¶1 — This car accident case went through mandatory arbitration and trial de novo, both times resulting in a damage award for the plaintiff. We conclude the trial court did not abuse its discretion by excluding defense biomechanical testimony about the physical forces involved in the collision, and accordingly the defendant’s request for a new trial is denied. However, the court’s determination that the defendant failed to improve her position on trial de novo is inconsistent with the Supreme Court’s recent decision in Niccum v. Enquist, 175 Wn.2d 441, 286 P.3d 966 (2012). The award of attorney fees to the plaintiff is reversed and remanded with directions to enter judgment on the jury verdict only.
FACTS
¶2 One afternoon in January 2006, Patricia Stedman was driving her Ford Aerostar van westbound on a residential street in Seattle. Her sister was with her. Stedman noticed a car was in a recessed parking spot on her right side, with its front end pointed east toward oncoming traffic. As Stedman drew nearer, the car pulled forward out *13of the parking spot. The two vehicles collided along their right sides. The impact cracked the right front bumper of Stedman’s van, dislodged the hubcap of her front wheel, and marked up the passenger side door.
¶3 The driver of the car was Stacey Cooper. She and Stedman exchanged contact information. Stedman was able to drive the van back home. The next day, she sought medical attention. Over the following months, Stedman saw medical doctors, chiropractors, and physical therapists.
¶4 Stedman sued Cooper in October 2008. The case went to mandatory arbitration. The arbitrator found Cooper at fault and awarded Stedman $16,300.00 in special damages and $7,000.00 in general damages, for a total of $23,300.00. Cooper filed a request for trial de novo. Stedman offered to settle for $23,299.99. Cooper declined the offer, and the case went to trial.
¶5 Before trial, Stedman learned that Cooper planned on calling Dr. Allan Tencer, a University of Washington professor of mechanical engineering, to testify about the severity of the force involved in the car accident. The court granted Stedman’s motion to exclude Tencer. On Stedman’s motion for summary judgment, the court ruled that Cooper was negligent as a matter of law, but there were issues of fact for trial as to whether Stedman was contributorily negligent and whether any injuries were caused or medical treatment necessary as a result of the accident.
¶6 At trial, Stedman described the accident as a sudden jolt that pulled her to the right, followed by her seat belt jerking her back into the seat and the movement of her steering wheel causing her to hit her face on the driver’s side window. She testified about pain in her shoulder and lower back.
¶7 A doctor testified that the mechanism of injury was a jarring impact from the side that caused hyperextension and hyperflexion injuries to the joints in her neck and mid and low back, and injury to her sacroiliac joint. A chiroprac*14tor testified to a similar diagnosis. In the defense case, a doctor who had conducted a CR 35 examination of Stedman thought she had suffered a “pretty minor injury” and that not all of the treatment she received was related to the accident. A chiropractor called by the defense testified that much of the chiropractic care was unnecessary, the accident was not serious, and Stedman had a soft tissue injury that probably resolved itself in about two months. The treatment providers agreed that Stedman’s obesity and general lack of fitness made her unusually susceptible to an impact injury.
¶8 Cooper, testified that she had been pulling out very slowly from where she had parked in front of her home. She estimated that Stedman was going more than 20 miles per hour when the two vehicles collided.
¶9 The jury found no contributory negligence by Stedman and awarded her $22,000.00 in damages. The court awarded statutory costs of $1,469.83. Stedman moved for attorney fees, contending that after costs were factored into the analysis, Cooper had failed to improve her position at trial. The court agreed and awarded Stedman $58,546.88 in attorney fees. Cooper appeals.
TIMELINESS OF APPEAL
¶10 A threshold issue raised by Stedman is whether Cooper’s appeal was untimely. It was not. The judgment in favor of Stedman against Cooper was entered on January 25, 2011. Cooper appealed on March 17, 2011. In general, a notice of appeal must be filed within 30 days. RAP 5.2(a). However, a timely motion for reconsideration in the trial court will extend this time until 30 days after entry of the order deciding that motion. RAP 5.2(e). That is what happened here. Cooper moved for reconsideration, and the trial court denied the motion in an order entered on February 17. The appeal filed on March 17 was within the 30-day time limit.
¶11 Stedman argues Cooper’s motion for reconsideration was actually a motion to vacate under CR 60 because it *15asked the trial court to “vacate” the judgment. The use of the term “vacate” to refer to the relief requested, however, does not change the fact that Cooper asked the court to reconsider the judgment in a timely motion brought under CR 59. We conclude the appeal is properly before us.
