This is a local property tax case wherein plaintiff seeks review of a judgment of the Hudson County Board of Taxation affirming the 1986 assessment on its property located at Harrison and Worthington Avenues in Harrison, New Jersey (Block 222, Lot 1). The assessment was:
Land $ 966,950
Improvements 1,467,100
Total $2,434,050
*163Plaintiff also seeks direct review, pursuant to N.J.S.A. 54:3-21, of the 1987 and 1988 assessments on the aforesaid property plus contiguous vacant lots. Those lots (Block 200, Lot 1, Block 199, Lots 1, 5, 9 and 13, Block 23, Lot 199) were withdrawn at the trial for tax year 1987, but they remained part of plaintiffs 1988 appeal. Plaintiffs expert, however, addressed only Block 222, Lot 1 for all years under review. No credible valuation evidence was introduced by either party for the vacant lots.
The 1987 and 1988 assessments on Block 222, Lot 1 (hereafter referred to as the subject property) were:
1987 1988
Land $ 966,950 $ 4,888,000
Improvements 1,467,100 9,518,000
Total $2,434,050 $14,406,000
A district-wide revaluation was put into effect for tax year 1988.
At issue are the true value of the subject property, whether plaintiff is entitled to assessment discrimination relief pursuant to N.J.S.A. 54:51A-6 (chapter 123 in the Tax Court) for tax years 1986 and 1987, whether the Director’s general average ratio for tax year 1986 should be recalculated by reason of excluding one sale and including another sale, both of which occurred during the relevant sampling period, and whether plaintiff is entitled to an assessment on the subject property at a level below 100% of true value for 1988, a revaluation year.
The subject property is an owner-occupied multi-building manufacturing complex constructed in 1902 and containing at least 748,223 square feet of usable building area. The improvements are located on 27.09 acres. In addition, the property includes an abandoned masonry foundry building containing 122,000 square feet. This building is in poor condition, having been exposed to the weather for many years; large portions of exterior walls have been removed. The building has dirt floors with no utilities and no heat.1
*164Ceiling heights in the usable buildings range from 40 to 70 feet, a condition which both experts agree results in substantial functional obsolescence. Five of the buildings have galleries containing 165,205 square feet. These galleries are of wood-plank construction and, by reason of their deteriorated condition, have limited utility as dead storage space. The galleries were utilized in years prior to World War I in the manufacture of condensers, 40- to 50-feet high, on a vertical plane. Workmen were stationed on the galleries to assemble the condensers. The size of the condensers accounted not only for the galleries but for the long and narrow configuration of the buildings themselves. Building # 1, for example, is 125 feet wide by 1,000 feet in length, while building # 3 is 125 feet wide and 523 feet long. After World War I the condenser manufacturing process changed, as condensers were manufactured on a horizontal plane and were much smaller.
Plaintiffs expert estimated the true value of the subject property to be $5,525,000 for tax year 1986, $5,895,000 for 1987 and $6,200,000 for 1988. Defendant’s expert estimated the true value of the subject property to be $9,500,000 for 1986, $11,-000,000 for 1987 and $13,500,000 for 1988.
Both experts utilized the income and sales comparison approaches to value. The parties stipulated economic rent for 1986 and 1987 with respect to all but the galleries and foundry; they agreed upon the capitalization rate for 1986 and 1987. They agreed upon nothing for 1988 and they differed substantially with respect to vacancy and loss allowance and expenses for all three years.
I find that the income approach is inappropriate for this superannuated, owner-occupied industrial property. Notwithstanding their agreement on some essential aspects, the experts are impelled to construct an amalgam of attenuated hypotheses *165concerning economic rent, projected expenses, vacancy and loss allowance, utility of space (e.g., the galleries and foundry) and whether the leases would be gross or net, not to mention whether the property is best suited for a single tenant or multiple tenancies. The sheer number of assumptions, some interdependent, precludes acceptance of the income approach utilized by both experts as a viable indicator of value. Shulton, Inc. v. Clifton, 7 N.J.Tax 208 (Tax Ct.1983), aff’d 7 N.J.Tax 220 (App.Div.1983).
