In this appeal, we consider breach of contract claims brought by Fleur Bresler (Fleur) and her son, Sidney Bresler (Sidney) (collectively, the plaintiffs), as Co-Personal Representatives of the Estate of Charles S. Bresler (the Estate). A jury determined that defendants Wilmington Trust Company and Wilmington Brokerage Services. Company (collectively, Wilmington) breached an agreement to lend money for the acquisition, maintenance, and certain investments relating to life insurance policies obtained for Charles S. Bresler (Charlie)1 and his wife, Fleur. The jury awarded the plaintiffs around $23 million in damages. The district court determined post-trial that Wilmington also had breached an agreement to return certain funds to the Estate upon Charlie’s death, and ordered Wilmington to return those funds in accordance with the parties’ agreement.
Wilmington appeals, arguing that: (1) the district court erred in admitting testimony from the plaintiffs’ expert witness; (2) the jury verdict, including the jury’s award of damages, was not supported by the evidence; and (3) additional terms of the district court’s order also were not supported by the evidence. Upon our review, we affirm the district court’s judgment.
I.
A.
We state the evidence in the light most favorable to the plaintiffs, the prevailing *184parties at trial. See King v. McMillan, 594 F.3d 301, 306 (4th Cir. 2010). The evidence at trial showed that the parties’ dispute involved an estate-planning strategy-known as “premium financing.” In the particular type of premium financing at issue here, an individual establishes an Irrevocable Life Insurance Trust (ILIT), which acquires one or more life insurance policies and pays the insurance premiums with loans obtained from a third-party lender. Each insurance policy consists of two components: (1) the face value of the policy, and (2) an investment component whereby the death benefit, or the amount that is paid to the insurance beneficiaries when the insured dies, increases if the policies retain excess funds above those required to cover the cost of the insurance and related expenses.
The process of investing in the insurance policies by making payments exceeding the minimum required amount is known as “overfunding.” Through this technique, the policy funds grow because the insurance company pays interest on the policies acquired by the ILIT at a specified crediting rate. The goal of overfunding is to borrow from the third-party lender at an interest rate that is lower than the crediting rate, which causes the value of the policies to grow more quickly than the amount of the debt incurred.
During this process of overfunding, the funds in the policies accrue without being subject to taxation.2 After the insured dies, the insurance company pays a portion of the death benefit from the policies to the third-party lender to repay the ILIT’s outstanding loans, and thereafter pays the remainder of the death benefit to the ILIT, also without being subject to taxation. After the loans are repaid, the funds remaining in the ILIT, known as the “net-in-trust,” pass tax-free to the ILIT’s beneficiaries.
B:
Charlie was a successful entrepreneur in the Washington, D.C. area, and was a co-founder and Chairman of the Board of Directors of Bresler & Reiner, Inc. (B&R), a publicly traded company engaged in real estate development and commercial property management.3 By 2003, Charlie was 75 years old and had established a net worth of $150 million.
In 2002, Wilmington began development of a premium financing product. Wilmington and B&R had a prior business relationship, and in 2003, a Wilmington employee who had been managing Wilmington’s relationship with B&R introduced Edmond Ianni, a Wilmington corporate vice president, to Charlie’s attorney, Larry Shaiman. In March 2003, aware of Charlie’s significant assets, Ianni and another Wilmington account executive approached Shaiman to discuss the possibility of Wilmington developing for Charlie and Fleur a “tax-saving wealth creation and preservation strategy.” .
■ Throughout Shaiman’s and Charlie’s discussions with Ianni, Charlie expressed reservations about entering into an arrangement that would require him to post a significant amount of collateral. During the course of their conversations, Ianni sent Charlie and Shaiman spreadsheets detailing net-in-trust projections to be derived from a premium financing arrangement. Unlike the spreadsheets Wilmington frequently used with other customers, the *185spreadsheets Ianni sent to Charlie omitted a column identifying payments of collateral. On November 10, 2003, Ianni sent Shai-man an email stating that any collateral Wilmington required from Charlie would be “minimal,” because “the value of the ... trust’s ... main asset (namely, the significant, growing cash value of the policy, as well as the increasing death benefit) is substantial; that, as you know, will serve as the significant source for satisfaction of the trust’s outstanding loan to [Wilmington].”
Eleven days later, on November 21, 2003, Ianni sent Charlie a letter (the November 21 letter) detailing a proposed premium financing arrangement, in which Wilmington would lend to an ILIT established by the Breslers (the Trust) “the annual premium plus allowable overfund-ing ($5.5 million) to acquire and maintain” several “second-to-die” life insurance policies4 for Fleur and Charlie with a combined face value of $50 million. The November 21 letter proposed a “blended fixed” crediting rate of 5.675 percent, and a “favorable” interest rate of 1.5 percent above LIBOR5 “on terms required by the lender.” Wilmington projected that after Fleur and Charlie both died, the net-intrust would be between “$40.6 million to over $45.5 million,” with tax savings of between “$24.5 million to over $115 million.”
In the November 21 letter, Ianni also attempted to clarify “any misunderstanding of the collateral pledge aspect” of the arrangement. According to Ianni, Wilmington would require from Charlie a collateral pledge amount of between $2.9 and $4.2 million in the first year. Ianni indicated that “[t]he amount of the needed collateral going forward obviously will depend in part on the aggregate cash surrender value of the trust’s assets (namely, of the insurance) and will be reviewed periodically.” Charlie rejected the requirements for collateral stated in the November 21 letter, based on his understanding that the parties already had agreed that ongoing payments of collateral would not be required.
After further negotiations, in January 2004, Wilmington and Charlie executed three written agreements: (1) an irrevocable life insurance trust agreement (Trust Agreement) establishing the Trust; (2) an investment management agreement establishing an investment management account; and (3) a collateral pledge agreement. The Trust Agreement named Wilmington as trustee and the five children of Fleur and Charlie, including Sidney, as the Trust’s beneficiaries.
The parties dispute whether Charlie and Ianni ultimately agreed that Charlie would make ongoing collateral payments. However, Charlie made an initial collateral payment to Wilmington in the amount of $3.7 million, for deposit into the investment management account. Wilmington later loaned funds to the Trust to acquire three “second-to-die” life insurance policies for Charlie and Fleur, with a combined face value of $50 million, and overfunded the policies in 2004.
C.
The events culminating in the present litigation occurred in 2005, after the first *186year of overfunding had concluded.6 At that time, Wilmington informed Charlie that he was required to post additional collateral before Wilmington would provide another $5.5 million loan to the Trust to cover the cost of the insurance premiums and overfunding contributions. Charlie rejected Wilmington’s demand, and maintained that the parties’ agreement required only that he provide the initial $3.7 million payment of collateral. Because the parties could not resolve their differences concerning the posting of collateral, Charlie provided an additional $1.3 million in collateral to prevent the policies from lapsing. Those funds were placed in the investment management account.
