MADSEN, J.
*282¶1 Ronald Brown appeals an unpublished Court of Appeals decision affirming his exceptional sentence for two counts of first degree robbery and one count of first degree burglary. At his first sentencing hearing, the trial court decided not to impose an exceptional sentence on his original convictions. On appeal, four of his seven original convictions were vacated. Upon resentencing, the trial court exercised its discretion and imposed an exceptional sentence above the sentencing range for his remaining convictions. Brown argues that the decision to impose an exceptional sentence on remand was collaterally estopped, that the exceptional sentence is the result of judicial vindictiveness, and that the State's recommendation for an exceptional sentence is the result of prosecutorial vindictiveness.
*965We affirm the Court of Appeals decision and affirm Brown's sentence.
FACTS
¶2 In 2012, Brown was charged with two counts of first degree kidnapping, two counts of first degree robbery, one count of first degree burglary, and two counts of second degree assault, all with a firearm. Brown proceeded to jury trial. At the close of trial, the jury convicted Brown of all seven counts.
¶3 At the sentencing hearing, the State recommended the high end of the sentencing range for Brown's convictions.
*283The State noted that an exceptional sentence upward would be warranted based on Brown's high offender score but did not recommend an exceptional sentence at the original sentencing. The trial court also declined to impose an exceptional sentence, citing the victims' statements that their lives would have been in danger if not for Brown being present. The trial court ultimately sentenced Brown to the high end of the sentencing range, 638 months in prison.
¶4 On appeal, the Court of Appeals reversed Brown's two kidnapping convictions on instructional error and his two assault convictions on double jeopardy grounds. The court then remanded for resentencing.
¶5 On remand, the State elected to dismiss the two kidnapping charges without prejudice because of the time and effort involved in relocating the victims and codefendants, and the resources spent by the prosecution in retrying Brown. At the resentencing hearing, the State initially recommended a sentence of 351 months-reflecting the high end of the standard sentencing range. However, the State later amended its recommendation to reimpose the original sentence as an exceptional sentence. The court declined to reimpose the original sentence but did impose an exceptional sentence of 399 months. At the resentencing hearing, the judge noted that he did not impose the exceptional sentence at the original sentencing because he felt the 638 months was "legally appropriate and within the law." Verbatim Report of Proceedings, Resentencing & Mot. Hr'g (VRP Mot.) (June 21, 2016) at 34. The judge also noted that imposing the high end of the sentencing range with the remaining charges would give Brown a "free crime[ ]," justifying the imposition of an exceptional sentence. Id. However, the judge stopped short of imposing the original sentencing range.1 Id.
*284ANALYSIS
Collateral Estoppel
¶6 Brown first argues the trial court is collaterally estopped from imposing an exceptional sentence at the resentencing hearing when it chose not to impose one at the original sentencing hearing.
¶7 For collateral estoppel to apply, (1) the issue in the prior adjudication must be identical to the issue currently presented for review, (2) the prior adjudication must be a final judgment on the merits, (3) the party against whom the doctrine is asserted must have been a party to or in privity with a party to the prior adjudication, and (4) barring the relitigation of the issue will not work an injustice on the opposing party. State v. Harrison, 148 Wash.2d 550, 561, 61 P.3d 1104 (2003). Courts should not apply collateral estoppel hypertechnically but, rather, with realism and rationality. State v. Tili, 148 Wash.2d 350, 361, 60 P.3d 1192 (2003).
¶8 Brown argues the issue in the prior adjudication is identical to the issue currently presented for review-whether to impose an exceptional sentence based on Brown's offender score. Suppl. Br. of Pet'r 6-7. Specifically, Brown argues because the judge chose not to impose an exceptional sentence at the initial sentencing despite being justified due to his offender score, collateral estoppel applies. Id. at 7.
*966¶9 His argument is similar to the one made in Tili. In that case, the defendant was initially sentenced to 417 months. 148 Wash.2d at 357, 60 P.3d 1192. The trial court did not impose an exceptional sentence because it treated his offenses as separate and distinct conduct. Id. However, the court indicated that if his ruling was reversed on appeal and the *285offenses should have been treated as the same criminal conduct, the court would impose the same sentence as an exceptional sentence. Id. The original sentence was reversed, to be treated as same criminal conduct, and the trial court imposed the same 417 month sentence as an exceptional sentence. Id.
