KUSKIN, J.T.C.
Plaintiff appeals the tax year 2001 property tax assessment on two parcels of vacant land in the Borough of Kinnelon, designated on the Borough Tax Map as Block 26, Lot 115 (“Lot 115”) and Block 30, L.1.03 (“Lot 1.03”) (hereinafter referred to together as the “subject property”). Lot 115 has a stipulated area of 454.06 acres, and Lot 1.03 has a stipulated area of 444.15 acres divided into two sections, Area C containing 262.35 acres and Area D containing 181.80 acres. Lot 115 was assessed at $7,316,000, and Lot 1.03 at $8,419,500. Kinnelon implemented a municipal-wide revaluation for tax year 2001 so that the Chapter 123 Ratio is inapplicable. See N.J.S.A. 54:51A-6d.
Plaintiff presented the testimony of three witnesses: one of its principals (Dennis Lam); a civil engineer; and an appraiser. Mr. Lam described the subject land and its development potential (based largely on the engineer’s analysis discussed below), and discussed the necessity for blasting and other difficulties and costs relating to his development of nearby land. Plaintiffs engineer described the topographic characteristics of the land, discussed maps showing those characteristics, and testified as to development potential. Plaintiffs appraiser testified as to the value of the subject property using the sales comparison approach and, as to Lot 115, also using a subdivision analysis as a check on his sales comparison approach. The appraiser determined a value for Lot 115 of $3,065,000 and a value for Lot 1.03 of $2,920,000. Defendant presented one witness, an appraisal expert, whose testimony was limited to a critique of the valuation analysis by plaintiffs appraiser. Defendant presented no affirmative evidence as to value.
The subject property is located within the gated community of Smoke Rise, and contains the following topographic features: [60]*60substantial areas of steep slopes, wetlands, soil types which are difficult for construction, and rock outcroppings. Few public roads provide access to the property, with access to Lot 1.03 being particularly limited. Entry into both Lot 115 and Lot 1.03 from some abutting public streets is impossible because of wetland areas or steep slopes. Housing prices in the area of the subject property are primarily in the $400,000 to $600,000 range, but some properties have sold somewhat below or significantly above those amounts. For Lot 115, plaintiffs appraiser determined a highest and best use consistent with the surrounding use, that is, for single family residential development of forty-eight lots. This lot quantity was based on the analysis by plaintiffs engineer discussed below. With respect to Lot 1.03, the appraiser concluded a highest and best use for Area C of limited residential development in conjunction with the development of Lot 115 and the extension of a road through Lot 115 to Area C. For Area D, the appraiser concluded, based on the engineer’s analysis, that the highest and best use was for six single-family residential lots. Although the appraiser expressed concern that the substantial development costs estimated by the engineer for Areas C and D might inhibit their use for residential development, the appraiser valued the lots for that use.
Plaintiffs engineer testified that he reviewed the location of slopes and wetlands on Lots 115 and 1.03, determined where access to each of the lots would be available, and, then, based on the Residential Site Improvement Standards (RSIS) promulgated by the New Jersey Department of Community Affairs, N.J.A.C. 5:21-1.1 to -8.1, he calculated the number of subdivided lots that could be developed on each Lot. The engineer testified that the RSIS establish a limitation on traffic which translates into a maximum of twenty-four lots located on a cul-de-sac street. If the street is an extension of an existing dead-end or cul-de-sac street, then the number of lots on the preexisting portion of the street must be subtracted from twenty-four to establish the maximum number of lots permitted on the extension. The engineer did not attempt to determine lot sizes or a subdivision layout, but merely made calculations based on the RSIS. Based on these calculations, [61]*61the engineer testified that, on Lot 115, the maximum number of lots would be forty-eight using access to the site from Green Hill Road (nine lots), Tower Hill Lane (twenty-two lots), Butternut Terrace (fifteen lots), Undercliff Road (one lot), and Mountain Road (one lot). The engineer further testified that development costs for these lots, including road improvements, would be approximately $3,000,000, and that wetlands, severe slopes, and access limitations would preclude development beyond the areas he identified. With respect to Lot 1.03, the engineer determined that six lots could be developed in Area D using the only available access, that is, from Orchard Road. He testified that his “rough estimate” of development costs for this Lot, including road improvements, was $1,000,000. The engineer determined that no development, other than in connection with Lot 115, would be possible in Area C because access from the only abutting road, Joanna Way, was prevented by wetlands.
