Deep Dive
Monthly number of privately owned housing units where construction has started (in thousands of units). This tracks realized homebuilding activity and is sensitive to financing conditions, builder sentiment, and demand in the housing market.
Source: HOUST via FRED, Federal Reserve Bank of St. Louis. See the FRED series notes for original source attribution.
Time Series
The range selector drives the backend statistics on the right. The slider below only zooms the chart slice, while fixed-horizon deltas stay anchored to the latest observation.
Showing 61 of 61 points.
Direction & Persistence
Direction stats use the selected 5Y window and persistence shows how sticky that move has been.
Acceleration
Acceleration compares the latest move with the prior move inside the selected window.
Forward Returns After Similar Readings
Compare what index benchmarks did after historical readings with the same regime and quantile setup as today in the selected 5Y window.
How To Read This
Use this tab for the original question: when old releases looked like today's reading, what happened next?
What counts as a match: It matches old releases by today's regime and quantile, so the comparison stays close to the existing forward-returns view.
Simple example: If today's payrolls are in a normal regime and middle quantile, this tab looks for prior payroll releases that were also normal and middle-range.
Read the card: A 3M avg of +1.2% means the benchmark averaged +1.2% over the next three months after those matching releases.
Current Setup Summary
This release maps to the Low regime and Bottom 20% historical bucket. The cards compare only prior releases with that same setup.
Regime compares the reading with its normal range using z-score bands: extreme low, low, normal, high, or extreme high.
Quantile ranks the reading against history using percentile buckets, such as bottom 20% or top 20%.
Use this to frame risk and upside; it is historical context, not a standalone forecast.
S&P 500 Index
Regime: Low
S&P 500 Index averaged +4.5% over the next 3M; 84.2% of 19 similar cases were positive.
S&P 500 Index
Quantile: Bottom 20%
S&P 500 Index averaged +3.3% over the next 3M; 73.9% of 23 similar cases were positive.
Nasdaq Composite
Regime: Low
Nasdaq Composite averaged +5.8% over the next 3M; 78.9% of 19 similar cases were positive.
Nasdaq Composite
Quantile: Bottom 20%
Nasdaq Composite averaged +4.1% over the next 3M; 69.6% of 23 similar cases were positive.
Example Calculation
If a matching macro release starts from an S&P 500 close of 5,000 and the 3M close is 5,250, the forward return is (5,250 - 5,000) / 5,000 * 100 = +5.0%.
If 22 of 26 matching 3M windows are above 0%, positive cases are 22 / 26 = 84.6%.
Regime
3M avg
+4.5%
Positive cases
84.2%
Samples
19
1M avg
+2.3%
12M avg
+18.8%
Worst 3M drawdown
-16.9%
Quantile
3M avg
+3.3%
Positive cases
73.9%
Samples
23
1M avg
+1.1%
12M avg
+17.2%
Worst 3M drawdown
-16.9%
This expands every historical bucket, not just the current one. Use it to compare today's setup against other macro states; the current bucket is marked.
Avg return
Mean market return after matching releases.
Median
Typical case, less sensitive to outliers.
Positive
Share of cases above 0%; tiny samples show counts.
Score
Shown under each bucket name: z-score for regimes, percentile for quantiles.
Volatility
Standard deviation of forward returns; needs at least two samples.
Worst DD
Worst intra-window drawdown among matching cases.
Samples
Number of complete historical windows.
