delivered the opinion of the Court.
In this wrongful death, automobile negligence case, a jury returned a verdict in favor of plaintiffs, finding defendants sixty percent negligent and the bankrupt tortfeasor dismissed before trial forty percent negligent. The Appellate Division reversed the verdict on apportionment of damages only and remanded for a new trial on that issue. This case raises three significant issues arising under the Comparative Negligence Act. First, whether the Act permits a jury to assign a percentage of fault to a joint tortfeasor dismissed from the case due to a discharge in bankruptcy. Second, whether a trial court may give an ultimate outcome charge to a jury explaining the implications of apportionment of fault among joint tortfeasors under the Act. Last, whether counsel is permitted in an opening or closing argument to state the specific percentages of fault that should be attributed to the parties.
I.
The essential facts are not in dispute. On February 16,1998, at approximately 6:15 a.m., defendant John Bennett was driving a forty-eight-foot tractor-trailer owned by defendant Grinnell Haulers, Inc. on a four-lane expanse of Route 80 when he changed lanes and crashed into a vehicle occupied by Bernard Brodsky and his wife, Gloria. The Brodskys’ car skidded out of control and came to rest facing oncoming traffic with the front end straddling the left shoulder of the road and the rear extending into the left lane. The Brodskys stepped from their disabled ear.
A few minutes later, William Horsman was driving in the far left lane of the highway when two ears in front of him swerved into the lane immediately to their right. Horsman observed the Brodsky vehicle directly in front of him, but was unable to change lanes because there were cars to his immediate right. Despite *107hitting his brakes, he slammed into Mr. Brodsky and then into the disabled vehicle, which struck Mrs. Brodsky, throwing her into a concrete divider. Both Mr. and Mrs. Brodsky suffered multiple, devastating injuries and were taken to St. Joseph’s Hospital and Medical Center. Mr. Brodsky died a short time after his arrival at the hospital. Gloria, his wife of forty-three years, survived, but is expected to suffer permanently from her injuries.
On her own behalf, Mrs. Brodsky filed a personal injury-negligence action, and, on behalf of her husband’s estate, she filed a survival action against Horsman and defendants, Bennett and Grinnell Haulers. Mrs. Brodsky and her three children also filed a wrongful death action against those parties. Defendants filed an answer to the complaint and a cross-claim against Horsman. Horsman, who was uninsured at the time of the accident, did not file an answer. Instead, he filed a bankruptcy petition in the United States Bankruptcy Court, identifying the Brodskys as potential judgment creditors. The bankruptcy court issued an order discharging Horsman from any debt arising from the accident.
The trial court granted summary judgment in favor of plaintiffs on the issue of liability and dismissed all claims and cross-claims against Horsman as a result of the bankruptcy court’s discharge order. At trial, there was no dispute concerning defendants’ negligence or plaintiffs’ lack of negligence. The only issues submitted to the jury were the extent of plaintiffs’ damages and the apportionment of fault between defendants and Horsman (even though any judgment against Horsman was uncollectable).
The jury found defendants sixty percent negligent and Horsman forty percent negligent, and awarded plaintiffs $1,640,000 in damages. The trial court denied defendants’ motion for a new trial and remittitur. Defendants appealed and plaintiffs cross-appealed. The Appellate Division found that the trial court erred in giving an ultimate outcome instruction to the jury and reversed and remanded for a new trial on apportionment of damages. Brodsky v. Grinnell Haulers, Inc., 362 N.J.Super. 256, 262, 284, *108827 A.2d 1104 (App.Div.2003). The Appellate Division affirmed both the trial court’s ruling allowing fault to be apportioned to the bankrupt Horsman and its ruling barring plaintiffs’ counsel from suggesting in his opening statement the specific percentage of fault to be allocated to Horsman. Id. at 263, 277, 827 A.2d 1104. We granted plaintiffs’ petition for certification. 178 N.J. 374, 840 A.2d 259 (2003).
II.
In the trial of a multi-defendant negligence action, the trier of fact must apportion fault among the parties it finds negligent by assigning each a percentage of fault on a scale of one to one hundred. N.J.S.A. 2A:15-5.2a. We must decide whether the trial court properly allowed the jury to assign a percentage of fault to Horsman—a party dismissed from the case before trial as a result of his discharge in bankruptcy. Plaintiffs argue that although Horsman was a named defendant in the complaint, he was no longer a party by the time of trial and, therefore, was not a “party” for the purpose of fault allocation under the Comparative Negligence Act. Plaintiffs press that argument because the assignment of a percentage of fault to Horsman may deny them a full recovery.
Defendants contend that the statute requires the jury to allocate a percentage of fault to each party whose negligence caused the accident, whether that party was dismissed due to a bankruptcy discharge or for some other reason, such as settlement. Defendants maintain that they should be accountable only for the damages assignable to them under the Comparative Negligence Act. We agree with the Appellate Division that “a bankruptcy discharge does not preclude the assessment of Horsman’s comparative liability,” even though a finding of fault by a jury will “not result in any personal liability to Horsman.” Brodsky, supra, 362 N.J.Super. at 277, 827 A.2d 1104.
