OPINION
STRAS, Justice.Macy’s Retail Holdings, Inc., challenges the Minneapolis Assessor’s valuation of its downtown Minneapolis property for the *4532008,2009, and 2010 tax years. Following a trial, the tax court reduced the valuation of the property, but not to the extent urged by the testimony and appraisal report of Macy’s expert witness. Macy’s appeals the tax court’s decision and argues that the tax court clearly erred in its determination of the property’s highest and best use and in its consideration of comparable-sales data. Macy’s also claims that the tax court abused its discretion when it declined to strike portions of the appraisal report and testimony of the County’s expert witness. Because the tax court did not commit reversible error on any of these grounds, we affirm.
FACTS
The parties in this case disagree about the taxable value of Macy’s downtown Minneapolis property. The property consists of three tax parcels located on Nicol-let Mall between 7th and 8th Streets in Minneapolis. The property is improved by a 12-story structure, originally built between 1902 and 1929, that housed a department store on the first 5 floors. The remaining floors included a combination of office space (6th, 7th, 9th, 10th, and 11th floors), storage (8th floor), and 2 restaurants (12th floor).
The Minneapolis Assessor’s Office valued the property at $22,700,000 for 2008, $19,500,000 for 2009, and $17,700,000 for 2010. Macy’s challenged the Assessor’s determination for all 3 tax years, claiming that it overstated the property’s market value. In the ensuing litigation, Hennepin County and Macy’s both retained experts who appraised the property. Todd Reid, who performed the appraisal on Macy’s behalf, valued the property at $12,800,000 for 2008, $10,000,000 for 2009, and $9,800,000 for 2010. Hennepin County’s expert, Paul Bakken, valued the property at $25,000,000 for 2008, $16,000,000 for 2009, and $17,000,000 for 2010.
Despite the significant differences in value placed on the property by the parties’ experts, their appraisal reports reflect that some of their underlying conclusions were the same. Both experts generally agreed that the highest and best use of the property was to operate it on an as-is basis in 2008, in anticipation of later developing the site to include an office tower along Nicol-let Mall and a high-rise residential tower on 8th Street.1 Both also concluded that renovation of the property was not financially feasible for any of the years in question. Additionally, each expert concluded that the appropriate way to value the property for 2009 and 2010 was through a market approach, which compares the property with the “prices paid in actual market transactions involving comparable properties.” Equitable Life Assurance Soc’y of U.S. v. Cty. of Ramsey, 530 N.W.2d 544, 552 (Minn. 1995). The experts, however, disagreed on three key points.
First, for 2008, the experts differed on which method would provide the most accurate value of the property. Bakken applied the income approach, “which is predicated on the capitalization of the income the property is expected to generate.” Id. In contrast, Reid applied the market approach because he determined that the income approach was not a rehable method of determining the property’s value. The tax court agreed with Reid and determined that the market approach was the appropriate method of valuing the property for each of the 3 years in question.
*454The second dispute between the two experts. was the date on which it would have been economically feasible to begin the development of a new downtown office tower. Reid’s view was that the Minneapolis rental market did not have an adequate number of available commercial tenants or sufficiently high market rents to justify the construction of the new office tower- until 2013 or 2014. Even so, Reid still believed that the highest and best use of the property during 2009 and 2010 would have been to tear down the existing structure in anticipation of future development. Because Reid believed that the property was best left vacant during 2009 and 2010, he deducted holding costs for those 2 years, which resulted in a lower overall valuation of the property. See Jack P. Friedman et al., Barron’s Dictionary of Real Estate Terms 80 (8th ed. 2013) (defining holding costs as “expenses necessary for holding property, such as taxes and interest on idle property or property under construction”).
Although Bakken acknowledged that the overall real-estate market in Minneapolis was “soft” during 2008 and 2009, he emphasized that the property had a number of characteristics that would have made it' an appealing development site, even as early as 2009. As he observed, the property is located in what he called “Target-ville,” a reference to its proximity to Target’s corporate headquarters. According to Bakken, commercial properties located in Targetville had lower vacancy rates and higher rents in 2008 and 2009 than non-Targetville properties. Bakken also pointed out that reports casting doubt on the viability of developing a new office tower in 2009 were specific to other available sites, none of which were located near Target’s corporate headquarters or directly on Nicollet Mall. Bakken was of the overall view that the Macy’s property was part of a sub-market that was resistant to general declines in the Minneapolis commercial-property market, which made it realistic to commence the construction of the office tower in 2009. Based on his view of the health of the property’s sub-market, Bakken disagreed that having the property sit as a vacant lot in 2009 and 2010 was the property’s highest and best use. Accordingly, unlike Reid, he did not deduct holding costs'for 2009 or 2010.