Regime Stats
| Bucket | Horizon | Avg | Median | Positive | Volatility | Worst DD | Samples |
|---|---|---|---|---|---|---|---|
| RegimeExtreme LowZ-score: -2.24 to -2.01 | 5D | +0.0% | +0.0% | 1/2 | 4.1% | -4.1% | 2 |
| 1M | -2.4% | -2.4% | 1/2 | 7.0% | -9.4% | 2 | |
| 3M | -1.1% | -1.1% | 1/2 | 7.0% | -16.9% | 2 | |
| 6M | +6.3% | +6.3% | 1/2 | 10.0% | -16.9% | 2 | |
| 12M | +13.0% | +13.0% | 2/2 | 11.4% | -16.9% | 2 | |
| RegimeLowZ-score: -1.88 to -1.08Current | 5D | +0.2% | -0.1% | 50.0% | 2.5% | -3.8% | 20 |
| 1M | +2.3% | +1.6% | 70.0% | 3.6% | -7.2% | 20 | |
| 3M | +4.5% | +4.4% | 84.2% | 4.6% | -16.9% | 19 | |
| 6M | +8.4% | +7.8% | 94.4% | 4.7% | -18.9% | 18 | |
| 12M | +18.8% | +17.1% | 100.0% | 6.6% | -18.9% | 17 | |
| RegimeNormalZ-score: -0.99 to +0.63 | 5D | +0.2% | +0.4% | 62.5% | 2.3% | -6.4% | 32 |
| 1M | +0.8% | +1.8% | 62.5% | 4.8% | -13.7% | 32 | |
| 3M | +3.5% | +4.2% | 77.4% | 6.8% | -18.9% | 31 | |
| 6M | +8.4% | +10.2% | 85.2% | 8.7% | -20.8% | 27 | |
| 12M | +18.0% | +19.2% | 91.3% | 10.6% | -22.8% | 23 | |
| RegimeHighZ-score: +1.00 to +1.68 | 5D | +0.3% | +1.2% | 80.0% | 2.3% | -4.0% | 5 |
| 1M | -2.7% | -2.0% | 0.0% | 2.0% | -12.2% | 5 | |
| 3M | -4.1% | -4.4% | 20.0% | 8.4% | -20.8% | 5 | |
| 6M | -12.2% | -12.2% | 20.0% | 7.4% | -23.6% | 5 | |
| 12M | -9.9% | -12.4% | 20.0% | 8.7% | -25.4% | 5 |
Quantile Stats
| Bucket | Horizon | Avg | Median | Positive | Volatility | Worst DD | Samples |
|---|---|---|---|---|---|---|---|
| QuantileBottom 20%Percentile: 1.8% to 20.0%Current | 5D | -0.0% | -0.1% | 50.0% | 2.8% | -5.4% | 24 |
| 1M | +1.1% | +0.4% | 58.3% | 4.4% | -9.4% | 24 | |
| 3M | +3.3% | +4.1% | 73.9% | 4.9% | -16.9% | 23 | |
| 6M | +7.9% | +7.8% | 90.9% | 5.4% | -18.9% | 22 | |
| 12M | +17.2% | +15.9% | 100.0% | 7.0% | -18.9% | 19 | |
| Quantile20-40%Percentile: 20.9% to 37.3% | 5D | +0.3% | -0.1% | 50.0% | 2.1% | -3.1% | 12 |
| 1M | +2.3% | +3.4% | 66.7% | 5.2% | -10.1% | 12 | |
| 3M | +6.0% | +4.9% | 81.8% | 6.7% | -18.9% | 11 | |
| 6M | +9.4% | +12.2% | 80.0% | 9.5% | -18.9% | 10 | |
| 12M | +22.4% | +21.4% | 100.0% | 9.1% | -18.9% | 10 | |
| Quantile40-60%Percentile: 41.3% to 60.0% | 5D | +0.0% | +0.8% | 75.0% | 2.3% | -6.4% | 12 |
| 1M | +0.5% | +1.8% | 75.0% | 4.4% | -12.1% | 12 | |
| 3M | +3.4% | +5.0% | 83.3% | 7.0% | -17.8% | 12 | |
| 6M | +7.4% | +12.0% | 81.8% | 9.6% | -20.8% | 11 | |
| 12M | +15.6% | +19.3% | 77.8% | 12.9% | -22.8% | 9 | |
| Quantile60-80%Percentile: 61.9% to 76.4% | 5D | +1.0% | +0.4% | 66.7% | 1.5% | -1.7% | 6 |
| 1M | +1.1% | +2.9% | 66.7% | 4.3% | -13.7% | 6 | |
| 3M | +1.3% | +4.0% | 83.3% | 7.0% | -18.9% | 6 | |
