Fleur Bresler v. Wilmington Trust Company

WYNN, Circuit Judge,

concurring in part and dissenting part:

I agree with the majority opinion that sufficient evidence supported the jury’s conclusions (1) that Defendant Wilmington Trust Company (“Wilmington Trust”) and Charlie Bresler (“Charlie”) entered into a “premium financing” contract; (2) that, pursuant to that contract, Wilmington Trust agreed to lend $5.5 million per year to an irrevocable trust established by Charlie to pay premiums on and “ov-erfund” three life insurance policies purchased by the trust; and (3) that the contract required that Wilmington Trust return collateral posted by Charlie upon Charlie’s death. I also agree with the majority opinion that disclosures and damages testimony by Plaintiffs’ accounting expert, violated Federal Rule of Civil Procedure 26 in numerous ways.

Notwithstanding Plaintiffs’ myriad violations of Rule 26, the majority opinion concludes that the district court did not abuse its discretion in admitting the damages testimony on grounds that Plaintiffs’ noncompliance was “harmless.” See Fed. R. Civ. P. 37(c)(1). But the district court never concluded that Plaintiffs failed to comply with Rule 26, and therefore never exercised its discretion to admit the evidence on harmlessness grounds. Accordingly, plenary, rather than abuse-of-discretion, review applies to the district court’s decision to admit the damages testimony. Indeed,' the application of deferential review is particularly unwarranted because Plaintiffs repeatedly mislead, intentionally or otherwise, the district court regarding information essential to determining whether Plaintiffs complied with Rule 26 and whether any violation of Rule 26 was harmless. In so doing, Plaintiffs deprived the district court of the opportunity to exercise its discretion in an informed manner, undermining the rationale for applying abuse-of-discretion review on appeal.

Under this Court’s precedent, the damages testimony — which Plaintiffs first disclosed after the close of expert discovery, after the court had ruled on the parties’ motions for summary judgment, and after the deadline for submitting pre-trial motions — was not harmless. See Wilkins v. Montgomery, 751 F.3d 214, 223 (4th Cir. 2014). On the contrary, Plaintiffs’ noncompliance with Rule 26 deprived Wilmington Trust of the opportunity to depose the expert regarding his damages opinion, to prepare rebuttal reports, and, therefore, to effectively challenge the expert’s damages testimony at trial — testimony from which the jury directly drew its multi-million dollar damages award. “[I]t would be a miscarriage of justice to allow [the] award to stand, where that award was brought about by plaintiffs’ misleading the court.” *203Diaz-Fonseca v. Puerto Rico, 451 F.3d 13, 37 (1st Cir. 2006). Accordingly, I dissent.

I.

A.

Charlie Bresler eo-owned a successful real estate development company. By 2003, Charlie was nearing the end of his life and had a net worth in excess of $150 million. At the same time, Wilmington Trust was trying to expand its personal trust, investment services, and wealth management businesses. As part of that effort, Wilmington Trust approached Charlie with various “premium financing” proposals, which Wilmington Trust touted as vehicles for Charlie to minimize taxes on his estate.

Under the premium financing plans that Wilmington Trust proposed, Wilmington Trust would make loans to a tax-sheltered irrevocable trust. The trust would then use the proceeds of the loans to purchase, and pay premiums on, second-to-die life insurance policies, which would pay out on the death of Charlie’s wife, Fleur. In addition to making the loans, Wilmington Trust also would administer the trust.

The life insurance policies at issue have two components: a face value and an investment account. The face value is a set amount that the policy will pay upon the insured’s death if the owner pays the premium each year. The owner of the policy can also “overfund” the policy by contributing money into the investment account above and beyond the annual premium, which then increases the death benefit. Overfunding contributions also increase the cash value of the policy at a faster rate than premium payment alone. Under Wilmington Trust’s proposed plan, the trust would use the proceeds from the death benefit and investment account, paid out to the trust upon Fleur’s death, to repay Wilmington Trust’s loans with interest.

In November 2003, Charlie and Wilmington Trust agreed on a multi-million dollar premium financing plan, but the parties never reduced their agreement to writing. The parties agree that under the oral arrangement Wilmington Trust would lend the trust money to pay premiums on three second-to-die life insurance policies with a combined face value of $50 million and payable to the trust on Fleur’s death. Plaintiffs asserted — and the jury found— that Wilmington Trust agreed to lend the trust $5.5 million each year to pay premiums on the policies and, if the annual premiums were less than the $5.5 million annual lending commitment, to overfund the policies.

In 2004, in advance of its first $5.5 million loan to the trust, Wilmington Trust requested, and Charlie posted, $3.7 million in collateral. In 2005, Wilmington Trust requested that Charlie post additional collateral before Wilmington Trust made its second $5.5 million loan to the trust. Charlie responded that the premium financing agreement only contemplated the original $3.7 million in collateral and threatened legal action if Wilmington Trust declined to lend to the trust. In order to avoid having the policies lapse, Charlie nonetheless posted an additional $1.3 million in collateral. The jury found that the premium financing agreement did not require Charlie to pledge this, or any, additional collateral.

On September 16, 2009, Charlie brought suit in Maryland state court against Wilmington Trust and several individuals and entities not party to this appeal. Wilmington Trust removed the case to federal court. The complaint asserted twelve causes of action, including claims for negligence, breach of fiduciary duty, breach of contract, fraud, and civil conspiracy. Only the breach of contract claim proceeded to trial.

*204B.

The district court entered the operative scheduling order on July 3, 2012. Under that order, Wilmington Trust had to file its Rule 26 expert reports by July 16, 2012. Plaintiffs designated Robert Pugh, CPA, as an expert in tax and present-value calculations and submitted his report on June 6, 2012. Pugh’s five-and-a-half page report provided four “calculations.” J.A. 701-07. The second and third calculations — entitled “present value” and “current valuation of underfunding” — estimated the value, as of the date of the report, of Wilmington Trust’s alleged $5.5 million annual lending commitment during Fleur’s anticipated lifespan. During his deposition, Pugh explained that these calculations involved “basically just taking money over time at a certain interest rate and calculating what it would be today.” J.A. 745-46.

To that end, the “present value” calculation set forth the formula for discounting money to present value and then, using an estimate of interest rates, calculated the present value of the $5.5 million lending commitment at $33.5 million for 2013 to 2019, the period running from the date of trial to the end of Fleur’s expected lifespan. The “current value of underfunding” formula “calculated the current value of annual payments of $5,500,000 commencing in January 2005 and concluding in January 2012.” J.A. 3262. Accordingly, Pugh’s calculations assumed that the entire $5.5 million loan would be used to overfund the policies, notwithstanding (1) that the premium financing agreement between Wilmington Trust and Charlie required that a portion of the loan be used to pay the premiums on the life insurance policies and (2) that Wilmington Trust never breached its obligation to lend the trust sufficient funds to pay the premiums required to keep the policies in force. Pugh’s report acknowledged that he had “not performed a calculation which takes into account various costs and expenses related to the insurance policies, nor [had he] taken into account the payments made to the insurance carriers, the commissions paid by the insurance carriers, or the effect of underfunding on the cash values or the cash surrender values of the policies.” Id. With these caveats, Pugh’s report calculated the current value of Wilmington Trust’s overfunding commitment at $51.7 million for the 2005 to 2012 period.

