---
title: "How We Rate Broker-dealers"
canonical: "https://papertrailfinance.com/broker-dealers/methodology"
as_of: 2026-11-09
updated: 2026-11-09
type: methodology
publisher: "The Paper Trail"
firm_type: "broker-dealers"
sources:
  - name: "FINRA BrokerCheck"
    url: "https://brokercheck.finra.org/"
---

# How We Rate Broker-dealers

The Paper Trail profiles 484 broker-dealers and scores them on version 1.9 of a written rubric; 379 have a published score. This page is the rubric as the rating program applies it: each factor and its weight, the evidence it uses, when a factor is left out, and the gate a score must pass. Records last read Nov. 9, 2026.

The scoring program and the second program that recomputes every score on each rebuild were written by the same research team, so the second program checks the arithmetic, not the reading of the rules (it had copied two of the rating program's faults). The check of the written rules is a re-implementation by a different reviewer, who built the rubric from this page alone without opening either program. When run against version 1.6 of this rubric (version dated Nov. 10, 2026), the re-implementation found 6 overall scores and 7 factor scores that differed; each was traced to the rating program and fixed before publication. When run against version 1.7 of this rubric (version dated Nov. 10, 2026), the re-implementation found 80 overall scores and 84 factor scores that differed; each was traced to the rating program and fixed before publication. When run against version 1.9 of this rubric on Nov. 11, 2026, the re-implementation matched every factor score, overall score, confidence level and publication decision for all 484 broker-dealers. Version 1.9, the one on this page, is the version that re-implementation checked. A patch to this version dated Nov. 12, 2026, which keeps the version number, is described in the change log below and came after that run, so a re-implementation has not re-checked it.

## What weights does the rating program use?

These weights are read from the scored profiles themselves, so they match every score on the site (379 published). A factor without enough evidence for a firm is left out and the remaining weights are scaled up to 100%.

| Factor | Weight |
| --- | --- |
| Regulatory record | 35% |
| Financial condition | 30% |
| Customer dispute record | 15% |
| Legal record | 10% |
| Operating history | 10% |


Follows the adviser model ([the adviser scoring method](/firms/methodology)): 0 to 10 factors with weights, N/A plus renormalise, outside-record factors, confidence, publication gate, change log. The score describes the **public regulatory and financial record of the firm**. It is not advice, not a rating of service, and never rates individuals (registered representatives, officers, owners). Pending matters and allegations are shown and never scored.

As-of date for every score: Nov. 9, 2026 (BrokerCheck capture date). Arithmetic: a program; recomputation of every score on each rebuild: a second program written by the same team, a consistency check of the arithmetic and not an independent check (the independent check is described at the top of this page). 

| Factor | Key | Weight | Inputs |
|---|---|---|---|
| Regulatory record | regulatory | 35% | Final regulatory matters against the firm on its BrokerCheck (CRD) record: FINRA, SEC, states, exchanges, other regulators |
| Legal record | legal | 10% | Final civil judicial, criminal, financial (bankruptcy) and judgment/lien matters on the firm's CRD record |
| Customer dispute record | disputes | 15% | FINRA arbitration awards with the firm as respondent (BrokerCheck arbitration section) |
| Financial condition | financial | 30% | Net capital against the required minimum, material-weakness flag, going-concern language, from the firm's latest electronic annual audited report (Form X-17A-5 on EDGAR) |
| Operating history | history | 10% | Years registered with FINRA (BrokerCheck registration table) |

Overall = weighted mean of the scored factors (weights renormalised) x 10, rounded once, half-up (0 to 100). Factor scores are rounded half-up to one decimal and stored as shown; the overall is computed from those one-decimal scores (v1.6).

## How are retail and non-retail firms treated?
A firm is **retail** when BrokerCheck shows a Form CRS on file (type BD, DU or IA), otherwise **non-retail** (institutional, proprietary, fund-distribution, investment banking). The label is shown on the profile. It changes one rule: Customer dispute record (below). Form CRS text is not retrievable under this program's access rules (BrokerCheck's file route refuses the request and investor.gov blocks it), so CRS clarity and CRS-versus-record consistency are **not scored in v1** (see Deferred).

