Defendant Paul Fitzpatrick was charged with rape and tried before two juries, both of which became deadlocked. After the second mistrial, the trial court, upon motion of defendant, dismissed the charges against defendant with prejudice, pursuant to Vermont Rule of Criminal Procedure 48(b)(2). The State, which seeks to try defendant a third time, appeals, claiming the court abused its discretion in dismissing the charges. Defendant cross-appeals, arguing that because of prosecutorial misconduct the court should have dismissed the case at the beginning of the second trial. We affirm without reaching the cross-appeal.
On September 4,1996, according to testimony at the two trials, the alleged victim met a friend for dinner and drinks. The two women, who knew each other through work, went to a restaurant at around 7:30 p.m. and stayed there for about an hour and a half. They then went to a bar, where they were acquainted with the owner and the bartender. At *113the bar they were approached by defendant, who introduced himself, and asked them to play pool. The three played pool together on and off, and struck up a conversation. At around 11:00 p.m., the alleged victim’s friend decided to go home. The friend tried several times to persuade the alleged victim to go with her, but she refused to leave.
According to the alleged victim’s testimony, she intended to finish the beer she was drinking and then drive home. However, at that point defendant approached her and said that he was having a party in his motel room nearby and invited her to come along. Believing that several people from the bar would be there, she accepted. When they arrived at the motel room, they were alone. The alleged victim claimed that when she realized there was no party and attempted to leave, defendant held her against her will and raped her. She testified that he slapped her and bit the earrings from her ears. She scratched his back with her fingernails, trying to resist. Eventually, he fell asleep, and she left the room and called the police to report the rape. The call was logged in at 3:31 a.m.
The police interviewed the alleged victim at the police station, and then took her to the hospital to have a rape kit examination performed to gather evidence. At about 7:00 a.m., the detective in charge of the investigation knocked on the door of defendant’s hotel room to question him. Defendant admitted that a woman had been there, but said that “nothing had happened.”
According to defendant, it was the alleged victim who asked to come to his motel room with him. He testified that while they were at thé bar they began flirting with each other. They were sitting at the bar together, talking, and she had her hand on his leg. They decided they were going to go back to his room, and she asked, “How do you want me to be for you?” When they got to his room, they began kissing and fondling each other. The alleged victim took off her earrings and placed them on the night stand. Defendant testified that, at one point, she ran her fingernails down his back, causing him pain. When he said that he did not like that, she' asked again, “How do you want me to be for you?” Defendant testified that he became uncomfortable with the situation, and decided to ask her to leave. He asked her several times to leave, but she kept saying that she wanted to stay. Finally, when he insisted, she became upset and stormed out of the room, leaving the door open as she left. He closed the door and went to sleep.
Defendant was charged with kidnapping, aggravated sexual assault and simple assault. The case was tried before a jury in February 1998. At the trial, the alleged victim and defendant both testified, as did *114several police officers who were involved in the investigation, medical personnel who performed the rape kit examination, two expert witnesses, and several other fact witnesses. A motel guest testified that he heard a commotion in another room during the night in-question. He heard a loud male voice that sounded upset and emotional, and then a sound like a child running across a room and hitting a coffee table. The doctor who examined the alleged victim at the hospital testified that she had scratches on her lower back, and a long scratch on her left interior thigh, and that there was a half centimeter fresh laceration in her perineum, but that no sperm had been found in her urine or vaginal mucous. The expert witnesses testified that there was saliva on one of the earrings the alleged victim had left in defendant’s room, and that there was a one in fifty chance it came from defendant. There was also a one in fifty chance that DNA material scraped from underneath the alleged victim’s fingernails came from defendant. On the third day of trial, the jury became deadlocked, and a mistrial was declared. The State claimed, and the trial court accepted, that the jurors were eleven to one for conviction.
After the first mistrial, the charges were amended to unlawful restraint, sexual assault, and simple assault, and the case was tried to a second jury in January 1999. Essentially the same witnesses testified, and the same medical and expert DNA evidence was presented. Another hung jury resulted, and the court declared a mistrial. Based on the State’s representation, the court found that the vote of the jurors was six for conviction and six for acquittal. Thereafter, defendant filed his motion to dismiss the prosecution -with prejudice under V.R.Cr.P. 48(b)(2).