Under the scheduling order, Plaintiffs had to file their Rule 26 rebuttal expert reports by August 10, 2012; both parties had to file their supplemental expert reports by August 21, 2012; and expert discovery closed on October 23, 2012. Pursuant to that order, Wilmington Trust submitted the expert report of Kevin Stephens, CPA, on July 16, 2012, more than a month before the deadline for submission of supplemental reports and three months before the close of expert discovery. Stephens’s report challenged several of Pugh’s assumptions regarding future interest rates and argued that Pugh’s report did not provide a valid damages calculation. Plaintiffs did not file a rebuttal to Stephens’s report or a supplemental report prepared by Pugh.

On September 20, 2012, Wilmington Trust deposed Pugh solely based on his June 6, 2012, report — the only report by Pugh that Plaintiffs disclosed. During his deposition, Pugh explained the present value and current value of underfunding calculations in his report as follows:

What I presented [in my report] is a model for which accomplishes a present and future value calculation of whatever variables go in. Those variables would be a combination of many things, not just the premium payments of $5.5 million but other things that I just didn’t have readily available to factor into this *205report. So I wasn’t asked to consider all factors and calculate a net result.

Pls.’ Resp. Opp. to Mot. In Lim. of Defendants Wilmington Trust Company and Wilmington Brokerage Services Company To Exclude the Expert Ops. of All Pls.’ Proffered Experts, Bresler v. Wilmington Trust Co., No. 8:09-cv-02957, Doc. No. 520-1, at 81 (D. Md. Dec. 3, 2013) (the “Motion In Limine Opposition”) (emphasis added).

On October 29, 2013, more than a year after expert discovery had closed and two months béfore trial, Plaintiffs stated in their proposed pre-trial order that they would be seeking damages based on the difference, with and without overfunding, in the “net in trust” — the difference between (1) the death benefit and investment account at the time of Fleur’s expected death and (2) the amount due to Wilmington Trust on the loans. Plaintiffs stated that they would rely on Pugh for an opinion as to this difference. A few days later, Plaintiffs disclosed a trial exhibit, PX-174, which was an Excel spreadsheet prepared by Pugh in which Pugh calculated his estimate of the alleged net-in-trust shortfall. In particular, the exhibit estimated an $11.2 million net-in-trust shortfall for the 2005 to 2013 period and an $8.7 million net-in-trust shortfall for the 2014 to 2019 period.

Wilmington Trust moved in limine to exclude PX-174 and to bar Pugh from testifying as to net-in-trust damages. Wilmington Trust argued that the proposed testimony and exhibit did not comply with Federal Rule of Civil Procedure 26 because it was outside the scope of Pugh’s expert report. In particular, Wilmington Trust asserted that Pugh’s report calculated only the “valuation of the purported underfunding” — not the net-in-trust shortfall — and that the report’s calculation did not incorporate — indeed, expressly declined to provide opinions as to the value of — several variables and factors necessary to estimate the net-in-trust shortfall. These variables included: “ ‘various costs and expenses related to the insurance policies’; ‘the payments made to the insurance carriers’; ‘the commissions paid by the insurance carriers’; and, ‘the effect of underfunding on the cash values or the cash surrender values of the policies.’ ” Mem. Supp. of Daubert Mot. Of Defs. Wilmington Trust Co. & Wilmington Brokerage Services Co. To Exclude the Expert Opinions of All of Pls.’ Proffered Expert Witnesses, Bresler v. Wilmington Trust Co., No. 8:09-cv-02957, Doc. No. 501-1, at 12 (D. Md. Nov. 1, 2013) (the “Motion In Limine Memorandum”) (quoting Pugh Report at 4). Wilmington Trust further argued that the net-in-trust exhibit and testimony regarding net-in-trust damages was inadmissible because Plaintiffs could not meet their burden to show that their noncompliance with Rule 26 was “substantially justified or harmless” under the five factors set forth in Southern States Rack & Fixture, Inc. v. Sherwin-Williams Co., 318 F.3d 592, 597 (4th Cir. 2003).

In their memorandum in opposition to Wilmington Trust’s motion to exclude, Plaintiffs maintained that Pugh’s report provided the “formula” Pugh used to estimate net-in-trust damages, and that PX-174 simply would allow “the jury[ ] to ‘populate’ or plug in the correct variables as of the trial to apply the formula and reach the correct value.” Motion In Limine Opposition, at 21. Plaintiffs further asserted that two of the variables Pugh relied on in PX-174 to calculate net-in-trust damages, LIBOR rates and annual cost-of-insurance figures, were discussed during Pugh’s deposition and introduced in Stephens’s report, respectively, and therefore that ‘Wilmington [Trustj’s assertion of ‘surprise’ is not supported by the discovery in this case.” Id. at 29-31. Plaintiffs also stated *206that Wilmington Trust could “satisfy none of the five factors enunciated in [Southern States],” but, with the exception of the “surprise” factor, Plaintiffs did not specifically address any of those five factors. Id. at 31.

At oral argument on the motion, Wilmington Trust said it was prejudiced by Plaintiffs’ failure to provide Pugh’s net-intrust damages calculation and exhibit PX-174 because Wilmington Trust never had a chance to depose Pugh regarding his opinion as to net-in-trust damages — and the variables and formulas underlying that opinion — or to submit a rebuttal expert report challenging Pugh’s net-in-trust damages estimate. In response, Plaintiffs again told the district court that Pugh’s disclosed report had provided the formula to calculate net-in-trust damages, and that exhibit PX-174 simply “plugged in” new numbers to that formula. See, e.g., J.A. 1390-91 (“[W]hat [Wilmington Trust] got in the report was the formula.... All this chart does is plug the new numbers or the old numbers into the formulas that he gave his opinion on. That’s all it does.”). Plaintiffs said that they could not have provided Pugh’s net-in-trust analysis earlier because they did not have access to key figures necessary to make those calculations, including cost-of-insurance information.

