concurring.

Although I agree with the opinion in most respects and concur in the result, I disagree with its discussion of Daubert and Kumho Tire in Part IV.C.2. Specifically, the opinion seems to misunderstand how Daubert and Kumho would apply in this case. In my view, the superior court’s evidentiary rulings can easily be sustained as correct applications of Daubert and Kumho. The opinion’s categorical refusal to extend these cases to experience-based expert testimony is unnecessary, overbroad, and unsound.

In'the court’s view, Daubert and Kumho need to be rejected if we wish to sustain the superior court’s ruling because, “[ujnder federal law, the manner in which the trial court admitted the testimony of Stirling and Dr. Rubenstein was probably erroneous,”1 and its “admissibility would be threatened under Kumho Tire.”2 But the court fails to provide any support for this prediction or to engage in any case-specific discussion of what Dau-bert and' Kumho would actually have required. The court instead chooses to restrict Daubert and Kumho, finding Professor Saltzburg’s views on this issue “especially persuasive,” and approvingly quoting selected passages from arguments expressed by the professor and two colleagues in an ami-cus brief they submitted in Kumho Tire.3

But while the amicus brief in Kumho opened by broadly questioning the utility of extending Daubert to experience-based expertise,4 its authors concentrated their main arguments on the extreme view of Daubert advocated by the petitioners in Kumho. The amicus briefs authors viewed the petitioners as arguing that Daubert categorically extends its four-factor test to all cases involving expert witnesses, including all cases of experience-based expert testimony; that its four-factor test (including the requirement of objective verification) is mandatory and exclusive; and that experience-based expert testimony must be presumed inadmissible unless its proponent can prove compliance with the Daubert test in a full Daubert hearing. It is this extreme view of Daubert that Professor Saltzburg and his colleagues so adamantly challenged:

But Daubert should not be extended to require the exclusion of all opinions drawing on any aspect of scientific or technical knowledge that in some measure involves the use of subjective judgment based on an expert’s experience. Such an expansion of Daubert would be inconsistent with Rule *1015702’s explicit acknowledgment of “experience” as a basis for expertise....

Nor is petitioners’ expansive reading of Daubert grounded in reality. In our everyday lives, we often obtain assistance from persons who, because of their past experience, have specialized knowledge and are able to reach reliable conclusions based on that experience. When such a person appears in court as an expert, the pertinent question should be whether, in light of such past experience, his or her opinions can assist the trier of fact — -not whether they have been validated by the factors set forth in Daubert bearing on the validity of an application of the scientific method. As Daubert emphasized, the inquiry under Rule 702 must be flexible. No bright-line test is feasible, because the circumstances under which an expert’s opinion can be deemed reliable depend on the specific field of expertise in question, and on the particular issue in dispute.[5]

As the underscored wording shows, the amicus briefs basic position was not that Daubert should never apply to experience-based expert testimony, but that Daubert should not always control, and automatically exclude, such testimony.

In keeping with this position, the amicus brief devoted most of its discussion to two practical points. In the first section of discussion, it argued that Daubert should be applied flexibly through the exercise of ease-by-case discretion by trial courts, with the four Daubert factors being treated as permissive and non-exclusive.6 In the second part of the discussion, the brief addressed the need to clarify the procedures required by Daubert, emphasizing the decision should be construed to require a full “Daubert hearing” only if the court was confronted with strong prima facie evidence of fundamentally flawed methodology.7

In unanimously holding that Daubert applies to experience-based expert testimony, the United States Supreme Court’s opinion in Kumho carefully addressed the concerns raised in Professor Saltzburg’s amicus brief; though ruling against the result the petitioners favored, the Court favorably viewed most if not all of the specific points that the amici pressed.

Kumho repeatedly emphasized that the Daubert test is a flexible one and draws no bright-line requirements; the opinion stressed that Daubert vests trial courts with broad discretion to use or ignore its four-factor analysis, depending on whether the court finds it helpful in deciding the specific issue at hand.8 Similarly, the Court stressed that the four factors articulated in Daubert are neither mandatory nor exclusive, and can be supplemented or disregarded when they are not useful.9 In addition, Kumho made it clear that a full Daubert hearing is needed only in exceptional cases, when case-specific evidence raises a genuine issue as to basic reliability — that is, when the “[expert] testimony’s factual basis, data, principles, meth*1016ods, or their application are called sufficiently into question.”10

At its core, then, Kumho views Daubert as a flexible, fact-specific, and non-exclusive approach that invites, rather than restricts, trial court discretion:

The conclusion, in our view, is that we can neither rule out, nor rule in, for all cases and for all time the applicability of the factors mentioned in Daubert, nor can we now do so for subsets of cases categorized by category of expert or by kind of evidence. Too much depends upon the particular circumstances of the particular case atissue.[11]

This flexible and individualized reading of Daubert answers almost every practical concern raised in the Kumho amicus brief. Of course, as today’s opinion notes, Kumho did not end the debate over Daubert; some commentators and a handful of cases continue to criticize Daubert and Kumho.12 The opinion quotes conclusory passages from these critics but fails to examine their conclusions for accuracy and merit; nor does it consider the contrary views advanced by other authorities concluding that these criticisms simply misunderstand Daubert and Kumho.13

Given the flexible nature of Kumho’s holding, a sufficient answer to the Daubert questions raised here is that the superior court properly applied the broad discretion given to it by Kumho in declining to require objective verification of the experience-based testimony offered by Stirling and Dr. Rubenstein. Both witnesses had abundant experience in long-recognized, widely practiced, and thoroughly vetted disciplines. To the extent that they proposed to testify on matters beyond their particular expertise, the superior court appropriately limited the scope of their testimony. And although Marrón purported to challenge their basic methodologies, her supporting pleadings suggested only that their opinions at most might reflect incorrect applications of accepted principles in reliable fields of specialized knowledge. The superi- or court correctly recognized this kind of alleged inaccuracy as garden variety impeachment for cross-examination, so the -record suggests no abuse of the broad range of trial court discretion granted by Kumho. I would dispose of this case on these narrower grounds.14

Instead, today’s opinion goes out of its way to disclaim Kumho’s useful elaboration of Daubert. While professing to constrain its ruling “to the situation before us,” the opinion expansively describes “the situation” as encompassing all cases involving “the admission of expert testimony based on accrued wisdom and personal experience.”15 This category could easily cover all expertise except pure theoretical science. The breadth of this ruling is especially striking because the main source the opinion cites to support such a categorical restriction on Daubert•— the Kumho amicus brief — repeatedly warned — as does Kumho itself — against attempts to draw categorical lines in this difficult area of the law.