| 6M | +9.8% | +8.3% | 4/4 | 4.6% | -18.9% | 4 | |
| 12M | +16.6% | +16.2% | 4/4 | 5.5% | -18.9% | 4 | |
| QuantileTop 20%Percentile: 100.0% | 5D | +1.9% | +1.9% | 2/2 | 0.5% | -1.2% | 2 |
| 1M | -3.4% | -3.4% | 0/2 | 2.9% | -12.2% | 2 | |
| 3M | -3.9% | -3.9% | 1/2 | 13.0% | -20.8% | 2 | |
| 6M | -5.7% | -5.7% | 1/2 | 6.5% | -20.8% | 2 | |
| 12M | -1.8% | -1.8% | 1/2 | 8.6% | -22.8% | 2 |
Regime
3M avg
+5.8%
Positive cases
78.9%
Samples
19
1M avg
+2.5%
12M avg
+25.2%
Worst 3M drawdown
-21.4%
Quantile
3M avg
+4.1%
Positive cases
69.6%
Samples
23
1M avg
+1.3%
12M avg
+23.7%
Worst 3M drawdown
-21.4%
This expands every historical bucket, not just the current one. Use it to compare today's setup against other macro states; the current bucket is marked.
Avg return
Mean market return after matching releases.
Median
Typical case, less sensitive to outliers.
Positive
Share of cases above 0%; tiny samples show counts.
Score
Shown under each bucket name: z-score for regimes, percentile for quantiles.
Volatility
Standard deviation of forward returns; needs at least two samples.
Worst DD
Worst intra-window drawdown among matching cases.
Samples
Number of complete historical windows.
Regime Stats
| Bucket | Horizon | Avg | Median | Positive | Volatility | Worst DD | Samples |
|---|---|---|---|---|---|---|---|
| RegimeExtreme LowZ-score: -2.24 to -2.01 | 5D | +0.3% | +0.3% | 1/2 | 5.8% | -5.5% | 2 |
| 1M | -1.6% | -1.6% | 1/2 | 10.2% | -11.9% | 2 | |
| 3M | -1.7% | -1.7% | 1/2 | 12.8% | -21.2% | 2 | |
| 6M | +10.7% | +10.7% | 1/2 | 18.6% | -22.1% | 2 | |
| 12M | +21.6% | +21.6% | 2/2 | 20.0% | -22.1% | 2 | |
| RegimeLowZ-score: -1.88 to -1.08Current | 5D | +0.0% | +0.2% | 50.0% | 3.2% | -4.6% | 20 |
| 1M | +2.5% | +2.2% | 60.0% | 4.7% | -10.1% | 20 | |
| 3M | +5.8% | +6.5% | 78.9% | 7.3% | -21.4% | 19 | |
| 6M | +11.0% | +12.5% | 88.9% | 7.8% | -23.9% | 18 | |
| 12M | +25.2% | +28.2% | 100.0% | 9.6% | -24.3% | 17 | |
| RegimeNormalZ-score: -0.99 to +0.63 | 5D | +0.3% | +0.7% | 59.4% | 3.1% | -8.3% | 32 |
| 1M | +1.2% | +2.1% | 68.8% | 6.4% | -16.6% | 32 | |
| 3M | +4.5% | +5.6% | 74.2% | 10.0% | -23.9% | 31 | |
| 6M | +10.3% | +12.5% | 85.2% | 11.7% | -27.2% | 27 | |
| 12M | +22.4% | +23.6% | 91.3% | 14.2% | -30.1% | 23 | |
| RegimeHighZ-score: +1.00 to +1.68 | 5D | +0.3% | +0.1% | 60.0% | 3.4% | -5.6% | 5 |
| 1M | -4.2% | -4.4% | 0.0% | 1.1% | -13.6% | 5 | |
| 3M | -7.4% | -8.5% | 20.0% | 11.8% | -27.2% | 5 | |
| 6M | -18.2% | -15.9% | 0.0% | 9.9% | -32.9% | 5 | |
| 12M | -17.0% | -22.3% | 20.0% | 15.1% | -35.7% | 5 |
Quantile Stats
| Bucket | Horizon | Avg | Median | Positive | Volatility | Worst DD | Samples |