Accordingly, in 2005, Wilmington lent the Trust $702,338 to cover only the cost of the life insurance premiums, and did not overfund the policies after 2004. However, the parties continued in their efforts to resolve the issue regarding additional collateral payments until 2007, when Wilmington stopped making any payments for the policies. At that time, Wilmington and Charlie entered into a series of tolling agreements memorializing their dispute “regarding the process by which the structure and proposal of the insurance program was portrayed by agents of [Wilmington],” and agreeing to maintain their rights to take legal action if negotiations proved unsuccessful.
Around the same time, Ianni sued Wilmington in Delaware state court, seeking allegedly outstanding commissions. In that suit, Wilmington filed a counterclaim asserting that Ianni had misrepresented to certain Wilmington customers, including Charlie, the requirements for posting collateral related to the life insurance trust agreements.
In September 2009, Charlie and his son, Sidney, filed suit against Wilmington in Maryland state court alleging breach of contract, negligence in managing the Trust and the life insurance policies, breach of fiduciary duty, negligent misrepresentation, fraud, and violation of the Delaware Consumer Fraud Act relating to Wilmington’s failure to overfund the policies.7 Wilmington removed the case to federal district court in November 2009, based on diversity jurisdiction under 28 U.S.C. § 1332.
At around the same time, in October 2009, Fleur and Charlie obtained term life insurance policies on Fleur’s life (the replacement policies), because Fleur and Charlie were concerned that Wilmington would allow their original policies to lapse. The replacement policies carried a fixed death benefit of $17.5 million, had no investment component, and required an annual premium payment of between $1.4 and $1.5 million.
In early 2010, Wilmington resumed lending the Trust enough money to make the minimum premium payments. Charlie died in October 2010, and Sidney and Fleur, as personal representatives of Charlie’s estate, became the plaintiffs in the present litigation. In February 2012, Sidney sent a letter to Wilmington demanding that it terminate the investment management agreement and return to Charlie’s estate *187the collateral being held in the investment management account. Sidney relied on a provision in the investment management agreement stating that “the agreement shall be terminated upon [Wilmington]’s receiving written notice of the death of [Charlie]. Upon such termination, [Wilmington] shall deliver to ... the executor or Administrator of [Charlie]’s estate, ... all of the property held by it hereunder, if any.”
Wilmington refused to return the collateral held in the investment management account. In September 2012, Fleur and Sidney, as co-personal representatives of Charlie’s estate, initiated a second lawsuit against Wilmington in Maryland state court, alleging that Wilmington breached an agreement to return funds held in the investment management account to Charlie’s estate upon receiving notice of Charlie’s death. Fleur and Sidney sought a declaratory judgment, specific performance of Wilmington’s contractual obligations, and monetary relief. Wilmington removed the case to federal court, again based on diversity jurisdiction under 28 U.S.C. § 1332. The district court consolidated the two cases, and scheduled a jury trial to resolve only the breach of contract claims against Wilmington.8
D.
The consolidated trial began in January 2014, and lasted for three weeks. During the trial, the plaintiffs argued that under an oral agreement between Charlie and Ianni, acting on Wilmington’s behalf, the final premium financing arrangement required Charlie to post as collateral only the initial amount of $3.7 million. The plaintiffs asserted that the policies themselves served as the primary collateral for the arrangement. According to the plaintiffs, the parties had agreed that when Fleur and Charlie died, Wilmington’s loans would be repaid in full from the proceeds of the death benefit. The plaintiffs contended that Wilmington had received millions of dollars in commissions from the insurance companies, as well as management fees, and had charged the Trust several million dollars in interest. Thus, the plaintiffs argued that Wilmington breached its agreement with Charlie when it demanded additional collateral payments and refused to lend the Trust $5.5 million annually to pay the premiums and overfund the policies through Fleur’s life without the payment of further collateral.
In its defense, Wilmington argued that Charlie and Shaiman were experienced businessmen who understood that funds in excess of Charlie’s initial pledge of collateral would be required to support the premium financing arrangement. Wilmington asserted that the November 21 letter noted the possibility of future collateral payments, and that Charlie eventually agreed that the amount of additional collateral would be assessed on a yearly basis. Wilmington contended that it never would have agreed, particularly in the absence of a written contract, to lend the Trust $5.5 million every year until Fleur’s demise while requiring merely $3.7 million in collateral funding.
In addition to contesting liability, the parties also disputed the existence and proper calculation of damages. In particular, the parties disagreed whether the plaintiffs’ expert witness, Robert E. Pugh, had accurately calculated the amount of the net-in-trust.
*188Pugh, an accountant, presented two sets of damages calculations. In the first set, he calculated the “present shortfall” of the net-in-trust, or the difference between (1) what the net-in-trust would have been at the time of trial in January 2014 had Wilmington lent the $5.5 million annually, and (2) what the existing amount of the net-intrust was at the time of trial, given that Wilmington had failed to overfund the policies since 2004.9 Pugh ultimately determined that the value of the “present shortfall” of the net-in-trust was around $10.7 million.
In his second set of damages calculations, Pugh calculated the “future” net-intrust shortfall over the period between 2014 and the projected end of Fleur’s life (future shortfall), which, according to mortality tables developed by the Social Security Administration, and the parties’ stipulation, would occur in 2019. Pugh’s future shortfall calculation included two scenarios: (1) the future shortfall if Wilmington began overfunding, or lending, the full $5.5 million per year, between 2014 and 2019; and (2) the future shortfall if Wilmington lent only the minimum amount necessary to maintain the life insurance policies. Accordingly, Pugh made two total net-intrust shortfall calculations, combining the present shortfall calculation with each of the two different future shortfall estimates. These two total net-in-trust shortfall calculations amounted to: (1) $17.8 million if Wilmington began overfunding the policies immediately and continued to do so through 2019, and (2) around $19.5 million if Wilmington made only the minimum life insurance premium payments through 2019.
Wilmington argued in response that even if it had breached an agreement to overfund the policies, Charlie did not suffer any damages because he had not made any payments himself to maintain the policies. Wilmington also vigorously challenged the accuracy of Pugh’s calculations, asserting that they were “riddled with mistakes” and were wholly unreliable.
In a special verdict, the jury concluded that: (1) Wilmington had agreed to lend funds to pay the life insurance policy premiums for the duration of Charlie’s and Fleur’s lives; (2) Wilmington’s commitment was not contingent on Charlie agreeing to post additional annual collateral; (3) Wilmington was not authorized by the parties’ agreement to establish an annual amount that Charlie and Fleur would be required to pay as collateral; (4) Wilmington was required to overfund the policies; (5) Wilmington breached the agreement in March 2005 by requiring that Charlie pledge an additional $1.3 million in collateral; (6) Wilmington breached the agreement by failing to make loans each year to pay the premiums on the life insurance policies and by failing to overfund the policies beyond the first year; and (7) Wilmington was obligated to make premium payments for each year that the policies remained in effect.
The jury adopted Pugh’s damages calculations, determining that Wilmington owed the plaintiffs either $17.8 million if Wilmington began overfunding the policies immediately, or $19.5 million if Wilmington were to make only the minimum premium payments through 2019. The jury also concluded that Wilmington owed the plaintiffs $3.9 million to reimburse the plaintiffs for part of the costs they incurred in obtaining the replacement policies.