## How is firm size taken into account?
Raw event counts are size-confounded (Wedbush 149 regulatory matters, most firms 0 to 4). FINRA's registered-person count is not in the BrokerCheck firm JSON or search hit (checked Nov. 9, 2026: the hit carries branch count and a count of approved FINRA registrations, not representatives) and X-17A-5 reports do not state it consistently, so it is not available at scale. Proxy: the larger of the floor and the square root of the number of branch offices, floor 1.0 Small, 2.0 Medium, 4.0 Large (FINRA size class), branch offices from the BrokerCheck search hit. Square-root scaling because compliance infrastructure grows with size, so events grow less than linearly. If the size class is missing (inactive firms) the event factors are N/A.

## How is regulatory record scored (35% of the score)?
Usable when the BrokerCheck detailed report was parsed and its firm-scope disclosure counts equal BrokerCheck's own disclosure counts per type (reconciliation; all 484 profiled firms reconcile (the matching run counted 491 firms, which also included 7 firms later left out because FINRA lists their main office outside the West)).
- Matters = one per "Disclosure N of M" block (the regulator's and the firm's version of one event are one matter).
- Only matters with status **Final**, firm scope, and an outcome against the firm score (v1.2, see "What counts as a scored matter" below). Pending, On Appeal and control-affiliate matters are shown, not scored.
- Tier from the CRD record's structured fields (never free text): money = fine + restitution amounts.
 - Tier 1 (3.0 points): money >= $250,000; or a Suspension, Expulsion, Revocation, Bar or Disqualification sanction; or a non-SRO initiator (SEC, state, federal) with money >= $100,000 or a Cease and Desist/Injunction.
 - Tier 2 (1.5): money > $10,000, or any sanction type other than Censure, Monetary/Fine and Costs (undertaking, restitution order, restriction); an order to reimburse costs is money shown and labelled on the row, not a sanction type, and does not by itself make a matter Tier 2 (v1.9.1).
 - Tier 3 (0.5): otherwise (censure, fine <= $10,000, no sanction). A FINRA membership **cancellation** under Rule 9553 for unpaid fees (sanction type "Cancellation", v1.9.2) is administrative: it is not a revocation, expulsion or denial and not a sanction type for Tier 2, so with no money it is Tier 3.
- Age weight on the resolution date (date initiated if none): <= 5 years 1.0, 5 to 10 years 0.5, older 0.2.
- Score = max(0, 10 - D), where D = max(3.0 x sum(points x age weight) / units, F) and F is the **absolute floor (v1.3, amended v1.4)**: F = min(10, 0.5 x (final firm-level Tier 1 or Tier 2 regulatory actions resolved in the last 5 years) + 0.25 x (the money in all final firm-level actions resolved in the last 5 years, $ millions)). A minor Tier 3 matter counts only through the size-scaled deduction (v1.4). Size scaling divides a firm's record by up to sqrt(branch offices); for a network of thousands of one- or two-person offices that erased a large absolute record (Osaic Wealth, 6,915 branches: 51 final actions, 9 in the last five years with $18.3 million, scored 8.6 of 10 on this factor). The floor keeps scaling for ordinary differences in size but makes each recent final action and each recent $1 million cost the firm a minimum regardless of size. It binds for 6 firms.
- Self-regulatory initiators are FINRA, NASD, NYSE, Nasdaq, Cboe, MSRB, CME and other exchanges/clearing bodies (list in a program).

- **Joint and several amounts and an individual's own sanctions.** A fine, restitution or cost ordered jointly and severally with an individual (Alpine Securities, CRD 14952: fines of $1,000 imposed jointly and severally with an individual) is the firm's liability for the whole amount: it is counted once, in full, as the firm's money and the row says "jointly and severally with an individual (not named)". Sanctions the order imposes on the individual alone (a bar, a suspension, a separate fine) are stated as on "an individual", never named, and are never scored or added to the firm's money. The firm's own sanctions score as the firm's.