The tax court thought it was a “close call” whether development of the new office tower could have commenced as early as 2009. Ultimately, however, the court was persuaded by Bakken’s theory that the property belonged to a desirable sub-market that was less vulnerable to declines in the overall commercial-property market. Because the tax court concluded that it was feasible to have commenced the construction of the office tower in 2009, the court rejected Reid’s proposed deduction for holding costs for 2009 and 2010.
' The third and final disagreement between the two experts was the choice' of comparable sales under the market approach. Both experts divided the property into separate office-tower and residential components, which reflected the property’s highest and best use. The tax court relied exclusively on Reid’s analysis to value the residential component. For the.office-tower component, Reid analyzed five commercial properties, but he placed particularly heavy emphasis on just two of those properties. Bakken considered seven comparable sales, but relied on only three of them in his report. The experts relied on two common sales: 50 South 6th Street and 800 Nicollet Mall, both in downtown Minneapolis.
In valuing the office-tower component, the tax court focused on the two common sales, and did not discuss any of the other properties considered by the experts, in-*455eluding a sale of a nearby commercial building located at 1000 Nicollet Mall. Based on its analysis of the two comparable sales, the tax court valued the property at . $20,100,000 for 2008, $15,800,000 for 2009, and $15,400,000 for 2010. These figures are lower than the Minneapolis Assessor’s original valuation of the property for each of the years in question.
Based on its disagreement with several of the tax court’s findings and conclusions, including the failure to analyze 1000 Nicol-let Mall as a comparable sale and the conclusion that construction on the new office tower could have commenced in 2009, Macy’s filed a motion for amended findings. After the tax court denied the motion in its entirety, Macy’s timely appealed the tax court’s judgment by filing a petition for a writ of certiorari with this court. See Minn. Stat. § 271.10, subd. 1 (2016).
ANALYSIS
We have recognized that property valuation is an “inexact” science and therefore we will “defer to the decision of the tax court unless the tax court has either clearly overvalued or undervalued the subject property, or has completely failed to explain its reasoning.” Equitable Ufe, 530 N.W.2d at 552. As a result, “we will sustain the tax court’s valuation determination[s] on appeal unless [they are] clearly erroneous.” Eden Prairie Mall, LLC v. Cty. of Hennepin, 797 N.W.2d 186, 192 (Minn. 2011). A finding is clearly erroneous only if “the evidence as a whole does not reasonably support the decision,” Lewis v. Cty. of Hennepin, 623 N.W.2d 258, 261 (Minn. 2001), and we are “left with a definite and firm conviction that a mistake has been committed.” Berry & Co. v. Cty. of Hennepin, 806 N.W.2d 31, 33 (Minn. 2011) (citation omitted) (internal quotation marks omitted).
I.
The first question presented by this case is whether the tax court clearly erred when it agreéd with Bakken that market conditions would have allowed construction of the office tower to begin in 2009. Although the parties agree that the highest and -best use of the property was to operate it on an as-is basis in 2008 and then demolish it a year later in anticipation of constructing an office tower and separate residential tower, Macy’s relies on its own expert report to argue that the Minneapolis property market could not have supported the development of the new office tower until at least 2013. The logical implication of Macy’s argument is that the tax court’s failure to deduct holding costs overestimated the market value of the property in 2009 and 2010.
As noted above, Hennepin County argued at trial based on Bakken’s report and testimony that the 2009 commercial-property market could have supported the construction of a new office tower in Tar-getville.2 The tax court ultimately agreed with Bakken’s- theory that the location .of the property in a healthy sub-market of downtown Minneapolis made it financially feasible to begin construction of the new office tower in 2009. •
■ Macy’s argues that this conclusion was clearly erroneous because it • presented overwhelming evidence that pointed to the opposite conclusion: it would not have been *456possible to begin construction on the office tower in 2009. To support its position, Macy’s makes several interrelated arguments about the alleged unavailability of anchor tenants, the failure of the tax court to consider feasibility rents, and the overall unreliability of Bakken’s report. None of Macy’s arguments are persuasive.
A.