The district court denied Wilmington’s post-trial motion for judgment as a matter *189of law, as well as Wilmington’s alternative motions to amend the verdict, to set a new trial, and to require a remittitur. Ultimately, the court entered final judgment ordering Wilmington to continue lending “sufficient funds to pay the minimum premiums required to maintain in force the Policies until the death of Fleur Bresler,” to pay the plaintiffs about $23 million in damages, and to return about $5 million in collateral to the Estate.10 Wilmington noted a timely appeal from the district court’s judgment.
II.
On appeal, we first consider Wilmington’s challenges to the admissibility of Pugh’s testimony. Wilmington argues: (1) that the district court was required under Rule 37 of the Federal Rules of Civil Procedure to exclude Pugh’s testimony, because the plaintiffs violated certain provisions governing expert witness disclosures; and (2) alternatively, that the court should have excluded Pugh’s testimony as analytically invalid under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). We address both of these arguments in turn.
A.
Wilmington asserts that the district court abused its discretion in admitting Pugh’s testimony and in allowing his use of a particular exhibit setting forth his calculation methodology. Wilmington primarily contends that the court disregarded the requirements of Rule 26: (1) by permitting the plaintiffs to use the exhibit depicting Pugh’s net-in-trust formula and calculations, when this information was submitted after the deadline for such disclosures and was not included in Pugh’s expert witness report; and (2) by permitting Pugh to testify regarding his updated calculations.
According to Wilmington, the district court’s decision to admit Pugh’s testimony disrupted the trial and deprived Wilmington of an adequate opportunity to cross-examine Pugh. Wilmington thus argues that it is entitled to a new trial or, alternatively, to the exclusion of Pugh’s testimony and entry of judgment in Wilmington’s favor. We disagree with Wilmington’s arguments.
We review a district court’s discovery rulings, as well as its decision to admit particular expert testimony, for abuse of discretion. Anderson v. Westinghouse Savannah River Co., 406 F.3d 248, 260 (4th Cir. 2005) (admission of expert testimony); Am. Chiropractic Ass’n v. Trigon Healthcare, Inc., 367 F.3d 212, 236 (4th Cir. 2004) (discovery rulings). In the absence of a stipulation or court order stating otherwise, Rule 26 requires litigants to provide opposing counsel with a written report prepared and signed by an expert witness who may testify at trial. Fed. R. Civ. P. 26(a)(2)(A)-(B). The expert witness’ report must contain, among other things, “a complete statement of all opinions the [expert] witness will express and the basis and reasons for them,” “the facts or data considered by the witness in forming them,” and “any exhibits that will be used to summarize or support them.” Fed. R. Civ. P. 26(a)(2)(B)(i)-(iii). A party must make required expert witness disclosures “at the times and in the sequence that the court orders.” Fed. R. Civ. P. 26(a)(2)(D).
A litigant also is required to supplement information provided in its expert witness report and during the expert witness’ de*190position “if the party learns that in some material respect the disclosure or response is incomplete or incorrect, and if the additional or corrective information has not otherwise been made known to the other parties during the discovery process or in writing.” Fed. R. Civ. P. 26(e)(1)(A), (e)(2). Unless the court orders otherwise, a party must supplement or correct such information regarding the expert witness’ opinion and report at least thirty days before trial. Fed. R. Civ. P. 26(e)(2), (a)(3)(B).
The purpose of Rule 26(a) is to allow litigants “to adequately prepare their cases for trial and to avoid unfair surprise.” Russell v. Absolute Collection Servs., Inc., 763 F.3d 385, 396 (4th Cir. 2014). Accordingly, a party who fails to comply with the expert witness disclosure rules is prohibited from “us[ing] that information or witness to supply evidence ... at a trial, unless 'the failure was substantially justified or is harmless.” Fed. R. Civ. P. 37(c)(1).
District courts are accorded “broad discretion” in determining whether a party’s nondisclosure or untimely disclosure of evidence is substantially justified or harmless. Wilkins v. Montgomery, 751 F.3d 214, 222 (4th Cir. 2014) (quoting S. States Rack & Fixture, Inc. v. Sherwin-Williams Co., 318 F.3d 592, 597 (4th Cir. 2003)). In making this determination, district courts are guided by the following factors:
(1) the surprise to the party against whom the evidence would be offered; (2) the ability of that party to cure the surprise; (3) the extent to which allowing the evidence would disrupt the trial; (4) the importance of the evidence; and (5) the nondisclosing party’s explanation for its failure to disclose the evidence.
S. States, 318 F.3d at 597. The first four factors listed above relate primarily to the harmlessness exception, while the last factor, addressing the party’s explanation for its nondisclosure, relates mainly to the substantial justification exception. Id. The party failing to disclose information bears the burden of establishing that the nondisclosure was substantially justified or was harmless. Wilkins, 751 F.3d at 222 (citations omitted).
Applying these principles, we agree with Wilmington that the plaintiffs did not timely disclose Pugh’s net-in-trust formula and calculations. Nevertheless, we conclude that the district court did not abuse its discretion in allowing use of the exhibit and in admitting Pugh’s testimony. We ultimately reach this result based on our conclusion that the untimely nature of the plaintiffs’ disclosures was harmless and did not materially affect Wilmington’s defense in the litigation, including during the trial. The relevant facts in this dispute chart our path leading to this conclusion.
In June 2012, the plaintiffs served on the defendants a copy of a report (the June 2012 report) drafted by Pugh, their expert witness on accounting issues. In his report, Pugh presented an opinion regarding the present value11 of an investment mechanism that would have received an annual contribution of $5.5 million between January 2013 and January 2019, and was subject to a 3.57 percent crediting rate. Pugh qualified his opinion by stating that his calculations were subject to amendment based on information regarding an actual award of damages or other court rulings, as well as on additional data revealed during the litigation about the applicable crediting rates.
*191Pugh also stated in the June 2012 report that he could provide an opinion regarding the “current valuation of ‘underfunding.’ ” He explained that:
Plaintiffs seek damages equal to the loss caused by the failure of Wilmington to overfund the three insurance policies .... I have been asked to perform calculations commencing in 2005.1 have not performed a calculation which takes into account various costs and expenses related to the insurance policies, nor have I taken into account the payments made to the insurance carriers....
(emphasis added). Pugh clarified that, instead, he had “calculated the current value of annual payments of $5,500,000 commencing in January 2005 and concluding in January 2012.” Additionally, Pugh stated that at trial, he might “use a chart which depicts the variables and the conclusion described in [his] opinion.”
On July 3, 2012, the court granted the parties’ joint motion requesting an extension of certain deadlines regarding their expert witness disclosures. Under the court’s order, the parties were required by August 21, 2012 to provide supplemental expert witness reports and corrections to existing expert witness reports. The order further provided that discovery concerning expert witness opinions would conclude on October 23, 2012. Under the court’s pretrial procedures, the parties also were required to provide an “update of damages claimed or relief sought” no later than five days before the first day of trial.