EXCLUSION OF EXPERT WITNESS
¶12 Cooper assigns error to the exclusion of Dr. Allan Tencer. Tencer has received government grants for research into the underlying mechanisms of injury to the cervical spine, and he has conducted tests aimed at developing improved car seat head restraints for prevention of impact injuries. He has written many peer-reviewed articles and book chapters related to the biomechanics of the human skeletal system. He has the training and experience to calculate forces between colliding objects.
¶13 The trial court reviewed Tencer’s opinion letter and his declaration, and later heard his testimony as an offer of proof in support of Cooper’s motion for reconsideration. Tencer’s opinion letter stated that since the forces generated by the impact and transmitted to Stedman were “low, relative to forces experienced in daily living,” the accident was “not a likely source of significant forces” acting on her body.1 The forces acting upon Stedman were “in the range of those found tolerable in testing we have performed on volunteers of a variety of ages, gender, and levels of fitness, related to development of improved head restraints for whiplash protection.”2
¶14 Tencer’s testimony for the offer of proof described the methodology he used to develop his opinion. First, using photographs, he sized the area of deformation of the bumper on Cooper’s vehicle. Then, using his knowledge of the weights of the two vehicles and the crush strength of the *16bumper, he calculated the forward speed of Cooper’s car and the sideways speed of Stedman’s van.3 From that, he determined “the jolt,” or acceleration.4 From that, he determined that the approximate force acting on the occupants of the van was 1.1G. “G force” is a term that refers to “acceleration of gravity.”5
¶15 According to Tencer, 1.1G is a “jolt level” comparable to what would be generated by hitting a curb at three to five miles per hour while parking a car.6 Forces of 3 to 5 G are “fairly common” in daily activities.7 Tencer testified that in a study he was doing on injuries in soccer players, he found that the participants generated a force of 3G while just walking and running.8 Tencer also testified that in his opinion, it was unlikely that Stedman hit her head on the driver side window because her body should have moved in the opposite direction.9
¶16 In reviewing a trial court’s exclusion of expert testimony under the rules of evidence, the standard is abuse of discretion. State v. Yates, 161 Wn.2d 714, 762, 168 P.3d 359 (2007), cert. denied, 554 U.S. 922 (2008). To be admissible, expert witness testimony must be relevant and helpful to the trier of fact. Anderson v. Akzo Nobel Coatings, Inc., 172 Wn.2d 593, 606, 260 P.3d 857 (2011). Conclusory or speculative expert opinions lacking an adequate foundation will not be admitted. Miller v. Likins, 109 Wn. App. 140, 148, 34 P.3d 835 (2001). When ruling on somewhat speculative testimony, the court should keep in mind the danger that the jury may be overly impressed with a witness possessing the aura of an expert. Miller, 109 Wn. App. at 148.
*17¶17 Tencer has testified as an expert witness in many similar cases. Attached to his declaration are 12 orders entered by various superior courts in Washington from 2004 to 2008, most of them denying motions to exclude his testimony.10 Two of the orders imposed limitations. One court permitted him to testify that the forces in the accident were “in the range of those experienced in normal activities (3-5g),” but barred him from expressing the opinion that the accident was “no more likely than other sources of daily forces to have resulted in the injuries claimed.”11 This limitation was imposed in part because the trial court did not view Tencer’s testimony as helpful to the jury on that point.12 Another court prohibited him from opining as to whether the specific forces involved in the collision were sufficient to cause injury. The order allowed him to testify about the range of forces found tolerable by the subjects in his tests but prevented him from using the volunteers in his studies as a basis of comparison for the injuries allegedly suffered by the plaintiff.13
¶18 Other Washington superior courts have granted motions to exclude Tencer’s opinion. Stedman attached three such orders, entered from 2008 to 2012, in reply to Cooper’s response to the motion to exclude.14 In one of the orders, the court concluded that Tencer’s testimony improperly invited the jury to infer that the plaintiff could not have been injured in the accident:
Dr. Tencer is very careful to state that he is not testifying to what specific injuries the accident caused to this plaintiff. But that is exactly the inference that the defense wants the jury to draw from his testimony: that because, on average, the forces in such an accident would not injure a vehicle occupant, the *18plaintiff in this case must not have been injured by this accident. If the jury does not draw this inference, Dr. Tencer’s testimony, while interesting, is irrelevant to the proceeding before the court.[15]
Another order excluded Tencer’s testimony about his experiment “as the court has concluded that the conditions were not substantially similar to the event at issue.”16
¶19 One Washington case in which Tencer was allowed to give expert testimony reached the appellate level in Ma’ele v. Arrington, 111 Wn. App. 557, 45 P.3d 557 (2002). Ma’ele involved a rear-end collision; the only issue at trial was damages. Tencer opined that “the maximum possible force in this accident was not enough to injure a person.” Ma’ele, 111 Wn. App. at 564. There was a defense verdict. The plaintiff appealed, and the decision to admit Tencer’s testimony was affirmed against a challenge under ER 702. “His testimony about the force involved in low-speed collisions and the impact on the body helped the jury determine whether Ma’ele got hurt in this accident.” Ma’ele, 111 Wn. App. at 563.