The district court orally denied Wilmington Trust’s request to exclude PX-174 and limit Pugh’s testimony to the opinions expressed in his report. The district court did not find that Pugh’s proposed net-intrust testimony violated Rule 26. On the contrary, the district said it “d[id]n’t know that the rule about providing expert testimony says that every exhibit that he’s going to use when he gives his testimony has to be provided at or about the time that he gives his report.” J.A. 1393. The district court likewise did not expressly hold that any failure of Plaintiffs to comply with Rule 26 was substantially justified or harmless. Rather, the court noted that he would have allowed Wilmington Trust to depose Pugh if Wilmington Trust had so requested and that. Wilmington Trust would have a chance “chop th[e] exhibit apart” on cross-examination, including by establishing that the exhibit was not based on the opinions provided in the report. J.A. 1393-94.

The parties tried the case before a jury. Notwithstanding Wilmington Trust’s renewal of its motion to exclude, Pugh testified as to his opinion regarding Plaintiffs’ net-in-trust damages. Pugh’s opinion changed several times during the course of his direct testimony and cross-examination. Pugh first estimated an $11.2 million net-in-trust shortfall for the 2005 to 2013 period and a $6 million net-in-trust shortfall for the 2014 to 2019 period (assuming Wilmington Trust immediately began over-funding), approximately $2.7 million less than the estimate included in the exhibit when it was first disclosed. During cross-examination, Pugh conceded several errors and offered to revise his estimates “on the fly.” J.A. 2049. At that point, the district court cut off Pugh’s testimony, stating:

This case is in rather hopeless confusion right now with this witness. And what I’m going to let him do, we can pass him now and pick up another witness and let’s see what he can calculate.... [F]rankly, it’s just not clear where he is. He’s sort of stammering on what to do with these new numbers. I think he may not be able to come up with any number at all.... From my standpoint, I can’t follow this now.

J.A. 2049-50. Five days later, Plaintiffs called Pugh back to the stand and introduced a new exhibit, PX-174B, in which Pugh offered revised estimates of net-intrust damages. Notwithstanding that he added data from 2004, Pugh revised his *207estimate for the pre-2013 period downward to approximately $10.7 million as a result of errors in his original estimate identified on cross-examination. Pugh revised upward his estimate for the 2014 through 2019 period to approximately $7.1 million (assuming Wilmington Trust immediately began overfunding). Plaintiffs did not provide Wilmington Trust with the data and formulas Pugh used to generate PX-174B.

The jury returned a verdict in favor of Plaintiffs, awarding more than $23 million in damages. The jury drew its damages award — and the calculation of the net-intrust shortfall, in particular — directly from Pugh’s estimates in PX-174B.

Wilmington Trust filed a Rule 50 motion to set aside the verdict on numerous grounds, including Plaintiffs’ failure to comply with Rule 26. In their memorandum in opposition, Plaintiffs again asserted that they had fully complied with Rule 26. And Plaintiffs again maintained that Pugh’s June 6, 2012, report “provided formulas for his calculations, and sample calculations” and that Pugh’s calculations in the exhibits and in his testimony at trial “merely utilized the numbers provided by Stephens and [Wilmington Trust’s] counsel to execute the calculations he described in his report.” Mem. Law in Opp. to Mot. of Wilmington Trust Co. for Judgment as a Matter of Law, or Alternatively To Amend the Verdict, Set a New Trial and/or Require Remittitur, Bresler v. Wilmington Trust Co., No. 8:09-cv-02957, Doc. No. 608-1, at 32-33 (D. Md. April 3, 2014) (emphasis added).

This appeal followed. In their brief to this Court, Plaintiffs continued to assert that they had complied with Rule 26. Ap-pellee’s Br. at 25 (“[Plaintiffs] properly disclosed Pugh’s basic methods and approach in discovery and timely disclosed his damages calculations and detailed illustrative exhibits under applicable deadlines in this case.”). In particular, Plaintiffs again maintained that Pugh’s report complied with Rule 26 because it “showed his formulas and assumptions for calculating the past shortfall and projecting the future shortfall.” Id. at 27. Plaintiffs also reasserted that Pugh could not have provided his estimates earlier because Wilmington Trust had not made the necessary information available. Id. at 27-29. Plaintiffs further argued that the district court did not err in refusing to exclude Pugh’s damages testimony because exclusion is mandatory “only ‘if a party fails to provide information ... as required by Rule 26(a) or (e).’ ” Id. at 29 (quoting Fed. R. Civ. P. 37(c)(1)) (alteration in original). According to Plaintiffs, PX-174 “fully complied with Rule 26(e),” which provides for admission of “supplemental” expert reports, “obviating] any need even to consider sanctions, much less complete exclusion.” Id.

When pressed during oral argument, however, Plaintiffs conceded that — contrary to their repeated representations to both the district court and this Court— Pugh’s report did not provide a formula for calculating the alleged net-in-trust shortfall, and therefore that the “numbers” provided by Stephens and Wilmington Trust could not have been “plugged in” to any formula in Pugh’s report to determine his estimate of net-in-trust damages.

II.

Wilmington Trust argues that the district court improperly allowed Pugh to testify as to net-in-trust damages when (1) Plaintiffs did not provide Pugh’s net-intrust damages calculation in their Rule 26 disclosures, (2) Pugh’s expert report did not address net-in-trust damages, let alone provide a formula for calculating net-intrust damages, and (3) Plaintiffs submitted exhibit PX-174, which first provided information on Pugh’s net-in-trust damages cal*208culation, more than a year after expert discovery closed. I agree.

A.

As the majority opinion correctly states, we generally review for abuse of discretion a district court’s decision regarding whether a party violated a discovery rule. Russell v. Absolute Collection Servs., Inc., 763 F.3d 385, 396 (4th Cir. 2014). We also generally review for abuse of discretion a district court’s decision regarding whether to impose sanctions for a discovery violation. S. States, 318 F.3d at 595. We subject a district court’s discovery decisions to deferential review because the district court “has an intimate familiarity with the relevant proceedings,” Houston v. C.G. Sec. Servs., Inc., 820 F.3d 855, 858 (7th Cir. 2016), and, therefore, is in a “superior[] ... position to supervise the litigants and assess their good faith,” Lolatchy v. Arthur Murray, Inc., 816 F.2d 951, 957 (4th Cir. 1987) (Wilkinson, J., dissenting) (internal quotation marks omitted).

But the rationale for such deferential review does not exist when a district court “suffer[s] under a misconception that prevented a genuine application of [its] discretion to all facets of th[e] case,” particularly when that misconception is attributable to the conduct or representations of the party that benefitted from the district court’s misinformed exercise of its discretion. Dakota Indus., Inc. v. Dakota Sportswear, Inc., 988 F.2d 61, 63 (8th Cir. 1993) (refusing to defer to factual findings of district court supporting denial of motion for preliminary injunction when party against whom injunction was sought “misrepresent[ed], willful[ly] or otherwise” key facts and thereby “prevented the court from fully and fairly deciding whether a preliminary injunction should be issued”); Mangini v. United States, 314 F.3d 1158, 1161 (9th Cir. 2003) (subjecting district court’s denial of a disqualification motion to plenary — rather than abuse-of-discretion — review because party opposing disqualification motion “fail[ed] to provide [district court] with all the facts[,] depriving] him of the opportunity to exercise his informed discretion”). Put differently, when a party’s misrepresentations deprive a district court of the opportunity to exercise its discretion in an informed manner, the justification for appellate deference to the district court’s decision no longer exists.