Also in July 2012, Wilmington provided the plaintiffs with a copy of a report from Wilmington’s expert witness, Kevin Stephens, an accountant. Wilmington had retained Stephens to review the June 2012 report submitted by Pugh. Stephens criticized the June 2012 report for, among other things, its failure to incorporate calculations regarding the financing costs and costs of insurance associated with the premium financing arrangement. Stephens also provided exhibits in his report, which included: (1) financing costs between 2005 and 2020, under various scenarios applying differing financing rates; (2) the “actual cost of insurance and other policy expenses” for all three policies between the years 2005 and 201212; (3) the “guaranteed maximum cost of insurance rates” under each of the three policies; and (4) the projected cost of insurance for the different policies.13 According to Stephens, these figures should have been deducted from Pugh’s figures to calculate the account value, death benefit, and net-in-trust.
On October 29, 2013, the parties filed a Joint Proposed Pre-trial Order (Pretrial Order), in which the plaintiffs set forth their estimates of the present and future net-in-trust shortfalls caused by Wilmington’s failure to overfund. The Pretrial Order also indicated that Pugh would testify regarding “the reasonableness of the net in trust calculations,” and stated that Pugh would “present opinion testimony on the shortfall ... using the figures provided by [Stephens].” The next day, the court canceled the December 2013 trial date, and set the trial to begin on January 7, 2014.
In early November 2013, the plaintiffs provided Wilmington with a “thumb drive” containing one of the plaintiffs’ exhibits, labeled “PX174.” Exhibit PX174 depicted Pugh’s calculations of the net-in-trust shortfall, or the difference between the net-in-trust allegedly promised by Wil*192mington and the reduced net-in-trust resulting from Wilmington’s breach. Exhibit PX174 also included an interactive spreadsheet. An individual using the spreadsheet could enter values for certain variables to produce an outcome representing the net-in-trust. In addition to crediting rates, the formula in the spreadsheet contained estimates of variables such as the costs of insurance and other expenses, which Pugh had based in part on information provided by Stephens’s report. The plaintiffs concede that the formula was not included in the June 2012 report submitted by Pugh.
During a hearing that took place three days before trial on January 4, 2014, the district court declined Wilmington’s request to exclude Pugh’s testimony on the ground that exhibit PX174 contained a new formula that had not been disclosed before the deadlines for expert witness disclosures. The court noted that Wilmington had use of exhibit PX174, and the formula contained within that exhibit, for well over one month before the hearing, and that the court would have permitted Wilmington to depose Pugh again had Wilmington requested to do so.
The case proceeded to trial. Before Pugh was called to testify in the plaintiffs’ case-in-chief, the plaintiffs presented the testimony of Patricia Wilson, a Wilmington employee. Wilson testified regarding the value of the premium financing investment, stating that the actual loan balance on the investment was $13.6 million, and that the current insurance death benefit was around $50 million.
After this testimony, the plaintiffs presented Pugh as their expert witness on Friday, January 17, 2014. When Pugh stated that he was capable of updating his spreadsheet to incorporate Wilson’s figures, Wilmington objected. At that point, the district court limited Pugh’s testimony to the information he had provided before trial, but allowed him to inform the jury that application of Wilson’s figures would result in a decrease of the amount of damages Pugh had calculated.
As Pugh’s testimony unfolded, however, it became clear that Pugh needed to make additional changes to his calculations in light of Wilson’s testimony, in order to calculate more accurately the net-in-trust shortfall. When Pugh started to testify regarding future shortfall estimates, the district court stopped the questioning, stating that the testimony had devolved into “hopeless confusion.”14 The court ruled that Pugh needed to make changes to his spreadsheet by entering the information provided during Wilson’s testimony and making any other necessary updates. The court told counsel that the court was “concerned about [the] jury understanding what’s happening in [this] case.”
Accordingly, after Wilmington did not object to Pugh being excused as a witness to “work on [his] numbers” over the weekend before resuming his testimony, the plaintiffs proceeded in their case-in-chief and called their next witness to testify. After the weekend, the plaintiffs recalled Pugh to present his revised calculations. Pugh explained the changes he had made to his figures in response to some errors identified by defense counsel, and in response to financial information presented by Wilson in her trial testimony, including the actual loan balance and death benefit identified by Wilson.
Wilmington subjected Pugh to lengthy and rigorous cross-examination on both his original calculations and his amended figures, questioning him about matters including his choice of values for interest rates and the costs of maintaining the *193insurance policies. Wilmington did not present its own damages calculation.
Given this sequence of events, we agree with Wilmington that the plaintiffs violated the disclosure requirements for expert witnesses in Rule 26. See Fed. R. Civ. P. 26(a)(2)(D), (a)(3)(B). Exhibit PX174 contained a net-in-trust formula and associated calculations that Pugh considered in forming his ultimate opinion, which therefore should have been included in the June 2012 report. See Fed. R. Civ. P. 26(a) (2) (B) (i) - (iii). Even if we were to view exhibit PX174 as being a supplement to the June 2012 report, that exhibit was not provided to Wilmington until November 2013, well beyond the parties’ agreed deadline. See Fed. R. Civ. P. 26(a)(2)(D), (a)(3)(B).
Nonetheless, upon review of the Southern States factors, we conclude that the district court did not abuse its discretion in admitting Pugh’s testimony, because the plaintiffs’ noncompliance with Rule 26 was harmless in the context of the events that transpired.15 See S. States, 318 F.3d at 597. With regard to the first Southern States factor, we observe that any surprise resulting from the plaintiffs’ belated disclosure of exhibit PX174 and the net-in-trust calculation was minimal. Wilmington had been notified in a timely manner in the June 2012 report that Pugh would testify regarding the value that the policies would have yielded if Wilmington had met its overfunding commitments through 2019. The June 2012 report also stated that depending on additional information Pugh could obtain at a later date, he might use an updated chart or spreadsheet in the course of the litigation. Thus, Wilmington was aware from the June 2012 report that damages related to the plaintiffs’ anticipated net-in-trust was a central issue in the case, and that Wilmington’s decision whether to overfund the policies necessarily would impact the value of the policies and, consequently, the amount of the net-in-trust.
Also, it was undisputed that the very goal of the premium financing arrangement, which Wilmington emphasized in its proposals to Charlie, was to produce a sizable net-in-trust upon Fleur’s and Charlie’s deaths. Pugh ultimately updated his calculations from the June 2012 report. Some of these updates were reflected in exhibit PX174, in which Pugh incorporated certain cost estimates and other information included in Stephens’s report. In addition, Pugh’s testimony reflected changes made to incorporate information that Wilson, Wilmington’s own employee, presented in her testimony.