First, Macy’s points to an alleged lack of available “anchor tenants,” which were described at trial as larger prospective commercial tenants that would be willing to commit to leasing space in advance of construction. Macy’s argues that the high vacancy rates in comparable commercial buildings and the lack of demand in the commercial-property sector in 2009 would have made it impossible to secure a sufficient number of anchor tenants for an office tower of the type contemplated by the parties’ experts.
Although acknowledging that the issue was a “close” one, the tax court agreed with Bakken that the desirability of the office tower’s location could have attracted an adequate number of anchor tenants to financially justify building the office tower in 2009. For example, the court observed that the reports submitted by both experts contained lists of office tenants that could have potentially anchored the development. The tax court also credited Bakken’s testimony that new tenants could have been drawn from the suburbs or, if necessary, acquired through lease buyouts. These possibilities, which are supported by evidence in the record, are sufficient to support the tax court’s conclusion that anchor tenants would have been available for the office tower.
Macy’s further challenges the size of the potentially available anchor tenants, estimating that at least one of the anchor tenants would have needed to lease at least 350,000-450,000 square feet of space to have allowed construction to begin in 2009. Of Bakken’s list of potential tenants that had leases set to expire in 2011 and 2012, the earliest possible dates of occupancy, Macy’s argues that nearly all of them were too small to anchor a large office building.
The primary problem with Macy’s argument is that the available evidence does not support it. Nothing in the record suggests that construction of the office tower depended on a single anchor tenant willing to lease 350,000-450,000 square feet of space. To the contrary, Macy’s own expert testified that a total of 50 percent of the office space would need to be leased to justify construction, but he did not say, either in his report or in his testimony, that the space would need to be leased by a single tenant. Based on this evidence, it was sufficient for the County to show, as it did, that there were a sufficient number of potential anchor tenants available to have allowed construction on the office tower to begin in 2009.
As a final critique of the tax court’s anchor-tenant analysis, Macy’s contends that, even if a sufficient number of potential anchor tenants were available, Bakken failed to identify specific tenants that could have been drawn from the suburbs or lured through lease buyouts. This argument simply asks for an unrealistic level of detail about a purely hypothetical scenario. The County cannot be expected to identify specific anchor tenants that would have definitively agreed to lease space in a hypothetical office tower that could have theoretically been constructed beginning in 2009. All that the County was required to do was present sufficient evidence for the tax court to conclude, by a preponderance of the evidence, that the office tower could have attracted a sufficient number of tenants to allow construction to begin in 2009. Given that the County satisfied its burden, *457the tax court did not clearly err in its anchor-tenant analysis.
B.
Second, and closely related to its discussion of anchor tenants, Macy’s argues that the tax court “wholly ignored” the issue of feasibility rents, which are “[t]he rent[s] necessary to justify new construction.” Appraisal Institute, The Appraisal of Real Estate 342 (14th ed. 2013). Whereas anchor tenants address the need to fill the building’s space in advance of construction, feasibility rents are the levels of rent that those tenants would need to pay to justify development. If the rents were too low, then constructing an office tower would not have been financially feasible even if potential occupancy of the space were high. Macy’s argument is that the tax court addressed only half of the equation by evaluating the potential number of anchor tenants without also considering whether those tenants would have been willing to pay sufficiently high rents.
Despite Macy’s current focus on feasibility rents, neither of the parties’ experts insisted that feasibility rents needed to be analyzed separately from the question of anchor-tenant availability. In fact, if anything, the testimony at trial and the appraisal reports treated the availability of anchor tenants and the necessity for feasibility rents together, not separately. Moreover, the issue of feasibility rents played a small role at trial, so it is not surprising that the tax court considered the anchor-tenant issue as a whole without isolating its analysis of feasibility rents.
Even so, the tax court’s analysis was sufficient to support its conclusion that construction of the new office tower could have commenced in 2009. Both Bakken and the tax court analyzed the timeframe for beginning construction by considering the overall financial feasibility of the project at various points in time. For example, Bakken’s testimony that the commercial-property rental market in Targetville was strong supported the notion, perhaps implicit in the court’s decision, that adequate rental income would have been available to support the office tower’s construction in 2009. Bakken also explained that the evidence of rents paid on leases in effect in 2009, which Macy’s now faults him for failing to consider, difi not reflect then-current market rates because many of those leases were signed years before the proposed construction of the office tower was set to begin. In light of the evidence about market strength in Bakken’s report and his explanation of market rents, we cannot say that the tax court clearly erred when it found that beginning construction on the office tower in 2009 would have been financially feasible.