In my view, the opinion’s bright-line rejection of Daubert and Kumho is needless, and *1017it is sure to stir more trouble than it settles.16 I thus concur in the result reached by the opinion on the Daubert/Kumho issue but decline to join in its unnecessary rationale. I agree in all other aspects with the opinion.

OPINION

I. INTRODUCTION

Teva Marrón and Lyle Stromstad were involved in an automobile accident, for which Stromstad admitted fault. Marrón sued for damages, particularly compensation for medical treatments, that she claims arose from the accident. The only issue at trial was causation, and the trial consisted largely of expert testimony. The jury found for Strom-stad and the court awarded him costs and fees. On appeal Marrón claims the superior court made a variety of discovery, evidentia-ry, and procedural errors. We affirm the superior court’s decisions on all issues except its award of attorney’s fees to Stromstad, on which we remand.

II. FACTS AND PROCEEDINGS

A. Facts

Marrón was a passenger in a car that Stromstad rear-ended with his own vehicle at a traffic light in October 1998. By her own admission, Marrón was a “chronic pain patient” prior to this accident, and had herniated a disc in a previous automobile accident in 1991. She suffered additional pain following yet another automobile accident in 1994. An MRI performed on Marrón in March 1998 showed her thoracic region to be, in her words, a “train wreck.” However, with the help of chiropractic manipulation under anesthesia, Marrón described feeling better before the accident with Stromstad. Although that accident occurred at a very low speed,1 Marrón reported a severe headache and facial numbness at the scene of the accident. These symptoms worsened in the days following, to include vomiting, spotting in front of her eyes, and pain from the waist up.

Marrón, distrusting Alaskan doctors, flew to California to see Dr. John White, an orthopedic surgeon recommended by a friend. Dr. White performed an examination on Mar-rón, designed to test back pain and response to certain stimuli. Based at least in part on the results of that examination, Dr. White refused to operate on Marrón. Marrón then visited Dr. Sunny Uppal, another California doctor. Based upon the results of an MRI performed in Spring 1998, Dr. Uppal decided that Marrón likely had a herniated disc. Dr. Uppal decided to perform a discogram on Marrón, a diagnostic procedure admittedly “controversial” and “incredibly painful.” The discogram allegedly proved that surgery was an appropriate treatment for Marron’s pain. Marrón agreed to undergo surgery. She asserts that the surgery was successful, in that it left her feeling roughly as well as she did before the accident with Stromstad.

*997B. Proceedings

Marrón sued Stromstad for negligence and gross negligence, claiming that the accident exacerbated her previous back injuries, causing her C 4-5 disc to become herniated. She claimed damages for lost earnings, and past and future pain and suffering, loss of enjoyment of life, and medical care. Stromstad admitted he was at fault for the accident, but argued that the accident was not the cause of Marron’s injuries or surgery.

Marrón filed several motions in limine to: (1) compel Dr. Richard Rubenstein and James Stirling, two of Stromstad’s expert witnesses, to produce “all 1099 income tax forms for all insurance companies ... or other entities from whom [they] obtained monies or fees [for] performing forensic expert services for the past five (5) years,” in order to show the bias of each witness; (2) strike Dr. White’s expert deposition testimony, because Stromstad failed to timely produce an expert report; (3) exclude Dr. Ru-benstein’s testimony, because he was not a qualified expert, because his opinions were speculative, and because his opinions did not satisfy the requirements of Daubert v. Mer-rell Dow Pharmaceuticals2 and as adopted by State v. Coon;3 (4) exclude the testimony of Stirling, offered as an expert accident reeonstructionist, because his credentials were insufficient, his investigation was too limited, and his opinions failed the Daubert test; and (5) exclude various pieces of evidence, including post-accident photographs of the vehicles involved. Superior Court Judge William F. Morse denied all five of the motions. Marrón appeals these decisions.

A four-day jury trial was held in September 2002. Before and during the trial, Mar-rón opposed the introduction of post-accident photographs of Marron’s car and an insurance company appraisal of the accident damage. Marrón argued that this evidence portrayed the accident “as a mere fender-bender,” and was thus “irrelevant and prejudicial.” The superior court admitted the evidence. Judge Morse denied Marron’s request to introduce evidence of Stromstad’s insurance coverage to rebut a statement by defense counsel that Stromstad would be personally liable for any judgment. Additionally, Marrón sought to introduce several of Dr. Rubenstein’s medical examination reports. According to Marrón, this eyidence would “show Rubenstein’s bias against plaintiffs and towards the insurance defense industry.” The court refused to allow Mar-rón to introduce the reports themselves into evidence, though it did allow Marrón to impeach Dr. Rubenstein on cross-examination with some of the contents of the reports. Marrón appeals all of these decisions.

Following trial, the jury unanimously concluded that Stromstad’s actions were not a legal cause of injury to Marrón. Marrón moved for a new trial on the grounds that: (1) Stromstad untruthfully implied that Stromstad was not insured and would be personally responsible for any judgment; (2) the superior court failed to exclude Dir. Ru-benstein’s testimony as to lack of causation; and (3) Stromstad’s attorney violated a court order precluding Dr. Rubenstein from testifying as to how much force is required to herniate a disc when he stated in closing that “Dr. Rubenstein testified that it would take more than 1-3 mph to herniate a disc.” Marrón appeals the superior court’s denial of this motion.

Stromstad moved for attorney’s fees following trial, and Marrón opposed the motion. Stromstad had made an offer of judgment pursuant to Civil Rule 684 less than sixty days after both parties had made their respective initial disclosures under Civil Rule 26, which Marrón rejected. Based on the timing of Stromstad’s offer, and the fact that it exceeded Marron’s eventual jury verdict (of $0), the superior court held that Civil Rule 68(b) entitled Stromstad to receive seventy-five percent of his post-offer attorney’s *998fees from Marrón. The court awarded fees incurred in opposing Marron’s motion for a new trial. Stromstad did not itemize the total amount of fees he claimed to have incurred. Marrón protested the failure to include a detailed listing of services provided, claiming that it prohibited her from evaluating the reasonableness of the fees sought. The court stated that, without itemization, it could not discern whether the fees were “either actually or reasonably incurred,” and held the requested amount to be “much higher than one would expect in a minor accident.” The court therefore decided to reduce the base amount of the fees by one-third, before awarding seventy-five percent of that reduced amount to Stromstad.5 Mar-rón appeals the fee award.