Here, Plaintiffs repeatedly misled, intentionally or otherwise, the district court regarding information essential to its disposition of Wilmington Trust’s motion to exclude Pugh’s net-in-trust damages testimony. For example, Plaintiffs repeatedly told the district court that Pugh’s report disclosed the “formulas” Pugh used to calculate his opinion as to net-in-trust damages — and to estimate the variables underlying that calculation — and that exhibit PX-174 simply “plugged” newly available data into those formulas. Yet when pressed at oral argument, Plaintiffs conceded — as the record demonstrates — that Pugh’s report did not include the formulas for calculating net-in-trust damages and, therefore, that PX-174 did not simply involve “plugging in” newly available data to previously disclosed formulas.

Likewise, Plaintiffs repeatedly told the district court that Pugh could not have provided his estimates of net-in-trust damages before the close of expert discovery because the data necessary to make those calculations — including data necessary to calculate future interest and crediting rates, expenses, financing costs, and cost-of-insurance — was unavailable. But Pugh’s net-in-trust damages opinion relied on actual expense figures and premium financing and cost-of-insurance assumptions included in Stephens’s report, which Wilmington Trust timely filed well before the *209close of expert discovery and before the scheduling deadline for rebuttal and supplemental expert reports, i.e., more than a year before Plaintiffs disclosed Pugh’s damages estimate. Likewise, PX-174 and Pugh’s trial testimony involved interest rate projections using historical LIBOR rates and projections using historical crediting rates, which Pugh could have made at the time he submitted his initial report, but did not. And most glaringly, Plaintiffs maintained — both before the district court and this Court — that actual cost-of-insurance information from the three insurance carriers was not available to Plaintiffs, and therefore Pugh, before the close of expert discovery. Yet, again contrary to Plaintiffs’ assertions, Appel-lee’s Br. at 28-29, Wilmington Trust granted Plaintiffs access to cost-of-insurance information from all three carriers years before Pugh submitted his initial report, J.A. 3284, 3287, 3290 (“This authorization shall be in force and effective until the later of December 31, 2009 or notification by us.” (emphasis added)).

By misrepresenting the formulas and opinions contained in Pugh’s 2012 report and the timing of the availability of the data Pugh used to prepare his net-in-trust estimates, Plaintiffs deprived the district court of information necessary to determine whether Plaintiffs failed to comply with Rule 26 and the district court’s expert discovery scheduling order. Likewise, these misrepresentations deprived the district court of information necessary to determine whether any noncompliance with either Rule 26 or the scheduling order was substantially justified or harmless — the standard for determining whether a court must exclude noncompliant expert testimony. Because Plaintiffs’ misrepresentations deprived the district court of information necessary to exercise its informed discretion, we review de novo — rather than for abuse of discretion — (1) whether Plaintiffs violated Rule 26 and (2) whether any such violation warranted exclusion of Pugh’s testimony as to net-in-trust damages. See Dakota Indus., 988 F.2d at 63; Mangini, 314 F.3d at 1161.

B.

Regarding whether Pugh’s damages testimony violated Rule 26, Rule 26(a)(1)(A)(iii) requires that a party provide “a computation of each category of damages claimed by the disclosing party— who must also make available for inspection and copying as under Rule 34 the documents or other evidentiary material, unless privileged or protected from disclosure, on which each computation is based.” A plaintiff must disclose its damages calculation within 14 days after the parties’ Rule 26(f) conference, “unless a different time is set by stipulation or court order.” Fed. R. Civ. P. 26(a)(1)(C).

Rule 26(a)(2)(B) requires that expert witnesses provide a written report that “must contain,” among other things: “(i) a complete statement of all opinions the witness will express and the basis and reasons for them; (ii) the facts or data considered by the witness in forming them; [and] (iii) any exhibits that will be used to summarize or support them.” And under Rule 26(e)(1)(A), “[a] party who has made a disclosure under Rule 26(a) ... must supplement or correct its disclosure or response ... in a timely manner 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.”

“Rule 26 disclosures are often the centerpiece of discovery in litigation that uses expert witnesses.” Saudi v. Northrop Grumman Corp., 427 F.3d 271, 278 (4th *210Cir. 2005). “The purpose of Rule 26(a)(2) is to provide notice to opposing counsel— before the deposition — as to what the expert witness will testify.” Ciomber v. Coop. Plus, Inc., 527 F.3d 635, 642 (7th Cir. 2008). Likewise, the Rule “prevent[s] unfair surprise at trial and [ ] permit[s] the opposing party to prepare rebuttal reports, to depose the expert in advance of trial, and to prepare for depositions and cross-examination at trial.” Minebea Co. v. Papst, 231 F.R.D. 3, 5-6 (D.D.C. 2005). “The Rule also prevents experts from ‘lying in wait’ to express new opinions at the last minute, thereby denying the opposing party the opportunity to depose the expert on the new information or closely examine the expert’s new testimony.” Id. at 6.

To that end, the Rule “mandates a complete and detailed report of the expert witness’s opinions, conclusions, and the basis and reasons for them.” Ciomber, 527 F.3d at 642. “Expert reports must not be sketchy, vague or preliminary in nature” and “must include ‘how* and ‘why’ the expert reached a particular result, not merely the expert’s conclusory opinions.” Salgado by Salgado v. Gen. Motors Corp., 150 F.3d 735, 741 n.6 (7th Cir. 1998). “A party that fails to provide these disclosures unfairly inhibits its opponent’s ability to properly prepare, unnecessarily prolongs litigation, and undermines the district court’s management of the case.” Saudi, 427 F.3d at 278.

The majority opinion correctly concludes that Plaintiffs violated Rule 26. Ante at 193. But in reaching this conclusion the majority opinion omits any discussion of the numerous ways in which Pugh’s net-intrust testimony violated that Rule. First, Pugh’s report did not provide Wilmington Trust with “a complete statement of all opinions the witness will express,” as Rule 26 requires. Fed. R. Civ. P. 26(a)(2)(B)(i). In particular, Plaintiffs did not identify Pugh as a damages expert, nor did Pugh’s report state that he would provide a damages opinion. Notably, Pugh’s five-and-a-half page report did not include the term “net-in-trust,” let alone provide — or even allude to — Pugh’s opinion regarding an estimate of net-in-trust damages.