*109A.
We begin by enunciating certain key principles that govern New Jersey’s modified comparative negligence system. A plaintiffs contributory negligence does not bar a recovery so long as that negligence “was not greater than the negligence of the person against whom recovery is sought or was not greater than the combined negligence of the persons against whom recovery is sought.” N.J.S.A. 2A:15-5.1. In other words, a plaintiff who is found to be more than fifty percent at fault is entitled to no recovery. A plaintiff who is found to be fifty percent or less at fault is entitled to a recovery, but any award of damages is diminished by the percentage of negligence attributed to her. Ibid.; Ostrowski v. Azzara, 111 N.J. 429, 445—46, 545 A.2d 148 (1988).
In a ease in which more than one defendant is found negligent, the trier of fact must then determine the amount of damages suffered by the plaintiff and each party’s percentage of negligence. N.J.S.A. 2A:15-5.2a. Based on the percentage of fault attributed to each party, the trial court then “mold[s] the judgment” and computes the amount of damages owed by each defendant. N.J.S.A. 2A:15-5.2d. A plaintiff is entitled to recover the full amount of the damages from a defendant found to be sixty percent or more at fault. N.J.S.A 2A:15-5.3a. A plaintiff, however, may recover only that percentage of damages directly attributed to a defendant found to be less than sixty percent at fault. N.J.S.A. 2A:15-5.3c. A defendant who pays “more than his percentage share” of an award is entitled to “seek contribution from the other joint tortfeasors” for the amount he has overpaid. N.J.S.A. 2A:15-5.3e.
B.
Our overriding goal in interpreting a statute is to determine the Legislatures intent. James v. Bd. of Trustees of the Public Employees’ Retirement Sys., 164 N.J 396, 404, 753 A.2d 1061 (2000). To divine that intent, we first look to the plain *110meaning of the words of the statute. Nat’l Waste Recycling, Inc. v. Middlesex County Improvement Auth., 150 N.J. 209, 223, 695 A.2d 1381 (1997). When the words of the statute do not provide a clear answer, then the context of those words in relation to other provisions of the statute may reveal their meaning. State in re G.C., 179 N.J. 475, 481-82, 846 A.2d 1222 (2004). When faced with an ambiguous statute, we also rely on legislative history to gain further insight into the probable intent of the Legislature. State v. Brannon, 178 N.J. 500, 507, 842 A.2d 148 (2004). Here, the issue to be decided is whether, under the Comparative Negligence Act, a defendant who has been dismissed from a case as a result of a bankruptcy discharge is still a “party” to whom a percentage of fault may be allocated. Under the Act, the trier of fact must first determine “the full value of the injured party’s damages” and then the extent of each party’s negligence “in the form of a percentage,” with the total equaling 100 percent. N.J.S.A. 2A:15-5.2a.
The Act also provides that the plaintiff may recover:
a. The Ml amount of the damages from any party determined by the trier of fact to be 60% or more responsible for the total damages.
[and]
c. Only that percentage of the damages directly attributable to that party’s negligence or fault from any party determined by the trier of fact to be less than 60% responsible for the total damages.
[N.J.S.A. 2A:15-5.3a, c.]
So, for example, a plaintiff who is injured by two defendants may collect all his damages from a defendant found to be sixty percent or more negligent. A defendant compelled to pay more than his percentage of fault may seek contribution from a joint tortfeasor. N.J.S.A. 2A:15-5.3e. In the illustration above, if the defendant were sixty percent negligent and made to pay all the damages, he could seek contribution of the forty percent he overpaid from the other tortfeasor. On the other hand, a defendant found to be fifty-nine percent or less negligent is liable to the plaintiff only for the percentage of damages he caused.
Neither the plain language of N.J.S.A. 2A:15-5.3a and c nor the legislative history addresses how to apportion fault when a “party” *111has received a bankruptcy discharge and been dismissed from the case before commencement of the trial. We, therefore, look at N.J.S.A. 2A:15-5.3a and c within the wider compass of the Comparative Negligence Act. By comparing those provisions to others within the Act, we can better understand how the Legislature intended fault to be apportioned between a non-settling defendant and a bankrupt tortfeasor in an automobile negligence case.
In 1995, as part of a comprehensive tort reform package, the Legislature amended the Comparative Negligence Act, allowing joint and several liability in an environmental tort action involving an insolvent defendant.1 L. 1995, c. 140, § 2 (amending N.J.S.A. 2A:15-5.3). As a result of the amendment, the plaintiff in an environmental tort action, with few exceptions, may recover the percentage of fault attributable to an insolvent party from the financially sound defendants. N.J.S.A. 2A:15-5.3d. In particular, N.J.S.A. 2A:15-5.3d provides that a plaintiff may recover “the percentage of compensatory damages attributable to a non-settling insolvent party’s negligence or fault[ ] ... from any non-settling party[,] in proportion to the percentage of liability attributed to that party.” N.J.S.A. 2A:15-5.3d(2). The amendment was “intended to enable an injured party in an environmental tort action to recover 100% of the compensatory damage award, notwithstanding a non-settling party’s insolvency.” Statement to Assembly Floor Amendment to Senate Bill No. 1494, at 9 (June 1, 1995).