## What counts as a scored matter?
Every firm-scope matter carries class of score in its fact value, computed from structured CRD fields only (a program classify, re-implemented in a second program). Anything not scored is shown with reason a factor was not scored and never enters a factor:
| Class | Rule | Effect |
|---|---|---|
| arbitration-claim | A row in the regulatory section whose initiator is FINRA, whose docket has the FINRA Dispute Resolution case-number form YY-NNNNN, and that records no sanction, no fine and no restitution | A customer arbitration claim, not a regulatory action. Not scored as regulatory. It reaches the Customer dispute record only if it is an award of money (it never is here: no award is recorded). |
| no-outcome | A final regulatory row resolved "Dismissed", "Withdrawn", "Vacated", "Other" or blank, with no sanction, no fine, no restitution | No outcome against the firm is recorded. Not scored. |
| private-suit-no-outcome | A civil row brought by a private party (not a regulator, government body, court or self-regulatory organisation) with no sanction and no money recorded against the firm | Private suits with no recorded outcome are not scored. A private suit that ended in a money judgment against the firm still scores under Legal. |
| pending | Pending / On Appeal | Shown, not scored (unchanged). |
| affiliate, individual | Rows in the report's control-affiliate section (v1.3: concerns = "affiliate", or "individual" when the affiliate named is a natural person) | Shown as the affiliate's record, never as the firm's; an individual's name is withheld; never scored. |
| duplicate (v1.3) | A second BrokerCheck listing of the same event: same docket, kind and money, resolved within 30 days of the first | Shown, counted once. |
Every matter carries concerns ("firm", "affiliate" or "individual"), decided from the BrokerCheck section that holds the row. A reviewed outside-record order (SEC or FINRA, read at the regulator's site) whose file, release or case number equals a firm matter's docket is the same matter (v1.3): it is merged into that row (sources, count of outside-record items merged into a row) instead of being listed twice; the regulator's order amount becomes the matter's money (amount in dollars) and the CRD figure is kept as figure the firm filed. Customer arbitration awards keep their own rule (Customer dispute record): only an award of money to the claimant counts; a claim is never a regulatory action. A private party's name is never published: the matter's who brought the matter reads "a private party", the "Initiated By" quote line is dropped from the published quote, and the raw value stays in the unpublished value as first read / quote as first read. A regulator's staff contact (a name and phone number BrokerCheck prints after the agency) is cut the same way.

## How is legal record scored (10% of the score)?
Final matters of kind criminal (4.0 points), civil (2.0), financial (2.0), judgment/lien (1.0), age-weighted as above. Score = max(0, 10 - 3.0 x sum / units). Fidelity-bond payout records ("Civil Bond") are shown, not scored. BrokerCheck reports every such event the firm or a regulator filed, so no event is a clean record, scored 10.

## How is customer dispute record scored (15% of the score)?
BrokerCheck lists FINRA arbitration awards where the firm was respondent. Only an award of money to the claimant counts (relief awarded > 0): 2.0 points if >= $1,000,000, 1.0 if >= $100,000, else 0.5, age-weighted on the disposition date. Denied claims ("AWARD AGAINST PARTY" with $0 awarded) do not count. Score = max(0, 10 - 5.0 x sum / units). Settled cases are not public, so a clean record does not mean no complaints (shown on the page).
- Retail firm: always scored.
- Non-retail firm: scored only when at least one award record exists; with none the factor is **N/A** (no retail customers, so silence carries no information).