In his sales comparison approach, the appraiser determined a per acre value for the subject property, relying on the following five sales of vacant land, all of which involved purchases for purposes of open space preservation:
Sale Number 1 took place on January 7, 1997. The sale property was located in Kinnelon Borough and contained 881.393 acres. The seller was the Miklos Felkay Family Limited Partnership and the purchaser was the New Jersey Department of Environmental Protection (DEP), pursuant to the New Jersey Green Acres land acquisition statutes. N.J.S.A. 13:8A-1 to -55. Plaintiffs appraiser estimated that the property had a development potential of forty-eight to fifty lots.
Sale Number 2 took place on December 20, 2001. The sale property was located in Roxbury Township and contained 100.173 acres. The sellers were Clifford Johnson, George Johnson, and Elizabeth Moses and the purchaser was the DEP under the Green Acres statutes. The property sold without municipal approvals but after submission of a preliminary subdivision plan showing fourteen building lots ranging in size from two to thirteen acres.
[62]*62Sale Number 3 took place on October 11, 2002, although the contract date was September 15, 2001. The sale property was located in Roxbury Township and contained 99.863 acres. The seller was Three Investors Associates, Inc. and the purchaser was the DEP under the Green Acres statutes, acting in conjunction with the Morris Parks and Land Conservancy. The Conservancy estimated a development potential for the property of approximately fifteen lots.
Sale Number 4 took place on July 14, 1998. The sale property was located in Jefferson Township and contained 96.79 acres. The seller was the Boys’ and Girls’ Club of Clifton and the purchaser was Jefferson Township. Plaintiffs appraiser estimated that the property had a development potential of twenty to thirty lots.
Sale Number 5 took place on January 12, 1999. The sale property was located in Kinnelon Borough and contained 29.53 acres subdivided from a 69 acre site. The seller was Joseph Piccone and the purchaser was the Morris Parks and Land Conservancy, which resold the property to Kinnelon Borough for $1.00 on February 8,1999. Plaintiffs appraiser estimated that the property had a development potential of two to three lots.
Plaintiffs appraiser testified that, although each of the sales was to a governmental or land preservation entity, he deemed the sale prices, after appropriate adjustments, to be reflective of the fair market value of the subject property for a highest and best use of residential development. He determined that the sales price for each comparable property was based on appraisals, and that each appraisal used residential development as the highest and best use for the property. Accordingly, the appraiser opined that, even though the actual use to which the respective buyers intended to put the various properties was for open space preservation, the price paid for each property reflected its value for residential development.
Plaintiffs appraiser acknowledged that the development potential for each of the comparable sale properties entered into the negotiation of the sale price. He stated, however, that the exact size of the subdivided lots on the sale properties or on the subject [63]*63property was not critical to his per acre valuation because, based on the topography and soil types of the sale properties and the subject property, land in excess of the necessary building envelope could not be used for grassy backyards or for pools or tennis courts and, therefore, would contribute only a marginal increment to the value of a lot.
It is well established that assessments are presumed valid, and a taxpayer has the burden to establish a right to a reduction under the following standard:
The presumption in favor of the taxing authority can be rebutted only by cogent evidence, a proposition that has been long settled. The strength of the presumption is exemplified by the nature of the evidence that is required to overcome it. That evidence must be “definite, positive and certain in quality and quantity to overcome the presumption.”
[Pantasote Co. v. City of Passaic, 100 N.J. 408, 413, 495 A.2d 1308, 1310 (1985) (citations omitted).]
After a careful review and weighing of the testimony and exhibits presented by both parties and after considering the credibility of the witnesses, I conclude that plaintiffs proofs are inadequate to warrant a reduction in the assessment on either Lot 115 or Lot 1.03.
In reaching my conclusion, I recognize that, if I were to apply the standard of proof appropriate under R. 4:37-2(b) (applicable to a motion at the end of the plaintiffs case) and under R. 4:40-1 (applicable to motions for judgment at the conclusion of the ease), I would hold that the proofs, when viewed under that standard, are sufficient to overcome the presumption of validity attaching to the assessments under appeal. However, when analyzing and weighing the evidence, as I must do at the conclusion of a trial, I am not required to accept the plaintiffs testimony as true or accord all favorable inferences to that testimony, but must determine the credibility and probative value of the evidence. See MSGW Real Estate Fund, LLC. v. Mountain Lakes Bor., 18 N.J.Tax 364, 373-79 (Tax 1998). After completing my analysis and weighing of plaintiffs proofs, I have several significant concerns as to the comparable sales valuation analysis by plaintiffs appraiser. Any one of these concerns, by itself, might not be sufficient to deny plaintiff relief. Cumulatively the concerns are [64]*64more than sufficient to render plaintiffs proofs inadequate in quality and quantity to establish the value of the subject property and overcome the presumption of validity.