Plaintiffs expert relied upon 12 sales of allegedly comparable properties for tax year 1986 and 14 sales for 1987, which included the 12 utilized for 1986. For 1988 he simply added 5% to his 1987 estimate of value. He made adjustments for time for all 14 sales, ranging from a negative 5% to a positive 25%; he made a negative adjustment of 5% for age and condition for five sales; he made size adjustments ranging from a negative 5% to a positive 20% for all but one sale; and he made utility adjustments ranging from 5% to 30% for 10 of the 14 sales.
These adjustments resulted in adjusted sales prices ranging from $3.13 a square foot to $11 a square foot, land and buildings merged. The expert’s unadjusted sales prices ranged from $2.61 a square foot to $14.70 a square foot. The sales occurred between December 16, 1980 and December 16, 1987, with five sales transacted between March 1, 1984 and September 13, 1985. The sale properties, all industrial properties of comparable age and condition to the subject, are located in Harrison, Kearny, Jersey City, Linden, Elizabeth, Perth Amboy, Garfield, Hillside and Belleville.
Defendant’s expert utilized nine sales of allegedly comparable properties. These properties sold between December 16, 1980 and June 7,1989 for unadjusted prices ranging from $6.22 a square foot to $15 a square foot. The expert quantified adjustments (for time and utility) only with respect to one sale, the December 16, 1980 sale of the Otis Elevator property in Harrison, a sale also utilized by plaintiff’s expert.
*166Of the nine sales relied upon by defendant’s expert, four are simply not comparable by reason of size. The subject contains at least 748,223 square feet of usable buildings, while the usable square footage in three of the expert’s comparables is only 140,000 and another sale property contains only 124,600 square feet of building area.
The size differentials between the four comparables and the subject tend to vitiate comparability. See Thomas J. Lipton, Inc. v. Raritan Tp., 10 N.J.Tax 202, 208 (Tax Ct.1988), aff’d o.b. per curiam 11 N.J.Tax 100 (App.Div.1989).
The same can be said of a fifth comparable utilized by defendant’s expert; except that that property, the former Western Electric plant in Kearny, was exponentially larger than the subject, both in building size (14 buildings with a total of 2,684,342 square feet) and land area (144.902 acres).
A sixth allegedly comparable sale, that of a manufacturing and warehousing facility of Owens-Illinois, Inc. while similar in size and utility, is light years removed from the subject in terms of age, land area, and coverage ratio. The Owens-Illinois property was built in 1959 and expanded between 1962 and 1967; the subject was built in 1902. The land area of the alleged comparable is 57.83 acres; the subject’s land area is 27.09 acres. The land-to-building coverage ratio of the sale property is 2.70 to 1; the subject’s ratio is 1.3 to 1 (including the foundry, which defendant’s expert included in his valuation). The Owens-Illinois plant is simply not comparable to the subject.
A seventh allegedly comparable sale, a property in Elizabeth,'New Jersey, is smaller than the subject in land area (10.14 acres) and building size (206,982 square feet). This property also has an average ceiling height of 25 feet compared to the 40- to 70-foot ceiling heights in the subject. This property is not comparable.
Evidence of comparable sales is effective in determining property value for tax purposes only where there is substantial *167similarity between the sale properties and the subject so as to admit of reasonable comparison. Kearny Leasing Corp. v. Kearny, 6 N.J.Tax 363 (Tax Ct.1984), aff’d o.b. per curiam. 7 N.J.Tax 665 (App.Div.1985).
This leaves two comparables, a property on West Side Avenue in Jersey City which sold for $2,800,000 in September 1982 (sale # 8) and the Otis Elevator Manufacturing plant in Harrison which sold on December 16, 1980 for $6,000,000 (sale # 9). The latter sale was also utilized by plaintiffs expert. The unadjusted unit price for the West Side Avenue property, sold by Ball Glass Container to Union Minerals and Alloys, was $6.22 a square foot, while the unadjusted unit price for the Otis Elevator property was $10.98 a square foot.