III. STANDARD OF REVIEW

We generally review a trial court’s discovery rulings for abuse of discretion. We will find an abuse of discretion only when left with a definite and firm conviction after reviewing the whole record that the discovery ruling was erroneous.6 However, we review de novo whether a trial court weighed the appropriate factors in issuing a discovery order.7

We review a trial court’s decision to admit or exclude evidence for abuse of discretion, and will reverse such a decision only if the error affected the substantial rights of a party.8 Similarly, we generally review a trial court’s decision to admit expert'testimony for abuse of discretion. But when the admissibility of evidence or expert testimony turns on a question of law, we apply our independent judgment.9

The decision to grant or deny a new trial is also within the trial court’s sound discretion.10 We will “disturb the trial court’s exercise of discretion only ‘in the most exceptional circumstances to prevent a miscarriage of justice,’ ” and will find an abuse of discretion only “ “when evidence to support the verdict was completely lacking or was so slight and unconvincing as to make the verdict plainly unreasonable and unjust.’ ”11 In reviewing an order denying a new trial, we view the evidence in the light most favorable to the non-moving party.12 However, whether the trial court applied the correct legal standard in granting or denying a new trial is a question of law that we review de novo.13

We review for abuse of discretion a trial court’s fact-based decisions as to whether attorney’s fees are reasonable and should be awarded.14 An award of attorney’s fees constitutes an abuse of discretion only when it is manifestly unreasonable.15 However, we review de novo whether the trial court applied the law correctly in awarding attorney’s fees.16

IV. DISCUSSION

A. The Superior Court Did Not Abuse Its Discretion in Denying Marron’s Motion To Compel Production of Dr. Rubenstein’s and James Stirling’s Income Tax Returns.

The superior court denied Marron’s pre-trial motion to compel two of Stromstad’s *999expert witnesses, James Stirling and Dr. Ru-benstein, to disclose their tax returns. Both witnesses had refused Marron’s request to produce the returns at their depositions. Although both originally claimed to have testified or worked as “litigation consultants” roughly as many times for civil plaintiffs as for civil defendants, Stromstad concedes that their deposition testimony revealed that in fact both witnesses had worked predominantly as defense experts.

Marrón sought to compel discovery of their income tax returns to prove what she -believed to be a strong financial motivation on the part of both Stirling and Rubenstein to provide testimony that favored the defense.

The superior court denied Marron’s motion. The court held that revealing opposing witness bias was an important interest, and one that Marrón was entitled to reveal through cross-examination at trial. However, the court stated that “[p]laintiff s right to discovery must be balanced with the expert’s right of privacy.” The court held that the experts would “retain a right of privacy in their tax returns, but may be questioned about the type of information that may be contained in or referenced from the returns.” Marrón claims this ruling constitutes an abuse of discretion.

There is no absolute right to privacy from discovery orders to produce tax returns.17 Generally, a litigant may discover an opponent’s tax returns for the sake of determining a party’s damages.18 But we have never specifically determined whether a witness’s tax records are discoverable for the purpose of showing that witness’s bias, or impeaching that witness’s credibility at trial.

Alaska Civil Rules 26(b)(2)(i) and (iii) allow a court to limit discovery where the information sought is’ obtainable from “some other source that is more convenient, less burdensome, or less expensive,” or if “the burden or expense of the proposed discovery outweighs its likely benefit.” We interpret this rule in light of Marron’s purpose in seeking the tax returns: to prove the experts’ alleged pro-defense bias. Under cross-examination by Marrón, both Dr. Rubenstein and Stirling admitted that they offer their litigation services primarily to defendants. Dr. Ruben-stein also discussed the amount Stromstad compensated him for testifying. Because Marrón elicited the information that she sought — that the experts worked primarily for defendants — the superior court did not abuse its discretion in not allowing Marrón to discover the witnesses’ tax records.19

*1000Because the superior court acted well within its discretion, we uphold its refusal to compel discovery of the experts’ tax forms.

B. The Superior Court Did Not Abuse Its Discretion in Refusing To Strike Dr. White’s Testimony.

Marrón appeals the superior court’s refusal to strike Dr. White’s testimony that disco-gram procedures such as the one performed by Dr. Uppal are unreliable and that Mar-ron’s reaction to the “Waddell maneuver” indicated her unfitness for surgery.

1. The superior court was not required to consider Dr. White, as Marron’s treating physician, to be an expert witness, and was thus within its discretion to admit his testimony without a Rule 26(a)(2)(B) expert report.

Marrón argues that the superior court abused its discretion in admitting the testimony of Dr. White, Marron’s treating physician, pertaining to discograms, because he did not provide an expert witness report and because Marrón was unfairly surprised by the testimony. The issue turns on the proper interpretation of Alaska Civil Rule 26(a)(2)(B), which provides that “[ejxcept as otherwise stipulated or directed by the court,” a party must provide a written report of each of its expert witnesses containing “a complete statement of all opinions to be expressed and the basis and reasons therefor; the data or other information considered by the witness in forming the opinions ... [and] the qualifications of the witness.”

Stromstad does not dispute that expert testimony is inadmissible without a preliminary report. Rather, Stromstad explains that the superior court did not require the disclosure of Dr. White’s opinions prior to his deposition “because Dr. White was Marron’s treating doctor, not a retained defense expert.” Stromstad cites Fletcher v. South Peninsula Hospital20 and Miller ex rel. Miller v. Phillips21 for the proposition that treating physicians need not be listed as experts, and can testify without meeting the disclosure requirements of Rule 26(a)(2)(B). We agree with Stromstad.22

Marrón asserts that her principal argument is based on fairness: “A treating physician is allowed to testify regarding anything previously produced in his medical records because there is no surprise to the opposing party.” Marrón states that she was “completely surprised” by Dr. White’s discogram testimony at deposition, even if he was her “treating physician.” Undoubtedly, an overall aim of evidentiary, discovery, and procedural rules is to prevent unfair surprise to litigants. The critical question is thus whether Marrón was unfairly surprised.

The superior court acknowledged that Dr. White’s medical report (prepared prior to the deposition) contained no information on dis-cograms or why Dr. White elected not to perform this procedure on Marrón to deter*1001mine her fitness for surgery. However, Marrón next consulted Dr. Uppal, who did utilize a discogram in deciding to operate on Marrón. Accordingly, the court felt it was “foreseeable that Stromstad would explore why Uppal and not White found Marrón suitable for surgery, ... [and] that Strom-stad would ask White about the use of a discogram or why he had not used the technique used by Uppal.” Considering that Marrón “knew exactly what techniques Dr. White used and did not use,” the court concluded that “Marrón should not have been surprised that White discussed discograms at his deposition.” Furthermore, the court reasoned that if Dr. White was not allowed to explain at trial why he did not use a disco-gram, then Dr. Uppal would be able to argue that his use of the discogram explained the difference in the two opinions regarding Marron’s need for surgery. This, the court concluded, “would give the jury an inaccurate picture of what happened.” A trial court has wide discretion in managing the discovery process23 and admitting expert testimony.24 In light of the above considerations, we find that the superior court did not abuse its discretion on this point.