My concerns are the following:
1. None of the five comparable sales sold for use at its highest and best use. Even though the properties were appraised for residential use (see discussion in Item 2), the intended use of the purchaser may have affected the price. In Jersey City v. Parsippany-Troy Hills Tp., 16 N.J.Tax 504 (Tax 1997), aff'd, 17 N.J.Tax 538 (App.Div.1998), I discussed the impact of highest and best use on sale price as follows:
In the marketplace, land having a highest and best use which is higher on the scale of uses than a particular buyer’s intended use would not be sold to that buyer. The price warranted by the highest and best use would be unacceptably high to the buyer, and the price warranted by the buyer’s intended use would be unacceptably low to the seller. If the buyer purchased such land for a price based on the land’s highest and best use, the buyer’s use would not sustain or support the price paid.
[Id. at 519.]
If, therefore, land having a highest and best use for residential development was purchased for a lower and lesser use, open space, either the seller or purchaser may have had a motivation not typical for the open market. This concern as to motivation is reinforced by the statutory provision stating that, in Green Acres purchases, the Commissioner of Environmental Protection shall “[i]nsofar as practicable, limit acquisition to predominantly open and natural land and minimize the cost of acquisition and the subsequent expense necessary to develop such land for recreation and conservation purposes” N.J.S.A. 13:8A-39b (emphasis added). See also N.J.S.A. 13:8A-5(b) and 13:8A-23b (predecessor Green Acres statutes containing the same provision).
The significance of the actual use of a property is recognized in The Appraisal of Real Estate, the appraisal text on which the Tax Court has frequently relied.
Any difference in the current use or the highest and best use of a potential comparable sale and the subject property must be addressed. The appraiser must recognize the difference and determine if the sale is an appropriate comparable and, if so, whether an adjustment is required. In most eases the buyer or buyer’s agent must confirm the ultimate use for which the comparable was purchased.
[65]*65For example, an apartment complex purchased for conversion to condominiums may reflect a sale price above the market level for apartment properties. This property would not be an appropriate comparable for the “as is” valuation of an apartment complex for which no change in use is intended or one for which the highest and best use remains apartment use.
[Appraisal Institute, The Appraisal of Real Estate, 436-37 (12th ed.2001)J
See also Riegel Prods. Corp. v. Milford Bor., 13 N.J.Tax 546, 561 (Tax 1994) (concluding that sales of large manufacturing properties for use by the purchaser for warehouse or multi-tenant occupancy were not indicative of the value of a fully utilized manufacturing plant containing features suitable for heavy manufacturing).
Motivations that might have affected the sale price of some or all of the comparable sales can be inferred from facts relating to four of the comparables used by plaintiffs appraiser. The deed for sale Number 1 contained the seller’s express reservation of the right to develop “a maximum of sixty-seven (67) detached single family homes” on adjoining property which the seller retained. Consequently, the sale may have been partially motivated by a desire to preserve development rights and enhance the value of the remaining lots by surrounding them with open space, and these factors may have constituted a form of consideration which the seller deemed valuable but which was not reflected in the deed. A similar situation existed with respect to the appraiser’s sale Number 5, a conveyance to the Morris Parks and Land Conservancy of approximately thirty acres out of an approximately sixty-nine acre tract. The remaining acreage was subsequently sold to a developer for a per acre price almost twice the price paid by the Conservancy. The owner of the appraiser’s sale Number 3 had proposed a golf course development on the property and not residential development. Therefore, in selling for open space use, the grantor may not have been motivated to obtain a price for the property reflecting its value as a site for residential development. Similarly, the Boys’ and Girls’ Club of Clifton, the owner of the appraiser’s sale Number 4, was not a residential developer and may have been motivated by a desire to dispose expeditiously of property it no longer needed in order to generate funds for its other activities.
[66]*662. The sales prices for the comparable sales were determined by appraisal and not by competitive bidding in the marketplace. Defendant’s appraiser testified that, in his opinion, a sale based on appraisals is not a market transaction, and, therefore, none of the sales used by plaintiffs appraiser was a reliable, or even usable, comparable. Further doubt as to the reliability of the sales as indicators of market value results from uncertainty as to whether sellers, as well as purchasers, obtained appraisals. The testimony of plaintiff’s appraiser established that the purchaser in each sale obtained one or two appraisals. Under the regulation applicable to Green Acres purchases, the DEP or other acquiring agency must obtain two appraisals if the estimated land value is $250,000 or more. N.J.A.C. 7:36-6.6(a). However, the appraiser’s testimony did not establish that the seller in each sale also obtained an appraisal.
