concurring in the judgment:
In light of Brisker v. District of Columbia, 510 A.2d 1037, 1040 (D.C.1986), I believe that I am compelled to vote to affirm the decision of the trial court. I do so with considerable reluctance, however, for I have serious concerns as to whether the *1313taxpayer was treated fairly or even rationally by the District’s “revenue men,” here the officials of the Department of Finance and Revenue (DFR). Accordingly, although there is much in Judge Ferren’s opinion for the court with which I can agree, I join only in the judgment.
I
The sequence of events in this case would surely startle a reasonable person not versed in the technical intricacies of the assessment process or familiar with the strictures of the “clearly erroneous” rule. For tax year 1985, DFR assessed the property at $14,620,500, an increase of more than four million dollars over the previous year. Upon the taxpayer’s appeal, the Board of Equalization and Review (the Board) determined that the property had been over-assessed by $1,080,478, and reduced the assessment to $13,539,022.
When the time came for DFR to assess the value of the taxpayer’s property for 1986, however, something occurred which I, at least, find quite remarkable. DFR’s new assessment was $14,620,500 — exactly the same amount which the Board had disapproved the previous year, not a penny more, not a penny less.1
Theoretically, there could be two different explanations for this denouement. The first is that DFR accepted the decision of the appellate tribunal which had reduced the assessment the prior year, but that by some strange coincidence there were new market developments which had increased the value of the property by $1,080,478— exactly the amount by which the Board had reduced the prior assessment. I suppose that anything is possible, but the problem with this hypothesis is that it roughly resembles the notion that the toss of an unbiased coin came out tails a few hundred times in succession. Most reasonable peo-pie would, I think, share my skepticism about this being what happened.
The second possibility is that the folks at DFR did not much care for the Board’s ruling and decided, in effect, to ignore it. We would hardly need the transcript in this case to hazard a guess, based on the comparative rarity of bizarre coincidences, that this second scenario seems a great deal more plausible than the first. But the implausibility of the first hypothesis becomes quite pronounced when we learn of some truly astonishing testimony by Troy Davis, the District’s line assessor. Mr. Davis explained that after he had analyzed all of the assessments, sales, and other pertinent data, he agreed with the view of his superi- or, Robert L. Klugel, Chief of DFR’s Real Property Tax Section, that “the Board results were ludicrous.” When the judge asked him whether he had “overruled the Board,” Mr. Davis answered with a simple “yes.”2 In other words, Mr. Davis told the Board, though not to its face, where it could take its “ludicrous” ideas about the value of the taxpayer’s property.
This is surely rather troubling testimony. Under our statute, the Board is an impartial arbiter to which the taxpayer is entitled to appeal if he or she contends that the property has been over-assessed. See D.C.Code § 47-825 (1990). An equalized assessment by the Board is entitled to the same treatment, for purposes of future valuation of the property, as a judicial decision. District of Columbia v. Burlington Apartment House Co., 375 A.2d 1052,1056 (D.C.1977) (en banc). It is one thing to say that the Board’s 1985 assessment does not bind DFR in 1986 because the facts have changed. It is quite another to legitimize the notion that DFR can lawfully ignore the Board because it finds that agency’s opinion to be “ludicrous.”
*1314In Mutual Benefit Life Ins. Co. v. City of Newark, 35 N.J.Super. 113, 116, 113 A.2d 185, 186 (1955), the court said:
True value cannot be established for tax purposes upon an assumption that a pri- or assessment, unappealed from, is correct. ... For if the earlier assessment is erroneous, then a decision as to the later year, which is predicated upon that assessment, would only continue the error.
In the present case, in which the appeal from the 1985 assessment resulted in what amounted to a reversal, the foregoing principles apply a fortiori.
“It is of the essence of an assessment that it fixes value as of a certain time. Each annual proceeding is separate and distinct from every other.” People ex rel. Hilton v. Fahrenkopf, 279 N.Y. 49, 52, 17 N.E.2d 765, 766 (1938). It has been held that a judgment in an earlier tax appeal does not conclusively establish the value of the property at the earlier date, even as between the parties to the earlier appeal. See In re Net Realty Holding Trust, 128 N.H. 795, 799-800, 519 A.2d 313, 316-17 (1986) (Souter, J.). Some courts, on the other hand, have applied the doctrine of collateral estoppel under such circumstances. See Uniroyal, Inc. v. Board of Tax Review, 182 Conn. 619, 634, n. 9, 438 A.2d 782, 789 n. 9 (1981); cf. Lethin v. Department of Revenue, 278 Or. 201, —, 563 P.2d 687, 689 (1977). In any event, if the 1986 assessment was to be the same as, or even based upon, the previous year’s, logic surely requires that it should conform to the amount approved by the Board, and not to the higher amount rejected by it. I am inclined to agree with the Supreme Court of New Jersey in Pitney v. State Bd. of Tax Appeals, 136 N.J.L. 157, 159, 55 A.2d 6, 7 (1947), that at least as between DFR’s 1985 assessment and the Board’s,
[sjince there was no evidence indicating that there had been a change of value for the years in question from that adopted and affirmed for the previous year, we think that the parties are bound by the results of the contest.