2. Because Dr. White was Marron’s treating physician, the superior court did not err in admitting his testimony without subjecting it to a Daubert analysis.

Marrón additionally asserts that Dr. White’s testimony should have been excluded because his decision to disqualify Marrón from surgery was based on a mistaken application of the Waddell test, one inadmissible under the Daubert standard for admitting or excluding expert testimony. The superior court disagreed, declaring that “White’s use of the Waddell factors in making his evaluation is not the type of expert testimony that must be subjected to Daubert,” because Dr. White “used those factors in making his decision not to perform surgery.” Stromstad agrees, explaining that Dr. White’s testimony was based on his application of the Waddell test in treatment, and was not the testimony of a retained expert.

As discussed below in Part IY.C, Daubeyt v. Merrell Dow Pharmaceuticals, Inc.25 governs the admissibility of scientific expert testimony in federal courts. We have adopted Daubert,26 but have never explicitly determined whether a treating physician’s testimony must meet the Daubert criteria.

Some federal courts have applied Daubert to exclude the testimony of treating physicians. However, they have done so in cases where the issue was whether the physician should be allowed to testify as to causation— that is, to state a scientific conclusion.27 As Stromstad notes, Dr. White applied the Wad-dell test to Marrón in the course of determining treatment, not in preparation for litigation, and not to determine the causation of her alleged injuries. At least one federal decision has explained that “testimony about the type of treatment performed on [a patient] is factual in nature and is not subject to exclusion under a Daubert analysis.”28 *1002As such, this testimony should be rebutted by cross-examination of the treating physician, not preemptively excluded.29 This notion finds support in our decision in Miller, which upheld a superior court ruling that the treating physician “would be forbidden from testifying in general terms about the appropriate standard of care, [but] he would be allowed to testify as to his own opinion as to what he observed.”30

The superior court followed this logic in denying Marron’s motion to strike Dr. White’s testimony. The court stated that even if Dr. White’s use of the Waddell factors was “unsupported by scientific theory or empirical research, White’s use of those factors is a historical fact that partially provides the basis for the decision not to perform surgery.” We now hold that when a treating physician testifies regarding a course of treatment, the physician’s testimony need not be subjected to a Daubert analysis. We therefore hold that the- superior court correctly refused to apply Daubert to exclude Dr. White’s testimony.

C. The Superior Court Was Not Required To Exclude the Testimony of James Stirling or Dr. Rubenstein.

Stromstad sought to introduce testimony from James Stirling as an “accident reconstruction expert,” to rebut Marron’s testimony suggesting the accident involved a forceful impact on her car, and the likely accompanying inference that the accident caused her injuries. Stromstad also sought to introduce expert testimony from Dr. Rubenstein that (1) Marrón was not a proper candidate for surgery; (2) there was no objective basis for Marron’s pain complaints; (3) her complaints and treatment were caused by other factors like “drug seeking behavior, secondary gain and/or some sort of histrionic psychological makeup,” and (4) a rear-end accident at less than five miles per hour could not have caused Marron’s injury. Marrón moved in limine to exclude both experts’ testimony. The superior court allowed Dr. Rubenstein to testify on the above topics, although it prohibited him from testifying about some of his specific sources or conclusions that the court determined were “too speculative.” The superior court allowed Stirling to testify as a general accident reconstructionist, which allowed him to testify “about the speed each car was going at impact, the difference in the relative speeds of the cars, the damage done to the car in which Marrón was riding ... and the effect of impact on the forward movement of [Marron’s] vehicle.”

On appeal, Marrón asserts that the superi- or court erred in admitting Stirling’s testimony, because he was not a qualified expert, his investigation of the accident and basis of his opinions were insufficient, his testimony does not satisfy Daubert, his testimony did not assist the trier of fact, and the probative value of his testimony was outweighed by its prejudicial effects. Similarly, Marrón claims that the superior court should have excluded Dr. Rubenstein’s testimony because he was not qualified to testify as an expert in this action, his opinions were speculative and did not meet the standards of medical certainty, and his opinions did not satisfy the Daubert standard adopted by Alaska. None of Mar-ron’s claims has merit.

1. Stirling and Dr. Rubenstein were properly qualified as experts.

The superior court did not abuse its discretion in qualifying Stirling and Dr. Rubenstein as expert witnesses. Trial judges have wide discretion to determine whether to qualify witnesses as experts.31 As explained earlier, our “liberal admissibility standard” for expert testimony32 allows any person with specialized knowledge to serve as an expert witness, so long as that knowledge is relevant, in that it can help the trier of fact understand evidence or determine facts in issue. No specific training or formal *1003education is required.33

Marrón argues that Stirling was unqualified to be an expert witness because he “does not have a Bachelor’s degree. He has no degree of any kind in engineering. He is not a biomechanic.” Stirling’s formal training is limited to courses in accident reconstruction, but the record indicates that this coursework was both extensive and highly specialized. Stirling is also a member of several professional societies, is certified by this state as a police instructor in accident reconstruction, and has been working in his field since the late 1970’s. By Stirling’s estimation, he has assisted in or investigated over 4,500 accidents. The superior court was thus within its discretion in holding that whatever the limitations of Stirling’s expertise in determining the force of the accident, “it is a more informed evaluation than could be made by a jury looking at the same evidence without the assistance of an expert.” This is fundamentally all that Alaska Rule of Evidence 702 requires.

Marrón argues that as a neurologist, Dr. Rubenstein should not have been qualified as an expert witness, because he is not a surgeon, biomechanics specialist, psychiatrist or psychologist, yet his testimony touched on those areas of expertise. But we have “specifically rejected a reading of Evidence Rule 702 that would require expertise ‘in precisely the area upon which the expert proposes to comment.’ ”34 Despite his limited knowledge of surgery, psychology, or biomechan-ics, as a neurologist Dr. Rubenstein testified that he was an “expert in eliciting objective findings via the neurological examination,” and an “expert in recommending alternative forms of treatment.” Dr. Rubenstein also performed a five-hour exam on Marrón, explained the procedures performed by Dr. Uppal, and correlated Marron’s MRI results with his own examination. Dr. Rubenstein’s testimony was thus certainly helpful to the jury in making a more informed evaluation of the evidence.