And the “present value” and “current valuation of underfunding” calculations in the report materially differ from the various amounts of net-in-trust damages Pugh testified to at trial ($51.7 million in report vs. $11.2 million at trial for January 2005 to January 2013 period and $33.5 million in report vs. $8.7 million at trial for January 2014 to January 2019 period), which is unsurprising given that each calculation measured different things (the value, as of trial, of Wilmington Trust’s annual $5.5 million lending commitment vs. differences in net-in-trust value as a result of Wilmington Trust’s failure to comply with the premium financing agreement). J.A. 704, 3080. As Pugh explained, the calculations in Pugh’s report simply determined the aggregate value, as of 2013, of a stream of $5.5 million loans from 2005 to 2019. J.A. 745-46 (explaining that the calculations in his report involved “basically just taking money over time at a certain interest rate and calculating what it would be today”). These calculations did not factor in, among other things, the costs of obtaining the loans, the trust’s obligation to repay the loans, and the returns the trust would, or would not, obtain through use of the loaned funds — all of which are essential to estimating damages in a breach-of-contract-to-lend case, like the instant case. See Restatement (First) of Contracts § 343 (1932) (“Damages for breach of a contract to lend money are measured by the cost of obtaining the use of money during the agreed period, less interest at the rate provided in the contract, plus compensation for other unavoidable harm that the defendant had reason to foresee when the *211contract was made”). For this reason, the present value and current valuation of underfunding amounts were not estimates of damages at all, which is why Pugh’s report did not characterize those amounts as damages estimates and why Plaintiffs did not designate Pugh as a damages expert.

Second, not only did Pugh’s report fail to provide an opinion as to net-in-trust damages — or, for that matter, any opinion as to damages — it also did not discuss most of the “facts” and “data” Pugh ultimately relied on in rendering his opinion as to net-in-trust damages. Pugh’s damages testimony — and the trial exhibit upon which that testimony relied — used figures and assumptions that were never discussed in his report, such as projections of “cost of insurance,” “other expenses,” “crediting rates,” and “premium financing rates.” J.A. 704, 3067-79. Pugh’s report expressly did not address, much less provide estimates of, those variables. Id. Additionally, Pugh’s report did not include the “formula” Pugh used to calculate net-intrust damages, as Plaintiffs now concede. And even if Pugh’s report had included his “formula” for calculating net-in-trust damages — which it did not — the report would not have complied with Rule 26 because it would not have provided an opinion as to net-in-trust damages, and the data Pugh “plugged in” to formula to reach that opinion. Indeed, Pugh’s report is devoid of any discussion of “how” he would calculate net-in-trust damages and does not provide any of Pugh’s “bas[e]s and reasons” for concluding that the facts, data, variables, and methods he ultimately used were appropriate for calculating such damages. In sum, Pugh’s report was precisely the “sketchy, vague[, and] preliminary” report Rule 26 forbids. Salgado, 150 F.Sd at 741 n.6; see also R.C. Olmstead, Inc. v. CU Interface, LLC, 606 F.3d 262, 271 (6th Cir. 2010) (concluding expert report did not comply with Rule 26 due to “lack of reasoning” and only “cursory support” for conclusions).

Third, Plaintiffs did not submit PX-174 — their only disclosure estimating net-in-trust damages — as required by Rule 26(a)(1)(A) until more than two years after the court’s deadline for submission of Rule 26(a) disclosures. See Joint Mot. for Entry of Sched. Order, Bresler v. Wilmington Trust Co., No. 8:09-cv-02957, Doc. No. 229, at 1. (D. Md. June 10, 2011) (stipulating July 14, 2011, deadline for submission of Rule 26(a) disclosures). And Plaintiffs never provided Wilmington Trust with the formulas and data or other “bases and reasons” underlying Pugh’s analyses in PX-174B, notwithstanding that Rule 26(a)(2)(B) required that Plaintiffs provide such information.

Fourth, Plaintiffs did not disclose exhibit PX-174 or PX 174B with Pugh’s report, contrary to Rule 26’s requirement that a party submit all of an expert’s trial exhibits with the expert’s report. Fed. R. Civ. P. 26(a)(2)(B)(iii).

Fifth, contrary to Plaintiffs’ contentions, PX-174 did not constitute a valid Rule 26(e) “supplement” to Pugh’s report. Rule 26(e) requires supplementation in a “timely manner.” As explained above, virtually all of the additional information Pugh used to calculate net-in-trust damages was available more than a year before Plaintiffs submitted PX-174 — and before expert discovery closed. See supra Part II.A. Accordingly, Plaintiffs did not submit PX-174 in a timely manner. See Salgado, 150 F.3d at 743 (“Because discovery was closed in the case, the information contained in the supplemental report must have been available before the missed deadline.”). Indeed, Plaintiffs did not submit the exhibit within the deadline for supplemental reports established in the court’s scheduling order.

*212More significantly, far from being a “supplement,” PX-174 included opinions outside the scope of Pugh’s original report, including Pugh’s: (1) formulas for calculating net-in-trust damages and estimating the variables necessary to calculate those damages; (2) projections of “cost of insurance,” “other expenses,” “crediting rates,” and “premium financing rates”; and (3) estimates of damages. But Rule 26(e) “permits supplemental reports only for the narrow purpose of correcting inaccuracies or adding information that was not available at the time of the initial report.” Minebea, 231 F.R.D. at 6 (emphasis added). Indeed, “[t]o construe [Rule 26(e)] supplementation to apply whenever a party wants to bolster or submit additional expert opinions would [wreak] havoc in docket control and amount to unlimited expert opinion preparation.” Campbell v. United States, 470 Fed.Appx. 153, 157 (4th Cir. 2012) (first alteration added). Accordingly, because PX-174 included numerous wholly new opinions outside the scope of Pugh’s original report and relied on data and information that was available prior to the close of expert discovery, it did not constitute a valid supplement. See Mine-bea, 231 F.R.D. at 6 (excluding “supplement” that was a “substantial ‘refinement’ of the original report, containing new or different material and providing additional information to support specific elements of [the proponent’s] case”). Indeed, treating PX-174 as a supplement “would contradict the purpose of Rule 26(a)(2) and Rule 37(c)(1), which specifically prevent further disclosures of expert testimony as trial approaches.” Id.

Even if the rendering of an entirely new opinion constituted a valid supplement — as Plaintiffs wrongly claim — PX-174 did not constitute an adequate Rule 26(a)(2)(B) disclosure for numerous additional rea-' sons: it does not provide the “basis and reasons” for the opinions and methods Pugh used to calculate net-in-trust damages or the “how and why” of his calculations — such as why he adopted certain of Stephens’s assumptions and why he used a particular method or formula for estimating other variables. R.C. Olmstead, 606 F.3d at 271. Wilmington Trust was forced to learn the answer to these questions during cross-examination at trial, precisely the outcome Rule 26 is designed to prevent. See Saudi, 427 F.3d at 278; Minebea, 231 F.R.D. at 5-6.