## How is financial condition scored (30% of the score)?
Source: latest electronic X-17A-5 filing on EDGAR (greatest period end; paper filings cannot be read), annual report PDF text, plus the structured cover page (material weakness, accountant).
- Usable when the report states net capital and the required minimum, the period end is no more than 21 months before the as-of date, and no newer filing is paper-only. Otherwise N/A with the reason.
- ratio = net capital / required minimum. Base score by linear interpolation: ratio < 1.0 is 0; 1.0 is 2; 1.2 is 4; 1.5 is 6; 2.0 is 8; >= 3.0 is 10.
- Deductions (floor 0): material weakness = Y: -2; sentence stating substantial doubt about the ability to continue as a going concern: -4.
- Amounts stated "in thousands" or "million" are converted; "approximately" figures are used as stated.
- A figure the extractor cannot read (scanned PDF, unusual wording) is N/A, never a guess.

## How is operating history scored (10% of the score)?
Years since the FINRA "Date Effective" in the registration table: >= 25 years 10; >= 15 is 9; >= 10 is 8; >= 5 is 6;
>= 2 is 4; < 2 is 2. N/A without a date. Weakest signal, lowest weight; a new registration is a fact about age, not conduct.

## When is each factor not scored?
| Situation | Effect |
|---|---|
| No BrokerCheck report, or counts do not reconcile | Regulatory, Legal, Disputes N/A; held |
| Non-retail, no arbitration awards | Disputes N/A |
| No electronic X-17A-5, unreadable net capital, or stale report | Financial N/A |
| Size class missing | Regulatory, Legal, Disputes N/A |
| No registration date | History N/A |

## How is the outside record scored?
Regulatory record, Legal record and Customer dispute record rest on records regulators and FINRA publish, not on the firm's own statements. Financial condition rests on the firm's own filing (audited, but filed by the firm) and does not count for the gate.

## How is confidence set?
Points: +1 if Regulatory or Legal is scored; +1 Customer dispute; +1 Financial condition; +1 Operating history; +1 if a second-pass verification covered every fact with no unresolved mismatch; +1 if the outside-record sweep is complete (every SEC litigation-release and administrative-proceeding name hit and every web search lead on a primary domain or scoring >= 0.7 that names the firm was reviewed against primary records; a firm with none has nothing to review); -1 if the verifier's mismatch rate exceeds 5% (the mismatches still unresolved in the firm's verification block after review, divided by the facts the verifier checked; a mismatch the first comparison found and that was then fixed at the source does not count; no firm is at that rate today); -1 when the size proxy moved a score (any event points > 0). High >= 6, Medium 4 to 5, Low <= 3. High therefore needs the reviewed outside-record sweep (without it the maximum is 5, Medium). **v1.2 cap:** a firm registered with FINRA for less than 5 years (or with no registration date) is capped at Medium (confidence ceiling states it). An empty record over a short window is weak evidence: the event rules give full weight to the last 5 years, and a younger firm has not been observed over that window or filed five annual reports. Before v1.2, 63 of the 66 High firms were Small firms and 14 of them were under 5 years old (the newest registered Aug. 13, 2025); those 14 move to Medium.

**v1.3 coverage rule (shared across firm types).** Every record states share of the rubric's weight that was scored (share of the rubric's weight that was scored) and the factors left unscored, each with its reason (each with its reason). When a factor of 25% weight or more is not scored, or less than 75% of the weight is scored, confidence is at most Medium and note on confidence says why; under 60% the rating is held ("not enough of the rubric could be scored"). For broker-dealers the heavy factor is Financial condition (30%): 197 of 456 rated firms had it N/A under v1.2. A firm with both Financial condition and Customer dispute record N/A (non-retail, no electronic net-capital figure) has 55% of the weight scored and is held: 77 firms.

## When is a score published?
A firm publishes only when **all** hold; otherwise the page says why:
0. At least 60% of the rubric's weight scored (v1.3).
1. At least 3 of the 5 factors scored.
2. At least one outside-record factor scored.
3. Confidence Medium or High.
4. BrokerCheck status ACTIVE and Approved (not pending withdrawal, termination requested, postponed or inactive) and not currently suspended.
5. FINRA main office in one of the 13 West states. Firms the SEC lists in the West whose FINRA main office is elsewhere are **out of scope**: no firm file, listed in the underlying records, not published, with the reason "headquartered outside the West". (7 firms with a known outside state; where BrokerCheck has no office address the SEC-listed state is used and the file says so.)
6. Reconciliation passes.
7. Independent verification complete with no unresolved mismatch. a rating is published only when the gate passes.