2. Daubert did not apply to Stirling or Dr. Rubenstein.

As noted earlier,35 we adopted Daubert in State v. Coon.36 Daubert requires the trial judge to make a preliminary determination that “ ‘the reasoning or methodology underlying [expert] testimony is scientifically valid and ... properly can be applied to the facts in issue.’ ”37 In other words, at its most basic level Daubert contains two essential requirements for the admission of scientific expert testimony: it must be reliable and it must be relevant.38 In reviewing trial court Daubert decisions, we have adopted the abuse of discretion standard used by federal courts, in light of Daubert’s goal of allowing trial courts “greater flexibility in determining the admissibility of expert testimony.”39

In this case, the superior court expressed serious misgivings about the reliability of Stirling’s specific methods and con-*1004elusions. The court noted that “there is no reason to believe that Stirling has done any-testing to determine what damage, to a specific model of car, a collision of a particular configuration and speed will produce,” and the court saw “little reason to believe that Stirling can discern [the] quantity of damage from photos and repair estimates and translate it into speed differential.” Nonetheless, the superior court decided to allow Stirling to testify because “[t]here is an enormous body of experimentation concerning the damage done to vehicles in an accident,” and “[h]aving seen thousands of accidents, one assumes that Stirling has developed a pretty good sense of the amount of damage that occurs in accidents at different speeds.”

Similarly, the superior court did not conduct a Daubert-style analysis of Dr. Ruben-stein’s specific methodology in this case. Based only on his credentials as a neurologist, Dr. Rubenstein was allowed to testify as to the basis for Marron’s pain (whether psychological, “drug-seeking,” or the result of the accident with Stromstad) and whether Marrón was a proper candidate for surgery.40 Thus, the superior court admitted the testimony of Stirling and Dr. Rubenstein based on the reliability of their expertise in general, rather than its application in this particular case. We must determine whether this was proper.

Under federal law, the manner in which the trial court admitted the testimony of Stirling and Dr. Rubenstein was probably erroneous. Daubert by itself does not apply to the testimony of Stirling and Dr. Ruben-stein — -its holding was limited to testimony based strictly on “scientific knowledge,” that is, knowledge that has been “derived by the scientific method.”41 But the Supreme Court later extended Daubert to cover all “ ‘technical’ or ‘other specialized’ knowledge” in Kumho Tire Co. v. Carmichael.42 In other words, all expert testimony in federal courts must now meet the Daubert requirements of reliability and relevance. Thus, under Kumho Tire and Daubert, the testimony of Stirling and Dr. Rubenstein could not have been presented to the jury until the trial judge first evaluated the reliability of the evidence by, for example, determining whether Daubert’s four factors in assessing scientific testimony were applicable or dis-positive to the present testimony.43 But we have never adopted Kumho Tire’s extension of Daubert to all expert testimony,44 and we now explicitly decline to do so. Instead, we limit our application of Daubert to expert testimony based on scientific theory, as opposed to testimony based upon the expert’s personal experience.

Alaska’s rules of evidence are similar to, and were modeled after the Federal Rules of Evidence.45 This gives the evidentiary decisions of federal courts, particularly the United States Supreme Court, considerable persuasive weight. However, like other states, we are not bound by the Federal Rules of Evidence, and federal decisions interpreting federal rules do not govern state court decisions based on state rules.46 *1005We noted in Coon that in adopting Daubert “we are not bound by the Supreme Court’s conelusion[s],”47 but “its analysis of the corresponding federal rules is helpful.”48 And as the Indiana Supreme Court has stated, “Rule 702 of the Federal Rules of Evidence is not a minimum constitutional requirement applicable to the states.”49 In short, we are not bound by Kumho Tire, even if we do apply the Daubert standard to our own rules of evidence.

Kumho Tire’s expansion of Daubert to cover all expert testimony has been widely criticized.50 Daubert was formulated with an eye toward the “permissive backdrop” of the Federal Rules of Evidence, and was intended to overturn the overly “austere” and “rigid” “general acceptance” test of expert testimony then favored by many federal courts.51 The Court noted that such a test was “at odds with the ‘liberal thrust’ of the Federal Rules and their ‘general approach of relaxing the traditional barriers’ ” to the admission of expert testimony.52 We too have a liberal standard favoring the admissibility of expert testimony,53 and we cited the Court’s language approvingly in Coon.54 We have adopted a liberal standard for admitting evidence to increase the information available to the fact-finder, whose role in the adversarial process is crucial. Yet the concern that Daubert would “usurp[ ] the jury’s duty to determine the credibility of expert testimony,”55 only increased after Kumho Tire. Critics argued that the new federal standard “ ‘presuppos[es] that the traditional adversary process is insufficient to enable opposing counsel and their experts to ferret out inaccuracies and bias in expert testimony.’ ”56 Moreover, as noted by Professor Saltzburg, these “reforms in the law of evi*1006dence that were intended to liberalize the admission of expert testimony have now been interpreted to constrict the flow of expert opinion in trials.”57 Daubert’s requirements can also be easily exploited by litigants, leading to mini-trials, a prolonged discovery process, and prohibitive costs to both parties and the court.58 Expanding Daubert’s scope to include all expert testimony seriously exacerbates these problems. Several states have agreed, and have declined to adopt the expansion of Daubert that Kumho Tire accomplished.59

In maintaining Daubert’s distinction between “scientific” and “other technical or specialized knowledge,” we look to the definitions provided by Daubert itself. As the Daubert Court explained, “ ‘scientific’ implies a grounding in the methods and procedures of science,” and “ ‘knowledge’ connotes more than subjective belief or unsupported speculation.”60 Science itself “represents a process for proposing and refining theoretical explanations about the world that are subject to further testing and refinement,” and “scientific knowledge” refers to inferences or assertions derived from the scientific method.61 In short, scientific testimony is based on theory, and may be subjected to objective testing.

It is true, as the Supreme Court suggested, that there is often “no clear’ line” dividing scientific from other technical or specialized knowledge, and technical but non-scientific evidence may often be vetted in the same fashion as strictly scientific evidence. But much if not most expert testimony is based on “specialized knowledge” derived only from experts’ personal experience and intuition.62 This evidence is not empirically verifiable or objectively testable. The Daubert factor-driven test “is useless as a criterion for the admissibility of other types of expert testimony.”63 As Professor Saltzburg and his colleagues noted, “[mjany fields of expert testimony ... simply lack the intellectual rigor of Newtonian experimental science.”64 Instead of “using methods that enjoy a high level of objective verifiability, they depend on a more subjective application of the expert’s practical experience to the particular facts of the case.”65 But “[njothing in the language or legislative history of the [federal rules] justifies any presumption against admitting expert testimony”66 in such cases. Rather, “the text and legislative history of the Rules affirmatively support the admission of such evidence.”67 Expanding Daubert to cover testimony of this sort is “inconsistent with Rule 702’s explicit acknowledgment of ‘experience’ as a basis for expertise,” contradicts the prior application of the evidence rules and the common law, and “would endanger a vast array of expert testimony that has been *1007accepted ... for decades.”68 We agree with Professor Saltzburg and his colleagues that experience-based expert testimony was traditionally favored over theory-based expert testimony, and that “[a]dopting the idea that ‘subjective’ expert testimony based on specialized knowledge derived from experience ... should be distrusted, and targeted for exclusion ... would turn evidence law on its head.”69

