The scoreboard, kept honestly.

Everywhere else on FDVS shows you a current snapshot — today's fixtures, this week's results, right-now ratings. A report is different on purpose: it's a fixed summary of a specific, closed period, published once and left as a permanent record rather than continuously updated. Last week's accuracy report describes last week; it doesn't quietly change if this week goes badly, which is exactly what makes it a genuine record rather than a moving target. A snapshot can drift with every new result; a report is a fixed point you can always look back at, and that permanence is the entire reason this page exists as its own section rather than folding into the results archive.

Weekly versus monthly versus season reports

Weekly reports track consensus accuracy at a granular level — useful for spotting a short-term calibration wobble early. Monthly reports zoom out to league-wide trends, which need a larger sample to say anything meaningful about tempo or scoring shifts. Season reports, like the Prime Pick performance summary, aggregate an entire period's settled picks into one place. Each cadence exists because the right sample size is different for the question being asked — a single week is enough to flag a band running hot, but not enough to say anything reliable about a full league's scoring trend, and a full season is what it actually takes to judge whether a curated pick list like Prime Pick is delivering on its own stated standard. Reading only one cadence gives a partial picture; reading all three together is closer to the complete one.

Why some numbers are free and the full export isn't

The headline metrics on every report — hit rate, fixtures tracked, the standout result — are visible to everyone, because the core claim of accountability shouldn't be locked behind a paywall. What's gated on Pro is the granular export: every individual fixture's grading, every market-level breakdown, in a downloadable format. The summary tells you the honest headline; the full export is for anyone who wants to audit the number themselves, fixture by fixture, rather than take the summary's word for it. That split mirrors how the rest of the platform is structured — free access to the honest headline, deeper layers unlocked as you go further.

How this connects to AI Research

AI Research explains the methodology — what calibration means, how bands are checked, why a 70% call losing three times in ten is expected rather than a failure. These reports are that methodology applied to a specific, dated period, with real numbers attached. If a weekly report flags a band running hot or cold, that's the kind of finding the calibration process on AI Research is designed to catch and act on — the research page explains the process, the reports are the process actually running, week after week, on real fixtures rather than in the abstract.

How this connects to the results page

Every fixture behind these reports is a result already shown individually on the results page, verdict and all — reports don't introduce new grading criteria, they aggregate the same honest wins-and-misses record into a period summary. If a specific result surprises you in a weekly report, the results archive is where you'd go to see that exact fixture's full breakdown, including the pre-match consensus it was checked against and the market-level chips that hit or missed alongside the headline verdict.

Why reports don't get revised after publishing

Once a weekly or monthly report is published, it stays as originally written — we don't quietly edit a past report to look better in hindsight. If a later review finds something worth correcting, that correction gets published as its own dated note rather than silently rewriting history. A report archive only holds value if you can trust that what you're reading today is what was actually said at the time, not a version that's been quietly smoothed over since a rough week made the original look worse than a later edit would.

Coverage and cadence

Weekly reports publish every Monday covering the previous seven days; monthly reports publish in the first days of the following month; season reports update as picks settle throughout the year. Coverage matches the rest of the platform — every league currently tracked on the fixtures board feeds into these reports as its sample size allows, and a newly added league won't appear in a monthly trends report until it has enough tracked matchdays behind it to say something reliable. That same discipline — waiting for a real sample before publishing a number — runs through every page on FDVS, and this one is no exception.

Why a periodic report matters alongside a live board

A live, constantly-updating board is excellent for right-now decisions, but it's a poor tool for judging accuracy over time — by the time you'd want to check whether last month's numbers held up, the board has already moved on. A report exists specifically to freeze a period in place so it can actually be judged, which a page that's always current structurally can't do on its own.