¶10 The defendant in Tili argued that the trial court was collaterally estopped from imposing an exceptional sentence on resentencing because it chose not to do so at the original sentencing hearing. Id. at 361, 60 P.3d 1192. This court was not persuaded. We noted that separate and distinct conduct for multiple offenses resulted in a fundamentally different sentence from same criminal conduct-the former resulting in consecutive sentences, while the latter results in concurrent sentences. Id. at 362-63, 60 P.3d 1192. Thus, the issue at resentencing was fundamentally different.
¶11 Despite this, Brown attempts to distinguish Tili because, here, the only relevant change was the dismissal of four charges resulting in a lowered offender score. Suppl. Br. of Pet'r at 7. This is a distinction without effect. Under RCW 9.94A.535(2), a trial court may impose an aggravated exceptional sentence without a finding of fact by a jury under certain circumstances, one of them being that "[t]he defendant has committed multiple current offenses and the defendant's high offender score results in some of the current offenses going unpunished." RCW 9.94A.535(2)(c). To justify an exceptional sentence upward, a trial court must first calculate or otherwise determine the defendant's offender score, and based on that factor, the trial court has discretion to impose an exceptional sentence if it deems the defendant's sentence will result in "free crimes."
¶12 Based on Brown's offender score in addition to the firearm enhancements, Brown's original sentence totaled 638 months. The trial court had a question before it at that time: Based on the defendant's high offender score, should it impose an exceptional sentence based on the current range? The court decided the sentence at the time *286was fair and opted not to impose an exceptional sentence. On appeal, based on the reversal of two convictions on the basis of double jeopardy and the State's decision not to retry two other convictions, Brown was left with three convictions. The question upon resentencing thus fundamentally changed: Was the newly computed sentencing range sufficient based on Brown's offender score? The trial court had a new offender score and a new sentencing range to consider when it decided to impose the exceptional sentence.
Finality under Collateral Estoppel
¶13 Next, Brown argues that the issue of imposing the exceptional sentence was "final" for purposes of collateral estoppel. In support of this position, he cites to State v. Kilgore, 167 Wash.2d 28, 216 P.3d 393 (2009), for the proposition that an unchallenged exceptional sentence on appeal is final and has preclusive effect on remand should any other portion of the judgment and sentence be reversed. Suppl. Br. of Pet'r at 9.
¶14 In Kilgore, the trial court imposed an exceptional sentence of 560 months on seven counts. 167 Wash.2d at 32, 216 P.3d 393. On appeal, two counts were reversed and the other five affirmed. Id. The trial court did not elect to resentence him and instead signed an order striking the two reversed counts and changing his offender score accordingly. Id. at 34, 216 P.3d 393. We held that finality occurs when " 'the availability of appeal' [has] been exhausted." Id. at 43, 216 P.3d 393 (emphasis omitted) (quoting In re Pers. Restraint of St. Pierre, 118 Wash.2d 321, 327, 823 P.2d 492 (1992) ). "[A] case has no remaining appealable issues where an appellate court issues a mandate reversing one or more counts and affirming the remaining count, and where the trial court exercises no discretion on remand as to the remaining final count."
*967Id. at 37, 216 P.3d 393. We noted that "[a]lthough the trial court had discretion ... to revisit Kilgore's exceptional sentence on the remaining five convictions, ... it was not reconsidering the exceptional sentence imposed on each of the remaining counts." Id. at 41, 216 P.3d 393.
*287¶15 Brown's reliance on Kilgore is misplaced. It does not stand for the proposition that all exceptional sentences are final when they are not appealed. Rather, when a trial court does not exercise its discretion on remanded issues, those issues become final for purposes of reviewability. Here, because the trial court did exercise its discretion on remand to determine whether an exceptional sentence was appropriate, the issue became reviewable and is not "final."
