclusion—that defendant was denied a fair trial. Indeed, it is difficult to imagine a scenario more fraught with prejudice, more stacked against a fair trial, than to wrongly join two highly incendiary cases "with one that should have been tried alone to a jury. The majority concedes that the K.G., L.R., and S.P. cases each should have been tried separately, see ante at 102, 71 A.3d at 807, and that most of the evidence presented in each case was inadmissible. Yet, the majority somehow finds that one lapping wave of prejudice after another was harmless in the L.R. and S.P. cases. See ante at 102-06, 106-08, 71 A.3d at 807-10, 810. After two wrongly obtained convictions are laundered through our harmless-error jurisprudence, they are ironed and pressed and upheld as fair convictions. But, in my view, those convictions are still soiled because they cannot be justified without completely compromising our harmless-error jurisprudence. However strong the State’s proofs may be, a defendant is entitled to a fair trial. I will not, as the majority does, sit as a thirteenth juror, ignore the basic unfairness of the proceedings, and then pronounce, “Well, he was guilty anyway.”
For these reasons, I respectfully dissent.
I.
Defendant was charged with the June 9, 2003 sexual assault of twenty-five-year-old K.G. in her New Brunswick apartment. At a lineup, two years after the brutal attack, K.G. was unable to make a positive identification of defendant. The State’s primary evidence against defendant was mitochondrial DNA taken from K.G.’s pants. Defendant had the same DNA sequence that is *110found in not more than .06% of North Americans. Almost fifty percent of the evidence presented at trial was irrelevant to the sexual assault of K.G.
Defendant was charged with the January 18, 2005 sexual assault of thirty-nine-year-old L.R. in her Edison apartment. Three weeks after the vicious assault, L.R. viewed a lineup of six men and pointed out one. She was “ninety percent” certain he was her attacker, but that man was not defendant. On May 31, 2005, L.R. viewed another six-man lineup. This time, when defendant came out to take his spot, she stated, “it’s him, it’s him.” Saliva left on L.R.’s body during the sexual assault matched defendant’s nuclear DNA profile—the DNA sequence occurring only in approximately one in 30.8 quadrillion of African Americans. Defendant argued that the DNA samples taken from L.R. may have been contaminated. At least sixty percent of the testimony elicited during the trial was irrelevant to the sexual assault of L.R.
Defendant was charged with the May 27, 2005 burglary of the North Brunswick apartment in which thirty-one-year-old S.P. lived with her fiancé. At 2:40 a.m. on that day, a man had approached S.P. outside of her apartment. A short while later, he was observed by S.P.’s fiancé attempting to gain entry into their apartment through a window. The police arrested defendant a short distance from S.P.’s apartment. Both S.P. and her flaneé identified defendant at the scene. At trial, at least eighty-five percent of the testimony presented was irrelevant to the apartment burglary.
II.
As mentioned earlier, the majority concedes that these three separate crimes were wrongly joined. Ante at 102, 71 A.3d at 807. The chief evil in wrongly joining separate offenses is the admission of other-crime evidence that poisons the impartiality of the jury. See State v. Orlando, 101 N.J.Super. 390, 394, 244 A.2d 506 (App.Div.) (noting that “in view of the abhorrent nature of the offense[s],” wrongful joinder “multiplied] the chances that defen*111dant would be convicted”), certif. denied, 52 N.J. 500, 246 A.2d 457 (1968). We recognize that “other-crime evidence has a unique tendency to turn a jury against the defendant” and that “[t]he erroneous admission of evidence of other crimes ... carries such a high risk of prejudice as ordinarily to call for reversal.” State v. G.V., 162 N.J. 252, 262, 744 A.2d 137 (2000) (internal citation and quotation marks omitted). Acknowledging the powerful capacity of other-crime evidence to deny a defendant a fair trial, “[t]his Court has sought to prevent overuse of the ‘harmless error’ doctrine” when such wrongful evidence is admitted. See ibid. Indeed, such “ ‘[ejrrors impacting directly upon ... sensitive areas of a criminal trial are poor candidates for rehabilitation under the harmless error philosophy.’” Ibid, (quoting State v. Simon, 79 N.J. 191, 206, 398 A.2d 861 (1979)); see also State v. Pitts, 116 N.J. 580, 601, 562 A.2d 1320 (1989) (noting that “ ‘[tjhere is indeed always a danger when several crimes are tried together, that the jury may use the evidence cumulatively’ ” and that “ ‘the sum of it will convince them as to all.’ ”) (quoting United States v. Lotsch, 102 F.2d 35, 36 (2d Cir.), cert. denied, 307 U.S. 622, 59 S.Ct. 793, 83 L.Ed. 1500 (1939)).
At issue here is the application of the harmless-error doctrine. That doctrine holds that an error of constitutional magnitude is harmless only if the court is “ ‘able to declare a belief that it was harmless beyond a reasonable doubt.’ ” See State v. Cabbell, 207 N.J. 311, 338, 24 A.3d 758 (2011) (quoting Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 828, 17 L.Ed.2d 705, 710-11 (1967)). The presentation of inadmissible evidence “which possibly influenced the jury adversely to a litigant cannot ... be conceived of as harmless.” Chapman, supra, 386 U.S. at 23-24, 87 S.Ct. at 828, 17 L.Ed. 2d at 710. An error that denies a defendant his substantial rights, or a fair trial, cannot be deemed harmless. See ibid.
These principles, when applied to the facts of this case, led the Appellate Division to reverse defendant’s convictions.
*112III.
A.