In the 1995 tort reform package, the Legislature did not provide to plaintiffs in cases other than environmental torts protection from insolvent defendants. See L. 1995, c. 140. In an automobile *112negligence case, unlike an environmental tort ease, there is no provision similar to N.J.S.A. 2A:15-5.3e that allows plaintiffs to seek a full recovery from financially sound defendants when a joint tortfeasor is insolvent. The question, therefore, arises whether the Legislature intended to exclude allocation between a defendant and an insolvent party in an automobile negligence case.
The canon of statutory construction, expressio unius est exclusio alterius—expression of one thing suggests the exclusion of another left unmentioned—sheds some light on the interpretative analysis. Chevron U.S.A. Inc. v. Echazabal, 536 U.S. 73, 80, 122 S.Ct. 2045, 2049, 153 L.Ed.2d 82 (2002); Allstate Ins. Co. v. Malec, 104 N.J. 1, 8, 514 A.2d 832 (1986). That the Legislature, when it amended the Comparative Negligence Act in 1995, expressly permitted a plaintiff to recover an insolvent defendant’s portion of fault from the remaining defendants in an environmental tort action, but omitted such a provision for other causes of action implies that the omission was intentional, not an oversight. See GE Solid State, Inc. v. Dir., Div. of Taxation, 132 N.J. 298, 308, 625 A.2d 468 (1993) (finding that Legislatures use of words in one section of statute indicated that omission of same words in another section was intentional). The Legislature clearly knew how to impose full responsibility on a defendant joined with an insolvent tortfeasor. A textual analysis of the statute, therefore, strongly suggests that, with the exception of an environmental tort action, the Legislature did not intend to expand the right of recovery of a plaintiff in the ease of an insolvent tortfeasor.
We now address the significance of Horsman’s dismissal from the case with regard to fault apportionment. Horsman clearly was no longer a defendant in the negligence action at the time of trial because all claims against him had been dismissed. But did he remain a “party” for purposes of allocation of fault?
Whether to apportion fault to a defendant dismissed from a case is not a novel issue. In Young v. Latta, 123 N.J. 584, 585, 589 A.2d 1020 (1991), we addressed a case involving a defendant who had settled with the plaintiff before trial. We found that under *113the Comparative Negligence Act the trier of fact must allocate the percentage of fault among the settling and non-settling defendants to enable the court to calculate the percentage attributable to the non-settlers. Id. at 592, 594, 589 A.2d 1020. That was so even though the defendant who had not settled could not pursue a claim for contribution against one whose case had settled and been dismissed. Ibid. In Young, we observed that the Comparative Negligence Act does not address the effect of a settling tortfeasor on the apportionment of fault among the remaining joint tortfeasors. Id. at 589-90, 589 A.2d 1020. The Act, we noted, simply “deelare[s] the right to contribution and leave[s] most questions to the courts.” Id. at 589, 589 A.2d 1020 (internal quotation marks omitted). We held in that case that a non-settling defendant should not be accountable for the percentage of fault of a settling defendant. Id. at 592, 589 A.2d 1020. We found that result to be “a logical incident of the created right of contribution and no provision expressly stating that effect was necessary.” Id. at 591, 589 A.2d 1020 (quoting Judson v. Peoples Bank Trust Co. of Westfield, 17 N.J. 67, 93, 110 A.2d 24 (1954), aff'd on reconsid., 25 N.J. 17, 134 A.2d 761 (1957) (internal quotation marks omitted)). Implicit in our construction of the Act in Young was our recognition that a defendant who settles and is dismissed from the action remains a “party” to the case for the purpose of determining the non-settling defendant’s percentage of fault.
Another example of a case in which a dismissed defendant remained on the verdict sheet for purposes of allocation of fault is Burt v. W. Jersey Health Systems, 339 N.J.Super. 296, 771 A.2d 683 (App.Div.2001). In Burt, the Appellate Division held that the Comparative Negligence Act required the trier of fact to allocate a percentage of fault to a defendant who had been dismissed from a medical malpractice case as a result of the plaintiffs failure to timely serve an affidavit of merit. Id. at 306-07, 771 A.2d 683. The panel found that to do otherwise would “deprive the [remaining co-]defendants of their right to contribution and their right to respond in damages only to the extent they are found negligent.” Id. at 307-08, 771 A.2d 683.
*114In other contexts, the Appellate Division has noted that a defendant is allowed to prove that a non-party was the sole proximate cause of the plaintiffs harm—the so-called “empty chair” defense in which a defendant shifts blame to a joint tortfeasor who is not in the courtroom. See, e.g., Fabian v. Minster Mach. Co., Inc., 258 N.J.Super. 261, 276-77, 609 A.2d 487 (App.Div.) (stating that it would not be improper for defendant manufacturer in products liability case to shift causal blame to another who is not legally liable), certif. denied, 130 N.J. 598, 617 A.2d 1220 (1992); Kane v. Hartz Mountain Indus., Inc., 278 N.J.Super. 129, 145, 650 A.2d 808 (App.Div.1994) (stating in personal injury case that “regardless of whether an employer is joined in a plaintiffs suit by a third-party complaint, and despite the fact that the employer is immune from suit by an employee, the employer’s negligence may under certain circumstances be placed in issue during the trial of plaintiffs suit”), aff'd o.b., 143 N.J. 141, 669 A.2d 816 (1996). The practical effect of a defendant proving that the “empty chair” was responsible for the accident is that the plaintiff will receive no recovery.