The Effect of Collicott
¶16 Brown also argues that collateral estoppel applies based on State v. Collicott , 118 Wash.2d 649, 827 P.2d 263 (1992). Brown asserts the lead opinion's discussion on collateral estoppel is applicable as it is factually analogous to the instant case.2 In Collicott , the trial court, in its original sentencing, could have imposed an exceptional sentence but did not. Id. at 652, 827 P.2d 263. The defendant appealed, and the case was remanded to the trial court for resentencing. At the resentencing, the trial court learned that the defendant had a stayed charge pending until resolution of the case. Id. at 653, 827 P.2d 263. The trial court imposed an exceptional sentence at rehearing based on deliberate cruelty. Id. at 653-54, 827 P.2d 263. This court, in its lead opinion, reversed, holding that the defendant's offender score was not properly calculated and the trial court was collaterally estopped from imposing the exceptional sentence. It is the court's discussion that collateral estoppel applies to exceptional sentences that Brown relies on.
¶17 But Brown's reliance on this is misguided. First, the collateral estoppel discussion did not command a majority of the court.3 As subsequent case law has held, the collateral *288estoppel analysis is dicta and is not binding on this court. See Harrison, 148 Wash.2d 550, 61 P.3d 1104 ; Tili, 148 Wash.2d 350, 60 P.3d 1192. Second, the facts in Collicott are quite distinct from the instant case.
¶18 In Collicott, the trial court imposed an exceptional sentence upon resentencing based on deliberate cruelty. 118 Wash.2d at 654, 827 P.2d 263. The trial court could have imposed an exceptional sentence based on deliberate cruelty at the initial sentencing but chose not to. Id. at 653, 827 P.2d 263. The lead opinion stated that collateral estoppel applied where the court could have imposed an exceptional sentence based on the same factor that it relied on at resentencing. Id. at 661, 827 P.2d 263.
¶19 Brown argues that the same situation is presented here; the trial court could have imposed an exceptional sentence based on the "free crime" rule but chose not to, triggering collateral estoppel. But Brown overlooks another part of the lead opinion in Collicott in which the court suggested that a trial court may impose an exceptional standard based on the "clearly too lenient" standard (now the "free crime" rule) upon resentencing. Id. at 659-60, 827 P.2d 263.
Judicial Vindictiveness
¶20 Brown next argues that the exceptional sentence imposed by the trial court at the resentencing hearing is presumptively vindictive. Generally, a trial judge may impose a new sentence that is greater or less than the sentence originally imposed based on events subsequent to the first trial that may throw new light on the defendant's life, health, habits, conduct, and mental and moral propensities. North Carolina v. Pearce , 395 U.S. 711, 723, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969). But the "imposition of a *968penalty upon the defendant for having successfully pursued a statutory right of appeal or collateral remedy [is] a violation of due process of law." Id. at 724, 89 S.Ct. 2072. The Court in Pearce held that "whenever a judge imposes a more severe *289sentence upon a defendant after a new trial, the reasons for ... doing so must affirmatively appear." Id. at 726, 89 S.Ct. 2072. Such reasons must be based on "objective information concerning identifiable conduct on the part of the defendant occurring after the time of the original sentencing proceeding." Id.
¶21 For years, Pearce seemed to stand for a sweeping rule that applied a presumption of judicial vindictiveness whenever a new sentence was harsher than the original sentence imposed. However, the scope of the rule set out in Pearce has been substantially narrowed over the years. In Alabama v. Smith , the defendant originally was sentenced based on a guilty plea. 490 U.S. 794, 795, 109 S.Ct. 2201, 104 L.Ed.2d 865 (1989). The defendant successfully had his guilty plea vacated, and the case proceeded to trial. Id. After the trial the defendant was convicted and resentenced, this time to a life sentence. Id. at 796-97, 109 S.Ct. 2201. The defendant argued there was a presumption of vindictiveness under Pearce. The Supreme Court disagreed, stating that "subsequent cases have made clear that [Pearce' s] presumption of vindictiveness 'do[es] not apply in every case where a convicted defendant receives a higher sentence on retrial.' " Id. at 799, 109 S.Ct. 2201 (second alteration in original) (quoting Texas v. McCullough , 475 U.S. 134, 138, 106 S.Ct. 976, 89 L.Ed.2d 104 (1986) ). Rather, the presumption applies only in "[s]uch circumstances ... in which there is a 'reasonable likelihood' that the increase in sentence is the product of actual vindictiveness on the part of the sentencing authority." Id. (citation omitted) (quoting United States v. Goodwin, 457 U.S. 368, 373, 102 S.Ct. 2485, 73 L.Ed.2d 74 (1982) ). When there is no reasonable likelihood, the defendant must prove actual vindictiveness. Id. at 799-800, 109 S.Ct. 2201. The Court later went on to hold that the presumption does not apply where a greater penalty is imposed after trial than was imposed after the guilty plea because "the judge may gather a fuller appreciation of the nature and extent of the crimes charged [at trial]." Id. at 801, 109 S.Ct. 2201.