In sum, Plaintiffs violated Rule 26 in numerous ways with respect to Pugh’s expert report and damages testimony.

C.

Having concluded that Plaintiffs failed to comply with Rule 26, it is next necessary to determine whether that violation warranted exclusion of Pugh’s testimony as to net-in-trust damages. Under Rule 37(c)(1), “[i]f a party fails to provide information ... as required by Rule 26(a) or (e), the party is not allowed to use that information ... to supply evidence on a motion, at a hearing, or at a trial, unless the failure was substantially justified or is harmless.” The advisory committee notes to Rule 37(c)(1) characterize exclusion as an “automatic sanction.” To that end, if a party’s failure to comply with Rule 26(a) or (e) is not substantially justified or harmless, Rule 37(c)(1) “requires exclusion.” Campbell, 470 Fed.Appx. at 156; Salgado, 150 F.3d at 742.

1.

The majority opinion concludes that the district court did not abuse its discretion in concluding that Plaintiffs’ noncompliance with Rule 26 was substantially justified or harmless under the Southern States factors. Ante at 195. But, as explained above, Plaintiffs’ repeated misrep*213resentations deprived the district court of the opportunity to exercise its discretion in an informed manner and, therefore, mandate de novo, rather than deferential, review. See supra Part II.A.

Plenary review of the district court’s decision not to apply the “automatic sanction” of exclusion is doubly warranted because the record provides no indication that the district court actually exercised its discretion to find that Plaintiffs’ failure to comply with Rule 26 was substantially justified or harmless, as Rule 37(c)(1) requires to avoid mandatory exclusion. It is axiomatic that “[djeference to an exercise of discretion requires discretion actually to have been exercised.” Jebian v. Hewlett-Packard Co. Emp. Benefits Org. Income Prot. Plan, 349 F.3d 1098, 1106 (9th Cir. 2003). To that end, “[i]f the record reveals that the trial judge has failed to exercise the ‘sound discretion’ entrusted to him, the reason for ... deference by an appellate court disappears.” United States v. Sloan, 36 F.3d 386, 394 (4th Cir. 1994) (quoting Arizona v. Washington, 434 U.S. 497, 510 n.28, 98 S.Ct. 824, 54 L.Ed.2d 717 (1978)) (alterations in original). Accordingly, when there is no evidence that a district court exercised its discretion, “a de novo standard of review applies.” Jebian, 349 F.3d at 1106.

Here, the record provides no evidence that the district court concluded that Pugh’s damages testimony violated Rule 26 and therefore was admissible only if the noncompliance was substantially justified or harmless. On the contrary, in its only statement regarding whether Plaintiffs complied with Rule 26, the district court said it “d[idn]’t know that the rule about providing expert testimony says that every exhibit that he’s going to use when he gives his testimony has to be provided at or about the time that he gives his report,” J.A. 1393, indicating that the district court incorrectly concluded that Pugh’s proposed damages testimony complied with Rule 26.1 And the district court never stated that any violation of Rule 26 was substantially justified or harmless, let alone addressed whether the Southern States factors supported admission of Pugh’s net-in-trust testimony.2 Indeed, the district court’s apparent conclusion that Plaintiffs did not violate Rule 26 rendered it unnec*214essary for the district court to consider substantial justification or harmlessness.

Because the district court (1) never concluded that Pugh’s proposed damages testimony failed to comply -with Rule 26, and therefore was subject to exclusion absent a finding that the noncompliance was substantially justified or harmless, and (2) never expressly addressed whether any noncompliance was substantially justified or harmless, there is no basis for this Court to conclude that the district actually exercised its discretion to find that Plaintiffs’ noncompliance with Rule 26 was substantially justified or harmless.' Accordingly, this Court must review de novo whether Plaintiffs’ numerous violations of Rule 26 were substantially justified or harmless — the question to which I now turn. See also Sanders v. Venture Stores, Inc., 56 F.3d 771, 773 (7th Cir. 1995) (applying de novo review to district court’s denial of leave to amend complaint — a question typically reviewed for abuse of discretion — because “district court did not indicate that its denial of Plaintiffs’ motion ... resulted from an. exercise of its discretion”).3

2.

Because Plaintiffs failed to comply with Rule 26, they bear the burden “to show that the failure to comply was either substantially justified or harmless.” Carr v. Deeds, 453 F.3d 593, 602 (4th Cir. 2006), abrogated on other grounds by Wilkins v. Gaddy, 559 U.S. 34, 130 S.Ct. 1175, 175 L.Ed.2d 995 (2010). We have identified five factors for courts to consider in determining whether a party’s failure to make a disclosure required by Rule 26(a) or (e) was substantially justified or harmless: “(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.

As an initial matter, the majority opinion correctly recognizes that Plaintiffs have never offered any justification for their noncomplianee with Rule 26. Ante at 194. Accordingly, Plaintiffs’ failure to comply with Rule 26 was not substantially justified.

Regarding harmlessness, Plaintiffs do not argue on appeal that they met their burden to establish harmlessness under the Southern States factors. And other than their conclusory assertion that Wilmington Trust could “satisfy none of the five factors enunciated in [Southern States],” Motion In Limine Opposition, at 31 — an argument that wrongly suggested that Wilmington Trust, rather than Plaintiffs, bore the burden to establish harmlessness — Plaintiffs did not argue harmlessness before the district court, instead electing to maintain that Pugh’s damages testimony complied with Rule 26. This failure alone warranted exclusion of Pugh’s damages testimony, which did violate Rule 26. R.C. Olmstead, 606 F.3d at 270 (concluding that expert report that did not comply with Rule 26(a) was properly excluded because proponent “did not argue *215harmlessness before the district court. Instead, [proponent] argued only that [the] expert report met the requirements of Rule 26(a)”)

Nonetheless, the majority opinion concludes that the Southern States factors weighed in favor of admitting Pugh’s damages testimony. I disagree.

Regarding the first factor — surprise— Plaintiffs first disclosed that Pugh would testify to net-in-trust damages after the deadlines for submitting expert reports and supplemental and rebuttal expert reports, after the close of discovery, and after the court had ruled on the parties’ motions for summary judgment. Indeed, Plaintiffs did not disclose exhibit PX-174, which contained Pugh’s first opinion as to net-in-trust damages, until after the deadline for filing pre-trial motions. This Court has found surprise in virtually identical circumstances. Wilkins, 751 F.3d at 223 (“The [expert] disclosure was made after the agreed-upon expert disclosure date, after discovery was closed, after Appellee filed a motion for summary judgment, and on the very date set by the court for the filing of motions to exclude experts. It is hard to accept that these events would not serve as a surprise to Appellee, or that Appellee could easily cure such a surprise.”).