And, decisively, the collective result of these changes was a decrease in the amount of damages calculated. Thus, the present case does not involve a situation in which a defendant was “blindsided” by an expert witness’ testimony that damages would be greater, or from a different source, than the witness earlier had indicated. Pugh’s updated calculations actually decreased his estimates regarding the net-in-trust shortfall. Accordingly, we conclude that any surprise to Wilmington caused by the plaintiffs’ belated disclosure of exhibit PX174, and Pugh’s related testimony regarding the net-in-trust shortfall, was inconsequential. See Howe v. City of Akron, 801 F.3d 718, 748-50 (6th Cir. 2015) (concluding that plaintiffs’ late disclosure of expert witness’ back-pay calculations was harmless, in part because defendant possessed information relevant to calculations *194and knew plaintiffs were reconsidering their calculations, which plaintiffs had realized might be flawed).
We next consider the second Southern States factor, and address Wilmington’s ability to cure any purported surprise. See 318 F.3d at 597. We observe that Wilmington had access to exhibit PX174 and its associated net-in-trust formula for nearly two months before the trial began. During that period, Wilmington did not seek to depose Pugh or to take any other steps to mitigate the purported surprise caused by the plaintiffs’ delayed disclosure of the net-in-trust formula. Moreover, the record before us does not indicate that an earlier disclosure of exhibit PX174 and Pugh’s updated calculations would have enabled Wilmington to conduct additional cross-examination of Pugh or to introduce competing evidence at trial.16 See Davis v. U.S. Bancorp, 383 F.3d 761, 765 (8th Cir. 2004); see also Reese v. Herbert, 527 F.3d 1253, 1265 (11th Cir. 2008) (“[T]he expert disclosure rule is intended to provide opposing parties [a] reasonable opportunity to prepare for effective cross examination and ... arrange for expert testimony from other witnesses.” (internal quotation marks and citations omitted)). Thus, the record does not show that Pugh’s supplementary calculations and their timing affected Wilmington’s ability to conduct its defense in any material respect.
The third Southern States factor is “the extent to which allowing the evidence would disrupt the trial.” 318 F.3d at 597. The record before us does not indicate any significant disruption caused by the district court’s decision to admit exhibit PX174 and to permit Pugh to testify concerning his updated calculations. Cf. S. States, 318 F.3d at 598 (agreeing with district court’s finding that allowing “continuance to accommodate [expert witness’] third opinion would have significantly disrupted the trial,” because continuing and presenting case anew would render “much of the parties’ trial preparation ... obsolete”).
We observe pursuant to the fourth Southern States factor that Pugh’s testimony was vital evidence, because it provided the basis for the jury’s damages award. See id. at 597. And, with regard to the fifth Southern States factor, the plaintiffs have not provided a sufficient explanation for their delay in disclosing exhibit PX174 in a timely manner. See id. at 598-99. These two factors weigh in Wilmington’s favor in the Southern States analysis. Nonetheless, given that any surprise to Wilmington at trial was minimal, and that Wilmington had an opportunity to cure any surprise from exhibit PX174 that may have affected the preparation of its defense, we cannot say that the district court abused its discretion in declining to exclude exhibit PX174 and Pugh’s testimony under Rule 37. See id. at 597 (recognizing the district court’s “broad discretion to determine whether a nondisclosure of evidence is. substantially justified or harmless” under Rule 37(c)(1)).
B.
We next address Wilmington’s contention that the district court should have excluded Pugh’s testimony under the principles expressed by the Supreme Court in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 *195L.Ed.2d 469 (1993). Wilmington argues that Pugh’s testimony was inadmissible because his calculations: (1) erroneously incorporated cost of insurance values from Stephens’s model to calculate the “with overfunding” net-in-trust shortfall up to the date of trial; (2) used “an invalid interest spread” to project the future net-intrust shortfall; and (3) improperly discounted the future net-in-trust shortfall to present value.
We find no merit in Wilmington’s arguments. We review for abuse of discretion a district court’s evidentiary rulings regarding the reliability of an expert opinion. Bryte ex rel. Bryte v. Am. Household, Inc., 429 F.3d 469, 475 (4th Cir. 2005). We conclude that although Pugh’s testimony was unclear at various points during the trial, the district court did not abuse its discretion in declining to exclude Pugh’s testimony under Daubert.17
Under Rule 702 of the Federal Rules of Evidence, an expert witness’ testimony must, among other things, be “based on sufficient facts or data,” and must be “the product of reliable principles and methods.” Fed. R. Evid. 702(a)-(d). Courts are required to act as “gatekeepers” to ensure that expert testimony is relevant and reliable. Cooper v. Smith & Nephew, Inc., 259 F.3d 194, 199 (4th Cir. 2001) (citing Daubert, 509 U.S. at 588, 113 S.Ct. 2786).
In fulfilling its gatekeeping function, a district court “must conduct a preliminary assessment” to determine whether the methodology underlying the expert witness’ testimony is valid. Id. (internal quotation marks omitted) (quoting Daubert, 509 U.S. at 592-93, 113 S.Ct. 2786). District courts have “considerable leeway” in determining the manner in which they evaluate an expert witness’ reliability. Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). In assessing the validity of the methodology employed by a proposed expert witness, a court may consider whether the expert witness’ theory or technique: (1) “can be or has been tested”; (2) “has been subjected to peer review and publication”; (3) “has a high known or potential rate of error”; and (4) is generally accepted “within a relevant scientific community.” Cooper, 259 F.3d at 199.
To determine whether an opinion of an expert witness satisfies Daubert scrutiny, courts may not evaluate the expert witness’ conclusion itself, but only the opinion’s underlying methodology. TFWS, Inc. v. Schaefer, 325 F.3d 234, 240 (4th Cir. 2003). Moreover, “questions regarding the factual underpinnings of the [expert witness’] opinion affect the weight and credibility” of the witness’ assessment, “not its admissibility.” Structural Polymer Grp. v. Zoltek Corp., 543 F.3d 987, 997 (8th Cir. 2008).
In the present case, Wilmington’s Daubert challenge amounts to a disagreement with the values Pugh chose to assign to certain variables, including the cost of insurance and future interest rates. While Wilmington may have preferred that Pugh use higher costs of in-*196surance in his formula, along with other values more favorable to Wilmington, Pugh’s failure to do so did not require the district court to exclude Pugh’s opinion under Daubert. Cf. Tyger Constr. Co. v. Pensacola Constr. Co., 29 F.3d 137, 143 (4th Cir. 1994) (holding that district court abused its discretion in admitting expert opinion that conflicted directly with un-controverted record evidence). Rather, such challenges to the accuracy of Pugh’s calculations “affect the weight and credibility” of Pugh’s assessment, not its admissibility. See Zoltek Corp., 543 F.3d at 997.
III.
Wilmington also raises several arguments challenging the sufficiency of the evidence. Initially, Wilmington asserts that the evidence was insufficient to support the jury’s finding that Charlie did not agree to provide additional collateral after the initial $3.7 million payment. According to Wilmington, there could not have been a “meeting of the minds” between Charlie and Wilmington on this issue, because an agreement to lend Charlie $5.5 million annually without requiring him to make additional payments of collateral was “too good to be true.” Wilmington also maintains that the evidence did not support the jury’s conclusion that Wilmington agreed to overfund the policies by lending the Trust $5.5 million annually. Finally, Wilmington contends that the evidence failed to support the court’s order requiring Wilmington to return to Charlie’s estate the initial $3.7 million collateral payment, as well as the second collateral payment of $1.3 million made in 2005. Again, we disagree with Wilmington’s arguments.