See also 16 Stephen M. Flanagan, McQuillin on the Law of Municipal CORPORATIONS § 44.111, at 392-94 (3rd rev. ed. 1984).
Aside from DFR’s disregard of the action of the Board, there remains the question how it can be, if DFR studied the value of the taxpayer’s property anew, that the result in 1986, ostensibly based on a consideration of extensive newly-acquired information, could be identical to the 1985 assessment. “It should be obvious that absent [attention to intervening developments], the carrying over of assessments each year from one general revaluation to the next is not the proper discharge of the assessor’s function.” Tri-Terminal Corp v. Borough of Edgewater, 68 N.J. 405, 414, 346 A.2d 396, 401 (1975), cert. denied, 425 U.S. 958, 96 S.Ct. 1739, 48 L.Ed.2d 203 (1976).
Both Mr. Klugel and Mr. Davis testified that they had studied new data for 1986 which had not been available for 1985. During the part of his testimony when he was so assuring the court, Mr. Klugel described the information which DFR studied in 1986, but which was unavailable in 1985, as quite extensive. He insisted that the new data included information about the income and operating expenses of some 500 office buildings, about leases, about interest rates, about land sales, about construction costs, and about “every aspect of valuation as far as we’re concerned.” Nevertheless, this newly-acquired knowledge did not, at least initially, result in the change of the 1985 assessment by a single penny.
Both Mr. Klugel and Mr. Davis were also asked to explain why the figures for the two years were identical; neither, at least in my judgment, was able to provide even a reasonably plausible answer. What Mr. Davis did say in his deposition, and to some extent reiterated in court, was that Mr. Klugel “recommended” that he use the 1985 figure, and that he agreed. Curiously, in attempting to explain why the assessments for the two years were the same, Mr. Klugel stated, contrary to the implications of his earlier testimony summarized above, that “there was a very limited amount of available additional market evi-*1315denee.” He decided that “we should hold primarily the existing assessments intact to the extent that we had again reviewed all of this [sic] data....” This is pretty close to carrying over to 1986 the assessment which the Board had disapproved for 1985.3
Neither Mr. Davis’ overruling of the Board nor the identity between DFR’s 1985 and 1986 assessments appears to have troubled either the trial judge or my colleagues in the majority. The Corporation Counsel devoted only one page of his brief to the issue whether DFR in fact conducted an authentic new assessment in 1986. The problems which I have discussed are not mentioned at all in the trial judge’s decision, and are treated quite summarily in the majority opinion. I find this all quite perplexing.
Coincidences do happen, but not all that often. Where a party’s evidence is believable only if an otherwise unlikely coincidence has occurred, the trier of fact ought in my view to be wary of crediting it, especially where, as here, no intelligible explanation of the coincidence has been provided. Mr. Klugel’s testimony was characterized by a major contradiction; he said he had lots of new information in 1986 (which purportedly tended to show that a new assessment had been made), but subsequently claimed that there was hardly anything new (so that the identity of the two assessments was supposedly not surprising). If courts should be “earthy” and practical in their evaluation of street encounters between the gendarmerie and the citizenry, see Cooper v. United States, 368 A.2d 554, 557 (D.C.1977), then a similar adherence to hard-nosed realism is surely á propos when the clash is between taxpayers and revenue agents.
In Burlington, supra, a five-member majority of this court, sitting en banc, appeared to share my present skepticism after a disapproved assessment for one year came up unchanged for the next:
Prior to the hearing on the contested 1973 assessment, Burlington received a notification from the District of the assessment on the property for fiscal year 1974. The figure thus conveyed was exactly the same as that which had been set for the previous fiscal year by the Board of Equalization and Review, which then was in the process of being disputed in Superior Court. It is clear that the 1974 figure was not based upon a reassessment utilizing updated sources of information, but rather was simply a routine repetition of the challenged 1973 assessment.