Additionally, during his deposition, Pugh explained (accurately) that the present value and current valuation of overfunding calculations in his report involved “basically just taking money over time at a certain interest rate and calculating what it would be today.” J.A. 745-46. And Pugh further testified at his deposition that he “wasn’t asked to consider all factors and calculate a net result.” Motion In Limine Opposition, at 31. Based on these representations, Wilmington Trust was — and should have been — surprised that Plaintiffs intended for Pugh to provide an opinion as to net-intrust damages (or, for that matter, any other type of damages). See S. States, 318 F.3d at 598 (holding that opponent of evidence was “surprised” by expert opinion provided in untimely supplemental report because, prior to submission of supplemental report, expert had stated in his deposition that “he had completed his opinions”). That surprise was exacerbated by Pugh’s use of numerous facts and figures to calculate his opinion as to net-in-trust damages that were never discussed in his report or his deposition.

The majority opinion nonetheless maintains that the surprise resulting from Plaintiffs’ noncompliance with Rule 26 was “minimal” because Pugh’s report made Wilmington Trust “aware ... that damages related to the plaintiffs’ anticipated net-in-trust was a central issue in the case.” Ante at 193. But a party does not satisfy Rule 26 simply by making the opposing party “aware” of “a central issue in the case.” Rather, Rule 26 requires that expert reports include “a complete statement of all opinions the witness will express and the basis and reasons for them” and “the facts or data considered by the witness in forming them,” Fed. R. Civ. P. 26(a)(2)(B)(i)-(ii), so as to “permit the opposing party to prepare rebuttal reports, to depose the expert in advance of trial, and to prepare for depositions and cross-examination at trial,” Minebea, 231 F.R.D. at 5-6. Accordingly, even if Wilmington Trust was not surprised that Pugh would provide an opinion on net-in-trust damages—which is doubtful given that Pugh testified during his deposition that he was not asked to render an opinion as to net-intrust damages and therefore did not provide one—Wilmington Trust still was — and should have been—surprised as to the particular formulas, methods, facts, and data Pugh used to calculate net-in-trust damages, many of which Pugh did not explain *216until he took the stand at trial. Indeed, if the majority opinion is correct that a party cannot be surprised by an opponent’s expert testimony so long as the testimony relates to a “central issue in the case,” then, without subjecting the opposing party to “surprise,” a plaintiffs expert medical report in a medical malpractice action could omit any discussion of the mechanism by which the malpractice harmed the plaintiff, or the plaintiffs expert damages report could omit any discussion of the methods and assumptions used to calculate damages. Rules 26 and 37 do not contemplate such results.

The majority opinion further maintains that any surprise was “minimal” because some of the “updates ... reflected in exhibit PX174” to the calculations in Pugh’s report were based on data included in Stephens’s report. Ante at 193. But PX-174 did not “update” Pugh’s report, it provided wholly new opinions as to net-intrust damages and the formulas, methods, facts, and data necessary to estimate those damages. See supra Part II.B. And even if PX-174 did amount to an “update” of Pugh’s report, Wilmington Trust still would have been surprised by Pugh’s use of information in Stephens’s report because Stephens’s report was submitted months before the deadline for submitting supplemental and rebuttal expert reports, which Plaintiffs declined to submit. Because Plaintiffs elected not to timely supplement Pugh’s report or rebut Stephens’s report, Wilmington Trust reasonably would have expected that Plaintiffs did not intend for Pugh to use the information in Stephens’s report to calculate net-in-trust damages, or for any other purpose.

Finally, the majority opinion states that the surprise was minimal because the “updates” in PX-174 resulted in “a decrease in the amount of damages calculated.” Ante at 193. But the present value and current value of underfunding figures provided in Pugh’s reports were not estimates of damages, net-in-trust or otherwise, and therefore Pugh’s opinions of net-in-trust damages did not “decrease” a previously disclosed estimate of damages — they amounted to wholly new opinions. See supra Part II.B. Moreover, even if PX-174 had decreased a pre-existing estimate of damages, it still would have resulted in prejudicial surprise. Errors in Pugh’s analysis that Wilmington Trust identified during cross-examination reduced Pugh’s net-in-trust damages estimates. If Wilmington Trust had had an opportunity to depose Pugh on his ultimate net-in-trust damages opinion, submit a rebuttal report, and prepare testimony challenging that opinion, it may have been able to identify further problems with Pugh’s analysis and therefore further reduce Wilmington Trust’s liability. Indeed, under the majority’s position, a party could insulate its expert damages analyses from exclusion simply by submitting a conclusory report with an inflated damages estimate and then providing a reduced estimate prior to trial. Such a result contradicts Rule 26’s purpose of “preventing] experts from ‘lying in wait’ to express new opinions at the last minute, thereby denying the opposing party the opportunity to depose the expert on the new information or closely examine the expert’s new testimony.” Mi-nebea, 231 F.R.D. at 6.

The second factor — the ability to cure the surprise caused by the Rule 26 violation — also supports excluding Pugh’s damages testimony. Plaintiffs did not disclose that Pugh would testify as to,, damages and provide exhibit PX-174 until nearly a year after Wilmington Trust deposed Pugh and after the close of expert discovery. More specifically, Plaintiffs waited to disclose Pugh’s damages testimony until only a few weeks before trial. “The purpose of Rule 26(a)(2) is to provide notice to opposing *217counsel—before the depositions—as to what the expert witness will testify,” Ciomber, 527 F.3d at 642 (emphasis added); “to prepare rebuttal reports,” Minebea, 231 F.R.D. at 5 (emphasis added); and “to allow an opponent to examine an expert opinion for flaws and to develop counter-testimony through that party’s oum experts,” S. States, 318 F.3d at 598 (emphasis added). Plaintiffs’ noncompliance with Rule 26 deprived Wilmington Trust of the opportunity to depose Pugh with the benefit of his net-in-trust damages opinion, prepare a rebuttal report responding to Pugh’s opinion, and to develop counter-testimony.

