Skip to content

Scoring

Every forecast is scored. Misses first.

Instead of quotes, a promise you can check: each forecast is written down before the record shows it, then scored against what the record shows next.

What is scored

A forecast is a claim you can check.

Each one names five things on the day it is made. Those five things are what the scorecard holds it to.

The forecast

Company, matter, practice, timing, strength.

  • Company. One organisation, by its registered name.
  • Matter and practice. What it will need counsel for, and which practice that falls to.
  • Timing. A range in months from the day of the forecast, never a single date.
  • Strength. Strong, moderate or early, with the number of records behind it.
  • What would confirm it. The records that would settle it, written down in advance.

Forecast · Coldbrook Refrigeration, which no filing names yet. Strong signal, five records, expected in 2–4 months.1

The four outcomes

Every forecast ends in one of four places.

The shaded bar is the timing range, here 1–3 months. What happens on the record decides the outcome.
01

Missed

036

The range closed and no record showed the matter within the grace period3, or a later record ruled it out. Matters the record showed before any forecast are counted here too.

02

Too early

036

Right company, right matter, but the record came after the range closed. A partner acting on the timing would have been early. Never counted as confirmed.

03

Confirmed

036

A public record names the company in the forecast matter before the range closes. A record that arrives sooner than expected still confirms it, and is marked as early.

04

Still open

today036

The range has not closed. Open forecasts are listed by name but not counted either way until they are settled.

What settles a forecast

  • A docket entry that names the company in the forecast matter
  • A lien or charge registered against it, or released
  • A filing at a companies registry
  • A regulator notice or final rule that names it

What does not

  • News coverage or a company’s own announcement without a filing
  • Whether the firm was instructed, or won the work: that is confidential, and not what is being measured
  • Another forecast, ours or anyone’s
  • A record found after the scorecard date, which goes into the next one

Strength and timing

No likelihood percentages. Ever.

A percentage on a single company suggests a precision nobody has. Monition shows what it actually knows: how many independent records point the same way, how directly they point, and a range of months in which the matter is expected. The scorecard then reports outcomes by strength, so a firm can see whether “strong” has earned its name.

Board · every card carries its strength, record count and timing range. Those three are what get scored.1

Strength

How many independent records support the forecast, and how directly. The record count is always shown beside it.

  • StrongSeveral independent records, at least one direct
  • ModerateTwo or more records, mostly indirect
  • EarlyOne leading record so far: worth watching

Timing range

Months from the day of the forecast, for example 2–4. Shorter ranges are harder to hit; the scorecard does not reward a range for being wide, because every range is shown next to its outcome.

Forecast or record

A forecast comes before any filing names the need. When a record already shows it, Monition identifies it instead, and the scorecard counts that as a miss for the forecast.

Sample scorecard

Week 8, misses first.

What a firm receives in week 8: 25 forecasts, every one accounted for. The firm and companies are fictional.2
Week 8 scorecardWrenfield Tate LLP · forecasts made in weeks 1–8IllustrativeFictional firm and companies
5Missed: 2 ruled out, 3 not forecast
0Too early: no range has closed yet
5Confirmed inside the range
18Still open, not scored yet

Missed · ruled out (2)

A later record shows the forecast cannot happen.

  • Vitra-Form GlassSupply contract renegotiationCorporate & M&AEarly· 1 rec3–6 moRegulator notice · week 6The revised rule drops glass from scope. The forecast cannot happen.
  • Brisk HaulageUnpaid sub-haulage invoicesLitigationEarly· 1 rec1–3 moCourt filing · week 4The haulier paid its sub-hauliers in full. The forecast cannot happen.

Missed · not forecast (3)

The record showed the need first. Monition identified it, but did not predict it.