|---|---|---|---|---|---|---|---|
| QuantileBottom 20%Percentile: 1.8% to 20.0%Current | 5D | -0.2% | -0.4% | 45.8% | 3.5% | -5.5% | 24 |
| 1M | +1.3% | +0.7% | 50.0% | 5.7% | -11.9% | 24 | |
| 3M | +4.1% | +4.3% | 69.6% | 8.2% | -21.4% | 23 | |
| 6M | +10.5% | +11.7% | 86.4% | 9.1% | -23.9% | 22 | |
| 12M | +23.7% | +21.6% | 100.0% | 10.9% | -24.3% | 19 | |
| Quantile20-40%Percentile: 20.9% to 37.3% | 5D | +0.4% | +0.4% | 66.7% | 3.0% | -5.1% | 12 |
| 1M | +3.2% | +3.1% | 83.3% | 7.3% | -13.7% | 12 | |
| 3M | +7.7% | +5.8% | 81.8% | 9.6% | -23.9% | 11 | |
| 6M | +11.7% | +13.9% | 80.0% | 12.5% | -24.3% | 10 | |
| 12M | +27.9% | +25.0% | 100.0% | 11.1% | -24.3% | 10 | |
| Quantile40-60%Percentile: 41.3% to 60.0% | 5D | +0.2% | +1.0% | 66.7% | 3.0% | -8.3% | 12 |
| 1M | +0.6% | +2.1% | 75.0% | 5.7% | -16.6% | 12 | |
| 3M | +4.9% | +7.3% | 75.0% | 10.0% | -22.3% | 12 | |
| 6M | +9.0% | +15.7% | 81.8% | 12.5% | -27.2% | 11 | |
| 12M | +19.3% | +29.1% | 77.8% | 18.1% | -30.1% | 9 | |
| Quantile60-80%Percentile: 61.9% to 76.4% | 5D | +0.9% | +0.1% | 50.0% | 2.7% | -3.5% | 6 |
| 1M | +1.4% | +2.5% | 66.7% | 5.3% | -16.4% | 6 | |
| 3M | +1.6% | +3.2% | 83.3% | 10.2% | -23.9% | 6 | |
| 6M | +12.8% | +8.9% | 4/4 | 8.3% | -23.9% | 4 | |
| 12M | +21.0% | +22.3% | 4/4 | 7.6% | -23.9% | 4 | |
| QuantileTop 20%Percentile: 100.0% | 5D | +2.2% | +2.2% | 2/2 | 2.0% | -2.3% | 2 |
| 1M | -3.8% | -3.8% | 0/2 | 1.7% | -13.6% | 2 | |
| 3M | -4.1% | -4.1% | 1/2 | 17.7% | -27.2% | 2 | |
| 6M | -9.0% | -9.0% | 0/2 | 6.9% | -27.2% | 2 | |
| 12M | -1.7% | -1.7% | 1/2 | 12.6% | -30.1% | 2 |
Macro Release Event Study
Measure S&P 500 and Nasdaq behavior immediately before, on, and after genuine macro release dates in the selected 5Y window.
Each card groups releases by one feature of the stored selected-metric history. It does not reconstruct every historical first print.
Events count releases assigned to the bucket.
Samples count releases with enough market history to complete that particular window.
Alignment uses the first market session on or after the release date.
Direction
Up, down, or flat versus the previous stored metric reading.
Magnitude
Small, medium, or large absolute changes within the selected range.
Regime
Low, normal, high, or extreme z-score backdrops.
Revisions
First published base actuals later revised up, down, or unchanged.
How To Read This
Use this tab to compare market behavior after releases where the indicator rose, fell, or stayed flat versus the prior release.
What counts as a match: It only looks at the selected metric's release-to-release direction. It does not decide whether the release was good news or bad news.