C.
Finally, Macy’s broadly asserts that the tax court’s rejection of some of Bakken’s analysis required it to disregard the entirety of his testimony and report. The unstated premise of Macy’s argument is that, because the tax court rejected a portion of Bakken’s expert analysis, it was clearly erroneous for the tax court to accept any of it, including the evidence that supported the feasibility of constructing the office tower in 2009.
Our role, however, is not to reweigh the evidence, and it is firmly within the tax court’s province, as the factfinder, to weigh the evidence and assess the credibility of the witnesses. See Lewis, 623 N.W.2d at 262 (“[T]he tax court is in the best position to evaluate the credibility of witnesses.”); see also KCP Hastings, LLC v. Cty. of Dakota, 868 N.W.2d 268, 275 (Minn. 2015) (“The tax court need not accept an appraiser’s valuation in its entirety; instead, it may adjust the calculations *458based on evidence in the record and its own expertise.”). In fact, as we have acknowledged, “[rjarely is-one appraiser’s methodology and opinion accepted in its-entirety by the [cjourt.” Montgomery Ward & Co. v. Cty. of Hennepin, 450 N.W.2d 299, 308 (Minn. 1990). The tax court was therefore free to accept the opinions of one expert on certain issues and those of another expert on others.
Taken tc its logical end, moreover, Macy’s argument proves too much. To be sure, the tax court disagreed with Bakken’s analysis on a number of points. But the court was equally critical of Reid’s opinion, which would have required the tax court, under Macy’s theory, to disregard the opinions of Macy’s expert as well. A fair reading of the tax court’s order reveals that the court was not totally persuaded by the analysis of either expert, leading it to rely on each at various points in its analysis. This is precisely what we have asked the tax court to- do: to value “the property by exercising its own independent skill and judgment.” Beck v. Cty. of Todd, 824 N.W.2d 636, 639 (Minn. 2013). By exercising its independent judgment and incorporating the opinions of both experts into its own' analysis, the tax court did not clearly err in determining that the construction of the office tower could have commenced in 2009.
II.
The second question presented by this case is whether the tax court abused its discretion in declining to strike portions of Bakken’s report and testimony as a sanction for a discovery violation. Before trial, the tax court ordered both experts to produce their reports and work files for the other party. During its cross-examination of Bakken, however, Macy’s discovered that Bakken had withheld certain lease documents and rent information relating to his list of potential anchor -tenants. Bakken claimed that he could not disclose the documents and information, which he described as confidential, because he had obtained them from previous clients. As Macy’s points out, however, the tax: court had already entered a protective order that should have addressed Bakken’s concerns about confidentiality. .
Upon learning that Bakken had withheld the lease documents and rent information, Macy’s orally moved to strike portions of Bakken’s report and testimony discussing the possible development date for the office tower. Macy’s also argued that Bakken’s actions violated his duty- to disclose the “facts or data” underlying-his expert opinion, as required by Minn. R. Evid. 705. The tax court denied the-motion, but allowed Macy’s to renew it in written form after trial. After reviewing the written submissions of the parties, the tax court once again denied the motion, concluding that striking portions of Bakken’s report was inappropriate under the standard from Dennie v. Metropolitan Medical Center, 387 N.W.2d 401 (Minn. 1986). Macy’s raised the issue for.a third time in its motion for amended findings, but the tax court denied the request on the grounds that the request constituted an unauthorized motion for reconsideration and that, in any event, the proposed sanction did not fit the violation,
“The decision to admit or exclude evidence rests with the tax court, and its rulings will not be disturbed absent an error of law or abuse of discretion.” Cont'l Retail, LLC v. Cty. of Hennepin, 801 N.W.2d 395, 399 (Minn. 2011). Additionally, “[tjrial courts must have discretion to determine the sanction appropriate to a violation of the discovery rules, for they are in the best position to assess the degree of prejudice that will arise from, the violation and the efficacy of the remedies *459available that may prevent prejudice from resulting.” Cornfeldt v. Tongen, 262 N.W.2d 684, 697 (Minn. 1977).
