Analysis Date: July 27, 2026 · Projection Horizon: 24 Months to July 2028
Executive Summary
Rolling Hills is one of the thinnest, highest-value residential markets in Los Angeles County — a gated city of roughly 700 homes that produces about 21 closed sales in a typical year. Over the 21-year record contained in this dataset, the 180-day moving average of price per square foot has climbed from roughly $803 to $1,160.54, a gain of approximately 45% in nominal $/SqFt terms. But that headline understates the volatility: this series has swung as violently as any submarket on the Peninsula, with year-over-year moves ranging from +55.74% to -26.94%.
The current reading is a market that has plateaued and just begun to roll over. The moving average sits at $1,160.54 — up 12.91% year over year, but essentially unchanged over the last nine months, oscillating in a tight band between $1,142 and $1,178. Underneath that flat surface, the calculus tells a more specific story.
The first derivative (velocity) is negative at -$12.54 per SqFt per month, and the second derivative (acceleration) is also negative at -$4.71 per SqFt per month squared. Both derivatives pointing the same direction — down — is the technical signature of a market in the early stage of a decline, not the late stage. Prices are falling, and as of the current reading they are falling slightly faster each month rather than slower.
The model projects the moving average to trough near $1,017 in late December 2027 before flattening and turning marginally positive, ending the 24-month window at $1,024 — a total decline of 11.8%, or an average of -$5.70 per SqFt per month. That is a meaningful correction, but it is comfortably inside the historical downside bound of -26.94% and it does not represent a collapse. It represents a return to the $1,000–$1,050 band this market defended repeatedly in 2022, 2023, and mid-2025.
Step 3 — Year-over-Year Extremes on the 180-Day Moving Average
Computed using only the Closed Sale Date and Price Per Square Foot columns, on the daily-resampled, forward-filled 180-day moving average, with year-ago values matched using a ±5-day tolerance window:
Both extremes are characteristic of a thin market. In a city that closes fewer than two sales per month on average, a single 12,000-square-foot estate trading at $650/SqFt or a 2,000-square-foot ranch trading at $1,600/SqFt can move the entire 180-day average by double digits. These are real, tradeable price levels, but they reflect changes in what sold at least as much as changes in what things are worth. That caveat matters for how much weight to place on any single month's reading.
Step 4 — Projection Boundaries Applied to This Analysis
Per the governing rule, all boundaries from any prior analysis are disregarded. The projection for this analysis is constrained exclusively by the two limits derived above from this dataset:
Methodology
- Moving average construction: Transactions were resampled to a daily series, gaps forward-filled across the full 7,657-day span, and a 180-day rolling mean applied with a 30-period minimum. The resulting MA carries 7,657 points against 448 transactions — a ratio of 17.1, confirming the forward-fill executed correctly.
- Derivatives: First and second derivatives were computed with Savitzky-Golay filtering (cubic, 181-day window for velocity, 361-day for acceleration) rather than naive differencing, which would amplify the noise inherent in a thin series.
- Steady-state slope: Ordinary least squares regression across the trailing 24 months yields +$2.06/SqFt per month (r² = 0.048) and across 60 months +$0.37/SqFt per month (r² = 0.006). Both r² values are effectively zero, which is itself the finding: there is no durable trend in this market's price per square foot — only cycles around a slowly drifting mean.
- Projection model: Exponential convergence with a linear momentum term. Current velocity and current acceleration are both honored as initial conditions, then decayed toward the OLS steady-state slope with a 120-day time constant. This guarantees the projected derivative curves are mathematically consistent with the projected price curve.
- Data integrity: All 448 rows carry Closed status with valid dates and price-per-square-foot values. No records were excluded. The dataset spans a 6.9x range in $/SqFt ($233 to $2,862) and a wide range in home size (median 3,792 SqFt), consistent with the heterogeneity of the Rolling Hills housing stock.
Step 5 — First Derivative: Velocity
The first derivative measures the slope of the moving average — how many dollars per square foot the market is gaining or losing each month. The current reading is -$12.54 per SqFt per month.
Context matters here. Twelve months ago this same measure sat at -$39.69/month. The market absorbed that decline, rallied through the autumn of 2025 to a peak velocity of +$45.58/month in November 2025, and has since rolled back over. Velocity crossed below zero in January 2026 and, apart from a brief positive blip in March, has stayed negative and drifted steadily lower: -$0.20 in February, +$8.25 in March, -$0.01 in April, -$4.12 in May, -$8.74 in June, -$12.54 in July.
That progression — a sequence of increasingly negative monthly readings — is the substance of the bearish case. It is not a single bad print. It is six consecutive months of deterioration.
The projection carries velocity to a trough of -$16.90 per SqFt per month around September 27, 2026, after which the decline begins to lose force. Velocity is projected to cross back above zero on December 25, 2027 — the calculus inflection point marking the end of the correction.
Step 6 — Second Derivative: Acceleration
The second derivative answers whether the current move is gaining or losing force. The current reading is -$4.71 per SqFt per month squared — negative.
With velocity negative and acceleration negative, this market is in the quadrant of accelerating decline. Prices are falling and the rate of fall is still increasing. This is the least favorable of the four possible derivative combinations for a seller, and it is the reason this report projects a further decline rather than an immediate base.