“Despite differing opinions on individual aspects of the market value definition, it is generally agreed that market value results from the collective value judgments of market participants.” The Appraisal of Real Estate, supra, at 21. The definition of market value adopted in The Appraisal of Real Estate is the following:
The most probable price, as of a specified date, in cash, or in terms equivalent to cash, or in other precisely revealed terms, for which the specified property rights should sell after reasonable exposure in a competitive market under all conditions requisite to a fair sale, with the buyer and seller each acting prudently, knowledgeably, and for self-interest, and assuming that neither is under undue duress. [Id. at 22.]
When a property is sold based on appraisals prepared by one or both sides, the sale is divorced from “a competitive market.” That the property may have been exposed to the market for a reasonable period of time before the “appraisal sale” does not eliminate the significance of the fact that, ultimately, the sale was not the result of offers in a competitive market.
The risk that a sale price based only on appraised value will not be reflective of the market conditions envisioned by the foregoing definition is illustrated by the history of one of the comparable sales used by plaintiffs appraiser in his appraisal report prepared for Jefferson Township in connection with its purchase of the appraiser’s comparable sale Number 4. In the report, the apprais[67]*67er relied on four comparable sales, the third of which was a sale on January 19, 1995 from Rock Pear, Inc. to Kinnelon Borough for a price of $1,179,250. The appraiser’s comments on this sale include the following: “The sale price was established through litigation after the Borough attempted to acquire the site through negotiation. An initial offer was made by the Borough at $710,000 ($4,063/AC) based on appraisals submitted to N.J. Green Acres.” The litigation resulted in a sales price 66% higher than the offer made by the Borough of Kinnelon based on appraisals. This suggests that a sale price based solely on appraisals may be significantly understated, notwithstanding the assumed desire of a seller to maximize the sales price based on a property’s highest and best use.
3. None of the sales is a usable sale for purposes of determining sales ratios. Under N.J.A.C. 18:12-1.1(a), which defines usable sales for purposes of establishing sales ratios pursuant to N.J.S.A. 54:1-35.1, Item 15 excludes “sales to or from the United States of America, the State of New Jersey, or any political subdivision of the State of New Jersey; including boards of education and public authorities.” The regulation provides that the presumption of non-usability may be overcome “if after full investigation it clearly appears that the transaction was a sale between a willing buyer, not compelled to buy, and a willing seller, not compelled to sell, and that it meets all other requisites of a usable sale.” N.J.A.C. 18:12-1.1(b). At the very least, the fact that all of the comparable sales used by plaintiffs appraiser are non-usable sales creates doubt as to the reliability of the sales as indicators of fair market value for tax purposes. In addition, for the reasons set forth in Items 1 and 2 above and in the discussion below, I am not satisfied that the appraiser’s investigation of his comparable sales satisfied the regulation’s standard for overcoming the presumption of non-usability.
4. Plaintiffs appraiser failed, to consider the significance of the number of potential subdivided, lots on each of his comparable sale properties. Plaintiffs appraiser acknowledged that the number of developable lots is a factor in the negotiation of a sales price for vacant land having potential for residential development. [68]*68However, he had no definitive estimate of lot potential for four of his five comparable sales, and only a preliminary indication for the fifth sale. As to three of his sales, Numbers 1, 4, and 5, the appraiser made his own estimate of development potential without the benefit of any engineering studies. As to Sale Number 3, the development potential used by the appraiser was estimated by the not-for-profit purchaser which was not in the business of residential land development. For Sale Number 2, a preliminary site plan had been submitted but not approved at the time of the sale. The significance of the lot potential of a parcel of vacant land is indicated by an appraisal prepared by plaintiffs appraiser for the Morris County Park Commission of an approximately thirty-four acre tract of vacant land in Kinnelon owned by Davanne Realty. After determining that the highest and best use for the site was for subdivision into six residential building lots, the appraiser valued the property based on its value per lot and not per acre. Cf. Emmis Broadcasting Corp. of N.Y. v. East Rutherford Bor., 14 N.J.Tax 524, 542 (Tax 1995), aff'd, 16 N.J.Tax 29 (App.Div.1996) (commenting on the significant difference in an appraiser’s valuation conclusion as a result of his analysis of comparable land sales using a per acre unit of comparison instead of developable square feet of office space, the unit of comparison he had used in other appraisals).