We review de novo the district court’s denial of Wilmington’s motion for judgment as a matter of law, and consider the evidence in the light most favorable to the plaintiffs, the prevailing parties at trial. See Myrick v. Prime Ins. Syndicate, Inc., 395 F.3d 485, 489-90 (4th Cir. 2005). We “draw all reasonable inferences in [the plaintiffs’] favor without weighing the evidence or assessing the witnesses’ credibility.” Dennis v. Columbia Colleton Med. Ctr., Inc., 290 F.3d 639, 645 (4th Cir. 2002). Entry of judgment as a matter of law is appropriate only if the evidence is legally insufficient to support the jury’s verdict. Id.; see Fed. R. Civ. P. 50(a)-(b). Thus, we must affirm the district court’s denial of Wilmington’s motion if reasonable minds could differ regarding the findings contained in the jury’s special verdict. Dennis, 290 F.3d at 645.
Under Delaware law, a breach of contract claim has three elements: (1) a contractual obligation; (2) the defendant’s breach of that obligation; and (3) damage to the plaintiff caused by the breach.18 HM Wexford LLC v. Encorp, Inc., 832 A.2d 129, 140 (Del. Ch. 2003). In interpreting contractual provisions, Delaware courts seek to uphold the parties’ intention, looking first to the four corners of the agreement to determine whether that intention is evident from the express language in the contract. Paul v. Deloitte & Touche, LLP, 974 A.2d 140, 145 (Del. 2009). Delaware courts accord unambiguous terms their ordinary meaning. Id.
We reject Wilmington’s claim that there could not have been a “meeting *197of the minds” between Charlie and Wilmington, because the premium financing arrangement was “too good to be true.” As an initial matter, we do not dispute the principle that if an agreement’s terms “are so vague that a [cjourt cannot determine the existence of a breach, then the parties have not reached a meeting of the minds,” and no agreement exists. Cont’l Ins. Co. v. Rutledge & Co., Inc., 750 A.2d 1219, 1230 (Del. Ch. 2000). However, Wilmington has not identified any terms of the premium financing arrangement that were “so vague” that they rendered the jury unable to determine whether an agreement existed and, if so, whether a breach had occurred.
We agree with the district court that the plaintiffs presented sufficient evidence for the jury to conclude that Wilmington had agreed that Charlie would not be required to post any collateral other than the initial $3.7 million. In his deposition received at trial, Charlie testified unequivocally that he and Ianni agreed orally that additional collateral payments would not be required.19 Given the special verdict returned by the jury, we conclude that the jury accepted Charlie’s testimony on this issue as credible, and we will not reassess that credibility determination on appeal.
Additionally, evidence from other witnesses supported Charlie’s testimony on this point. This evidence included testimony from a Wilmington employee that unlike the spreadsheets Wilmington used with other customers, the spreadsheets Ianni sent to Charlie omitted a column identifying collateral payments. Moreover, the collateral pledge agreement signed by Charlie eliminated certain provisions contained in Wilmington’s standard templates for such agreements, and lacked any provision subjecting Charlie to ongoing collateral payment obligations.
The record also contains documentary evidence demonstrating that Ianni’s proposals did not mention ongoing collateral payment obligations, and that Ianni was aware that Charlie was unwilling to agree to any arrangement that would require him to post significant additional collateral. The jury also heard unrefuted evidence that Wilmington had filed a counterclaim in a separate civil action Ianni had brought against Wilmington. In that counterclaim, Wilmington alleged, in part, that Ianni had misrepresented to Charlie the requirements for posting collateral. From this extensive evidence, the jury reasonably could have concluded that to secure Charlie’s business, a Wilmington vice president, Ianni, had agreed that Charlie would not have to make annual collateral payments as security for Wilmington’s commitment to overfund the policies. See Myrick, 395 F.3d at 489-90; Dennis, 290 F.3d at 645.
Our conclusion is not altered by Wilmington’s suggestion that Charlie should have known that Ianni lacked the authority to waive annual collateral payments. When Wilmington promoted Ianni to the position of vice president, Wilmington emphasized that Ianni should be “presented and perceived as ... someone with credibility and in whom [prospective clients] ... can place their confidence.” Wilmington also had recognized Ianni for his success in selling premium financing products, referring to him at an annual meeting as “Moses parting the Red Sea” because of his successful promotion of lucrative premium financing products. In light of this evidence, a reasonable jury could infer that Wilmington had conferred extensive discretionary authority on Ianni, and that Charlie had no reason to think otherwise. Cf. Limestone *198Realty Co. v. Town & Country Fine Furniture & Carpeting, Inc., 256 A.2d 676, 679 (Del. Ch. 1969) (concluding that no binding contract was created when offeree had cause to question offeror’s authority, and “accept[ed] an offer which he should have known was unintended and on its face was too good to be true”).
We also reject Wilmington’s contention that the evidence failed to support the jury’s conclusion that Wilmington agreed to overfund the policies by lending $5.5 million annually to the Trust. Documentary evidence at trial overwhelmingly proved that Wilmington had agreed to make annual loans in that amount to the Trust. A multitude of letters and spreadsheets that Ianni sent to Shaiman and Charlie, as well as Wilmington’s internal documents from November 2003, referenced annual loans to the Trust in the amount of $5.5 million. Also, Ianni’s November 21 letter, which detailed key terms of the premium financing arrangement that ultimately was executed, stated that Wilmington “would loan to the trust the annual premium plus allowable overfund-ing ($5.5 million) to acquire and maintain that insurance in [the Trust].” A letter that Ianni sent to Shaiman in February 2005 similarly referenced Wilmington’s obligation to lend $5.5 million annually to the Trust. Viewing this evidence in the light most favorable to the plaintiffs, the jury had an ample basis for concluding that Wilmington had agreed to overfund the policies by lending $5.5 million annually to the Trust. See Myrick, 395 F.3d at 489-90.
We next consider Wilmington’s challenge to the district court’s decision ordering Wilmington to return to the Estate Charlie’s initial collateral payment of $3.7 million, as well as his later payment of $1.3 million in collateral made to prevent the insurance policies from lapsing. Because the district court resolved this issue post-trial, we review the supporting factual findings for clear error and the court’s conclusions of law, including contract construction, de novo. See Roanoke Cement Co. v. Falk Corp., 413 F.3d 431, 433 (4th Cir. 2005).