The guiding principle of our State’s comparative fault system has been the distribution of loss “in proportion to the respective faults of the parties causing that loss.” Blazovic v. Andrich, 124 N.J. 90, 107, 590 A.2d 222 (1991). The comparative fault scheme serves to “implement New Jersey’s approach to fair apportionment of damages among plaintiffs and defendants, and among joint defendants.” Erny v. Estate of Merola, 171 N.J. 86, 99, 792 A.2d 1208 (2002); see also Governor’s Reconsideration and Recommendation Statement to Senate Bill No. 215, at 1 (Sept. 16, 1982) (endorsing change to Comparative Negligence Act as consistent with “policy of allocating responsibility among all negligent parties in proportion to their relative fault”). Those policies are the basis for the requirement that the trier of fact assess the negligence of a joint tortfeasor who has settled. Young, supra, 123 N.J. at 590, 589 A.2d 1020; Judson, supra, 17 N.J. at 92, 110 A.2d 24.
*115Plaintiffs argue that the Appellate Division’s decision requiring the trier of fact to assess the negligence of a bankrupt party conflicts with our decision in Ramos v. Browning Ferris Indus. of South Jersey, Inc., 103 N.J. 177, 510 A.2d 1152 (1986). We disagree. In Ramos, we held that a jury could not assign fault to an employer immune from suit under the Workers’ Compensation Act, thereby requiring fault to be apportioned entirely between the plaintiff and third-party defendant tortfeasor. Id. at 193-94, 510 A.2d 1152. That result followed because the Workers’ Compensation Act bars a plaintiff employee from suing a negligent employer for damages. The Workers’ Compensation Act
removes the employer from the operation of the Joint Tortfeasors Contribution Law. Because the employer cannot be a joint tortfeasor, it is not subject to the provisions of the Joint Tortfeasors Contribution Law, and a third-party tortfeasor may not obtain contribution from an employer, no matter what may be the comparative negligence of the third party and the employer.
[Id. at 184, 510 A.2d 1152.]
Stated differently, an employer cannot be a party to a negligence action and thus can never be considered a joint tortfeasor subject to the Comparative Negligence Act. See Arthur Larson, Third-Party Action Over Against Workers’ Compensation Employer, 1982 Duke L.J. 483, 488 (“The employer is not jointly liable to the employee in tort; therefore he cannot be a joint tortfeasor.”).
We agree with the Appellate Division that an employer’s immunity from suit under the Workers’ Compensation Act is different from a joint tortfeasor’s discharge in bankruptcy. As the Appellate Division correctly noted, “Horsman was not statutorily immune from a negligence suit at the time of the accident----[and] only became immune after he discharged his debt in bankruptcy.” Brodsky, supra, 362 N.J.Super. at 277, 827 A.2d 1104. A plaintiff, or for that matter a defendant seeking contribution, will be unable to collect from a joint tortfeasor who receives a bankruptcy discharge after trial, regardless of the apportionment of fault to that tortfeasor at trial. The fault of an insolvent joint tortfeasor who does not file for bankruptcy is apportioned in relation to the other parties, although in reality a judgment against that tortfea*116sor will not be collectable. We decline to carve out an exception in the case of a joint tortfeasor whose case is dismissed before trial because of a bankruptcy discharge. We decline to follow the approach advanced by plaintiffs, in which a defendant who is found to be one percent negligent would be held responsible for ninety-nine percent of the negligence caused by a joint tortfeasor dismissed from the case as a result of a bankruptcy discharge.
We hold that the trier of fact must determine the percentage of fault or negligence of a party dismissed from a negligence action following that party’s discharge in bankruptcy. If the jury finds Horsman forty percent negligent or less, plaintiffs can pursue a full recovery from defendants. N.J.S.A. 2A:15-5.3a, c. However, if Horsman is found to be more than forty percent negligent, then plaintiffs can collect from defendants only the percentage of fault allocated to those defendants. Ibid. That conclusion assures that defendants are not deprived of the benefits of the Comparative Negligence Act, namely their right to be held accountable only for their percentage of fault, provided that portion is less than sixty percent. N.J.S.A. 2A:15-5.3c.
III.
A.