¶22 The Supreme Court has since declined to apply the Pearce presumption in a number of cases. See, e.g., *290McCullough, 475 U.S. 134, 106 S.Ct. 976, 89 L.Ed.2d 104 (holding the presumption does not apply where retrial that resulted in the harsher sentence was initiated because the trial judge herself concluded a new trial was warranted based on prosecutorial misconduct); Chaffin v. Stynchcombe, 412 U.S. 17, 93 S.Ct. 1977, 36 L.Ed.2d 714 (1973) ( Pearce presumption does not apply when new jury at retrial imposes a harsher penalty than the original jury); Colten v. Kentucky, 407 U.S. 104, 92 S.Ct. 1953, 32 L.Ed.2d 584 (1972) (presumption does not apply where de novo trial in general criminal jurisdiction courts imposes a harsher sentence than inferior court did at original sentence).
¶23 Under Pearce, it appears the presumption of vindictiveness does not apply since the subsequent aggregate sentence was substantially lower than the original sentence.4 However, Brown points to federal *969circuit cases that consider the sentence imposed on each count, not the aggregate sentence. There are essentially two schools of thought in determining whether the presumption applies-whether we compare the overall length of the new sentence to the original sentence or whether we compare the remaining sentence once the dropped convictions are factored out *291to the new sentence.5 The overwhelming majority of federal circuits subscribe to the former analysis. See, e.g., United States v. Pimienta-Redondo, 874 F.2d 9 (1st Cir. 1989) (presumption does not apply when one count is dropped and resentencing judge increases the sentence on remaining count so the overall sentence remains the same);6 United States v. Nerius, 824 F.3d 29 (3d Cir. 2016) ( Pearce does not apply where the revised sentence is lower than that originally imposed); United States v. Gray, 852 F.2d 136 (4th Cir. 1988) (no possibility or appearance of vindictiveness when the second sentence is shorter overall than the first sentence);7 United States v. Cataldo, 832 F.2d 869 (5th Cir. 1987) (no presumption when the judge sentences a defendant to the same sentence even though there are fewer remaining convictions); United States v. Rivera, 327 F.3d 612, 615 (7th Cir. 2003) ("we compare the total original punishment to the total punishment after resentencing in determining whether the new sentence is more severe"); United States v. Horob, 735 F.3d 866, 870 (9th Cir. 2013) ("presumption of vindictiveness does not apply ... because ... the [ ] court considered his overall sentence at the time of his original sentence and again on remand, and because his overall sentence was not increased"); United States v. Sullivan, 967 F.2d 370 (10th Cir. 1992).
¶24 Brown advocates for the "aggregate remainder" approach taken by the Second and Eleventh Circuits in United States v. Markus, 603 F.2d 409 (2d Cir. 1979), and United States v. Monaco, 702 F.2d 860 (11th Cir. 1983). In Markus, the defendant was sentenced to a total of 15 years *292on seven counts. 603 F.2d at 411. On appeal, the judge vacated two counts " 'reluctantly.' " Id. The government filed an additional charge, on which the court sentenced the defendant to 5 years. Id. The defendant argued vindictiveness under the due process clause of the Fourteenth Amendment to the United States Constitution. The Court of Appeals held the appropriate analysis is to disregard the sentence originally imposed by the trial judge on the count dropped and then compare the total remaining sentence imposed to the current sentence. Id. at 413. Since there was no evidence on the record to justify the increase, the court found there to be vindictiveness. Id. at 414.