Of the 14 sales relied upon by plaintiff’s expert, the court rejects seven of them (sales #2, #5, #7, #8, #9, #11 and # 12) because the land area of the sale properties is either too small or too large in relation to the subject to permit a meaningful comparison. Sale # 6 is also rejected because the improvements are significantly newer than the buildings on the subject property.
The most significant area of disagreement is in the treatment of the abandoned foundry. Plaintiff’s expert excludes it entirely, while defendant’s expert includes it in his calculation of the aggregate square footage of improvements on the ground that it has marginal utility notwithstanding its deplorable, derelict condition. I agree with defendant’s expert in this regard. The foundry building still stands; it is used for storage of industrial refuse and contaminated soil, both of which items cannot be disposed of elsewhere except at considerable expense. Plaintiff’s expert makes substantial utility adjustments and eliminates the foundry altogether from his square foot calculations. Plaintiff cannot have the benefit of such elimination and still claim a utility adjustment. Congoleum Corp. v. Hamilton Tp., 7 N.J.Tax 436, 456 (Tax Ct.1985). A negative utility adjustment with respect to each sale property *168is more appropriate in determining value than eliminating the building altogether.
Accordingly, the court finds the usable square footage of buildings to be 870,223.2
The court finds that five sales utilized by the experts provide credible evidence of comparability to the subject.
The unit sales prices of sales # 1, #3, #4 and # 10 utilized by plaintiffs expert, adjusted for time, utility, age and condition, size and location, are $8.42, $6.48, $5.47 and $7.12 a square foot, respectively. The unit price of sale # 8 utilized by defendant’s expert, after adjustment for time (age, condition, and size being similar to the subject), was $8 a square foot. I conclude from the foregoing that the true value of the subject property on October 1, 1985 was $8 a square foot, land and building merged, or $6,961,800 (870,223 square feet x $8).
Both experts offered unsupported opinions of the rate of change in value from one tax year to the next. Plaintiff’s expert claimed the annual rate of increase to be 5%; defendant’s expert averred that the annual rate of increase was 10%. The only reliable indicator of the annual increase in property values in the defendant taxing district during the years in issue is the Director’s ratios, taken from the tables of equalized valuations, which show a uniform increase of approximately .20% annually for 1986 and 1987. Accordingly, I find the true value of the subject property to be $8,354,600 on October 1, 1986 and $9,746,500 on October 1, 1987.
Discrimination.
1986.
Plaintiff claims entitlement to assessment discrimination relief for 1986 for the reason that the ratio of assessment to true value, in the view of plaintiff’s expert, exceeds the upper *169limit of the common level range, as that term is defined in N.J.S.A. 54:l-35a(b). The general average ratio duly promulgated by the Director, Division of Taxation for Harrison for 1986 pursuant to N.J.S.A. 54:1-35.1, was 33.44% and the upper and lower limits of the common level range (plus or minus 15% of the average ratio) were 38.46% and 28.42%, respectively. If the ratio of assessed to true value exceeds the upper limit of the common level range the taxpayer is entitled to relief by the application of the average ratio to the true value as judicially determined. N.J.S.A. 54:51A-6(a).
Plaintiff contends, however, that the ratio for 1986 was incorrectly calculated in that one sale was improperly included in the category of usable sales which occurred during the relevant sampling period and another sale was improperly excluded, Plaintiff submits a proposed recalculation indicating that the correct ratio is 32.48% and the upper limit of the common level range is then 37.36%.
While plaintiff may challenge sales claimed to be erroneously included or excluded, Murnick v. Asbury Park, 95 N.J. 452, 471 A.2d 1196 (1984), the challenge in this case is irrelevant. The ratio of the 1986 assessment to the 1986 true value, as herein found, is 34.96% which is within the common level range no matter which calculation, the Director’s or plaintiff’s, is used. There will be no change in the 1986 assessment and the judgment of the Hudson County Board will be affirmed.
1987.
The general average ratio promulgated by the Director for Harrison for tax year 1987 is 26.46%; the upper and lower limits of the common level range are 30.43% and 22.49%, respectively. The ratio of the 1987 assessment to the 1987 true value as herein found is 29.13%.