The redeeming detail is that acceleration is projected to turn positive within approximately 60 days, peaking at +$1.77/SqFt/month² in January 2027. Negative velocity with positive acceleration is the classic late-correction signature: still falling, but with the downward force draining out. By the standards of this dataset the projected acceleration values are mild — the historical range runs from -$22.40 to +$13.70 — which argues for an orderly retracement rather than a disorderly one.
24-Month Projection Table
| Date | 180-Day MA $/SqFt | 1st Deriv ($/SqFt/mo) | 2nd Deriv ($/SqFt/mo²) | Cumulative Change | Momentum Reading |
|---|---|---|---|---|---|
| Jul 2026 | $1,161 | -$12.54 | -4.71 | 0.0% | Starting point — velocity negative, decline steepening |
| Oct 2026 | $1,113 | -$16.57 | +0.78 | -4.1% | Velocity trough (Sep 27, 2026); acceleration turns positive |
| Jan 2027 | $1,069 | -$12.19 | +1.77 | -7.9% | Peak deceleration — downward force fading fastest |
| Apr 2027 | $1,040 | -$7.15 | +1.48 | -10.4% | Decline losing more than half its original force |
| Jul 2027 | $1,025 | -$3.46 | +1.00 | -11.7% | Approaching the floor; slope nearly flat |
| Oct 2027 | $1,018 | -$1.09 | +0.61 | -12.3% | Basing zone |
| Jan 2028 | $1,017 | +$0.33 | +0.35 | -12.3% | Inflection point — velocity crosses zero (Dec 25, 2027) |
| Apr 2028 | $1,020 | +$1.13 | +0.20 | -12.1% | Early recovery, converging to steady-state slope |
| Jul 2028 | $1,024 | +$1.56 | +0.11 | -11.8% | 24-month terminal value |
Terminal value: $1,024 per SqFt in July 2028. Total change: -11.8%. Average rate of change: -$5.70 per SqFt per month.
Step 7 — Buyer Advice
Step 7 — Seller Advice
Step 8 — Devil's Advocate: Why This Projection Could Be Wrong
Intellectual honesty requires stating the case against my own conclusion. Here is the argument for why Rolling Hills $/SqFt could rise over the next 24 months rather than fall 11.8%.
1. The statistical foundation is weak, and I have said so myself.
The trailing 24-month OLS regression produced an r² of 0.048. The 60-month regression produced 0.006. In plain terms, a trend line explains almost none of the variance in this series. I have built a directional 24-month projection on top of a series that demonstrably has no reliable direction. If the r² is effectively zero, the honest confidence interval around $1,024 is wide enough to include $1,300 — and any forecast that admits this should be held loosely.
2. Nine sales in 2026 is not a market signal — it is a small sample.
The current negative velocity is being driven by a handful of transactions. The last two recorded sales came in at $1,069 and $788 per SqFt; the two before those at $1,166 and $1,152; and April produced prints at $1,277 and $1,606. That $788 print alone, on a 180-day window containing perhaps a dozen sales, can bend the entire moving average downward. The declining velocity I am projecting forward may be an artifact of which houses happened to close, not evidence that Rolling Hills values are falling. One large, older estate trading at a low per-foot number does exactly this.
3. This series has reversed sharply and without warning before — repeatedly.
Look at the record. Velocity was -$39.69/month in July 2025 and +$45.58/month by November 2025 — an 85-point swing in four months. The same thing happened in 2019, when the moving average ran +55.74% year over year and then gave nearly all of it back within twelve months. A market capable of a 55% annual gain and a 27% annual loss is a market where a 24-month projection of -11.8% carries enormous error bars in both directions. My model's 120-day convergence assumption is a reasonable engineering choice, but this series has never respected smooth convergence.
4. Rolling Hills supply is structurally, permanently constrained.
This is a gated city of roughly 700 homes with no meaningful new construction possible. Unlike almost any other submarket, supply cannot respond to price. When capital rotates toward trophy coastal property — as it did in 2019, 2021, and again in late 2025 — there is nothing to absorb it except higher prices. A single quarter of renewed demand from equity-rich or relocating buyers could re-run the late-2025 rally and invalidate this entire projection within two quarters.
5. The composition risk cuts both ways.
If the next four closings happen to be smaller homes on premium view lots, the per-square-foot average will jump mechanically, with no change whatsoever in underlying value. The same denominator effect that may be manufacturing my bearish signal could just as easily manufacture a bullish one. In a market this thin, price per square foot is a noisy proxy for value, and I am treating it as a precise one.
What would change my mind
If the next two quarters produce four or more closings above $1,200/SqFt, or if velocity crosses back above zero before the projected September 2026 trough, the correction thesis should be abandoned rather than defended. I would rather revise this projection on new data than be right about the method and wrong about the market.
45+ Years of Consecutive Daily CRMLS Market Tracking
Phone: (424) 722-9136
Email: george.fotion@homeispalosverdes.com
Web: PalosVerdesHomesBest.com · SearchHomesInPrivate.com
Schedule: calendly.com/george-fotion
Current 180-Day MA $1,161 as of Jul 22, 2026 |
24-Mo Projection $1,024 Jul 2028 target |
% Expected Change -11.8% over 24 months |
Avg Rate of Change -$5.70 per month |
Transactions in Pool 448 closed sales, 21 yrs |
(424) 722-9136 · george.fotion@homeispalosverdes.com
PalosVerdesHomesBest.com · SearchHomesInPrivate.com