¶25 Similarly, in Monaco, the Court of Appeals examined a sentence that remained the same after retrial, even though there were fewer counts. Monaco, 702 F.2d at 883. The court in Monaco also subscribed to the approach used in Markus. Id. at 885. Since the trial court did not state any reasons for the increase, the court applied the Pearce presumption. Id. However, more recently, the Eleventh Circuit discussed Monaco in United States v. Fowler , 749 F.3d 1010 (11th Cir. 2014). In Fowler , the court effectively repudiated the "aggregate remainder" approach taken in Monaco , stating, "Monaco's pre-guidelines approach for gauging the severity of a new sentence relative to an old one-the aggregate remainder approach-is not binding in the post-guidelines era, which presents materially different circumstances than those involved in that case." Id. at 1018. That court went on further to say that *970"[w]hile we are obligated to follow the holdings of an earlier decision, 'the holdings of a prior decision can reach only as far as the facts and circumstances presented to the court in the case which produced that decision.' " Id. at 1020 (quoting Anders v. Hometown Mortg. Servs., Inc., 346 F.3d 1024, 1031 (11th Cir. 2003) ). "Because Monaco arose and was decided before the sentencing guidelines existed, it could not, and did not purport to, decide what approach should be used to determine when the Pearce *293presumption applies to a new sentence imposed under the guidelines regime." Id.
¶26 Our court has never considered which approach to adopt in determining a Pearce presumption of vindictiveness. However, Division One of the Court of Appeals faced that issue in State v. Larson, 56 Wash. App. 323, 783 P.2d 1093 (1989). In that case, the defendant's original sentence was for 363 months as a consecutive sentence. Id. at 325, 783 P.2d 1093. After appeal, and upon resentencing, the trial court imposed a concurrent sentence of 360 months. Id. at 326, 783 P.2d 1093. The defendant raised a presumption of vindictiveness argument, and the Court of Appeals rejected it, stating the "revised aggregate sentence [was] less severe than his original aggregate sentence." Id. at 328, 783 P.2d 1093.
¶27 Similarly, Division Two, in State v. Ameline, addressed the Pearce presumption. 118 Wash. App. 128, 75 P.3d 589 (2003). In that case, the defendant was retried three times. The first conviction resulted in a sentence of 164 months. Id. at 130, 75 P.3d 589. After appeal, the second sentence resulted in a conviction, and the court reimposed the same 164 month sentence. Id. at 131, 75 P.3d 589. The defendant appealed a third time, this time on instructional error. The case was remanded for a third trial. After his third conviction, the trial court chose to impose an exceptional sentence of 240 months. Id. To justify it, the trial court made written findings of fact that could have been made at the other two trials. Id. The Court of Appeals found that because the third sentence was harsher than the previous two overall, the Pearce presumption applied. Id. at 133, 75 P.3d 589. The Court of Appeals also determined the presumption was not rebutted based on the justification the trial court made on record. Id. Division Three has not decided a case based on judicial vindictiveness.
¶28 Given that the overwhelming majority of the federal circuits subscribe to the "total aggregate" approach and that Divisions One and Two also adopted the same, we hold the Pearce presumption does not arise when the total *294sentence upon resentencing is not greater than the original sentence imposed.
Prosecutorial Vindictiveness
¶29 Finally, Brown argues that the State's request for an exceptional sentence on remand is presumptively vindictive. The due process clause is not offended by all possibilities of increased punishment upon retrial after appeal but only those that pose a realistic likelihood of vindictiveness. Blackledge v. Perry , 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628 (1974). The Pearce presumption of vindictiveness was extended to the prosecutorial context in Blackledge. In that case, the defendant was charged with a misdemeanor assault in an altercation with another inmate while serving another term. Id. at 22, 94 S.Ct. 2098. After his conviction, he appealed, which under North Carolina law automatically warranted trial de novo. Id. Prior to the trial, the State obtained a felony assault indictment that covered the same conduct as the misdemeanor assault. The defendant pleaded guilty to the felony assault charge. Id. at 23, 94 S.Ct. 2098. The defendant argued the Pearce presumption should apply in his case. Id. at 25, 94 S.Ct. 2098. The Court agreed and held that it was not constitutionally permissible for the State to respond to the defendant's appeal by bringing a more serious charge against him. Id. at 28-29, 94 S.Ct. 2098.