After hearing evidence and argument on the motion, the court, in a twenty-four page opinion, granted it, based on its evaluation of the factors set out in State v. Sauve, 164 Vt. 134, 140-41, 666 A.2d 1164, 1168 (1995), and State v. Abbati, 493 A.2d 513, 521-22 (N.J. 1985). The court relied upon the following factors: (1) no significant new evidence will be available in future trials; (2) the case is a credibility contest between .the victim and defendant, and the State’s evidence was unpersuasive in two trials and “will prospectively remain unpersuasive”; and (3) defendant continues to suffer personally and financially from the prosecution. The court recognized that there were factors weighing against dismissal, particularly the seriousness of the charges and the alleged victim’s desire to continue the prosecution, but it found that these factors, were' outweighed by those supporting dismissal. The court also evaluated whether dismissal would serve “the *115effective administration of the [cjourt’s business” and concluded it would because the backlog of untried felony cases in the Windham District Court meant that a retrial of defendant was “in contravention of the rights of other defendants and victims to have a speedy jury resolution.”
As the trial court recognized, this case calls upon us to interpret and apply our recent controlling decision in State v. Sauve, a case in which the trial court also dismissed the charges pursuant to V.R.Cr.P. 48(b)(2) following a hung jury. Four points from Sauve are critical to the decision in the case before us.
First, despite the involvement of “separation-of-powers principles” when a court dismisses a criminal prosecution supported by substantial evidence with no opportunity for the State to refile, Sauve, 164 Vt. at 139, 666 A.2d at 1167, the court nonetheless has the power to dismiss a case “when it would be fundamentally unfair to continue the prosecution.” Id. at 140, 666 A.2d at 1167. The power is recognized in V.R.Cr.P. 48(b)(2). This power extends to retrials following a mistrial because of a hung jury:
the repeated prosecution of a defendant for the same crime following hung juries where no new evidence exists raises issues concerning traditional notions of fundamental fairness and substantial justice. Repeated trials involving the same offense can frustrate the search for truth and the effective administration of justice by depleting the resources of the parties, by imposing hardships on witnesses, and by fostering the perfunctory presentation of stale testimony, the exaggeration of subtle differences in witnesses’ recollections to challenge their credibility, and the tailoring of testimony based on the jury’s perceived reaction in prior trials.
Id. at 142,666 A.2d at 1169.
Second, the power to dismiss is limited: “trial courts may dismiss prosecutions in furtherance of justice against the wishes of the prosecutor only in- rare and unusual cases when compelling circumstances require such a result to assure fundamental fairness in the administration of justice.” Id. at 140, 666 A.2d at 1167. Therefore, “the trial court must generally defer to the prosecutor’s decision to retry the case, but if fundamental fairness compels dismissal, the court is authorized to do so.” Id. at 142-43, 666 A.2d at 1169. Defendant has *116the burden of proof to show grounds to dismiss with prejudice following one or more hung juries. Id. at 144, 666 A.2d at 1170.
Third, Sauve contains a list of nonexclusive factors for the trial court to evaluate and apply to its determination whether dismissal with prejudice is appropriate. Id. at 140-41, 666 A.2d at 1168. The court “should consider such factors, which weigh the respective interests of the defendant, the complainant, and the community at large.” Id. at 141, 666 A.2d at 1168.
Fourth, the trial court’s decision whether to dismiss with prejudice under V.R.Cr.P. 48(b)(2) involves an exercise of the court’s discretion, reviewable in this Court only for abuse of discretion. We reverse a discretionary decision only if the trial court has “‘entirely withheld its discretion or where the exercise of its discretion was for clearly untenable reasons or to an extent that is clearly untenable.’ ” Brueckner v. Norwich University, 169 Vt. 118, 133, 730 A.2d 1086, 1097 (1999) (quoting Lent v. Huntoon, 143 Vt. 539, 552, 470 A.2d 1162, 1171 (1983)).
Applying these principles, we reversed the decision to dismiss with prejudice in Sauve, concluding that the grounds relied upon by the trial judge were inadequate. Specifically, the court dismissed the case because of the traumatic effect on the victim of having to testify again, the fact that the evidence would be the same in another trial, and the age of the evidence. We held that these factors were inadequate to justify the remedy because the victim wanted to go forward with another trial despite the difficulty in giving testimony, there had been only one trial so it was not clear what verdict another jury would reach, and the age of the evidence was of limited relevance without some showing of prejudice, especially since the crime involved had a very long statute of limitations. Sauve, 164 Vt. at 144-45, 666 A.2d at 1170. As the trial court emphasized, this is a very different case from Sauve, primarily because there have been two prior trials and the trial judge was not relying upon its view of the impact of further trials on the alleged victim, without evidence to support that view. Thus, the trial court noted “judicial discretion is exercised on a stronger factual and procedural basis [than was present in Sauve].”