Dennie, the case relied, upon by the tax court, makes clear that “suppression of expert testimony is a serious sanction and should be imposed only in the most compelling circumstances.” 387 N.W.2d at 406. When considering exclusion as a remedy for nondisclosure, Dennie instructs trial courts to weigh six factors: (1) the extent of preparation required by an opposing party for cross-examination or rebuttal of the expert witness; (2) when the expert agreed to testify; (3) when the opposing party was notified of the expert’s availability; (4) when the attorney calling the expert assumed control of the case; (5) whether the failure to disclose was intentional and willful; and (6) whether the opposing party sought a continuance or other lesser remedy. Id.
Some of the Dennie factors are context specific, which means that certain factors will apply in some cases.but not others. Consistent with this observation, the tax court adapted the Dennie factors to fit the facts of this case and reorganized them into three broad headings: (1) the culpability for the failure to disclose, essentially a restatement of the fifth Dennie factor; (2) whether the qpposing party sought a continuance or other .remedy, the equivalent of the sixth Dennie factor; and (3) “preparation for cross-examination,” applied by the tax court in a manner consistent with the first Dennie factor.3
The first factor applied by the tax court was whether Bakken’s failure to disclose the rent information and lease documents was intentional and -willful. The tax court gave a nuanced answer to its consideration of this factor, determining that Bakken withheld. the documents intentionally based on his sincere belief that client confidentiality applied, but that “it was not willful or in any way culpable” because, he “believed he had justification for excluding the lease information from his work file.” (Emphasis omitted.) The tax court primarily based these findings on Bakken’s demeanor and his forthright testimony at trial.
In response, Macy’s points out that Bakken must have been aware that his concerns about confidentiality were -unfounded in light of the court’s protective order, which specifically addressed the concerns that Bakken raised during his testimony. As the tax court noted in its denial of Macy’s motion for amended findings, however, this argument assumes that Bakken was fully aware of the protective order, understood its contours, and had been in “ongoing discussions” with the County Attorney about what needed to be disclosed. None of those assumptions are established, in the record. Moreover, the court’s .credibility finding is bolstered by Bakken’s willingness to readily admit. on cross-examination that he withheld the information because of his concerns about client confidentiality. Considering the deferential standard of review, the subjective nature of the court’s credibility finding, *460and the court’s reliance on its observations during trial, we cannot say that the finding on Bakken’s lack of willfulness was clearly erroneous.
The second factor applied by the tax court was whether Macy’s sought a lesser sanction, such as requesting a continuance, after learning of Bakken’s discovery violation. On this point, the record is clear: Macy’s did not request any remedy other than the exclusion of portions of Bakken’s opinions on the feasibility of constructing the office tower in 2009, which was one of the key contested issues at trial. The tax court concluded that Macy’s failure to request a lesser sanction weighed against suppressing Bakken’s testimony.
Rather than arguing that the tax court’s finding on this factor was clearly erroneous, Macy’s position is that it was not required to seek a lesser remedy because disclosure of the documents was already mandatory under the tax court’s order. Macy’s, however, conflates two distinct concepts: the requirement under the court’s order for disclosure and the remedy for violating the order. Just because disclosure is mandatory does not mean that the only remedy for violation is exclusion. As the tax court observed, if Macy’s was truly concerned about the prejudice from the withheld documents and information, it could have requested a continuance to allow it to effectively cross-examine Bakken, or even asked the tax court to award attorney fees to account for the extra preparation and additional motions from the nondisclosure. Because Macy’s did not request any other sanction, the tax court concluded that this factor weighed “heavily” against suppressing Bakken’s testimony. This conclusion is consistent with Denme’s suggestion that exclusion of expert testimony should be a last resort.
The third factor applied by the tax court was whether Macy’s was prejudiced by Bakken’s failure to disclose the lease documents and rent information in advance of trial. We have described “prejudice” as the “crucial question” in determining whether exclusion is the appropriate sanction for nondisclosure. Dennie, 387 N.W.2d at 405. In assessing prejudice, the tax court properly focused on the nature of the discovery violation.
Bakken’s list of potential anchor tenants in his report included the name of each tenant, the amount of space each tenant was renting, and the expiration month of the lease. What the report did not include were the documents from which Bakken extracted the information. Yet, as the tax court observed, Macy’s questioning of Bakken had little to do with the nondis-closed documents; instead, it focused on market and feasibility rents, issues that Bakken’s report did not directly discuss. Though it is true that the documents would have revealed the rents that Bakken’s proposed anchor tenants were then paying, this data was not something on which either Bakken or Reid had relied.4 *461The tax court’s finding of a lack of prejudice, therefore, was not clearly erroneous.