The majority claims that defendant received a fair trial in the S.P. burglary case even though defendant was compelled to defend against two separate, wrongfully joined aggravated sexual-assault cases in the same trial. See ante at 106-08, 71 A3d at 810. The majority finds harmless the fact that over eighty-five percent of the testimony was not only irrelevant, but also highly prejudicial in the S.P. case. In concluding that the wrongful-joinder error is harmless beyond a reasonable doubt, the majority presumably believes that the jury—by some superhuman feat—could remain untouched by the torrent of inflammatory and inadmissible evidence that overwhelmed the S.P. trial.
Because the wrongful joinder so fundamentally undermined defendant’s ability to receive a fair trial in the S.P. case, I would uphold the Appellate Division’s reversal of the convictions in that matter.
B.
The majority also holds that defendant’s convictions for “the assault of L.R. must be affirmed based on the nuclear DNA evidence tying defendant to the crime, coupled with the victim’s strong identification of defendant.” Ante at 102, 71 A.3d at 807. First, I find it difficult to conclude that this case involves a “strong identification,” considering that L.R. misidentified a suspect with ninety-percent accuracy just three weeks after the attack and identified defendant with one-hundred-percent accuracy four months after the attack. We know that memory does not improve with time—that it, in fact, degrades—and that the reliability of an identification therefore must be viewed critically. State v. Henderson, 208 N.J. 208, 267, 27 A.3d 872 (2011). The majority’s inflated confidence in this identification undoubtedly stems from the DNA evidence. A conviction based on L.R.’s eyewitness *113identification, standing alone, would never survive a harmless-error analysis.
The question remains whether a conviction predicated on DNA evidence—despite the wrongful joinder of two cases to a third one—v\lll withstand the harmless-error test. Although it may be that a rational jury can convict based solely on DNA evidence, it is another thing to say that after the introduction of DNA evidence, no amount of prejudice will deny a defendant a fair trial. I am not willing to presume that every jury will swoon over DNA evidence—or presume that the presentation of such evidence will close the book on a defendant’s case. The path cannot be from introduction of DNA evidence to imposition of sentence. In between lies the fundamental guarantee of a fair trial. That cannot be swept away by science. In short, the harmless-error doctrine should not subvert the centrality of the jury’s role in determining guilt.
IV.
Defendant’s guilt in the S.P. case was improperly bolstered by the wrongful joinder of the L.R. and K.G. cases, and his guilt in the L.R. case was improperly bolstered by the wrongful joinder of the K.G. and S.P. cases. Prejudice has so thoroughly infected the proceedings that led to the convictions on the S.P. and L.R. charges that the integrity of the jury verdicts cannot be saved by the harmless-error doctrine. If the evidence against defendant is as strong as the majority believes, the State should have little trouble convicting him after a fair trial.
I fear that by upholding the guilty verdicts in this case the majority is setting an exceedingly high tolerance level for prejudicial evidence that undermines the fundamental guarantee of a fair trial. This case may well become the new standard by which every trial error, however deeply it undercuts the defendant’s substantial rights, will become harmless error. In the end, if the wrongful joinder of three cases together, each of which should have been tried separately, is harmless error, then there can *114hardly be any trial flaw, no matter how great or grievous, that cannot be explained away to uphold an unjust verdict.
For these reasons, I respectfully dissent from the majority’s reversal of the judgment of the Appellate Division in the S.P. and L.R. eases. I concur in the remainder of the majority’s opinion.
For Affirmance in part/reversal in part/remandment—Chief Justice RABNER and Justice LaVECCHIA, HOENS, PATTERSON, and Judges RODRÍGUEZ (temporarily assigned) and CUFF (temporarily assigned)—6.
For dissent in part/concurrence in part—Justice ALBIN—1.
The charges against the fifth victim were dismissed without reaching trial after defendant was sentenced for the other crimes. Those charges included first-degree aggravated sexual assault, second-degree sexual assault, and second-degree burglary, stemming from crimes committed against T.G. on April 24, 2005.
Although not elicited at trial, the State alleged during discovery that the attacker wore a condom and that it bore on the joinder issue addressed pre-trial.
A separate indictment charged defendant with two additional counts. The first was a second-degree certain persons offense, N.J.S.A. 2C:39-7b(l), based on defendant's possession of a handgun on May 27, 2005, which was seized from his home on that date during a search conducted pursuant to a search warrant. The second count, third-degree receiving stolen property (a handgun), N.J.S.A. 2C:20-7, was later dismissed by the State.
After the jury returned the verdict for the charges relating to the assault of J.L., it also found defendant guilty of the charge on the certain persons not to possess weapons offense, which was charged in a separate indictment. The Appellate Division affirmed defendant’s ten-year sentence on that charge. This Court denied defendant's cross-petition for certification on issues relating to that conviction.
The panel specifically rejected the argument that the presence of DNA evidence rendered the other trial errors harmless. Relying on State v. Bradshaw, 195 N.J. 493, 950 A.2d 889 (2008), the panel noted that the existence of DNA evidence does not automatically render an error harmless because, when there are issues of contamination of the DNA testing process, as there were in Bradshaw, the jury may conclude that DNA evidence is not conclusive.
Had the State produced expert testimony to support that use of a condom in a sexual assault was unique or that it was even unusual among sexual assaults committed during burglaries, we may have viewed that discrete fact differently in this fact-driven analysis.
To the extent that defendant maintains that the specter of contamination during the DNA testing nullifies any ability to rely on the DNA evidence, citing Bradshaw, supra, 195 N.J. at 493, 950 A.2d 889, we reject that comparison. This case is not Bradshaw. The State's expert witness did not concede to knowledge of any contamination at her laboratory during her years working there. And, she testified to following standard operating procedures. Speculation about *105contamination, even if mentioned in summation, does not suffice to raise a reasonable doubt about the quality of the nuclear DNA results in this matter.