The State argues that the trial court misapplied the Sauve factors and could not find a compelling reason to overrule the prosecutor’s decision to go forward with a third trial. This argument gives us an opportunity to revisit the Sauve factors. We continue to stress that in cases involving retrials after hung juries, the trial court *117should consider all relevant factors in deciding whether to prevent a retrial by dismissing the case with prejudice under Rule 48(b)(2). The trial court discharged this obligation in this case. We believe, however, that the most important factor will be the likelihood that the State will prevail in a third trial. Indeed, in State v. Abbati, the primary precedent upon which our decision in Sauve is based, the New Jersey Supreme Court characterized the decision before it as turning upon whether “the chance of the State’s obtaining a conviction upon further retrial is highly unlikely.” Abbati, 493 A.2d at 521. Virtually all decisions which have authorized trial courts to dismiss cases to prevent retrials following hung juries have emphasized the primary role of this factor. E.g., United States v. Ingram, 412 F. Supp. 384, 385-86 (D.D.C. 1976); Ex parte Anderson, 457 So. 2d 446, 451 (Ala. 1984) (“trial judge’s evaluation of the evidence must also be given serious consideration”); State v. Moriwake, 647 P.2d 705, 713 (Haw. 1982); State v. Simmons, 752 A.2d 724, 729-30 (N.J. Super. Ct. 2000); State v. Witt, 572 S.W.2d 913, 917 (Tenn. 1978).
In this case, the trial judge who presided at the second trial also ruled on the motion to dismiss with prejudice. His thorough ruling indicates that he had read the transcript of the first trial. He characterized the State’s prospects in a third trial as grim and stated its evidence “will prospectively remain unpersuasive.” He is in the best position to make this assessment because he observed the testimony and demeanor of defendant and the alleged victim, and each juror’s decision must turn on which of those two witnesses to believe.
Two related factors particularly support the trial court’s decision. First, the jury votes were less favorable to the State in the second trial, as compared to the first, to the point where the jurors were equally divided after that trial. As we emphasized in Sauve, retrials tend to be less effective vehicles in finding the truth than the first trial. For whatever reason, the retrial was less persuasive to the jurors than the first trial.
Second, the State has not identified any new evidence that could be introduced into the third trial that would offer a significant chance of improving the result. See State v. Paige, 607 A.2d 164, 169 (N.J. Super. Ct. 1992) (third trial authorized under the Abbati standard primarily because new evidence emerged). Indeed, the State has conceded that this factor is against it.
We also believe it is significant that there have been two hung juries in this case as compared to only one in Sauve. As we said in Sauve, the number of hung juries is not determinative, but it is an important *118factor. 164 Vt. at 143, 666 A.2d at 1169; accord Moriwake, 647 P.2d at 713. Each inconclusive retrial heightens our concern about the fairness of the process and the continued ability of the defendant to muster the human and financial resources to present a thorough and effective defense. See Jones v. Hogg, 732 F.2d 53, 57 (6th Cir. 1984) (multiple prosecutions involve financial burdens and “emotional stress associated with an unresolved accusation of wrongdoing”).
Primarily for these reasons, we conclude that the trial court could find compelling grounds to prevent further retrial in this case. Whether this Court, or another trial judge, would have reached a different decision is not the question. The trial court acted within the range of its discretion, and we must affirm its decision.
In reaching this conclusion, we reject the State’s argument that the trial court dismissed the case primarily because of docket congestion, which it considers to be an irrelevant factor. We are not prepared to state that docket congestion is irrelevant because it may affect the ability of the trial court to provide a speedy trial to the defendant before it and to other defendants. See United States v. Rossoff, 806 F. Supp. 200, 203 (C.D. Ill. 1992) (tripling of criminal caseload a factor in court’s decision to dismiss with prejudice after a second hung jury). Whatever our general view, however, docket congestion was not a determinative factor in this case. See State v. Corchado, 512 A.2d 183, 187 (Conn. 1986) (although trial court alluded to “crowded dockets,” decision to dismiss was based on other proper factors). The trial court noted that Rule 48(b)(2) has two prongs, requiring it to look at whether dismissal would serve “the ends of justice” and also “the effective administration of the [cjourt’s business.” It viewed the second prong as requiring it to look independently at the effect of a retrial in this case on the disposition of other cases.
Although the wording of the rule sets forth two independent elements, our decision in Sauve did not frame the inquiry in that fashion. Instead it required the court to look at all relevant factors against a general standard of “fundamental fairness and substantial justice.” Sauve, 164 Vt. at 142, 666 A.2d at 1169. That standard embodies both prongs of Rule 48(b)(2) and does not require an independent analysis of the state of the court’s docket. Thus, the court’s analysis of this factor was superfluous.
Following the first trial, defendant moved to dismiss the prosecution alleging prosecutorial misconduct in the closing argument. By cross-appeal, he argues here that the court should have granted the motion under the Vermont Constitution, Chapter I, Article 10. Because of our *119disposition of the State’s appeal, we do not reach the cross-appeal issue.
Affirmed.