5. Plaintiff’s engineer and appraiser miscalculated the number of potential lots which could be located on the subject property. The number of subdivided lots which could be located on Lot 115 and Lot 1.03, as determined by plaintiffs engineer (using the RSIS) and adopted by plaintiffs appraiser, appears to be incorrect based on a clarification of the RSIS issued by the New Jersey Department of Community Affairs on October 26, 2000 and amended April 19, 2001. Under this clarification, when an existing stub street is extended for new development and the existing and extended street form a cul-de-sac, the maximum average daily traffic calculation is to be performed with respect to the extension only, so that the number of lots that can be developed is calculated for the extension only. Plaintiff contends that the zoning ordinance of Kinnelon Borough imposes even more stringent stan[69]*69dards than the RSIS and limits the number of lots to fewer than determined by plaintiffs engineer. This argument, if correct, casts doubt on the reliability of the engineer’s estimate which apparently did not take into account local zoning ordinances. Although the RSIS might control the municipal ordinance with respect to the number of lots permitted on a cul-de-sac street, see N.J.A.C. 5:21-1.5(a) and — 1.5(d), the extent to which a municipal zoning ordinance may supersede the RSIS has not been clearly defined. See New Jersey State League of Municipalities v. Department of Community Affairs 158 N.J. 211, 226-27, 729 A.2d 21, 29-30 (1999) (conflicts between the RSIS and local zoning policy are appropriately addressed in “as applied” challenges to particular regulations). In either event, however, it appears the estimate of lot potential by plaintiffs engineer is incorrect.
Based on the clarification to the RSIS, the lot potential for Lot 115 is seventy-four lots not forty-eight, and the lot potential for Lot 1.03 is twenty-four lots and not six. Neither plaintiffs engineer nor plaintiffs appraiser made any effort to lay out the proposed subdivision on the subject property or determine lot sizes. Therefore, I cannot determine whether the maximum number of lots permitted under the RSIS would fit on either of the subject lots, although there appears to be more than adequate acreage available. The mere fact that additional lots are permitted suggests some potential inaccuracy in the valuation analysis by plaintiffs appraiser.
In reviewing the maps submitted with respect to Lot 1.03, I note that South Glenn Road passes within approximately twenty-five feet of one corner of Area D of this lot. The intervening twenty-five feet is located on building lots not owned by plaintiff. It is conceivable, however, that an easement could be obtained over one or both of these lots or that one of the lots could be purchased in order to provide access to a large, relatively flat portion of Area D on which, under the RSIS, an additional twenty-four lots could be developed. Neither plaintiffs engineer nor plaintiffs appraiser considered this possibility. I raise it not because the proofs indicate, or that I conclude, that this potential access is likely. However, the failure by the appraiser or engineer [70]*70to consider this possible means of access suggests that they did not adequately consider the development potential of the subject property.
6. Plaintiffs appraiser failed to consider subdivided lot sizes on his comparable sale properties or on the subject property. As set forth above, plaintiffs appraiser valued the subject property on a per acre basis and, therefore, analyzed his comparable sales on that basis without taking into account their respective development potentials. As also set forth above, the appraiser acknowledged that development potential would enter into the negotiation of the sales price for a parcel of vacant land. He testified that, for the subject property, acreage of a lot in excess of the minimum required in the zone (60,000 square feet) would add only marginally to the value of the lot. This conclusion is belied by the comparable sales of finished or partially finished lots used by the appraiser in his subdivision analysis. There, the appraiser relied on four sales of lots suitable for development. The smallest lot, 1.46 acres, sold for $162,000; the largest, containing 5.17 acres, sold for $242,500. A lot containing three acres sold for $205,000. These price differentials can be partially explained by the locations of the lots. However, the price per lot appears to vary directly with the size of the lot. Accordingly, not only the number of lots but the size of the potential lots would be a factor in the negotiation of the sales price for the subject vacant land.
Neither plaintiffs appraiser nor plaintiffs engineer determined potential lot sizes on the subject property, and the appraiser provided no information as to lot sizes on any of his comparable sales except for a range of sizes for his Sale Number 2. The appraiser’s analysis, therefore, does not address adequately the development potential of the subject property, casting doubt on his opinion of value. See The Appraisal of Real Estate, supra at 331 (stating that “it is the economic use of a parcel of land that determines its value in a particular market.”).