We find no error in the district court’s determination. As the court explained, the plain language of the investment management agreement required that Wilmington return all collateral upon Charlie’s death, stating that once Wilmington received notice of Charlie’s death, Wilmington “shall deliver to [Charlie], or the executor or administrator of [Charlie’s] estate, ... all of the property held by [Wilmington] hereunder.” Also, the jury had determined in its special verdict that Charlie was not required by the parties’ agreement to make additional collateral payments. Thus, based on the court’s interpretation of the parties’ contract and the jury’s well-supported factual finding that Charlie was not required under the parties’ agreement to make additional collateral payments, the court properly ordered Wilmington to return to the Estate the collateral payments that Charlie made in the total amount of $5 million. See Paul, 974 A.2d at 145.
In sum, we reject Wilmington’s challenges to the sufficiency of the evidence adduced at trial. We also affirm the district court’s order requiring Wilmington to return to the Estate the $5 million in collateral payments that Charlie had made.
IV.
We next consider Wilmington’s challenges to the jury’s damages award. Wilmington contends that the evidence did not support the jury’s finding that the plaintiffs were entitled to receive: (1) the $19.5 million award of damages representing the net-in-trust shortfall through 2019, if Wil*199mington were to make only the minimum premium payments through that date; and (2) an additional award of $3.9 million to compensate the plaintiffs for costs incurred in their purchase of the replacement policies.
In considering these issues, we apply a well-established standard of review. We will affirm a district court’s decision confirming a jury’s award of compensatory damages unless the verdict was contrary to the clear weight of the evidence, was based on false evidence, or would result in a miscarriage of justice. McMillan, 594 F.3d at 313-14 (citation omitted).
A.
With regard to the damages award for the net-in-trust shortfall, Wilmington maintains that the plaintiffs had not suffered any damages at the time of trial and, thus, that any purported damages were merely speculative in nature. Wilmington asserts that “[a]ny Net-in-TrUst shortfall payable at death is fictional, as it is based upon the timing of a death of an individual still alive,” namely, Fleur. Wilmington also argues that even if it breached an agreement to overfund the policies, there were too many unknown variables preventing the jury from making a reasonable estimate of the ultimate net-in-trust amount. Accordingly, Wilmington maintains that the plaintiffs must wait until Fleur dies to recover damages for any shortfall to their net-in-trust.20 We are unpersuaded by Wilmington’s arguments.
Under Delaware law, a non-breaching party is entitled to those damages that arise naturally from the breach or that were reasonably foreseeable at the time the parties entered into the contract. Paul, 974 A.2d at 146. As the Supreme Court of Delaware recently explained, expectation damages, or damages representing the parties’ reasonable expectation of the value of the contract at the time they entered into the contract, constitute the standard remedy for a breach of contract claim. Siga Techs., Inc. v. PharmAthene, Inc., 132 A.3d 1108, 1130, 1132 (Del. 2015) (citations omitted). Expectation damages typically provide the non-breaching party with the amount of money that would put him in the same position as if the breaching party had fulfilled his contractual obligations. Id. at 1130.
The non-breaching party must prove expectation damages to a reasonable degree of certainty, and may not recover damages that are speculative or uncertain. Id. at 1130-31 (citations omitted). Once the non-breaching party establishes the fact of damages to a reasonable degree of certainty, the amount of damages may be established with less precision. Id. at 1131.
We conclude that the plaintiffs proved the existence of damages to a reasonable degree of certainty and adequately proved the amount of damages awarded.21 See id. Wilmington’s failure to overfund the policies caused immediate damages, because the account value of the policies *200and, therefore, the expected death benefit, failed to grow as anticipated. Pugh projected a $19.5 million net-in-trust shortfall if Wilmington paid the minimum cost of the premium payments through 2019. Pugh’s figures relied, in part, on estimates regarding variables such as crediting rates, interest rates, and costs of insurance, as well as mortality tables estimating that Fleur would die in 2019. While Wilmington may disagree with the values Pugh selected in generating his calculations, this disagreement did not render the plaintiffs’ damages speculative. Moreover, Wilmington stipulated to Fleur’s life expectancy, and did not object when the court took judicial notice of the mortality tables that Pugh used in his calculations.22 Given these concessions, Wilmington may not now assert that the plaintiffs’ damages are speculative because the date of Fleur’s death is uncertain.
Pugh’s estimates regarding such factors as interest rates, crediting rates, and costs of insurance also did not render his testimony speculative. Reasoned assumptions and estimates about factors such as these are permitted in a breach of contract ease involving expectation damages, and may be attacked by a defendant through cross-examination or through the presentation of contrary evidence. See Siga Techs., 132 A.3d at 1111, 1122-24, 1135-37. Here, Wilmington thoroughly cross-examined Pugh regarding his calculations, and declined to exercise its right to present contrary expert witness testimony.
We find no merit in Wilmington’s separate assertion that the holding in American General Corp. v. Continental Airlines Corp., 622 A.2d 1 (Del. Ch. 1992), requires us to conclude that the damages award for the net-in-trust shortfall was speculative. There, unlike in the present case involving the future event of Fleur’s death, the court was presented with a contingency that may never have occurred. The court in American General was required to consider the damages, if any, that American General Corporation (American), was entitled to receive resulting from the failure of Continental Airlines Corporation (Continental) to provide American the same stock options that Continental provided to .its own employees, in connection with a merger agreement between Continental and its majority stockholder, Texas Air Corporation (Texas Air). Id. at 3, 5.
A central issue in making this determination was whether damages should be measured from the date that Texas Air stockholders elected to approve the employee stock option plan, or from the merger date, which took place three months before Texas Air stockholders approved the employee option plan. Id. The court concluded that it was appropriate to measure damages from the date on which Texas Air stockholders approved the employee option plan, because American employees did not have any right to receive those benefits until after the contingent event of Texas Air stockholder approval had occurred. Id. at 7-8.
This reasoning in American General is inapposite to our present determination. While the contingency in American Gener*201al, namely, the stockholders’ approval of the employee option plan, may never have materialized, the event at issue in the present case, that of Fleur’s death, is certain to occur. Therefore, the net-in-trust shortfall at issue here could reasonably be estimated because it was not measured with reference to a speculative future event, but with regard to the occurrence of a certain event in a year stipulated by the parties. Accordingly, we disagree with Wilmington’s argument that the holding in American General prohibits the plaintiffs from recovering net-in-trust damages until Fleur dies.
We therefore conclude that the plaintiffs provided sufficient proof of the existence and amount of their net-in-trust shortfall damages, which represented the amount of money required to place the plaintiffs in the position they reasonably expected to have achieved had Wilmington overfunded the policies as promised. See Siga Techs., 132 A.3d at 1130, 1131 n.128 (“[E]stimates that lack mathematical certainty are permissible so long as the court has a basis to make a responsible estimate of damages.” (citation omitted)); see also 24 Williston on Contracts § 64:10 (4th ed. 2016) (“Most contracts are entered into with the view to future profits, and such profits are in the contemplation of the parties; and, so far as they can be properly proved, they may form the measure of damage.”). Accordingly, we affirm the portion of the damages award representing the net-in-trust shortfall, because that award was not reached against the clear weight of the evidence, and would not result in a miscarriage of justice. See McMillan, 594 F.3d at 313-14.
B.