¶20 The fact that an appellate court has affirmed a decision allowing Tencer’s testimony does not, of course, necessarily mean that the trial court erred by excluding his testimony in this case. The broad standard of abuse of discretion means that courts can reasonably reach different conclusions about whether, and to what extent, an expert’s testimony will be helpful to the jury in a particular case.
¶21 Here, the trial court excluded Tencer’s testimony as both irrelevant and cumulative. We agree with Cooper that the court erred in characterizing the testimony as cumulative. Tencer’s expertise as a biomechanical engineer was different from the expertise of the treatment providers. The closer question is whether the court erred in ruling that Tencer’s testimony was “logically irrelevant to the issue *19the jury must decide: the degree to which these particular plaintiffs were injured in this particular automobile accident.”17
¶22 Appellate courts in other states have concluded that defense biomechanical testimony in minor impact soft tissue cases is not helpful to the jury where it is based on information gathered in human volunteer crash testing. A New York court found the sample size of studies relied upon by a biomechanical expert was “too small to create a statistically significant inference.” Clemente v. Blumenberg, 183 Misc. 2d 923, 927, 705 N.Y.S.2d 792 (1999). The Supreme Court of Virginia reversed a trial court for admitting testimony similar to Tencer’s in Tittsworth v. Robinson, 252 Va. 151, 475 S.E.2d 261 (1996). There was “no proof that these experiments were conducted under circumstances substantially similar to those existing at the accident scene.” Tittsworth, 475 S.E.2d at 263. The court found that the expert’s testimony “is speculative, is founded upon assumptions lacking a sufficient factual basis, relies upon dissimilar tests, and contains too many disregarded variables. Consequently, we hold that the testimony is unreliable as a matter of law, and, therefore, the trial court erred in admitting it.” Tittsworth, 475 S.E.2d at 263-64 (footnote omitted).
¶23 The Colorado Court of Appeals concluded a trial court did not abuse its discretion by prohibiting a defense biomechanical expert from testifying that there is “a threshold force level below which a person probably could not be injured in a rear-end automobile collision.” Schultz v. Wells, 13 P.3d 846, 849 (Colo. App. 2000). The trial court “reasoned that tests used to ascertain safety for the purposes of doing a cost-benefit analysis with regard to the expense of designing the seat of a car were not applicable to prove that a particular person was unlikely to be injured in a specific accident.” Schultz, 13 P.3d at 851.
*20Additionally, the court questioned the validity of using a series of tests designed for one purpose (designing cars) for a different purpose (assessing a threshold of applied force for injury in rear-end car accident). Specifically, the court addressed the circumstances of the tests that did not correspond with the circumstances of a rear-end car accident. It noted the fact that the statistical sample in the tests was “extremely low,” and there were “no controls among and between the experiments with regard to age, physical conditions [and] actual position of the body.” Also, the court noted the “expectation factor” of knowing one is going to be hit, as opposed to being unaware of an impending collision. The court concluded that there “is great controversy in the field about the quality and comparability of these tests.”
Schultz, 13 P.3d at 852. The appellate court concluded the trial court had identified appropriate factors to review the likelihood that the evidence would mislead the jury. The court was also persuaded that it was within the trial court’s discretion to keep out evidence about the G forces that occur during daily human activities. Schultz, 13 P.3d at 852.
