Summarizing Adjusted Indicators and Weighting the Comps

Three adjusted sale prices are three answers to the same question, and they rarely match. Reconciliation is the reasoning that turns them into one supported range. It is not an average. An average treats every comparable as equally reliable, and they seldom are: one sale is closer to the subject than the others, one is more recent, one needed more correction to become comparable at all. This lesson sets out how the office weighs the indicators, how the weighting is written so a seller can follow it, and what is done with an indicator that does not belong.

Step one: list the indicators and look at the spread

The adjusted prices are listed from lowest to highest beside the facts that bear on their reliability: the comparable’s location relative to the subject, its date of sale, the number of adjustments it needed, and its gross adjustment, the total of its adjustments without regard to direction. The gross figure is the better measure of how much correction a sale needed, because two adjustments in opposite directions can produce a small net figure from a large amount of correction.

The spread between the lowest and highest indicator is the first reading. Where the comparables are genuine substitutes for the subject, their adjusted prices normally show meaningful convergence: corrected for their differences, sales that competed for the same buyer land close to one another. No fixed percentage marks the line, because how close depends on the market, the price tier and the evidence behind the adjustments. A wide spread is a signal to re-examine the grid: it suggests that at least one comparable was less similar than it appeared, or that an adjustment is too large or too small, and the grid is examined before the reconciliation proceeds.

Step two: identify and investigate any outlier

An indicator well above or below the others is investigated before it is used. The questions are specific. Did the comparable need unusually large adjustments, so that its adjusted price is more arithmetic than evidence? Is there an adjustment applied to this column at a different rate than to the others? Was the sale verified, or is there a concession or a non-market circumstance in it that the grid did not capture? Should the sale have been set aside at selection?

An outlier explained by one of those causes is corrected or removed, with the reason written into the analysis. An outlier that survives investigation, a verified, well-adjusted sale that simply landed high or low, is retained and given little weight, because a sale that meets every test is evidence even when it is inconvenient.

Step three: weight the indicators

Weighting assigns each indicator a share of the conclusion according to how much it can be relied on. The reasoning follows a consistent order. The most similar sale, the one in the subject’s immediate competition, needing the fewest and smallest adjustments, is the best evidence and receives the most weight. Recency comes next, because the most recent sale reflects the current market most directly and required the smallest market conditions adjustment. Verification counts: a sale confirmed with the listing agent for terms and condition is more reliable than one taken from the MLS record alone. Location within the market area counts, so that a sale on the subject’s own street outweighs a sale across the market area even when both qualify. A sale that needed the most correction, with the largest gross adjustment and the most lines, receives the least weight.

The office states the weighting in words before it states it in figures, because the words are what a seller can follow. “The sale on the subject’s street, needing only two small adjustments, is the best evidence and carries most of the weight” is a reconciliation. “Weights of 60, 25 and 15 percent” is the arithmetic that expresses it.

Why not the average

An average is a reconciliation that gives every comparable the same weight. When the indicators are close and the comparables are equally reliable, the average and the weighted figure land in nearly the same place, and the average does no harm. When the most similar sale sits at one end of the spread, the average is pulled toward the less reliable sales, and the seller hears a figure that the best evidence does not support. The median, the middle indicator, is a useful check on the weighted figure but is not a substitute for the reasoning, because it says nothing about which sale is most like the subject.

Application

The grid for 42 Ledgewood Drive describes a 2,000-square-foot colonial in good condition with a two-car garage. The three indicators, with the facts that bear on their reliability, are:

The spread is $466,900 − $452,300 = $14,600, which is 3.2 percent of the lowest indicator. No indicator is an outlier. A note on the grid reads “average $461,000, recommend $464,900.”

The weighting is reasoned through. Comparable A is on the subject’s street, matches it in condition and garage, needed two adjustments totaling $9,900, and is the best evidence on the grid. Comparable B is the most recent sale, but it is a remodeled house in excellent condition with a one-car garage, and it needed $34,900 of correction to stand in for the subject; its adjusted price is the highest on the grid and depends on a $15,000 condition adjustment, one of the two largest single corrections on the grid. Comparable C is the oldest sale, carries a concession and a location adjustment, and needed the most correction of the three. The weighting is 60 percent to A, 25 percent to B and 15 percent to C.

The arithmetic: $452,300 × 0.60 = $271,380; $466,900 × 0.25 = $116,725; $463,900 × 0.15 = $69,585. The sum is $457,690, which the analysis rounds to $457,700. The simple average of the three is $1,383,100 ÷ 3 = $461,033, about $3,300 higher, because the average gives the two less similar sales two-thirds of the conclusion. The reconciliation supports a range of $452,000 to $462,000, anchored on Comparable A and extending toward B and C without reaching B’s or C’s figure, and the sentence that goes to the seller reads: “The sale on your street, the closest match to your house, sets the lower part of the range; the two other sales, which needed more correction, support the upper part.” The recommendation of $464,900 is above the range the sales support, and the note is replaced by the range and that sentence.

Summary

Reconciliation lists the adjusted indicators beside the facts that bear on their reliability, reads the spread, investigates any outlier, and then weights the indicators so that the most similar sale carries the most of the conclusion, followed by recency and verification, and the sale that needed the most correction carries the least. The weighting is stated in words before figures, because the words are what a seller follows. An average is a reconciliation in which every sale counts equally, and it is generally used as a check.