We also agree with Professor Saltz-burg and his colleagues that experts can often help jurors “on the basis of their experience” by virtue of their “specialized knowledge that educates them about relevant factors to consider in determining an issue.”70 In such cases, “the threshold for admissibility need not be as high as when a sophisticated scientific question beyond the jury’s everyday world experience and ordinary mode of reasoning is at issue.”71 The testimony of Stirling and Dr. Rubenstein is clearly within the “jury’s everyday world experience and ordinary mode of reasoning.” It is also the sort of experience-based testimony the admission of which is encouraged by our rules of evidence and traditional evidence standards, yet whose admissibility would be threatened under Kumho Tire. We hold today that the admissibility of experience-based testimony should be governed by these traditional standards and rules of evidence, as further discussed below. Where the expert testimony is plainly derived from experience — not from the scientific method — and is not dependent on sophisticated scientific theory, Daubert does not apply. The admission of such testimony is not dependent on satisfying the Daubert requirements. Kumho Tire is inconsistent with this standard.

A trial court need not apply the Daubert standard to all expert testimony in order to ensure that it is relevant and reliable. As the Supreme Court itself noted in Daubert, there are numerous other rules of evidence that serve to ensure'the reliability of expert testimony.72 For example, both Federal and Alaska Rule of Evidence 702 require that all experts be properly qualified “by knowledge, skill, experience, training, or education.”73 In addition, the expert’s knowledge must “assist the trier of fact to understand the evidence or to determine a fact in issue.”74 Under Rule 703, the “facts or data in the particular case upon which an expert bases an opinion or inference” must be “of a type reasonably relied upon by experts in the particular field.”75 Under both Federal and Alaska Rule.of Evidence 706, a court may appoint its own independent expert witness to advise the court.76 Under both Federal and Alaska Rule of Evidence 403, the judge may exclude evidence whose probative value is outweighed by “the danger of unfair prejudice, confusion of the issues, or misleading the jury.”77 In extreme cases, where one side has proffered only an insufficient “scintilla” of expert testimony or other evidence, the court is “free to direct a judgment” or to “likewise grant summary judgment.”78 Perhaps most importantly, the Supreme Court noted that “[vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof’ — in short, the basic pillars of the adversary system — -“are the traditional and appropriate means of attacking shaky but admissible evidence.” 79

*1008This decision should not be read as a wholesale rejection of Kumho Tire. We offer no opinion on Kumho Tire’s clarification of the proper procedure and standards for performing a Dauberb analysis of scientific testimony. Our decision not to adopt Kumho Tire and to limit Dauberb is limited to the situation before us: the admission of expert testimony based on accrued wisdom and personal experience. Accordingly, we hold that the superior court did not err in admitting the expert testimony of Stirling and Dr. Ru-benstein without a Dauberb analysis.

3. Dr. Rubenstein’s opinions were not overly speculative and they met the standards of medical certainty.

Marrón also claims that the superior court should have granted her motion in limine to exclude Dr. Rubenstein’s testimony because “his opinions are speculative and do not meet the standards of medical certainty.” Marrón recites several specific facts related to her medical history and treatment of which Dr. Rubenstein was apparently unaware at the time of his deposition: whether Marrón had suffered pain prior to the accident with Stromstad, “when the last time prior to his examination of her she had taken any prescription pain medication,” and the half-life or length of effectiveness of her prescription drugs. We reject the argument that Dr. Rubenstein’s testimony should have been precluded because it was speculative.

In Maddocks v. Bennett,80 we stated that a medical expert’s opinion must be within a “reasonable medical certainty” or the equivalent “reasonable medical probability” to be admissible.81 Maddocks indicates that “reasonable probability” is determined by the expert’s own confidence in the expert’s testimony.82 Dr. Rubenstein himself stated that he believed his conclusions to be within “a reasonable degree of medical certainty” based on Marron’s medical records and MRI results.

Although, as Marrón notes, Dr. Ruben-stein admitted ignorance as to several details of Marron’s medical history or treatment, nothing in the language of Maddocks indicates that a medical expert must know every factual detail of a case before he can testify to a reasonable probability. Even where a medical expert is uncertain as to a specific fact, his overall conclusion is not necessarily speculative.83 Dr. Rubenstein testified at trial that traumatic disc herniation is “extraordinarily unusual” without fracture of the bony vertebral bodies, and that MRI films indicated that there was “certainly no evidence” of disc herniation or other damage to Marron’s “vertebral bodies.” Dr. Ruben-stein’s language and the evidence upon which it was based easily satisfied the Maddocks standard of medical certainty.

4. The superior court did not abuse its discretion in admitting Stirling’s testimony.

Marrón also claims that Stirling’s “investigation” of the accident, and thus the basis for his opinions, was insufficient; that Stirling’s testimony “[did] not assist the trier of fact”; and that the probative value of Stirling’s testimony was outweighed by its prejudicial effects. All of these remaining claims essentially call for this court to find that the superior court abused its sound discretion in admitting evidence. We decline to do so.

Accident reconstructionists, and James Stirling in particular, have previously been allowed to testify in similar cases.84 That the testimony of such experts assists the trier of *1009fact is self-evident. Although a jury might have found that Stirling’s investigation, or the basis of his opinions, was insufficient, it was Marron’s responsibility to expose these weaknesses through cross-examination and the presentation of countervailing expert testimony. Accordingly, we find that the superior court did not abuse its discretion in admitting Stirling’s testimony.

D. The Superior Court Did Not Abuse Its Discretion in Admitting Photographic and Insurance Appraisal Evidence of the Damage to Mar-ron’s Vehicle.

Marrón claims that the superior court should not have allowed Stromstad to introduce photographs and an insurance appraisal indicating the damage (or lack of damage) to Marron’s car. Marrón asserts that any probative value of this evidence was outweighed by the risk of prejudice. Marrón suggests that this evidence would have led the jury to believe the accident was so minor that it could not have caused her injuries. Stromstad responds that this evidence, in tandem with Stirling’s accident-reconstruction testimony, shows the low force of the impact.

Marron relies principally on Davis v. Maute,85 whose essential facts are quite similar to this case: The defendant’s conceded negligence led to an alleged minor impact in which the plaintiff allegedly sustained injury. The only issue was the extent to which the accident caused the plaintiffs injuries. The defendant had introduced photographs showing “light damage” to the plaintiffs car.86 The Supreme Court of Delaware held that without the support of competent expert testimony, a party generally may not claim a “correlation between the extent of the damage to automobiles in an accident and the extent of the occupants’ personal injuries,” even by implication.87 Since the defendant had not introduced expert testimony, the court reversed the verdict and remanded for a new trial.88 Stromstad asserts that Davis is inapplicable to this ease, because he introduced two expert witnesses. We agree. Moreover, we decline to adopt the rigid approach represented by that case.