The trial court granted summary judgment in favor of plaintiffs on the issue of liability. Defendants did not argue that plaintiffs were at fault in causing the accident. The jury was asked to apportion fault only between defendants and Horsman. The court gave an “ultimate outcome charge” consistent with Model Jury Charge (Civil) § 8.21(C),2 instructing the jury on the effect of *117apportioning fault between defendants and Horsman. The court advised the jury that if it were to find defendants and Horsman negligent, then it would allocate a percentage of fault to each and the allocation would determine how much of the award each would pay.3 The court then explained the practical effect of that allocation under the Comparative Negligence Act:
One who is found to be 60 percent of [sic] more responsible for the total damages is liable to the plaintiff for the total amount of the award. If one is found to be less than 60 percent responsible for the damages is [sic] liable only for the amount of damages directly attributable to his negligence or fault; therefore, you will attribute to Mr. Bennett/Grinnell and Horsman the percentage that describes or measures their contributions to the happening of the accident.
[ (Emphasis added.) ]
Defendants objected to the court informing the jury of the consequences of its allocation, fearing that the jury would then shape the numbers to achieve an outcome-oriented decision. In other words, once the jury knew that plaintiffs could collect fully from a party found at least sixty percent negligent, it might fashion a verdict apportioning fault differently than if it were blind to the consequences. The jury allocated sixty percent of fault to defendants and forty percent to Horsman, thus allowing plaintiffs to recover 100 percent of the damages from defendants. See *118N.J.S.A 2A:15-5.3a. Because the bankruptcy court discharged Horsman of any financial responsibility for his role in causing the accident, defendants were left with an illusory right of contribution under N.J.S.A. 2A:15-5.3e.
The Appellate Division found that it was improper to give an ultimate outcome charge “regarding allocation of fault among joint tortfeasors” in the circumstances of this case and remanded for a new trial on that issue. Brodsky, supra, 362 N.J.Super. at 262, 827 A.2d 1104. The panel distinguished between the ultimate outcome charge that we approved in Roman v. Mitchell, 82 N.J. 336, 413 A.2d 322 (1980)—a charge that explains the consequences of an allocation of fault involving a plaintiff and defendant—and the charge in this case involving only joint tortfeasors. Brodsky, supra, 362 N.J.Super. at 270-71, 827 A.2d 1104. In “conelud[ing] that Roman should not be extended to cover the circumstances of this case,” the panel noted that “[wjhether and how a plaintiff recovers a damage award is none of the jury’s concern, and should not be part of its deliberations.” Id. at 270, 271, 827 A.2d 1104. The panel found that defendants suffered clear prejudice from the ultimate outcome charge because the charge might have encouraged the jury “to manipulate its allocation of fault, at the expense of its actual apportionment analysis, in order to ensure that plaintiff[s] could recover the full damage award.” Id. at 271, 272, 827 A.2d 1104. We agree. We, therefore, affirm the Appellate Division’s decision to reverse and remand for a new trial on the apportionment of damages between defendants and Horsman.
B.
We begin our analysis by noting that those who favor and disfavor the ultimate outcome charge offer antithetical rationales to support their positions. Proponents of the ultimate outcome charge argue that allowing jurors to know the consequences of their decision will assure that jurors are not acting under preconceived, false assumptions regarding the operation of the law and will better enable them to render a fair and just verdict. Jordan *119H. Leibman et al., The Rise and Fall and Perhaps Rise Again of the “Blindfold” Rule in Modified Comparative Fault Cases: A Proposed Experiment, 102 Dick. L.Rev. 33, 35-36 (1997). Opponents argue that presenting the jury with information irrelevant to its deliberations will tempt jurors to manipulate the outcome and will lead to intellectually dishonest results. Ibid. Our Court in addressing at various times whether an ultimate outcome charge is appropriate in a particular case has not attempted to bridge that philosophical divide by adopting an overarching theory to apply to all statutes and all circumstances. Instead, we have evaluated whether the purpose of the particular statute or law in question as well as the interest of justice would be advanced in each case by either giving or not giving an ultimate outcome charge. Such an individualized approach may appear inconsistent when applied over a wide spectrum of cases. But individual cases arise in different settings and under different statutes and present unique challenges and problems. In this case, our focus must be the operation of the Comparative Negligence Act as it relates to the apportionment of fault among joint tortfeasors.
Plaintiffs contend that the trial court’s ultimate outcome instruction was consistent with this Court’s holdings in Roman v. Mitchell, 82 N.J. 336, 413 A.2d 322 (1980), Fischer v. Canario, 143 N.J. 235, 670 A.2d 516 (1996), and Wanetick v. Gateway Mitsubishi, 163 N.J. 484, 750 A.2d 79 (2000). However, of those three cases, only in Roman did we address the propriety of an ultimate outcome charge in the context of apportionment of fault under the Comparative Negligence Act. See Wanetick, supra, 163 N.J. at 496, 750 A.2d 79 (holding that trial courts, in cases arising under Consumer Fraud Act, must give jurors ultimate outcome charge explaining effect of treble damages); Fischer, supra, 143 N.J. at 254, 670 A.2d 516 (concluding in medical malpractice case that ultimate outcome charge regarding molding of verdict on lost-chance-of-survival claim should be given to jury). Accordingly, we do not find Fischer and Wanetick of much value in resolving the particular issues in this case.