  • Arvel Industrial LLCEnforcing the contractor's lienLitigationModerate· 2 rec1–2 moProperty register · week 2A contractor’s lien named the company first. No forecast came before it.
  • Stellan Linen ServicesTrade claim in the caseRestructuringModerate· 1 rec1–3 moSame docket · week 3The creditor list named it as unpaid. No forecast came before it.
  • Rennick Co-PackersProducer duties under the proposed ruleRegulatoryStrong· 3 rec2–4 moRegulator notice · week 1The proposed rule named it as an obligated producer. No forecast came before it.

Too early (0)

Right company and matter, but the record came after the range closed.

None can be scored yet. The first open range closes in week 13; it will be reported here either way.

Confirmed (5)

A public record showed the matter inside the timing range.

  • Kestrel Row Retail ParkLease rejection and rent claimReal estateModerate· 2 rec1–3 moSame docket · week 7The lease at Kestrel Row Retail Park is listed for rejection.
  • Lumen Fit-Out LeasingRecovering leased fit-out equipmentBanking & financeModerate· 2 rec1–2 moSame docket · week 6Motion to recover the leased fit-out equipment from closed sites.
  • Keelson Trade FinanceCovenant breach on receivables facilityBanking & financeStrong· 3 rec1–2 moCharges register · week 5Its lender registered a notice of default on the receivables facility.
  • Quayside Market HallLease rejection and re-lettingReal estateModerate· 2 rec1–3 moSame docket · week 6The lease at Quayside Market Hall is listed for rejection.
  • Wexley Fuel CardsTrade claim against the haulierRestructuringEarly· 1 rec1–3 moCourt filing · week 8It files a claim against the haulier for unpaid fuel cards.

Still open (18)

The range has not closed. Not counted either way until it does.

  • Millbrook Commons2–4 mo
  • Ardent Square Partners2–4 mo
  • Orchard Bay Dining2–3 mo
  • Corvane Bank2–3 mo
  • Marlowe Retail Group1–3 mo
  • Pell & Daughters Grocers1–3 mo
  • Tolland Industrial Estates3–6 mo
  • Corran Fleet Leasing2–4 mo
  • Ashgrove Depot Holdings2–5 mo
  • Coldbrook Refrigeration2–4 mo
  • Ferris Lane Properties2–4 mo
  • Harrowgate Stores1–3 mo
  • Northgate Eateries2–3 mo
  • Sorrel Gate Retail2–4 mo
  • Penmore Logistics1–3 mo
  • Fairholt Drinks2–4 mo
  • Greyline Recovery3–6 mo
  • Harlow Bridge Insurance2–4 mo

By strength

So you can see whether “strong” means anything.

StrengthMadeConfirmedRuled outOpen
Strong3102
Moderate9306
Early131210

Illustrative scorecard for the fictional firm used across this site. Counts only, by design.1

How we report misses

Six commitments, made before the first forecast.

They are written into every firm’s agreement, so a firm can hold us to them.
  1. Misses come first.

    Every scorecard opens with what went wrong, by company, with the record that settled it.

  2. Nothing is edited after the fact.

    A forecast is time-stamped when it is made. It is never withdrawn, re-dated or reworded.

  3. What we failed to predict counts against us.

    If the record shows a need before any forecast did, it is a miss, even though Monition then identified it.

  4. Every line can be checked.

    Outcomes are settled by public records only, so your own team can verify each one.

  5. The date does not move.

    Week 8 is fixed in each firm’s agreement before it starts.

  6. Counts, never percentages.

    Small numbers are shown as what they are: how many, out of how many.

Backtest

Published when the data is ready. Not before.

The backtest runs Monition on an earlier stretch of the public record: freeze it at a cut-off date, let it forecast using only what was published by then, and score those forecasts with the same four rules against what the record showed afterwards.

Company names are redacted. Misses are listed first, as here. If the results are poor, they are published anyway, with what we changed.4

Method
Cut-off date
[date]
Period scored
[months to confirm]
Record types
Dockets, registries, charges, notices
Forecasts
[published with the results]
Publication
[date]

Hold us to the scorecard. Join the waitlist.