The One Lab
SCORECARDARENASDISCOVERYFOUNDATIONSFOR AGENTS
AR-02 / ARENALOCKED UNDER HOLD · STANDING ON THE EVIDENCE

Historical Housing / Canada forecast record

This is preserved history, not the current Canada intelligence product.

These forecasts were issued under hold. They are excluded from the registered Canada intelligence report, its licensed report-input relations and calculations, and any future institutional product track record. Open the current Canada Housing Intelligence report →

This arena has issued forecasts, and it was not supposed to. On 5 August 2026 a scheduled job locked 24 Canadian forecasts while this arena was held, because the hold was written on pages like this one and not into the job that issues. The fault was ours. The data they were fit on has since been re-collected under stricter provenance, and each of these forecasts was re-fit against the corrected archive at the publisher vintage it declared: every one reproduced exactly, from the same window. So they stand, they will be graded under the same rules as everything else, and the circumstances of their issue stay on this page permanently. Nothing here has been graded yet, and nothing here is a claim about accuracy.

WHAT THESE
LOCKS CLAIMED
The preserved locks claimed changes in the New Housing Price Index for Canada and eight metropolitan markets at three, six and twelve months ahead. This describes the historical record; it does not authorize new issuance or make these locks inputs to the current report.
PROVENANCE DISCLOSURE — PROVENANCE — ISSUED UNDER HOLD · STANDS ON THE EVIDENCE

These 24 forecasts were issued on 2026-08-05 by a scheduled job while this arena was held — the hold was written on pages like this one and not into the job that issues, and the fault is ours. They were fit on a collection of the source data that has since been superseded: StatCan WDS returns no value for a suppressed period. The defective run wrote 0.0 for those points instead of discarding them, so 40 fabricated zeros entered the archive as measured index levels. HALIFAX was refused by the model's positivity check as a direct consequence, which is what first evidenced the contamination., superseded on 2026-08-06. The archive was re-collected and each of these forecasts was re-fit against it, at the same publisher vintage each one declared, and every one reproduced exactly — same fit window, same number. They therefore stand, graded under the same rules as everything else, and they carry this note permanently. They are also marked in the record so they can never be pooled with forecasts issued after the re-collection without that being visible.

RULING — THE 24 STAND, with the scar disclosed permanently. Two independent falsifiable paths to one answer: 4,874 period+vintage pairs corroborated across two collections BEFORE the supersession that would have made the measurement vacuous, and 24/24 locks re-fit IDENTICAL at exact double equality with zero fit-window movement, from a tool proven able to report DIVERGENT. The contamination was confined to HALIFAX, which issued no locks. The defect was in the pipeline and the pipeline is fixed; withdrawing sound forecasts to punish a process failure would put process above truth. · Director · 2026-08-06

EVIDENCE — DECISION LEDGER
2026-08-05Canadian refusal report BEFORE supersession — the baseline half of the comparison: 9 series, 4,914 observations, 1 refusal (HALIFAX, positivity), 8 would issue
136ed132-1ac8-46f0-8a8b-966814fdd47b
2026-08-06Canadian archive corroborated: 4,874 period+vintage pairs identical across two independent collections three days apart, the forty fabricated zeros the entire divergence
c233302e-faa3-4d9b-9fda-3d5349968815
2026-08-06Canadian refusal report AFTER supersession — the after-half: 9 series, 4,874 observations, ZERO refusals, and HALIFAX becomes eligible at 506 rather than falling below the floor
dd28a402-605f-4b63-ad8a-c9f7c200b48c
2026-08-06WINDOW DIFF, the 24 Canadian locks: 24 IDENTICAL, 0 DIVERGENT, 0 NO_REFIT, and 0 fit windows moved — every lock re-fits from the clean archive to the value it locked, at full double precision rather than merely to the 3dp bar. Evidence only; no disposition taken.
f76d64cc-1d1a-4d3d-92fb-9bd80b632a9e

How to read this

We said, on the record and in advance, that Canada new-home prices would be −0.21% over 12 months. Nobody can edit that now. In ~Jul 2027 the official figure arrives and we publish whether we were right — including when we were not.

What the words mean →
A forecast of −0.21% over 12 months
Our best single guess at how much the price index moves between now and then. If today's index is 100 and we say +3%, we are guessing it reads about 103 when the period we are forecasting is finally published. It is a guess about an official statistic, not about the price of any particular house.
What "12 months" is counted from
From Jun 2026 — the most recent month Statistics Canada had published when we locked this — not from the day we locked it. Official price statistics arrive a couple of months late, so the newest real figure is always somewhat behind today. Counting the horizon from the data is what makes the forecast a forecast: we predict a month whose figure does not exist yet, and cannot, until Statistics Canada publishes it.
"Locked"
Written down before the answer existed, in a place that cannot be edited or deleted — not by us, not by anyone. The date stamped on it is put there by the database at the moment of writing, so it cannot be backdated. This is the whole point: anyone can look accurate by adjusting a prediction after the fact, and the only defence against that is making it impossible.
The range around the forecast
A single number pretends to a confidence nobody has, so each forecast carries a range. Read it as: if our method is working as intended, the true figure should land inside this range about 95% of the time. A wider range is not us hedging — it is us telling you how much we actually know. Ranges get wider the further ahead we look, because they should.
When this gets its grade
On ~Jul 2027, when the statistics agency publishes the figure for the period we forecast. We do not grade against a number we produced, and we do not go back and choose a kinder one later. Every forecast on this page shows its own verdict date, and the wait is real: a twelve-month forecast cannot be graded early, and we would rather say so than fill the gap with something that looks like progress.
"PENDING", "PASS", "MISS"
PENDING means the answer does not exist yet. PASS and MISS are the verdict once it does. Misses stay on the page permanently, next to the hits. A record with no misses in it is not a good record — it is an incomplete one.