We are unaware of any other jurisdiction which has adopted a rule that collision evidence is per se inadmissible without expert testimony,89 and we decline to do so. The trial court properly has the discretion to weigh the prejudicial and probative value of photographs and other evidence of the severity of an accident. Evidence showing Mar-ron’s vehicle was undamaged can be probative of the force with which the accident occurred, and the likelihood that it caused serious harm to Marrón. Thus, the court did not abuse its discretion in admitting this evidence.

E. The Superior Court Did Not Abuse Its Discretion in Limiting Cross-Examination of Dr. Rubenstein and Excluding Reports of His Previous Examinees.

Marrón next asserts that, even assuming that Dr. Rubenstein’s testimony was properly admitted, the superior court abused its discretion in limiting her cross-examination of Dr. Rubenstein on medical examination reports that Dr. Rubenstein had prepared in earlier cases. Marrón also asserts that the superior court abused its discretion by refusing to admit those same reports into evidence. Marrón offered these reports “to show Dr. Rubenstein’s bias against plaintiffs and towards the insurance defense industry.” According to Marrón, the striking similarity between Dr. Rubenstein’s reports reveals “Dr. Rubenstein’s skill' in invalidating 'claimants,” and “tends to show'that Dr. Ruben-stein is hired because the insurance defense *1010industry already knows what his report will say.”

However, at trial Marrón focused primarily on a different basis for admitting the reports. She asserted that the reports would “impeach Dr. Rubenstein regarding the thoroughness of his examination of Teva Mar-ron.” Marrón explained that the reports show that Dr. Rubenstein performed much more comprehensive examinations on the subjects of the reports, including a set of specific tests administered only on the three prior plaintiffs. Accordingly, the superior court gave Marrón “a fair amount of leeway to impeach [Dr. Rubenstein] on what tests he used and what tests he didn’t use ... to somehow convince the jury that his evaluation is less credible.”

Marrón briefly presented a second theory for admissibility of Dr. Rubenstein’s reports of prior examinations. She argued that because Dr. Rubenstein offered the same opinion for the three prior plaintiffs, despite the possibility that they had “different constellations [of symptoms],” the reports were probative of Dr. Rubenstein’s bias:

MS. POWELL: Okay. And the conclusions on each of the [defense medical exams conducted by Dr. Rubenstein] were identical.

THE COURT: That I don’t see any reason for you to bring in. Because now we’re going to have to, I mean, who knows why are[,] or the other three people[,] I mean, those are three different patients with different constellations of problems.

MS. POWELL: With the same result.

THE COURT: Maybe — the result maybe the same.

MS. POWELL: Uh-hum.

THE COURT: But that doesn’t necessarily mean that the individual constellation of symptoms are the same.

MS. POWELL: Yeah, that — that’s why I believe it’s probative, because they are three different people, three different constellations and he has the same opinion about all of them. Actually four including [Marrón].

THE COURT: I’m not going to let you go into that. You can ask about the specific tests that he used or didn’t use or specific circumstances where he says he might not or would not use a test, you can impeach him by showing that he used them in those circumstances in the past, but you can’t bring in the three other case histories with the similar conclusion.

MS. POWELL: Okay.

THE COURT: Anything further?

MR. BURKE: No, Your Honor.

MS. POWELL: No.

This was the extent of Marron’s presentation on the relevance of the earlier reports. Marrón points out on appeal that Dr. Ruben-stein used identical language in his conclusions and recommendations in these reports and in Marron’s and that he simply “parrot[ed]” his conclusions from report to report. But this argument was not made to the trial court, and Marrón did not narrow her request by pointing to specific language from the reports that she sought to introduce.

The superior court did not explain its reasons for limiting Marron’s examination of Dr. Rubenstein. Stromstad suggests quite reasonably that the court’s decision was based on Alaska Rule of Evidence 403. That rule allows a court to exclude evidence, even if relevant, “if its probative value is outweighed by the danger of ... confusion of the issues ... or by considerations of undue delay [or] waste of time.” Marrón herself admits that cross-examination is “[s]ubject always to the broad discretion of a trial judge to preclude repetitive and unduly harassing interrogation.”

The record shows that Marrón mounted an effective cross-examination of Dr. Rubenstein within the court’s parameters. The cross-examination revealed evidence that Dr. Ru-benstein did in fact fail to thoroughly examine Marrón; in particular, Dr. Rubenstein admitted he neglected some aspects of the examination because he was fatigued. Mar-ron’s cross-examination also revealed the fact that Dr. Rubenstein’s litigation services are predominantly done on behalf of defendants. And Dr. Rubenstein admitted that he had received $25,000 in compensation for his pretrial services on behalf of Stromstad. Mar-*1011ron was therefore able to adduce substantial evidence tending to dimmish the effectiveness of Dr. Rubenstein’s testimony.

The superior court thus did not abuse its discretion in applying the balancing test inherent in Rule 403 to limit cross-examination of Dr. Rubenstein and exclude his reports of previous eases from evidence.

F. The Superior Court Did Not Abuse Its Discretion in Denying Marron’s Motion for a New Trial.

Following trial, Marrón moved for a new trial on various grounds. The superior court denied the motion and Marrón appeals. Marrón argues that she should receive a new trial because the superior court denied Mar-ron’s request to introduce evidence of Strom-stad’s insurance coverage, because the superior court erroneously admitted the expert testimony of Dr. Rubenstein, and because Stromstad’s closing argument violated the superior court’s protective order.90 The starting point of any inquiry into the request for a new trial is Alaska Rule of Civil Procedure 61, which provides:

No error in either the admission or the exclusion of evidence and no error or defect in any ruling or order or in anything done or omitted by the court or by any of the parties is ground for granting a new trial ... unless refusal to take such action appears to the court inconsistent with substantial justice. The court at every stage of the proceeding must disregard any error or defect in the proceeding which does not affect the substantial rights of the parties.

The party requesting a new trial has the burden of proving both error and prejudice.91 As noted above,92 a trial court’s grant or denial of a new trial is reviewed for abuse of discretion. Before we will find an abuse of discretion, we must have “a definite and firm conviction” of an error, based on the entire record, viewed in a light most favorable to the non-movant.93 This is a highly deferential standard of review, and we have expressed great reluctance to interfere with a superior court’s decision to deny a new trial,94 absent exceptional circumstances.95

1. The superior court properly denied Marron’s request to introduce evidence of Stromstad’s insurance coverage in response to his opening statement.