*120In Roman, supra, a twelve-year-old boy was seriously injured when the wheels of a truck came off and struck him while he was standing on the shoulder of the New Jersey Turnpike. 82 N.J. at 340, 413 A.2d 322. At the conclusion of the trial, the plaintiff requested an ultimate outcome instruction to inform the jury that if the plaintiff were found fifty-one percent or more at fault, the plaintiff would not be entitled to a recovery. Id. at 342-43, 413 A.2d 322. The trial court denied the request, and the jury subsequently found the plaintiff seventy-five percent at fault and the defendant twenty-five percent at fault. Id. at 343, 413 A.2d 322. In accordance with the Comparative Negligence Act, the plaintiff was denied an award of any damages—even the twenty-five percent allocated by the jury. Ibid. We reversed, holding that “a jury in a comparative negligence situation should be given an ultimate outcome charge so that its deliberations on percentages of negligence will not be had in a vacuum, or possibly based on a mistaken notion of how the statute operates.” Id. at 345, 413 A.2d 322. We were concerned that the jury in allocating fault may have acted under the mistaken belief that the plaintiff would recover twenty-five percent of his damages. Ibid. In Roman, “[w]e conclude® that, ordinarily, a jury informed of the legal effect of its findings as to percentages of negligence in a comparar tive negligence trial is better able to fulfill its fact finding function.” Id. at 346, 413 A.2d 322 (emphasis added). We added the caveat that “in a complex case involving multiple issues and numerous parties, the trial court, in the exercise of sound discretion, could withhold the instruction if it would tend to mislead or confuse the jury.” Id. at 346-47, 413 A.2d 322.
The Appellate Division, in interpreting Roman and distinguishing it from the present case, observed that “[i]n Roman, the ultimate outcome charge was deemed appropriate because it was important that the jury understand the potentially counterintuitive nature of New Jersey’s modified comparative negligence system, under which the plaintiff would recover no damages if defendants were held less than fifty percent at fault.” Brodsky, supra, 362 N.J.Super. at 270-71, 827 A.2d 1104. The panel saw no similar *121concern raised in the present case because N.J.S.A. 2A:15-5.3 “permits a plaintiff to recover the full amount of the damage award from a defendant held to be at least sixty percent” at fault and permits that defendant the right to seek contribution from any party that paid less than his percentage share. Id. at 271, 827 A.2d 1104.
In Roman, the statutory provisions at issue dealt with how a jury determines liability between a plaintiff and defendant. In this case, the statutory provisions deal with how a jury apportions fault among joint tortfeasors. The distinction between the two is of sufficient significance that we find that Roman does not govern this case.
An ultimate outcome charge explaining how the Comparative Negligence Act operates between joint tortfeasors will not advance any of the legislative purposes of the Act. The Act calls for the jury to make a good-faith allocation of the percentages of negligence among joint tortfeasors based on the evidence—not based on the collectability or non-collectability of a judgment. We cannot untether the jury from the dictates of the statute and allow it to determine the percentage of the defendants’ fault based on its own intuitive notion of equity. The determination of each defendant’s percentage of negligence must represent each defendant’s degree of fault for causing the accident. We cannot conceive that the Legislature intended the jury to be provided with information that would permit it to manipulate that evidence-driven paradigm in exchange for an outcome-based one dependent on the jury’s own innate sense of fairness.
In Weiss v. Goldfarb, 154 N.J. 468, 479-81, 713 A.2d 427 (1998), a medical malpractice case, we expressed similar concerns regarding the capacity of an ultimate outcome charge to divert jurors from the evidence and to encourage them to consider the collectability of a judgment. We held that jurors should not be given an ultimate outcome charge informing them that a defendant-hospital’s liability is limited by the Charitable Immunity Act. Id. at 469, 713 A.2d 427. In that case, the jury was asked to determine *122the plaintiffs damages and the percentage of negligence attributable to multiple defendants—the hospital and several doctors. Id. at 471, 713 A.2d 427. The trial court rejected the plaintiffs request to inform the jury that the hospital’s liability was limited to no more than $10,000 under the Charitable Immunity Act. Ibid. The jury found the hospital 100 percent negligent and awarded $150,000 in damages. Ibid. The court reduced the award to $10,000 in accordance with the Act. Ibid. We concluded that an ultimate outcome instruction would not only be “irrelevant” but also “highly prejudicial,” because it might induce a jury “to shift to other defendants some percentage of negligence that the jury thought should rightfully be assessed against the hospital.” Id. at 481, 713 A.2d 427.
In this case, the trial court informed the jury that apportionment of sixty percent or more of the liability to defendants would obligate defendants to pay 100 percent of plaintiffs’ damages. The jury’s duty was to determine the amount of damages and each party’s percentage of negligence. N.J.S.A. 2A:15-5.2a. The court’s duty was to “mold the judgment” in accordance with the jury’s findings and the Comparative Negligence Act. N.J.S.A. 2A:15-5.2d. As in Weiss, the ultimate outcome instruction in this case was “irrelevant” to the jury’s function of apportioning percentages of fault and determining damages, and was “highly prejudicial” to defendants. See Weiss, supra, 154 N.J. at 481, 713 A.2d 427. The Comparative Negligence Act does not authorize a jury to consider how a plaintiff will collect a damage award. See N.J.S.A. 2A15-5.1 to -5.8.