¶30 The Pearce presumption does not apply to all cases where a prosecutor brings more serious charges. In Bordenkircher v. Hayes, the Supreme Court determined that the presumption does not apply in the pretrial context during plea negotiations when the prosecution threatens and executes additional charges when plea negotiations do *971not result in a guilty plea. 434 U.S. 357, 98 S.Ct. 663, 54 L.Ed.2d 604 (1978). The Court noted the violation "lay not in the possibility that a defendant might be deterred from the exercise of a legal right, but rather in the danger that the State might be retaliating against the accused for lawfully attacking his conviction." Id. at 363, 98 S.Ct. 663 (citations omitted). *295Thus, the presumption was inapplicable where there is a " 'give-and-take negotiation common in plea bargaining between the prosecution and defense, which arguably possess relatively equal bargaining power.' " Id. at 362, 98 S.Ct. 663 (quoting Parker v. North Carolina, 397 U.S. 790, 809, 90 S.Ct. 1458, 25 L.Ed.2d 785 (1970) (Brennan, J., dissenting) ).
¶31 Similarly, in Goodwin, the Supreme Court held that a presumption of vindictiveness was not warranted when a prosecutor brings a more severe charge after a defendant has demanded a jury trial. 457 U.S. 368, 102 S.Ct. 2485, 73 L.Ed.2d 74. The Court noted, "[T]he mere fact that a defendant refuses to plead guilty and forces the government to prove its case is insufficient to warrant a presumption that subsequent changes in the charging decision are unjustified." Id. at 382-83, 102 S.Ct. 2485.
¶32 This court has decided the issue of prosecutorial vindictiveness in the pretrial context. See State v. Korum, 157 Wash.2d 614, 141 P.3d 13 (2006) (plurality opinion). Similar to Goodwin and Bordenkircher, Korum involved guilty pleas. In that case, the prosecution threatened 32 additional charges if he did not plead guilty. Id. at 620-21, 141 P.3d 13. This court held that adding 15 counts did not constitute prosecutorial vindictiveness, finding the case was similar to Bordenkircher and Goodwin, where the defendants chose to walk away from plea negotiations. Id. at 635-36, 141 P.3d 13. Later, in State v. Gamble, 168 Wash.2d 161, 187, 225 P.3d 973 (2010), this court noted that the Blackledge presumption was inapplicable where, as a result of case law, the conviction was necessarily vacated, requiring the prosecution to reevaluate what charges to bring against the defendant.
¶33 Unlike cases where the prosecution chooses to add charges after a defendant exercises his right of appeal and succeeds, this case involves a sentencing recommendation. Courts should be cautious when expanding the scope of prophylactic rules. While it is possible that the prosecution decided to recommend the original sentence as an exceptional sentence out of spite, the presumption does not apply simply because there is an opportunity for vindictiveness.
*296There must be a realistic likelihood of vindictiveness. Here, the State was faced with a decision-was the length of a standard range sentence sufficient given the facts of the case?
¶34 Unlike a charging decision, imposing a sentence does not fall under the core responsibilities of a prosecutor. Ultimately, a trial court determines what sentence is appropriate. The State merely recommends what it believes to be an appropriate sentence based on the crimes. Given this important distinction, we decline to extend the Blackledge presumption in this context.
CONCLUSION
¶35 We hold that collateral estoppel does not apply when a court imposes an exceptional sentence at resentencing based on the "free crime" aggravator when it chose not to impose an exceptional sentence at the first sentencing. Further, we hold that a presumption of vindictiveness is not triggered when a judge imposes a shorter overall sentence than the original or when a prosecutor recommends an exceptional sentence at resentencing when it did not recommend such a sentence at the original sentencing. We affirm the Court of Appeals.
WE CONCUR:
Fairhurst, C.J.
Johnson, J.
Owens, J.
Stephens, J.
Wiggins, J.
González, J.
Yu, J.