The longer version: how the scorecard works, and why we show our misses →

01 / PRESERVED FORECAST RECORD
LOCKED · LATEST HISTORICAL VINTAGE2026-08-05T09:40:47.249028+00:00
03be6caa · CANADA · 12M
120.347
95% interval 112.804 … 128.393
vintage 03be6caaFOR Jun 2027 · VERDICT ~Jul 2027
OTHER MARKETS · SAME VINTAGE
Winnipeg CMA152.762
Vancouver CMA120.582
Toronto CMA105.062
Québec CMA147.976
Montréal CMA162.066
Edmonton CMA106.952
Calgary CMA132.587
NEXT LOCKheld — schedule disabled pending proof of the issuance gate
02 / EVERY LOCK IN THIS CYCLE

Every forecast this arena has issued, in the order the run log prints them. Each row opens the locked record it came from. None has been graded, so none carries an outcome — a pending row shows what was committed, and nothing else.

DATA THROUGH is the last month Statistics Canada had published when the forecast was locked — the point the horizon is measured from. TARGET PERIOD is the month the forecast is about: data-through plus the horizon. EXPECTED VERDICT is roughly when Statistics Canada publishes that month's figure and the forecast can be graded, on their release calendar rather than ours, which is why it carries a tilde.

Read the three together. A horizon counts months of the index, not months on the wall clock.

EVERY ROW BELOW · PROVENANCE — ISSUED UNDER HOLD · STANDS ON THE EVIDENCE
VINTAGESERIESHORIZONFORECASTCHANGEINTERVALDATA THROUGHTARGET PERIODEXPECTED VERDICTORIGINGRADE
bd9aaa34Calgary CMA3M130.126+0.17%125.577 … 134.839Jun 2026Sep 2026~Oct 2026STANDPENDING
77579f1dCalgary CMA6M130.978+0.83%122.192 … 140.397Jun 2026Dec 2026~Jan 2027STANDPENDING
dfd721d0Calgary CMA12M132.587+2.07%116.073 … 151.451Jun 2026Jun 2027~Jul 2027STANDPENDING
016fcb80Canada3M120.382−0.18%118.474 … 122.320Jun 2026Sep 2026~Oct 2026STANDPENDING
849a1e4fCanada6M120.163−0.36%116.457 … 123.987Jun 2026Dec 2026~Jan 2027STANDPENDING
03be6caaCanada12M120.347−0.21%112.804 … 128.393Jun 2026Jun 2027~Jul 2027STANDPENDING
f4682b44Edmonton CMA3M105.760−0.23%102.502 … 109.122Jun 2026Sep 2026~Oct 2026STANDPENDING
4ff79aaeEdmonton CMA6M106.154+0.14%99.866 … 112.837Jun 2026Dec 2026~Jan 2027STANDPENDING
d3e8bcceEdmonton CMA12M106.952+0.90%95.080 … 120.306Jun 2026Jun 2027~Jul 2027STANDPENDING
b70d532cMontréal CMA3M159.684+0.49%156.635 … 162.792Jun 2026Sep 2026~Oct 2026STANDPENDING
74b2a639Montréal CMA6M160.272+0.86%155.326 … 165.375Jun 2026Dec 2026~Jan 2027STANDPENDING
334dd226Montréal CMA12M162.066+1.99%152.703 … 172.003Jun 2026Jun 2027~Jul 2027STANDPENDING
880e2b40Québec CMA3M145.008+0.56%142.278 … 147.791Jun 2026Sep 2026~Oct 2026STANDPENDING
7c8626beQuébec CMA6M146.011+1.26%141.581 … 150.580Jun 2026Dec 2026~Jan 2027STANDPENDING
dad4904dQuébec CMA12M147.976+2.62%140.460 … 155.895Jun 2026Jun 2027~Jul 2027STANDPENDING
16c3ccf2Toronto CMA3M105.480−0.49%102.646 … 108.392Jun 2026Sep 2026~Oct 2026STANDPENDING
3cd44fa7Toronto CMA6M105.143−0.81%99.995 … 110.557Jun 2026Dec 2026~Jan 2027STANDPENDING
69bad65fToronto CMA12M105.062−0.88%94.700 … 116.558Jun 2026Jun 2027~Jul 2027STANDPENDING
d67e1b39Vancouver CMA3M121.596−0.90%117.959 … 125.344Jun 2026Sep 2026~Oct 2026STANDPENDING
a1086239Vancouver CMA6M120.950−1.43%114.375 … 127.902Jun 2026Dec 2026~Jan 2027STANDPENDING
ff1ee75fVancouver CMA12M120.582−1.73%108.552 … 133.945Jun 2026Jun 2027~Jul 2027STANDPENDING
d6c82e3eWinnipeg CMA3M148.972+0.66%145.477 … 152.552Jun 2026Sep 2026~Oct 2026STANDPENDING
cedb0436Winnipeg CMA6M150.188+1.48%144.432 … 156.174Jun 2026Dec 2026~Jan 2027STANDPENDING
575a136eWinnipeg CMA12M152.762+3.22%143.085 … 163.094Jun 2026Jun 2027~Jul 2027STANDPENDING
03 / GRADED HISTORY