Finally, we address Wilmington’s challenge to the jury’s award of $3.9 million for costs incurred by the plaintiffs in obtaining the replacement policies. Wilmington asserts that the plaintiffs were not entitled to these damages, because the plaintiffs “incurred these costs to protect against losing their claims already in litigation.” 23 Wilmington’s contention is unavailing.
In reaching our conclusion, we are guided by the principle, noted above, that a non-breaching party is entitled to damages that were reasonably foreseeable at the time the parties entered into the contract. Paul, 974 A.2d at 146. In the present case, Fleur and Charlie obtained replacement life insurance policies on Fleur’s life in October 2009, because Wilmington had failed to overfund the policies since 2004, and Fleur and Charlie were concerned that Wilmington would allow their original policies to lapse. The replacement policies carried a fixed death benefit of $17.5 million. From 2009 through 2014, when the trial took place, Fleur had paid around $7.75 million in premiums for the replacement policies. A witness for Wilmington estimated that the market value of Charlie’s and Fleur’s policies was around twenty-two percent of the policies’ face value, or around $3.85 million. Thus, as the district court explained, subtracting the $3.85 million market value of the policies from the $7.75 million in premiums that Fleur paid from 2009 through 2014 yielded $3.9 million, which is the amount that the jury awarded to the plaintiffs for damages associated with obtaining the replacement policies.
We conclude that it was reasonably foreseeable at the time the parties entered into the premium financing arrangement that if Wilmington were to breach the agreement *202by failing to overfund the policies, the Breslers would need to seek replacement policies to secure the insurance benefits they had expected to realize but would no longer receive as a result of Wilmington’s breach. See id. at 146-47. We therefore hold that the jury properly awarded the plaintiffs $3.9 million in consequential damages, and that this award was neither contrary to the clear weight of evidence nor one that would cause a miscarriage of justice.24 McMillan, 594 F.3d at 313-14.
V.
For these reasons, we affirm the district court’s judgment.
AFFIRMED
. In accordance with the naming conventions in the plaintiffs’ brief, we refer to Charles Bresler as "Charlie.”
. The funds in the policies that exceed the cost of insurance constitute the "account value.”
. Charlie’s son, Sidney, held various officer-level positions at B&R, and became the company's CEO in 2005.
. In a "second-to-die" life insurance policy, the insurer pays the death benefit to the policy beneficiaries upon the death of both insureds.
. The London Interbank Offered Rate, or LI-BOR, is a fluctuating benchmark interest rate used by several banks around the world. In-vestopedia, http://www.investopedia.com/ terms/l/libor.asp (last visited Jan. 27, 2017).
. Also in 2005, Wilmington terminated Ianni, stating, among other things, that he had "lost credibility” and had "engaged in questionable sales practices.”
. Highland Capital Brokerage (Highland), a Highland representative, Ianni, and two other Wilmington employees also were named as defendants in the action. The district court later dismissed the claims against Ianni. The claims against the Highland representative eventually were dismissed, and Highland later was dismissed through a settlement agreement. The other two Wilmington employees remain parties to the litigation.
. The court held in. abeyance the non-contract claims pending the outcome of the breach of contract claims. Following the trial, the court decided to continue holding these claims in abeyance pending the outcome of the present appeal, and severed and assigned the claims a new docket number.
. We recognize that the net-in-trust will be disbursed to the ILIT’s beneficiaries after Fleur dies.
. After the jury rendered its verdict, the district court resolved in the plaintiffs' favor their claim that under the investment management agreement, Wilmington was required to return to the Estate the collateral that was being held in the investment management account.
. Pugh explained that a present value is "the current worth of a future sum of money or stream of cash flows given a specified rate of return.”
. Stephens noted that "[a]nnual statements for the various policies we[re] not available for each year.”
. At trial, Stephens testified that he intended his projected cost of insurance figures "to represent a floor amount,” rather than the actual costs of insurance.
. The court made this remark to counsel during a bench conference.
. In addressing whether to admit the plaintiffs' delayed disclosures, the district court did not discuss individually the factors enumerated in Southern States. However, the district court was not required to do so. Wilkins, 751 F.3d at 222.
. Wilmington did not present its own damages calculation to the jury, relying instead on its attempts to discredit Pugh’s calculations. Notably, Wilmington does not contend that its failure to present its own calculations was a result of the plaintiffs’ delayed disclosures, rather than a strategic decision that Wilmington made during the course of the litigation.
. Wilmington also asserts that, in forming his opinions, Pugh impermissibly relied on Randy Whitelaw, a professional in the insurance industry with experience in premium financing, because Whitelaw did not testify and Wilmington thus was unable to cross-examine him. We disagree. While an expert witness may not bolster the reliability of his own opinion by testifying about a non-testifying expert witness’ credentials and opinion, see United States v. Tran Trong Cuong, 18 F.3d 1132, 1144 (4th Cir. 1994), Pugh relied on Whitelaw for information about general premium financing concepts, and did not rely on the opinions Whitelaw prepared for the present litigation. See Fed. R. Evid. 703.
. The Trust Agreement contains a choice of law provision identifying Delaware law as governing contract disputes, and the parties agree that Delaware law governs our resolution of the breach of contract claims, including the parties’ disputes regarding damages. Accordingly, we apply Delaware law here. See Chorley Enters., Inc. v. Dickey’s Barbecue Restaurants, Inc., 807 F.3d 553, 563 n.11 (4th Cir. 2015).
. Because Charlie died before the trial began, portions of his videotaped deposition, taken in August 2010, were played for the jury at trial.
. Wilmington does not challenge the court’s order of specific performance, which requires Wilmington to lend to the Trust sufficient funds to pay the minimum premiums required to maintain the policies through Fleur’s life.
. We reject Wilmington's argument, raised for the first time on appeal, that the plaintiffs are not entitled to damages because the plaintiffs failed to demonstrate an "injury.” Despite Wilmington's attempt to insert an "injury” element into a breach of contract claim, under Delaware law a breach of contract claim does not include a distinct "injury” element. See H-M Wexford LLC, 832 A.2d at 140.
. In its brief, Wilmington relies on In re New York, N.H. & H.R. Co., 92 F.2d 428 (2d Cir. 1937), to argue that Pugh's use of mortality tables for his damages calculation "was speculative and improper.” As noted above, Wilmington did not object when the court took judicial notice of the mortality tables, and Wilmington also conceded at oral argument that it is appropriate for courts to use mortality tables in calculating damages. We therefore decline to address Wilmington’s challenge to Pugh's use of mortality tables in calculating the net-in-trust shortfall damages. See Weidman v. Exxon Mobil Corp., 776 F.3d 214, 220 (4th Cir. 2015).
. We decline to address Wilmington’s argument that these damages were improper under a theory of anticipatory repudiation. Wilmington raises this argument for the first time on appeal, and it is therefore waived. See Weidman, 776 F.3d at 220.
. We also note that Wilmington fails to cite any authority to support its contention that the plaintiffs were not entitled to the $3.9 million award for costs incurred obtaining the replacement policies.