7. The sizes and development potentials of the comparable sales are significantly different from the size and development potential of Lot 115 and of Lot 1.03. Plaintiffs appraiser adjusted [71]*71his comparable sales for the size differentials, but the comparability of the lots nevertheless remains questionable. “A major premise of the sales comparison approach is that the market value of a property is related to the prices of comparable, competitive properties.” The Appraisal of Real Estate, supra, at 417 (emphasis added). The appraiser’s sale Number 5, consisting of 29.53 acres, is less than seven percent of the size of each of the subject lots. This sale, therefore, is entitled to virtually no weight in valuing the subject property. The appraiser’s sale Number 1 is almost twice the size of each of the subject lots. His sales Numbers 2, 3 and 4 are each less than one-quarter the size of each of the subject lots. Each of the comparable sale properties has a development potential significantly different from the potential of Lot 115 and Lot 1.03. Therefore, each property may not be competitive with the subject property.
8. Plaintiffs appraiser made large gross adjustments (ignoring plus and minus signs) to three of his comparable sales. His gross adjustments to his comparable sale Number 1 were 30% plus 10% for time, his gross adjustments to his Sale Number 3 were 30% plus 10% for time, and his gross adjustments to his sale Number 4 were 45% plus 11% for time. The magnitude of these adjustments alone suggests a lack of comparability between these sales and the subject property. See, e.g., Global Terminal & Container Service v. City of Jersey City, 15 N.J.Tax 698 (App.Div. 1996) (affirming the Tax Court’s rejection of comparable sales because of the magnitude of gross adjustments). See also The Appraisal of Real Estate, supra, at 447 (“A net adjustment figure may be misleading because one cannot assume that any inaccuracies in the positive and negative adjustments will cancel each other out.”).
9. The appraisals used to establish the sales prices for the comparable sales were based on sales comparison approaches using non-market transactions. Defendant’s appraiser testified that his investigation of the appraisals used to establish the sales prices for the comparable properties used by plaintiffs appraiser revealed that the appraisals generally were based on analyses of Green Acres or other open space sales. Defendant’s appraiser [72]*72concluded from this that appraisals for Green Acres purchases were self-perpetuating at low-end valuations based on non-market transactions involving sales based on appraisals.
As discussed above, plaintiffs appraiser prepared an appraisal for Jefferson Township in connection with the Township’s purchase of the appraiser’s comparable sale Number 4. Of the four comparable sales included in that appraisal, one was a sale by AT & T to Roxbury Township. AT & T is not a developer of residential real estate and appears to have been disposing of excess land. See MSGW Real Estate Fund, LLC v. Mountain Lakes Bor., supra, 18 N.J.Tax at 401 (describing a situation in which a corporate decision to dispose of excess land “drove the real estate decision” and may have resulted in a sale below the maximum price attainable). A second sale was to the State of New Jersey consisting of property described by the appraiser as having “limited development potential.” The third sale by the Borough of Kinnelon for inclusion in park lands had its price determined by litigation. The fourth sale, also having “limited development potential,” was a sale at auction by the Resolution Trust Company. Accordingly, the appraisal of the Jefferson Township property was based primarily on sales for open space and one by the Resolution Trust Company at auction. These sales were used to establish a purchase price representing the value of the property for a highest and best use of residential development. The gross adjustments to the four comparable sales were in the following respective amounts: 35% plus 7% for time; 60% plus 18% for time; 55% plus 13% for time; and 55% plus 23% for time. The extent of these adjustments raises doubt as to the comparability of the sales. Consequently, the reliability of the appraisal for Jefferson Township as establishing fair market value for residential development is questionable.
10. Some of the comparable sales used by plaintiffs appraiser are either remote in time from the October 1, 2000 valuation date applicable to this appeal or occurred after that date, and the appraiser failed to make all appropriate adjustments. Comparable sale Number 1 used by the appraiser involved two deeds, one conveying approximately 881 acres plus roadway easements for a [73]*73price of $3,969,391.50, and the other granting a conservation easement encumbering a total of approximately ninety-six acres for a price of $433,075.50. The sale occurred three years and ten months before the October 1, 2000 valuation date applicable to plaintiffs appeal. The sale is somewhat remote in time, and plaintiffs appraiser did not discuss the basis for allocation of the purchase price between the two deeds. The appraiser’s sale Number 2 took place one year after the applicable valuation date, involved five acre minimum zoning, and had development potential for fourteen building lots ranging in size from two to thirteen acres. The Tax Court, although frequently considering post-valuation date sales, has expressed concern about their probative effect. See, e.g., Wedgewood Knolls Condo. Ass’n v. West Paterson Bor., 11 N.J.Tax 514 (Tax 1991); Newport Ctr. v. City of Jersey City, 17 N.J.Tax 405, 425-26 (Tax 1998). Even if I do not exclude sale Number 2 from consideration because of its date, the probative effect of the sale is limited by the appraiser’s failure to discuss or make any adjustments for the zoning differences between the sale property and the subject property or adjustments for the potential lot sizes on the sale property. The appraiser’s sale Number 3 is problematical as another post-valuation date sale, occurring two years after October 1, 2000 (the contract of sale was signed one year earlier), and is also problematical because, as discussed above, the seller intended to develop the property as a golf course and not for residential development, and, therefore, may not have sought to sell the property at its value for residential development.