The trial court’s ultimate outcome instruction in this ease was prejudicial to defendants because it may have led the jury to the forty-sixty allocation, shifting a percentage of fault from Horsman to defendants in order to assure plaintiffs a full recovery of their damages. For that reason, we affirm the Appellate Division and remand for a new trial on the allocation of fault between defendants and Horsman.
*123IV.
We have never addressed the question whether counsel may suggest in an opening or closing statement that the jury find a party responsible for a specific percentage of fault. In his opening statement to the jury, plaintiffs’ attorney stated, “I’m going to suggest to you that Mr. Horsman’s responsibility in this case is a fraction ... maybe a small fraction, 5 percent, 8 percent, maybe 10 percent.” The attorney’s objective was clear enough. So long as the jury found the negligence of the bankrupt Horsman to be forty percent or less and the negligence of defendants to be sixty percent or greater, plaintiffs could pursue a 100 percent recovery from defendants, otherwise the percentage of fault attributed to Horsman would be uncollectable. N.J.S.A. 2A:15-5.3. The trial court sustained defendants’ objection to the attorney’s statement quantifying the percentage of Horsman’s fault and gave a curative instruction to the jury. Brodsky, supra, 362 N.J.Super. at 277, 827 A.2d 1104. In affirming the trial court’s decision, the Appellate Division extended the sweep of this Court’s decision in Botta v. Brunner, 26 N.J. 82, 100, 103-05, 138 A.2d 713 (1958), which bars trial counsel from suggesting in an opening or closing statement a specific dollar amount as a measure of damages to compensate for an injured plaintiffs pain and suffering.4 The panel reasoned that counsel’s “assertion of specific percentages” as to fault was comparable to counsel’s quantification of unliquidated damages in Botta. Brodsky, supra, 362 N.J.Super. at 278, 827 *124A.2d 1104. The panel applied the Botta rationale and prohibited counsel from attributing “percentages of comparative fault among joint tortfeasors” on the basis that the “subject does not lend itself to precision.” Ibid.
We conclude that the jury’s determination of percentages of fault is different from its open-ended evaluation of damages for pain and suffering. We, therefore, disagree with the panel’s decision to extend the Botta rationale to limit counsel from suggesting to the jury a specific degree of fault to be attributed to a party.
In Botta, supra, the plaintiff was the passenger in a car that collided with another ear. 26 N.J. at 86, 138 A.2d 713. She sued both drivers, claiming that they operated their vehicles negligently and caused her injuries and monetary losses. Ibid. In his closing argument to the jury, the plaintiffs counsel suggested how the jury could quantify his client’s pain and suffering in terms of damages. Id. at 91-92, 138 A.2d 713. The attorney noted that the plaintiff had endured “125 weeks of pain and suffering,” and then asked the jurors, “how much do you think you should get for every day you had to go through that harrowing experience, or every hour? ... Would fifty cents an hour for that kind of suffering be too high?” Id. at 91-92, 138 A.2d 713. The trial court declared the argument to be improper and stopped counsel from proceeding any further on the subject. Id. at 92, 138 A.2d 713. The jury found the driver of the car in which the plaintiff was traveling negligent and awarded her a sum of money that she considered inadequate compensation for her injuries. Id. at 86, 138 A.2d 713. The Appellate Division ordered a new damages trial on the basis that the trial court improperly prohibited the 'plaintiffs counsel from suggesting a mathematical formula for computing damages for pain and suffering. Id. at 87, 91, 138 A.2d 713.
This Court parted with the Appellate Division and held that trial counsel may not in an opening or closing statement place a pecuniary value on the plaintiffs pain and suffering. Id. at 103, *125105, 138 A.2d 713. The Court observed that the “general standard” for measuring damages for pain and suffering has been “fair and reasonable compensation” because of the “universal acknowledgment” that its precise calculation is elusive. Id. at 92, 138 A.2d 713. It noted that “pain and suffering have no known dimensions, mathematical or financial” and that “[t]here is no exact correspondence between money and physical or mental injury or suffering.” Id. at 95,138 A.2d 713. The Court flatly stated,
[t]here is and there can be no fixed basis, table, standard, or mathematical rule which will serve as an accurate index and guide to the establishment of damage awards for personal injuries [because] ... there is no measure by which the amount of pain and suffering endured by a particular human can be calculated---The varieties and degrees of pain are almost infinite. Individuals differ greatly in susceptibility to pain and in capacity to withstand it.
[Id. at 92-93, 138 A.2d 713.]
Central to its holding was a belief that—based on the intangible quality of pain and suffering—an attorney’s discoursing on a specific monetary amount was nothing more than “sheer speculation” and possessed a serious capacity for misleading the jury by “instill[ing] in the minds of the jurors impressions, figures and amounts not founded or appearing in the evidence.” Id. at 99, 100, 138 A.2d 713. The Court concluded that to avoid “unwarranted intrusion into the domain of the jury,” trial counsel may not suggest a monetary amount or numerical standard for the evaluation of pain and suffering. Id. at 103, 138 A.2d 713.