The tax court, after carefully considering the evidence in the record and applying the relevant Bennie factors, concluded that exclusion was not the appropriate remedy for Bakken’s discovery violation. Much of the court’s discussion was focused on the interrelated factors of Macy’s failure to request a lesser remedy and the lack of prejudice suffered by Macy’s due to the nondisclosure. Because the court’s discussion of these factors, as well as the culpability factor, \yas consistent with Bennie and none of the court’s findings were clearly erroneous, the court did not abuse its discretion when it denied Macy’s serial motions to strike Bakken’s report and testimony.
III.
The third and final question presented by this case is whether the tax court was required to consider the sale of a nearby commercial property in determining the value of Macy’s property. The nearby property, located at 1000 Nicollet Mall, was considered by both experts, but the two experts disagreed on whether it was sufficiently similar to be comparable. Bakken concluded that the property was too small to be useful in his valuation analysis. Reid, by contrast, adjusted the sale based on its differing size and other characteristics, but did not. “actually rel[y] on this comparable” sale in conducting his calculations. Reid’s choice effectively gave dispositive weight to only two of the five comparable properties, the same two properties the tax court used in conducting its own analysis.
The tax court’s selection of comparable properties, like other factual determinations, is subject to a clearly erroneous standard of review. See Archway Mktg. Servs. v. Cty. of Hennepin, 882 N.W.2d 890, 893, 898 (Minn. 2016). The tax court declined to incorporate the sale of 1000 Nicollet Mall into its analysis because neither expert gave it significant weight. Although Reid analyzed the sale—at least to some degree—the record supports the tax court’s conclusion that Reid placed greater weight on two other sales. In addition, Bakken provided a rationale for the tax court to disregard the sale: it was too small to be considered comparable. Based on this evidence and the lack of emphasis placed by both experts on the 1000 Nicol-let Mall sale, we conclude that the tax court’s decision to disregard the sale in its own analysis was not clearly erroneous.
CONCLUSION
For the foregoing reasons, we affirm the tax court’s decision.
Affirmed.
LILLEHAUG, J., took no part in the consideration or decision of this case. MCKEIG, J., not having been a member of this court at the time of submission, took no part in the consideration or ■ decision of this case.. Property is appraised at its “highest and best use,” which is “the most profitable, competitive use to which the subject property can be put.” Appraisal Institute, The Appraisal of Real Estate 331 (14th ed. 2013).
.. Macy’s now argues for the first time that Bakken’s testimony lacked "foundational reliability” under Minn. R. Evid. 702. Macy’s did not raise this objection before the tax court and, therefore, this argument has been forfeited. Thiele v. Stick, 425 N.W.2d 580, 582 (Minn. 1988) (holding that arguments not raised in the trial court are forfeited on appeal).
. We reject Macy’s argument that, due to the unique circumstances of this case, Dennie is inapplicable. Nothing in Dennie suggests that the test is uniquely suited to only certain types of cases. To the contrary, the language in Dennie is broad, stating that the factors apply whenever a trial court is considering the "suppression of expert, testimony’’ as a possible remedy for a discovery violation, due to the fact that exclusion is a "serious sanction” that should be imposed only "in the most compelling circumstánces and then only after careful consideration of the [Dennie] factors.” 387 N.W,2d at 406. The all-encompassing inquiry contemplated by the Dennie factors belies the notion, advanced by Macy’s, that Den-nie should be confined to its facts.
. The tax court's analysis of prejudice is closely related to Macy’s argument that Rule 705 required Bakken to disclose the rents paid by the potential anchor tenants he had listed in his report. Rule 705, however, requires only the disclosure of "facts or data” underlying an expert witness’s opinion, and even then, only when the court orders the disclosure of such facts or data. Bakken, during his testimony, expressly disclaimed any reliance on rent data from the expiring leases because he believed they were too old to be a reliable reflection of market rents. Nothing in the record refutes his testimony, and Macy’s novel argument that Bakken’s answers during cross-examination about specific tenants somehow retroactively changed the nature of his original opinion is inconsistent with the plain language of Rule 705. See Minn. R. Evid. 705 (requiring “prior disclosure” of “underlying facts or data” only when the trial court orders it (emphasis added)). In any *461event, Rule 705 says nothing about exclusion of the expert witness’s testimony or report, so Macy’s argument does not entitle it to the relief it seeks any more than its discussion of Dennie does.