As a result of the questions and doubts discussed in the preceding numbered items, I conclude, after analyzing, weighing, and evaluating the evidence before me, that plaintiff has failed to present proofs sufficient to establish the value of the subject property and overcome the presumption of validity which attaches to the subject assessments. In reaching that conclusion, I am cognizant of the discussion in Ford Motor Co. v. Edison Township, 127 N.J. 290, 604 A.2d 580 (1992), in which our Supreme Court noted that, in order to be admissible in evidence, comparable sales need not be similar “in all comparative respects, so long as there [74]*74is sufficient similarity in some significant respects to permit the expert testifying, or the fact-finder, to draw rational probative valuation inferences from the sales cited.” Id. at 307, 604 A.2d at 588. I am also cognizant of our Supreme Court’s instruction in Glen Wall Associates v. Wall Township, 99 N.J. 265, 491 A.2d 1247 (1985) that “the volume of information that is required to support an expert’s opinion must be kept within practical and realistic limits.” Id. at 280, 491 A.2d at 1255. The Court also cautioned, however, that: “[w]e do not suggest that a court should accept an expert’s opinion that is unsubstantiated.” Ibid.
In applying the quoted language from Ford Motor, I conclude, as discussed above, that any one of the ten enumerated concerns, by itself, might not be sufficient to preclude reliance on the comparable sales used by plaintiffs appraiser. Cumulatively, however, the concerns result in my having substantial doubt as to the reliability of the sales, and, therefore, result in my conclusion that plaintiffs proofs are inadequate to prove the value of the subject property and overcome the presumption of validity. I accept as credible the testimony by the appraiser that the comparable sales he used were the best sales he found in the market he investigated. However, as I commented in Newport Center, supra, 17 N.J.Tax 405, a sale which provides an inadequate basis to determine whether the plaintiff has overcome the presumption of validity of an assessment or determine the value of the property under appeal “does not become more probative, more convincing, or more admissible, because of the paucity, or even non-existence, of data which would provide such a basis.” Id. at 423-24. With respect to Glen Wall, I note that the assessments in issue before me are substantial and that plaintiff seeks to reduce them by more than half. Under these circumstances, I cannot simply ignore deficiencies in plaintiffs case, speculate that adequate proofs would be disproportionately costly, and award plaintiff the relief it seeks.
Plaintiffs appraiser used a subdivision analysis as a “check” on his comparable sales approach valuation for Lot 115. The appraiser did not use this check for Lot 1.03. In the [75]*75subdivision analysis, the appraiser did not determine a specific value for Lot 115, but merely established a range of values and concluded that the value for the lot demonstrated by his comparable sales approach was within the range. Consequently, the subdivision analysis does not provide an independent basis for determining the value of Lot 115. In addition, I find the analysis inherently unreliable.
The Appraisal of Real Estate sets forth the following requirements for a subdivision analysis:
To estimate raw land value with subdivision development analysis an appraiser is required to
• Accurately determine the highest and best use of the land
• Create or affirm a supportable subdivision development plan
• Determine the timing and cost for approval and development (including mitigation needs and costs of obtaining development entitlements)
• Forecast a realistic pricing schedule over time
• Forecast accurately the lot absorption rate and price mix (including properly supported projections of community or market growth, over the absorption period)
• Estimate accurately the staging or phasing of land development and related expenses
• Forecast marketing and related holding expenses over the absorption period
• Estimate the annual real estate taxes
• Include overhead and an entrepreneurial profit allowance in the discount rate and/or line item allocation for entrepreneurial profit
• Estimate the appropriate discount rate consistent with the selection of the line item allocation for entrepreneurial profit
[The Appraisal of Real Estate, supra, at 343.]
When tested against these requirements, the appraiser’s subdivision analysis is deficient in the following respects:
A. The appraiser did not have available a “supportable subdivision development plan” because neither he nor plaintiffs engineer had determined the lot layout or lot sizes for the subject property.