Simple example: If payrolls were 180k last month and 230k this month, this release goes in Up. If the next release is 150k, that release goes in Down.
Read the card: An Up card with Post 1M avg +2.0% means the benchmark averaged +2.0% one month after releases where the indicator increased from the prior release.
Direction
Pre 1D
+0.2%
62.1% positive · 29 samples
Event 1D
-0.2%
41.4% positive · 29 samples
Post 5D
+0.0%
58.6% positive · 29 samples
Post 1M
+0.7%
55.2% positive · 29 samples
| Window | Avg | Median | Positive | Samples |
|---|---|---|---|---|
| Pre 5D | +0.4% | +0.3% | 65.5% | 29 |
| Pre 1D | +0.2% | +0.1% | 62.1% | 29 |
| Event 1D | -0.2% | -0.2% | 41.4% | 29 |
| Post 5D | +0.0% | +0.3% | 58.6% | 29 |
| Post 1M | +0.7% | +1.0% | 55.2% | 29 |
| Post 3M | +2.9% | +4.2% | 72.4% | 29 |
| Post 6M | +5.5% | +8.3% | 77.8% | 27 |
| Post 12M | +13.8% | +17.7% | 82.6% | 23 |
Direction
Pre 1D
-0.0%
51.6% positive · 31 samples
Event 1D
-0.2%
48.4% positive · 31 samples
Post 5D
+0.4%
60.0% positive · 30 samples
Post 1M
+1.1%
63.3% positive · 30 samples
| Window | Avg | Median | Positive | Samples |
|---|---|---|---|---|
| Pre 5D | +0.2% | +0.3% | 54.8% | 31 |
| Pre 1D | -0.0% | +0.0% | 51.6% | 31 |
| Event 1D | -0.2% | -0.0% | 48.4% | 31 |
| Post 5D | +0.4% | +0.3% | 60.0% | 30 |
| Post 1M | +1.1% | +1.0% | 63.3% | 30 |
| Post 3M | +3.1% | +4.1% | 75.0% | 28 |
| Post 6M | +7.2% | +7.3% | 84.0% | 25 |
| Post 12M | +16.3% | +16.5% | 91.7% | 24 |
Direction
Pre 1D
+0.2%
65.5% positive · 29 samples
Event 1D
-0.3%
48.3% positive · 29 samples
Post 5D
+0.1%
58.6% positive · 29 samples
Post 1M
+1.0%
58.6% positive · 29 samples
| Window | Avg | Median | Positive | Samples |
|---|---|---|---|---|
| Pre 5D | +0.4% | +0.5% | 55.2% | 29 |
| Pre 1D | +0.2% | +0.1% | 65.5% | 29 |
| Event 1D | -0.3% | -0.2% | 48.3% | 29 |
| Post 5D | +0.1% | +0.6% | 58.6% | 29 |
| Post 1M | +1.0% | +2.0% | 58.6% | 29 |
| Post 3M | +3.8% | +6.5% | 72.4% | 29 |
| Post 6M | +6.2% | +10.3% | 74.1% | 27 |
| Post 12M | +16.5% | +22.3% | 82.6% | 23 |
Direction
Pre 1D
+0.0%
48.4% positive · 31 samples
Event 1D
-0.4%
41.9% positive · 31 samples
Post 5D
+0.3%
53.3% positive · 30 samples
Post 1M
+1.1%
60.0% positive · 30 samples
| Window | Avg | Median | Positive | Samples |
|---|---|---|---|---|
| Pre 5D | +0.4% | +0.6% | 61.3% | 31 |
| Pre 1D | +0.0% | -0.0% | 48.4% | 31 |
| Event 1D | -0.4% | -0.1% | 41.9% | 31 |
| Post 5D | +0.3% | +0.4% | 53.3% | 30 |
| Post 1M | +1.1% | +1.5% | 60.0% | 30 |
| Post 3M | +3.5% | +3.7% | 67.9% | 28 |
| Post 6M | +9.6% | +11.4% | 80.0% | 25 |
| Post 12M | +21.7% | +23.2% | 91.7% | 24 |