## What does the page show from the outside record?
Every outside-record item carries published summary (published as summary): regulator or forum, matter type, docket / release / case numbers, sanctions and amounts, status, date, assembled from structured fields by a program (hand corrections in the underlying records), never copied from free text. The reviewer's text stays in the unpublished reviewer's original text. Every BrokerCheck pending or on-appeal matter is also listed as an outside-record row (marked as coming from the BrokerCheck report), so the page's pending count and the rating text count the same matters; a reviewed item for the same docket is merged into that row (count of reviewed items merged into a row). The audit checks every count stated in a factor's basis, its plain sentence, pending matters, shown and not scored and the search log against the rows it summarises.

## How are scores spread, ranked and turned into percentiles?
**Rank eligibility (v1.5).** A firm is rank-eligible only when it is rated and at least 85% of the rubric's weight was scored, and it does not reach the top tenth only because of a data gap (below). percentile within its type (share of rank-eligible firms with a lower overall plus half the ties) and any rank list are computed over rank-eligible firms only; every other firm has percentile within its type blank and a ranking note saying why. On Nov. 11, 2026: 379 rated, 258 rank-eligible, 121 rated firms not ranked because only 70% of the weight was scored (financial condition N/A; Somerset Securities, 2493, is one).

The score distribution: 379 rated firms; min 40, P10 83.0, P25 91.0, median 95.0, P75 99.0, P90 100.0, max 100; mean 92.64, standard deviation 9.75; 79.7% score 90 or more. Most broker-dealers have no final regulatory event, so the scale separates the minority with events and compresses the rest; a high score means a clean public record, not a better firm. Re-anchoring is deferred.

## How much do data gaps move the scores?
Every rated firm's overall recomputed with each N/A factor scored at the type median of that factor (median over rated firms that have it) instead of renormalising the other weights. 244 of 379 rated firms have at least one N/A factor (disputes 123, financial 121).

| Measure | Value |
|---|---|
| Spearman rank correlation, published vs median-filled | 0.991 |
| Firms whose overall moves 3 points or more | 48 |
| Largest change | 18 points |
| Median change among firms with a gap | 1.0 |
| Type median of the disputes and financial factors | 10 and 10 |
| Top-tenth cut, published / median-filled | 100 / 100 |
| Firms in the top tenth only because of a data gap | 0 |

Reading: the medians of both factors that are ever N/A are 10, so filling a gap with the median raises an overall or leaves it; no firm reaches the top tenth only because of a gap. The median test therefore cannot catch a firm whose high overall rests on a missing financial factor; the 85% rule does (the 121 firms with financial condition N/A are not ranked).

## How sensitive are the scores to the weights?
Each factor's weight moved by 5 points with the others unchanged (the weighted mean renormalises); compared with the published overall.
| Factor (weight) | Change | Spearman vs published | Median abs. change | Max abs. change | Firms moving >= 3 points |
|---|---|---|---|---|---|
| regulatory (35) | +5 | 0.9956 | 0 | 3 | 14 |
| regulatory (35) | -5 | 0.9944 | 0 | 4 | 16 |
| legal (10) | +5 | 0.9978 | 0 | 4 | 6 |
| legal (10) | -5 | 0.9941 | 0 | 5 | 10 |
| disputes (15) | +5 | 0.9986 | 0 | 3 | 4 |
| disputes (15) | -5 | 0.9942 | 0 | 4 | 4 |
| financial (30) | +5 | 0.9979 | 0 | 3 | 1 |
| financial (30) | -5 | 0.9989 | 0 | 4 | 6 |
| history (10) | +5 | 0.9832 | 1 | 5 | 56 |
| history (10) | -5 | 0.9724 | 1 | 5 | 85 |
Reading: the ranking is stable under every 5-point shift (Spearman >= 0.97). Operating history, the weakest signal, moves the most firms, because most firms score 10 on the event factors and differ only on age.