In his opening statement, Strom-stad’s counsel told the jury that this case was about “an unusual extreme overreaetion that Mr. Stromstad should not be held responsible for and should not have to pay for.” Marrón asserts that this remark “improperly and untruthfully implied that [he] was uninsured and would be personally responsible for any judgment.” Accordingly, at trial Marrón requested to “cure” this alleged impropriety by introducing evidence of Strom-stad’s insurance coverage. The superior court refused to allow her to do so. Marrón subsequently moved for a new trial, based partly on the court’s refusal to admit her evidence of Stromstad’s insurance. Marrón appeals the court’s refusal to admit the above insurance evidence as well as its refusal to grant her a new trial on this basis. Because both points on appeal require similar analysis, we address them together.

Trial courts should generally exclude evidence of parties’ insurance coverage, so as to prevent such information from distracting or prejudicing the jury. Improper admission of insurance evidence can constitute an abuse of discretion and grounds for a *1012new trial.96 And we have found that some comments can be understood as improperly suggesting that a defendant lacks insurance coverage.97 But even assuming without deciding that Stromstad’s counsel’s comment suggests a lack of insurance, we have held that “the mere inadvertent or incidental mention of insurance before a jury in the trial of a negligence action does not automatically call for a mistrial.”98 Rather, “[i]f there is any error at all in permitting a reference to be made to insurance ... it must appear that such reference had a prejudicial effect on the result of the trial in order for there to be reversible error.”99 No such prejudicial effect is apparent here.

The superior court was well within its discretion in finding that Stromstad’s comment did not prejudice Marrón. Marrón argues that Stromstad’s statement, coming “at the very beginning of the case, was prejudicial to the Appellant as the jury [subsequently] received all of the evidence with the misconception ... that the Appellee would have to pay the judgment personally.” According to Marrón, this misconception “thus predis-pos[ed] the jury to accept all of the evidence during the ensuing trial sympathetically to the Appellee.” However, subsequent testimony at trial should have sufficed to correct any misconceptions or predispositions. Stromstad testified that he had exchanged insurance information with Marrón after the accident.100 Stirling’s testimony was partly based on an auto repair estimate by a “field appraiser” for “Allstate Insurance.” Although the court excluded explicit testimony that Allstate was Stromstad’s insurer, the implication that Stromstad is insured was at least as apparent as any contrary implication from Stromstad’s opening statement. We thus uphold the court’s refusal to allow Mar-rón to cross-examine Stromstad about his insurance coverage as well within its discretion under Evidence Rule 403. Stromstad’s opening statement did not clearly prejudice Marrón, and is not grounds for a new trial.

2. Even if Stromstad’s closing argument violated the superior court’s protective order, Marrón failed to object timely.

The superior court issued a pretrial amended order in response to Marron’s motions in limine to limit various expert testimony and exclude certain evidence. This order stated that Dr. Rubenstein would be allowed to testify as to “whether a rear-end accident at less than five miles per hour could have caused Marrón to have herniated her cervical disc” but could not “express an opinion that low-speed car accidents are the equivalent of ordinary activities such as sitting down or being slapped on the back.” During his closing argument, Stromstad noted that Dr. Rubenstein had testified that it “would take much more than a one to three mile per hour impact” to cause a herniated disc. Marrón asserts that this was a violation of the pretrial order, and that “[violation of a court’s previous protective order by defense counsel in closing is ground for [a] new trial.”

We have never squarely held whether a summation that violates a previous protective order is in fact grounds for a new trial. But as a general rule, we have held that a party *1013waives the right to appeal improper summation arguments unless that party objects to them at trial.101 Regardless of what Dr. Rubenstein actually testified to, and whether Stromstad’s comment somehow violated the court’s pre-trial order, Marrón did not object to the purportedly inadmissible comment at any point during Stromstad’s closing argument, during Marron’s own rebuttal summation, or before or even after the final jury instructions were read. Accordingly, the superior court properly denied Marrón a new trial on this issue.

G. Stromstad’s Attorney’s Fees Must Be Itemized.

Marron claims that the superior court erred in awarding Stromstad attorney’s fees without analyzing the reasonableness of such fees.102 According to Marrón, this reasonableness analysis required the court to examine a “detailed listing of the services sought to be recovered,” which Stromstad failed to provide in response to Marron’s request.103 Marrón claims that awarding fees to Stromstad without itemization, based only on an estimate of what fee amount was reasonable, effectively denied her an opportunity to be heard on this issue.

Stromstad sought attorney’s fees pursuant to both Alaska Rules of Civil Procedure 68 and 82. The superior court’s decision to award fees to Stromstad was based, correctly, only on Rule 68.104 Nothing in the text of Rule 68 (or Rule 82, for that matter) specifically states that a prevailing party must itemize its fees before a court may award them. Rule 68(b) simply states that a settlement offeree “shall pay reasonable actual attorney’s fees incurred by the offeror.” We have never specifically defined “reasonable actual attorney’s fees,” or the process by which a court should determine what were “reasonable actual fees.” But we have suggested that a prevailing party must itemize any requested fees where his or her opponent has made “a specific cognizable request for itemization.”105 Marrón argues that because her demand for a “detailed listing of services” in her “Opposition to Motion for Attorney’s Fees” was sufficiently specific and cognizable, the" superior court should not have awarded Stromstad fees without itemization.106 We agree.

As Stromstad notes, we have stated that “[w]e will reverse an award of attorney’s fees only if the award is ‘arbitrary, capricious, manifestly unreasonable, or stems from improper motive.’ ”107 But when a litigant fails to provide an itemized explanation of fees, the trial court has no effective means of determining whether the amount of fees requested is arbitrary or unreasonable108 and *1014we are unable to review the trial court’s fee award for any abuse of discretion. We therefore hold that, where the rule authorizes. reasonable actual fees, a court may not award attorney’s fees to a party who has not itemized his or her requested fees, when the opposing party has requested such itemization. Accordingly, we reverse the superior court’s award of attorney’s fees to Stromstad, and remand this part of the case for further consideration.

V. CONCLUSION

Because we decline to apply Daubert to non-scientific expert testimony, we AFFIRM the superior court’s decision to allow the testimony of James Stirling and Dr. Ruben-stein. Because the superior court did not abuse its discretion in denying the motion to compel production of Dr. Rubenstein’s income tax returns, in refusing to strike Dr. White’s testimony, in limiting cross-examination of Dr. Rubenstein and excluding his reports of previous patients, and in denying the motion for a new trial, we AFFIRM the superior court in all of these respects. Because Stromstad was required to itemize his attorney’s fees in response to Marron’s request, we REMAND for a re-determination of attorney’s fees.