¶24 Tencer declared that he agreed with the ruling in Schultz and that his testimony was different because “I... have never described any threshold for injury in my opinions.”18 Emphasizing that he testifies from a biomechanical rather than a medical perspective, he disavowed any intention of giving an opinion about whether Stedman was hurt in the accident.19 Nevertheless, his clear message was that Stedman could not have been injured in the accident because the force of the impact was too small. Indeed, according to Cooper’s brief, Tencer’s conclusion was exactly that: the forces generated by the impact were not sufficient to cause the type of injuries Stedman was claiming.20
¶25 Schultz persuasively explains why a trial court may regard such an opinion as more likely to be misleading than *21helpful. Following Schultz, we conclude that excluding Tencer’s testimony was not an abuse of discretion.
ATTORNEY FEE AWARD ON TRIAL DE NOVO
¶26 Under MAR 7.3, the court “shall assess costs and reasonable attorney fees against a party who appeals the award and fails to improve the party’s position on the trial de novo.” ROW 7.06.060(1). Cooper contends the trial court’s award of attorney fees to Stedman was based on an erroneous determination that Cooper failed to improve her position at trial de novo. Because the issue involves application of a court rule, our review is de novo. Mee Soon Kim v. Pham, 95 Wn. App. 439, 441, 975 P.2d 544, review denied, 139 Wn.2d 1009 (1999).
¶27 The arbitrator awarded Stedman $23,300. After Cooper filed for trial de novo, Stedman offered to compromise her claim for a penny less than the arbitration award.
¶28 The amount of an offer of compromise replaces the amount awarded by the arbitrator for determining if the party appealing the award improved that party’s position at trial. RCW 7.06.050. Stedman’s offer stated that the “compromise is intended to replace the arbitrator’s award against defendant Stacey Cooper with an award of $23,299.99, inclusive of costs, statutory attorney fees, and attorney fees and sanctions.”21
¶29 Cooper rejected the offer and proceeded to trial. The jury awarded Stedman $22,000.00. The trial court entered judgment for Stedman in the amount of $22,000.00 along with costs of $1,469.83. Adding the two figures together meant Cooper was liable to Stedman for a total of $23,469.83. Using these figures, the court determined Cooper did not improve her position over the $23,299.99 offered by Stedman ($23,299.99 - $1,469.83 = $21,830.16). As a result, the court ruled that Cooper was obligated to pay Stedman’s attorney fees.
*22¶30 At first glance, one might think that Cooper did improve her position at trial because $22,000.00 is less than $23,299.99. In fact, Cooper makes precisely that argument on appeal. But she did not raise this argument in the trial court. Instead, she accepted that the controlling authority was the Court of Appeals decision in Niccum v. Enquist, 152 Wn. App. 496, 215 P.3d 987 (2009). At that time, the Supreme Court had accepted review of Niccum. Niccum v. Enquist, review granted, 168 Wn.2d 1022 (2010). In Niccum, the plaintiff’s offer of compromise was worded in substantially the same manner as Stedman’s. The Court of Appeals held that for purposes of determining whether the party who requested trial de novo had improved its position, court costs awarded to the prevailing party at trial would first be subtracted from the offer of compromise before comparing the offer of compromise and the verdict. Relying on the Court of Appeals decision in Niccum, Cooper presented her issue to the trial court only in terms of whether Stedman had a legitimate claim for costs sufficient to bring her judgment above $23,299.99:
Because the Plaintiff’s Offer of Compromise was inclusive of costs and statutory attorney fees, Plaintiff would be entitled to an attorney fee award only if her costs exceeded $1,299.99. Plaintiff has claimed $1,469.83 in costs. If the court awards these costs to her, the judgment amount will exceed the Offer of Compromise by a mere $169.84.[22]
¶31 On appeal, Cooper first argued that the Court of Appeals decision in Niccum was wrongly decided and costs should not have been factored into the comparison between the arbitrator’s award and the jury verdict. In our original opinion in this matter, we held that Cooper did not preserve her argument that the arbitrator’s award and the jury award are the only relevant figures to be used in determining whether she improved her position as the appealing party. We therefore declined to consider it, citing RAP 2.5(a).
*23¶32 The only argument of Cooper’s that we considered relative to the award of attorney fees was whether the court correctly determined the costs allowable to Stedman. Cooper assigns error to the trial court’s assessment of costs, arguing that the correct assessment — which would have lowered the judgment to an amount less than Stedman’s offer of compromise — would have omitted costs for records admitted at the arbitration. However, RCW 4.84.010(5) plainly allows costs for medical records so long as they are admitted into evidence, either in mandatory arbitration or at trial. The trial court did not err in allowing Stedman her costs for these records.