As the ratio of assessed to true value falls within the common level range plaintiff is entitled to no relief. The 1987 assessment will therefore be affirmed.
*170 1988.
A district-wide revaluation was put into effect in Harrison for tax year 1988; N.J.S.A. 54:51A-6 is thus inapplicable by its own terms.
Plaintiff argues, however, that the inapplicability of N.J.S.A. 54:51A-6 does not mean that its property must be assessed at 100% of true value. On the contrary, plaintiff goes on, the proofs demonstrate that a common level of assessment prevailed in the defendant municipality in 1988 and that level was 78%.
It is well settled that, when a taxpayer’s real property is assessed at a ratio to true value substantially higher than the common level of assessments prevailing in a taxing district, the taxpayer is ordinarily entitled to a reduction of his assessment to the valuation resulting from an application of the common level to the true value of the taxpayer’s property. In re Appeal of Kents, 2124 Atlantic Ave., Inc., 34 N.J. 21, 166 A.2d 763 (1961); Piscataway Associates, Inc. v. Piscataway Tp., 73 N.J. 546, 376 A.2d 527 (1977).
There is a presumption, however, that all property is uniformly assessed at 100% of true value in a revaluation year. Tri-Terminal Corp. v. Edgewater, 68 N.J. 405, 346 A.2d 396 (1975); Cantrell v. Upper Pittsgrove Tp., 1 N.J.Tax 508 (Tax Ct.1980); Highview Estates v. Englewood Cliffs, 6 N.J.Tax 194 (Tax Ct.1983).
In concluding that the 1988 common level of assessments was 78%, plaintiff relies upon 36 sales occurring in the first six months of 1988 and 59 sales transacted in the 12 months beginning July 1, 1988. The sales occurring in the first six months of 1988 disclosed an unweighted, unclassified ratio of 78%. The unweighted, unclassified ratio disclosed by the 59 sales occurring July 1,1988 to June 30,1989 was 78.7%. Of the 36 sales transacted in the first six months of 1988, 18 were clustered in the 61%-80% range. Only nine sales, however, were clustered in the 71%-80% range. Of the 59 sales occur*171ring in the 12 months beginning July 1, 1988, 30 were clustered in the 61%-80% range. Only 20 sales, however, were clustered in the 71%-80% range.
Plaintiff proferred no expert statistical testimony from which the court could find that the sales sample (95 sales in 18 months) was of sufficient size to permit a reasonable inference to be drawn concerning a common level of less than 100%. The same flaw inheres in plaintiffs cluster analysis. Proof of a common level requires a showing that the great bulk of sales is clustered around a particular assessment-to-value range. See In re Appeals of Kents 2124 Atlantic Ave., Inc., supra 34 N.J at 26, 166 A.2d 763. In this case plaintiff has shown only 29 sales out of a total of 95 sales occurring in a 18-month period to be clustered in the 10-point range in which lies plaintiffs claimed ratio of 78%.
In sum, plaintiff has failed to overcome the presumption that assessments in the revaluation year of 1988 were uniformly made at 100% of true value. This failure rests upon an inadequate and incomplete analysis of sales in the defendant municipality and the resulting failure to demonstrate the presence or absence of a common level of assessment below the presumed 100% of true value. See Skytop Gardens, Inc. v. Sayreville, 3 N.J.Tax 187 (Tax Ct.1981), aff’d o.b. per curiam 5 N.J.Tax 478 (App.Div.1983).
The 1988 assessment will be fixed at 100% of the true value herein found.
The Clerk of the Tax Court is directed to enter judgments indicating the assessments to be as follows:
1986 1987 1988
Land $ 966,950 $ 966,950 $3,313,800
Improvements 1,467,100 1,467,100 6,432,700
Total $2,434,050 $2,434,050 $9,746,500
Another building of some 10,000 square feet is unusable, in the view of both experts. The evidence shows that the building is boarded up and is unsafe to enter; it is ripe for demolition.