375 A.2d at 1056 n. 8. A three-member minority protested that
[a] majority of the en banc court, at the urging of appellee, concludes without basis that because the fiscal 1974 assessment of appellee’s property was the same as the equalized figure for fiscal 1973, no subsequent valuation of the property was made according to law. There is, of course, no testimony of record from which such a conclusion could be drawn.
Id. at 1059 (Kelly, J., dissenting) (footnote omitted). A reading of the two opinions together suggests that the members of the majority, just like the undersigned, were dubious about the plausibility of odd coincidences and were ready to infer from the identity of the numbers that there had been no genuine reassessment for the second year.
In Brisker, however, a division of this court interpreted Burlington differently:
Burlington did not hold, as taxpayers assert, that subsequent assessments that are identical to the assessment found invalid are themselves necessarily invalid. Rather, Burlington held that when *1316the trial court finds an assessment invalid and itself sets a valuation, that “valuation must constitute the continuing basis for taxation until there is a superseding valuation which has been made according to law.” Burlington, 375 A.2d at 1056. After an invalid assessment, the law requires the District make a new valuation in accordance with the statute. Id. Here, with respect to the 1984 assessments, the trial court found that the District had made such a new valuation based on comparable sales, notwithstanding the assessor’s testimony that in the course of doing so, he had looked at the 1983 assessments.
510 A.2d at 1040. Fairly or unfairly, Brisker seems to me to represent the graveyard of the taxpayer’s hopes in this case in light of the trial judge’s findings of fact, for the decision appears to legitimize such findings.
The present case arguably differs from Brisker in two respects. First, there is no suggestion that, in Brisker, the DFR agents acted out of a belief that the court’s action disapproving the prior assessment had been “ludicrous.” Second, there is no indication that the court was asked in Brisker to mistrust purported coincidences, or that it ever considered that question. Neither distinction, however, seems to me to be sufficient to take the present case out of Brisker’s reach. Even if DFR’s words in Brisker were less contemptuous of higher authority than those of Mr. Davis, its deeds were the same as those of DFR here. Moreover, the problem of coincidences was there in Brisker for anyone who cared to address it, and its presence apparently did not inhibit the court from affirming findings predicated on a coincidence similar to the one in the present record.
Moreover, when a taxpayer appeals to the Superior Court, the whole case, both facts and law, is subject to de novo evaluation. Washington Post Co. v. District of Columbia, 596 A.2d 517, 521 n. 2 (D.C.1991) (citations and internal quotation marks omitted). Both the Board and the trial judge sustained a 1986 assessment identical to that proposed by DFR in 1985. The contention that the court should have adopted for 1986 the Board’s reduced assessment for 1985, rather than DFR’s original proposed assessment for that year, is in some measure undercut by the Board’s own action with respect to 1986.4 Under all of these circumstances, and in light of M.A.P. v. Ryan, 285 A.2d 310 (D.C.1971), I am not prepared, despite my pronounced misgivings, to vote for reversal of the judgment.
II
DFR’s troubling initial handling of the 1986 assessment was compounded by what followed after the taxpayer sought judicial review. The District, as I have noted, eventually amended its answer and alleged that the correct assessment for 1986 should have been $17,830,000 rather than $14,620,-500. This was an increase of more than 3.2 million dollars, or about 22 percent. The taxpayer contends that by abandoning the original lower assessment, the District relieved him of the obligation to prove that the original assessment was incorrect. Although I agree with the District’s conclusion that this contention should be rejected, I cannot agree with its articulation of the issue.
Mr. Klugel testified that, in asking that the property be assessed at $17,830,000, he was not “backing off ... where we stood for tax year 1986.” The trial court accepted this formulation. But what the District is saying when it asserts its “no backing off” theory is essentially that it can have its multi-million dollar cake and eat it too. When Mr. Klugel testified that the correct assessment of the property was more than 17.8 million dollars, allegedly on the basis of new information5 provided by the tax*1317payer, he was necessarily opining that it was not worth just over 14.6 million dollars, as DFR had originally asserted. The correct value cannot be both 17.8 million and 14.6 million at the same time. To say that Mr. Klugel was not “backing off” his earlier assessment is to give that colloquialism a new meaning. I think that the taxpayer’s argument fails, however, for the reasons articulated by my colleagues. See ante, majority opinion at [1312],
But the District’s attempt to increase the assessment by more than three million dollars to 17.8 million dollars and change6 presents us with a new “coincidence” problem regarding the judge’s findings. The District’s ultimate submission was that incomplete information supported a valuation of 14.6 million dollars, but that on the basis of complete data (including material furnished by the taxpayer), the proper assessment was 17.8 million dollars. The trial judge, however, found on the basis of the complete information that the property was worth 14.6 million dollars. I suppose that the result is defensible on the theory that the original assessment, even if made on incomplete information, is presumed correct, and that neither side proved it wrong. Nevertheless, I am troubled by the intrinsic improbability characterizing the judge’s finding as to the value of the property.