¶33 Because we concluded that Cooper failed to preserve her argument that costs were irrelevant, and that the trial court properly calculated allowable costs, our original opinion upheld the trial court’s decision that Cooper failed to improve her position at trial de novo. Accordingly, we did not disturb the trial court’s award of attorney fees to Stedman against Cooper under RCW 7.06.060(1). Our opinion making this ruling was issued on August 13, 2012.
¶34 Cooper moved for reconsideration. That motion was still pending when, on September 20, 2012, the Supreme Court issued its decision in Niccum. Niccum, 175 Wn.2d 441. The Supreme Court reversed the Court of Appeals and held that it is “improper to subtract costs from an offer of compromise.” 175 Wn.2d at 448. The court held there is no right under the statute to include costs in an offer of compromise. 175 Wn.2d at 448.
¶35 Had the Supreme Court’s decision in Niccum been available to the trial court at the time of entering judgment in the present case, it would have prevented the trial court from awarding attorney fees to Stedman as the prevailing party after trial de novo. Under Niccum, Cooper did improve her position on trial de novo.
¶36 We asked the parties for supplemental briefs on the applicability of Niccum to the present case. The parties agree that Niccum applies “retroactively” or, perhaps better *24stated, that it applies from the date of the enactment of RCW 7.06.060. This is because of the “fundamental rule of statutory construction that once a statute has been construed by the highest court of the state, that construction operates as if it were originally written into it.” Johnson v. Morris, 87 Wn.2d 922, 927, 557 P.2d 1299 (1976). The only issue, therefore, is whether Niccum applies despite Cooper’s failure to challenge in the trial court the construction that the Court of Appeals had placed on the statute in its earlier (and now overruled) decision in that case.
¶37 The parties base their arguments on RAP 2.5:
(a) Errors Raised for First Time on Review. The appellate court may refuse to review any claim of error which was not raised in the trial court. However, a party may raise the following claimed errors for the first time in the appellate court: (1) lack of trial court jurisdiction, (2) failure to establish facts upon which relief can be granted, and (3) manifest error affecting a constitutional right. A party or the court may raise at any time the question of appellate court jurisdiction. A party may present a ground for affirming a trial court decision which was not presented to the trial court if the record has been sufficiently developed to fairly consider the ground. A party may raise a claim of error which was not raised by the party in the trial court if another party on the same side of the case has raised the claim of error in the trial court.
¶38 Citing RAP 2.5(a)(3), Stedman contends Cooper’s failure to raise the issue at trial precludes appellate review because an award of attorney fees not authorized by law is not a “manifest error affecting a constitutional right.” Cooper, on the other hand, cites RAP 2.5(a)(2) and argues that Stedman has failed “to establish facts upon which relief can be granted” under RCW 7.06.060 as construed by the Supreme Court in Niccum.
¶39 We agree with Cooper. To establish her right to attorney fees as the prevailing party at trial de novo, Stedman had to show that Cooper failed to improve her position. Under Niccum, facts establishing that right are *25simply not present. In this situation, Cooper is not precluded from raising the issue for the first time on appeal. And the statute, now correctly construed by the Supreme Court, may be enforced by this court even though the argument was not presented to the trial court. Batten v. Abrams, 28 Wn. App. 737, 742-43, 626 P.2d 984, review denied, 95 Wn.2d 1033 (1981).
¶40 Because Stedman was not the prevailing party on trial de novo, the award of attorney fees in the trial court will be reversed.
¶41 The denial of Cooper’s request for a new trial is affirmed. The award of attorney fees and costs is reversed. The judgment is remanded for entry of a judgment on the jury verdict only. No attorney fees are awarded on appeal.
Leach, C.J., and Spearman, J., concur.Clerk’s Papers at 25.
Clerk’s Papers at 27.
Report of Proceedings at 5-6.
Report of Proceedings at 5-6.
Report of Proceedings at 8.
Report of Proceedings at 8.
Report of Proceedings at 8.
Report of Proceedings at 8.
Report of Proceedings at 12.
Clerk’s Papers at 168-93.
Clerk’s Papers at 180-81.
Clerk’s Papers at 181.
Clerk’s Papers at 189-90.
Clerk’s Papers at 249-55.
15 Clerk’s Papers at 250.
Clerk’s Papers at 252.
Clerk’s Papers at 291.
Clerk’s Papers at 166.
Clerk’s Papers at 163, 166.
Clerk’s Papers at 68, 70 (Opposition to Motion to Exclude Testimony).
Clerk’s Papers at 587.
22 Clerk’s Papers at 625.