We find that Botta does not control the outcome of this case because of the significant distinction between a jury attributing percentages of fault among parties so that the total equals 100 percent and a jury assessing the amount of damages for pain and suffering without any limitation other than the standard of reasonableness. First, the quantification of a specific percentage of a party’s negligence is not “intrinsically and intractably subjective,” as is calculating the nature of pain and suffering. See Friedman, supra, 108 N.J. at 77, 527 A.2d 871. Determining which parties were at fault and the degree of their negligence must be grounded in the evidence presented at trial. We are confident that jurors *126can resolve those difficult issues by evaluating the evidence in the light of logic and their collective experience and common sense. Trial counsel generally has broad latitude to comment on the evidence and issues at trial. See State v. Bogen, 13 N.J. 137, 140, 98 A.2d 295 (1953). Constrained by the evidence concerning the nature and degree of a party’s fault, counsel will be less likely to engage in the “emotion, fancy and speculation” the Court deemed anathema to fair jury determinations in Botta, supra, 26 N.J. at 93, 138 A.2d 713.
Second, the concern in Botta that trial counsels’ quantification of pain and suffering could be infinite and unrestrained does not arise when assigning a percentage of fault. The allocation of fault among parties, although susceptible to varied combinations, will never exceed 100 percent. Thus, any evaluation of the evidence with respect to a party’s comparative degree of fault will be on a fixed scale. We have little doubt that a jury will view an argument by counsel suggesting a party’s percentage of fault no differently than any other argument—when based on the evidence, to be given weight, and when not, to be disregarded. Unlike arguments addressing pecuniary calculations of pain and suffering, there is little danger that a jury will interpret references to specific percentages of fault as evidence not in the record. See Botta, supra, 26 N.J. at 98, 138 A.2d 713 (disapproving of “statements calling attention to claims and amounts not supported by the evidence” which may “take[ ] the place of evidence” in jurors minds) (emphasis omitted). Moreover, in the adversarial setting of a trial, the arguments of opposing counsel will offer a balance, and may give greater focus to the issues that must be decided by the jury.
The appellate panel’s decision in this case bars counsel from suggesting a specific degree of fault, but allows counsel to refer to a party’s fault as “minimal” or “substantial.” In our view, however, the terms “minimal” and “substantial” are simply proxies for the specification of degrees of fault. When a plaintiffs counsel argues that the defendants are equally at fault the jury under*127stands that if there are two defendants each is fifty percent at fault and if there are four defendants each is twenty-five percent at fault. Yet no one suggests such a presentation is beyond the bounds of fairness or will distract, much less deceive, a jury. Because the argument that a party is “not at fault” or that the parties are “equally liable” have clear numerical analogues, a party may convey the same quantitative information whether numerically or qualitatively framed. We see no reason for concluding that jurors will be more swayed by arguments of counsel suggesting that a party is at fault by a specific percentage, than they will be by arguments suggesting a party is “minimally” or “substantially” at fault.
We have great faith that our jurors have the capacity “to digest complex evidence” and render fair verdicts. We do not view them as “rustics,” unsophisticated in the world and unable to discern a false from a genuine argument. See DeHanes v. Rothman, 158 N.J. 90, 99, 103, 727 A.2d 8 (1999) (allowing expert to testify to aggregate amount of economic losses and trial counsel to sum up total of such losses for jury).
Defendants urge us to limit trial counsel’s use of numerical figures to issues that can be “analyzed with mathematical precision,” as in the case of economic damages. We reject this invitation to circumscribe the scope of argument by counsel in opening and closing statements. Everyday, in courtrooms across the state, counsel argue to juries that a case has been proven beyond a reasonable doubt, by clear and convincing evidence, or by a preponderance of the evidence. Juries are required to make fine distinctions, and we do not restrict counsel in their opening and closing arguments from assisting them in making those distinctions. We expect a jury to discern between an argument that comports with the evidence and one that does not. We do not expect that a lawyer will be able to persuade a jury that has listened to the testimony and reviewed all of the evidence that a defendant minimally responsible for an accident should be found ninety-five percent negligent. We do not believe that allowing *128counsel to suggest that a specific percentage of fault should be attributed to a party will impinge on the exclusive domain of the jury or that the jury will uncritically accept an argument that has no basis in the record.
This was a typical negligence case with ample testimony of the parties’ respective degrees of fault. The jury was in the best position to determine whether the evidence supported counsels’ arguments. Because the jury must determine the degree of fault of the parties, we see no reason why that subject should be off-limits to the argument of counsel. In conclusion, we hold that in a case arising under the Comparative Negligence Act, counsel may argue the degree of fault that should be ascribed to a party, provided there is some evidence in the record to support the argument.
V.
We affirm the judgment of the Appellate Division reversing the trial court’s determination concerning apportionment of damages and remand for a new trial consistent with this opinion. We leave untouched the jury’s award of damages. All that remains is to apportion those damages between defendants and Horsman.
For affirmance in part/reversal in part/remandment—Chief Justice PORITZ and Justices LaVECCHIA, ZAZZALI, ALBIN and WALLACE—5.
Opposed—None.