B. The appraiser’s assumptions as to the time periods for obtaining approvals and for development, as well as his assumption as to the time period required for marketing the lots once developed, are based solely on his personal estimates. Although the appraiser testified that he conducted a supply and demand analysis with respect to the proposed lots, no details of that analysis were presented, and, therefore, I cannot evaluate the [76]*76extent or reliability of the analysis. See The Appraisal of Real Estate, supra, at 284-87 (describing a supply and demand analysis for a proposed residential subdivision).
C. The appraiser could not estimate a “realistic pricing schedule over time” because, without knowledge of the size of the lots being sold, he could not know what the likely price for each of those lots would be initially or as increased by inflation. In addition, the appraiser’s analysis of his comparable lot sales is questionable because he assumed, for purposes of comparison, without any engineering support and relying only on his estimate, that lots on Lot 115 would average four acres in size and adjusted his sales accordingly.
D. The development costs for Lot 115 used by the appraiser were a total of the engineer’s estimate, set forth in a letter, that the costs would be approximately $3,800,000, plus the appraiser’s estimate of soft costs (as to which he provided a limited breakdown). The engineer testified, however, that the costs would be $3,000,000. The appraiser characterized this testimony as a mistake. The engineer’s road cost estimate, which the appraiser accepted, was $425 per linear foot. In the Davanne Realty appraisal discussed above, plaintiffs appraiser used a road cost of $200 per linear foot based on a different engineer’s estimate involving terrain in Kinnelon Borough similar to the subject property, although for a roadway much shorter in length than the roadways required to develop Lot 115. The Davanne Realty engineer described this $200 per linear foot cost as resulting from “rock outcropping and the amount of cut and fill required to meet Borough standards.”
E. The appraiser’s estimate of the lot prices is not rehable because ah of the lot sales on which he relied were not fully improved building lots. The engineer’s development cost estimate included costs to produce fully improved lots. Thus, in order to establish a rehable value for unimproved lots, these costs would have to be deducted from the value of fully improved lots. However, each of lot sales two, three and four used by plaintiffs appraiser is described as “partially improved.” Only lot sale [77]*77number one is described by the appraiser as fully improved. A fully improved lot would sell for a price in excess of a partially improved lot. Thus, the improved lot price used by the appraiser in his subdivision analysis appears to be too low, and, therefore, after deduction of development costs, the unimproved lot value is too low.
Subdivision analysis is a type of discounted cash flow (DCF) analysis, a technique that generally has been viewed skeptically by the Tax Court. See University Plaza Realty Corp. v. City of Hackensack, 12 N.J.Tax 354 (Tax 1992), aff'd, 264 N.J.Super. 353, 624 A.2d 1000 (App.Div.), certif. denied, 134 N.J. 481, 634 A.2d 527 (1993) (describing the DCF method in tax appeal proceedings as “an amalgam of interdependent, attenuated assumptions of limited probative value.” Id. at 368.) The caution expressed by the Tax Court as to the use of DCF analysis is reflected in The Appraisal of Real Estate, supra, at 342, which states: “When used on its own without an abundance of reliable market data, [DCF] can be the least accurate raw land valuation technique.” See also Appraisals Standard Board, Uniform Standards of Professional Appraisal Practice 85 (2002 ed.) (“The results of DCF analysis should be tested and checked for errors and reasonableness. Because of the compounding effects in the projection of income and expenses, even slight input errors can be magnified and can produce unreasonable results.”)
The Tax Court has acknowledged that DCF analysis may be used if properly supported by market data. See Genola Ventures-Shrewsbury v. Shrewsbury Bor., 2 N.J.Tax 541, 551-52 (Tax 1981) In Tamburelli Properties Association v. Cresskill Borough, 15 N.J.Tax 629 (Tax 1996), aff'd, 308 N.J.Super. 326, 705 A.2d 1270 (App.Div.1998), the Tax Court accepted a DCF analysis when it was used by both appraisers and they were in substantial agreement as to the components of the analysis.
Because DCF analysis is reliable only when it is supported by abundant market evidence, and because the analysis provided by plaintiffs appraiser with respect to Lot 115 does not enjoy that support, I conclude that the analysis is not a reliable indicator of [78]*78value and, therefore, does not buttress the appraiser’s valuation conclusions based on the comparable sales analysis discussed above. The fact that two unreliable analyses lead to the same conclusion does not enhance the probative value of either.
Because of my concerns as to the reliability of the valuation proofs presented by plaintiff, I am constrained to affirm the assessments on the properties under appeal. Judgment will be entered accordingly.