Moreover, I am concerned by the implications of the District’s amendment of its answer. Our city is in a financial crisis which threatens the jobs of its employees, as well as many essential services to its citizens. If it is true, as Mr. Klugel asserted, that the correct value of the taxpayer’s property was more than $17.8 million, then it ought to have been assessed in that amount in the first place. If DFR needed more information to perform a proper assessment, it ought to have asked for it. Our citizens are being seriously shortchanged if commercial buildings are being substantially under-assessed because DFR lacks sufficient data.
But it is troubling, at least to me, that this large increase in DFR’s assessment— more than twenty percent — came about as a result of the taxpayer’s exercise of the right to request judicial review of some very shaky decisions by the bureaucracy. I do not know what the intent of the amendment of the District’s answer was,7 but its logical consequence was to send a rather frightening message — challenge us in court and we will try to make you feel it in your wallet! Another taxpayer who has a legitimate grievance may well think again before seeking judicial review if the result is likely to be a claim by the District that he owes tens of thousands of dollars more to the tax man than the tax man had previously contended.
I recognize that if DFR learns during litigation that a taxpayer’s property has been substantially under-assessed, it has an obligation to the citizens of our community to right that wrong. The trial judge’s findings in this case, though, indicate that there was no such under-assessment. If the judge was right, then the stern message emanating from DFR turns out to have been altogether gratuitous. If at all possible, in any event, the agency should obtain all necessary information before the original assessment, and avoid placing a formidable chill upon the taxpayer’s exercise of the right to seek judicial redress.
Ill
It is important not only that justice be done but that it appear to be done. I do not think that the taxpayer in this case believes that he has been fairly treated. I can well understand his point of view. If it were the court’s function to dispense its members’ personal notions of fairness, I would vote to reverse. But
*1318[o]ur duty, to paraphrase Mr. Justice Holmes in a conversation with Judge Learned Hand, is not to do justice but to apply the law and hope that justice is done.
Bifulco v. United States, 447 U.S. 381, 402, 100 S.Ct. 2247, 2259, 65 L.Ed.2d 205 (1980) (Burger, C.J., concurring). With only limited confidence that justice has been achieved, I concur in the judgment.
. Subsequently, after the taxpayer had the temerity to challenge DFR’s assessment in court, DFR amended its answer to claim that the value of the property was not 514,620,500 after all, but $17,830,000 — an increase of more than 3.2 million dollars!
. Mr. Davis’ responses were perhaps blunter than, but not out of line with, the testimony of Mr. Klugel, who proclaimed that he was "absolutely not” bound by the Board’s 1985 assessment. Mr. Klugel, however, justified his answer on the ground that "change is going on all the time."
. In District of Columbia Redev. Land Agency v. District of Columbia, 106 Daily Wash.L.Rptr. 2257, 2264 (D.C.Super.Ct. Dec. 15, 1978), (the DCRLA case), Judge Penn held that affidavits from the very same Mr. Kluge] and from another subordinate, stating that they had reviewed new data and that each "saw no reason to change his prior assessment [which had been disapproved by the court] and that he assigned the same value for the second year,” were insufficient to show that a new assessment had been made. Mr. Klugel changed his phrasing slightly in the present case but, unlike my colleagues, I see little difference, in the context of the real world, between the affidavits in DCRLA and Mr. Klugel’s testimony described above.
. I wonder if the Board would have approved the 1986 assessment if its members had known that it was predicated on Mr. Davis’ belief that the Board's action the previous year had been "ludicrous.”
. It is important to note that the “new information” on which Mr. Klugel predicated the proposed 3.2 million dollar increase in the assessment was not the same “new information” which DFR had available for its original 1986 *1317assessment but not for its 1985 assessment. The increase to $17,830,000 was based on documents made available by the taxpayer in connection with the lawsuit.
. If $30,000 can be described as "change.”
. But cf. Rabinowitz v. United States, 366 F.2d 34, 56-57 (5th Cir.1966) (en banc): "The [DFR officials] must be held to have intended the natural result which flowed from their conduct. ... The fruits of the harvest were clear to anyone who cared to look."