Every row links to the vintage's own page. Misses sit in the same table as passes, in issue order, never sorted away.

SEAM · NOTHING BELOW THIS LINE WAS FITTEDFORWARD RECORD OPENED UNDER HOLD · STANDING ON THE EVIDENCE
INSUFFICIENT DATA
No forecast has been graded yet. The forecasts above are locked and waiting on the official figures for the periods they name; the first verdict appears here on its own schedule, whether it is a pass or a miss.
04 / LIMITATIONS

Constitutional. This block is the same size class as the results panel above and is never collapsed, moved below the fold, or set in a smaller face.

WHAT THIS ARENA CANNOT TELL YOU
L-01
The 24 forecasts on this page were issued without authorisation. A scheduled job locked them on 5 August 2026 while this arena was held pending re-collection of its source data; the hold existed in this page's own copy and in our plans, but not in the job. The fault is ours and the forecasts stay visible because deleting them would be worse.
L-02
They were fit on a Canadian archive we had already declared unfit. An earlier collection turned periods Statistics Canada had withheld into recorded values of zero. Halifax is absent from these forecasts because the model refused to fit it on a non-positive value — which is direct evidence that the bad figures were still in the archive these were built from. That check proves no forecast here was fit on a zero. It does not prove the archive is clean: a fabricated figure that happens to be positive passes it untouched.
L-03
That comparison has now been made, and it is why these stand. The Canadian archive was re-collected under stricter provenance; the two collections were checked against each other before the old one was set aside and agreed on every overlapping figure except the fabricated zeros, which were confined to Halifax — a series that issued none of these forecasts. Each forecast here was then re-fit against the corrected archive at the publisher vintage it declared, and every one reproduced exactly, from an unchanged fit window. They are still marked in the record as issued from the superseded collection, so they can never be pooled with Canadian forecasts issued after the re-collection without that being visible.
L-04
No forecast here has been graded. Nothing on this page is a claim about accuracy.
L-05
The v1 Canadian backtest is quarantined and counted nowhere. It was loaded through an endpoint that accepted forecasts and grades as a request body and stored them as sent — no recomputation, no source binding, no manifest. It is kept visible because deleting a bad record is worse than showing it, and it is excluded from every figure on this site.
L-06
Our own backfilled Canadian history is not a real-time archive. StatCan's endpoint returns each period's current value stamped with the release time of that value, so for any period revised before we first pulled it we hold the revised figure with a late release date. Real-time coverage begins with periods first observed after this ingestor started; everything earlier is labelled honestly but is not a first print.
L-07
Ottawa–Gatineau is excluded, not missing. StatCan cube 18-10-0205 splits it into an Ontario part and a Quebec part with no combined metro series, and combining them needs weights the cube does not supply. It is named here rather than quietly approximated.
05 / CALIBRATION
INSUFFICIENT DATA
A calibration curve needs graded forecasts to plot. These forecasts are locked but none has matured, so there is no curve — not a flat one, and not a shape borrowed from anywhere else.
06 / PROVENANCE & ENDPOINTS
PRIMARY SOURCEStatistics Canada (WDS)
TABLENew Housing Price Index, monthly (18-10-0205)
SERIES RESOLUTIONresolved per run from cube metadata; no vector ID is pinned
COVERAGECanada national + 8 CMAs; Ottawa–Gatineau excluded
MODEL SPECv1.1 — AR(12) on log first differences, pre-registered 2026-08-04
ACTUALSfirst_print (headline) · settled_t_plus_12 (companion)
ISSUE CADENCEheld — schedule disabled pending proof of the issuance gate
ARCHIVEre-collected; the collection these locks were fit from is superseded and they are marked as such
THE v1 API SERVES THE RECORDIt reads the same rows this page renders, including these 24 locks and their provenance.
GET /v1/vintages?arena=housing_caThis arena's locked vintages, from the record, newest first.
GET /v1/calibration?arena=housing_caSample fixtures. Empty until grades exist in any case.
The One Lab · Toronto · Every number on this site resolves to a primary source or a graded model output. Where it can't, we say so.ARENA AR-02 · NEXT LOCK HELD — SCHEDULE DISABLED PENDING PROOF OF THE ISSUANCE GATE