The majority opinion nonetheless maintains that Wilmington Trust could have cured the surprise by deposing Pugh after Plaintiffs disclosed PX-174. But Rule 37(c)(1) provides for a single remedy: exclusion. Accordingly, Wilmington Trust cannot be held wanting for fading to pursue lesser relief, particularly since Plaintiffs’ disclosure came only weeks before trial. Cf. Campbell, 470 Fed.Appx. at 156 (“[T]he district court did not abuse its discretion in failing to consider less drastic sanctions than exclusion of Campbell’s expert witness, as Rule 37(c) requires exclusion unless the party establishes substantial justification or harmlessness.”). And even assuming Wilmington Trust could have deposed Pugh during that brief period, the purpose of Rule 26(a)(2) “would be completely undermined if parties were allowed to cure deficient reports with later deposition testimony.” Ciomber, 527 F.3d at 642. PX-174 was not a valid Rule 26 expert disclosure because it did not provide the “basis and reasons” and the “how and why” of Pugh’s net-in-trust damages opinions or the formulas, methods, facts, and data underlying that opinion. See supra Part II.B. Accordingly, deposition testimony could not have cured Plaintiffs’ noncompliance. And regardless of whether Wilmington Trust could have deposed Pugh, Plaintiffs’ late and insufficient disclosure precluded Wilmington Trust from preparing and filing a rebuttal report to Pugh’s net-in-trust damages opinion. Indeed, when Wilmington Trust sought to introduce post-trial a declaration by Stephens rebutting Pugh’s net-in-trust testimony, the district court struck the declaration as untimely. Accordingly, the surprise caused by Plaintiffs’ noncompliance was not — and could not have been — cured.

Regarding the third factor — disruption — not only did Pugh’s net-in-trust damages testimony have the potential to disrupt the trial, it did disrupt the trial. In particular, the district court had to adjourn Pugh’s testimony so he could recalculate his estimates of net-in-trust damages after cross-examination revealed numerous errors in Pugh’s original opinion. J.A. 2049-50 (“[Fjrankly, it’s just not clear where he is. He’s sort of stammering on what to do with these new numbers. I think he may not be able to come up with any number at all.... From my standpoint, I can’t follow this now.”). In cutting off Pugh’s testimony, the district court said Pugh’s shifting testimony left the district court “confused and the jury [] triply confused about where [Pugh] is coming out.” J.A. 2051. This disruption, delay, and confusion is precisely what the expert discovery rules are designed to prevent. Saudi, 427 F.3d at 278-79. Had Plaintiffs timely disclosed Pugh’s net-in-trust damages opinion, and provided a fulsome report supporting that opinion, then Wilmington Trust could have deposed Pugh with the benefit of his opinions and the data and analyses underlying those opinions. And Wilmington Trust could have prepared a rebuttal report explaining the errors in Pugh’s opinion. Therefore, the parties would have aired the issues with Pugh’s analyses before he took the stand, and the disruption, delay, *218and confusion at trial would not have occurred.

As the majority opinion correctly finds, the final factor relating to harmlessness— the importance of the evidence — also supported excluding Pugh’s damages testimony. This Court has emphasized that “importan[ce].... must be viewed from the perspective of both parties.” S. States, 318 F.3d at 598 (emphasis added) (internal quotation marks omitted). To that end, the more “important” the evidence, the more important it is for the proponent to “disclose[ the evidence] in a timely manner” because it is more likely to prejudice the opposing party. Id. at 599 (internal quotation marks omitted). That is precisely the case here. Pugh’s net-in-trust testimony was Plaintiffs’ principal evidence of damages. By depriving Wilmington Trust of the opportunity to “prepare rebuttal reports, to depose the expert in advance of trial, and to prepare for depositions and cross-examination at trial,” Minebea, 231 F.R.D. at 5-6, Plaintiffs deprived Wilmington Trust of the opportunity to rebut what the majority opinion concedes was “vital” evidence in the case, see Ante at 194-195.

In sum, even if Plaintiffs had argued that their noncompliance with Rule 26 was substantially justified or harmless, Plaintiffs could not have satisfied their burden under Rule 37(c)(1). On the contrary, all of the Southern States factors supported excluding Pugh’s damages testimony.

III.

“Compliance with the Federal Rules of Civil Procedure in all actions embraced within the scope thereof is required and is essential to the orderly administration of justice.” Smith v. United States, 369 F.2d 49, 55 (8th Cir. 1966). Likewise, “[c]ounsel have a duty to be candid ... and the trial court, bearing a heavy caseload, relies on counsel to meet that duty.” Diaz-Fonseca, 451 F.3d at 37. Plaintiffs’ inadequate expert disclosures and Pugh’s net-in-trust damages testimony at trial ran roughshod over the requirements of Rule 26 and materially prejudiced Wilmington Trust’s ability to refute Pugh’s damages testimony. And Plaintiffs’ repeated misrepresentations deprived the district court of the opportunity to exercise its discretion to determine in an informed manner whether to admit Pugh’s damages testimony. Affirming the admission of Pugh’s damages testimony only serves to reward Plaintiffs for disregarding the Rules and making misleading statements to the district court. Because I cannot join in sanctioning such conduct, I dissent as to the majority opinion’s affirmance of the jury’s damages award.

. Even if we reviewed the district court’s admission of the testimony for abuse of discretion, this statement is, by definition, an abuse-of-discretion because it runs contrary to the plain language of Rule 26(a)(2)(B), which says that "any exhibits ... must” be included with the report. Hunter v. Earthgrains Co. Bakery, 281 F.3d 144, 150 (4th Cir. 2002) ("Of course, an error of law by a district court is by definition an abuse of discretion.”).

. The majority opinion correctly notes that a district court need not expressly reference the Southern States factors in order to constitute a valid exercise of discretion to conclude that a party's failure to comply with Rule 26(a) or (e) was substantially justified or harmless. Ante at 193 n.15 (citing Wilkins, 751 F.3d at 222). But the absence of discussion of the Southern States factors, coupled with a district court’s failure to expressly find that a party did or did not comply with Rule 26(a) or (e), is relevant to determining whether the district court actually exercised its discretion to conclude that a party's failure to comply with Rule 26 was substantially justified or harmless. Wilkins is illustrative. There, the district court unambiguously held that an untimely and "preliminary” expert report failed to comply with Rule 26. 751 F.3d at 219-21. The district court further concluded that the failure to comply with Rule 26 was not “harmless.” Id. at 222-23. Although the district court did not expressly reference the Southern States factors, this Court held that the district court did not reversibly err, in part because its analysis of whether the noncompliance was harmless “implicitly addressed some of the Southern States factors.” Id. at 222. Unlike in Wilkins, the district court here never expressly concluded that Plaintiffs violated Rule 26.

. Even if abuse-of-discretion review was warranted, the district court's failure to exercise its discretion to determine whether Plaintiffs' noncompliance with Rule 26 was substantially justified or harmless amounts to an abuse of discretion. James v. Jacobson, 6 F.3d 233, 239 (4th Cir. 1993) ("Perhaps [the] most obvious manifestation [of an abuse of discretion] is in a failure ... either express or implicit, actually to exercise discretion.”); Ray v. Robinson, 640 F.2d 474, 478 (3d Cir. 1981) ("If a district court fails to exercise its discretion, that is itself an abuse of discretion.”).