## Which factors are deferred, and why?
- Form CRS clarity and CRS-versus-record consistency: text not retrievable under the access rules.
- FINRA Disciplinary Actions Online and Arbitration Awards Online searches: FINRA's robots rules disallows both; the same FINRA actions and awards arrive through BrokerCheck.
- Registered-representative count (events per rep): not available at scale.
- Predecessor and merged-firm actions not on the firm's own CRD record: shown as outside-record items, unscored.
- Control-affiliate matters (parent or sister company events in the report's affiliate section): shown, unscored.

## How are inactive firms handled?
A broker-dealer whose BrokerCheck scope is INACTIVE (FINRA expelled it, cancelled its membership, or its registration never became effective) is a **historical profile, not rated**. It is stored as rating status: "withdrawn-not-active" with a status object (state expelled, cancelled, inactive or registration-not-effective; effective month as BrokerCheck's own message states it; brokercheck message verbatim), the same way a closed bank or credit union is. It is never ranked, never counted among active or rated firms, and every factor says "No score: this firm is no longer an active broker-dealer". A cancellation for unpaid fees under Rule 9553 is a cancellation of membership; it is not a revocation, an expulsion or a denial, and a matter filed that way carries the sanction type "Cancellation". A firm whose status is Pending Withdrawal or Termination Requested is still active on the day it was read: it is not rated (gate 4), and the status says what is pending (status.pending event). Tool: a program (re-applied after every producer run by a program).

## Why do so many scores sit high?

Most broker-dealers in the record have no serious regulatory, legal or financial problem on file, so published scores bunch near the top: the median is 95 of 100, 302 of the 379 scores are 90 or higher and 54 are 100. The score separates firms that have a record of enforcement, disputes or weak finances from firms that do not; it does not rank one clean-record firm above another, and a difference of a few points between two high scores is not meaningful. Where firms share a score, the lists of highest scores on the type's page put the larger firm first and say that ties are ordered by size.

## How complete are the fields on these profiles?

For each field, how many of the 484 broker-dealer profiles carry a value; how many have the source read and the field blank or not reported there ("not disclosed in the source"); and how many have no such fact at all ("not collected").

**Field completeness, 484 broker-dealer profiles**

| Field | Present | Not disclosed in the source | Not collected |
| --- | ---: | ---: | ---: |
| Name | 484 (100.0%) | 0 (0.0%) | 0 (0.0%) |
| City | 484 (100.0%) | 0 (0.0%) | 0 (0.0%) |
| State | 484 (100.0%) | 0 (0.0%) | 0 (0.0%) |
| CRD | 484 (100.0%) | 0 (0.0%) | 0 (0.0%) |
| Firm size class | 476 (98.3%) | 8 (1.7%) | 0 (0.0%) |
| Branch offices | 484 (100.0%) | 0 (0.0%) | 0 (0.0%) |
| FINRA registration date | 469 (96.9%) | 0 (0.0%) | 15 (3.1%) |
| Form CRS (retail) | 261 (53.9%) | 223 (46.1%) | 0 (0.0%) |
| Disclosure counts | 484 (100.0%) | 0 (0.0%) | 0 (0.0%) |
| Audited annual report (X-17A-5) | 453 (93.6%) | 0 (0.0%) | 31 (6.4%) |
| Auditor | 452 (93.4%) | 0 (0.0%) | 32 (6.6%) |

Why audited annual report (X-17A-5) is missing: no electronic annual audited report on EDGAR: the firm's X-17A-5 reports are filed on paper (18); no EDGAR submissions record for the firm (no CIK filing history could be retrieved) (13).

Why auditor is missing: no electronic annual audited report on EDGAR: the firm's X-17A-5 reports are filed on paper (18); the electronic annual report's cover page names no accountant in a readable field (1); no EDGAR submissions record for the firm (no CIK filing history could be retrieved) (13).

## What changed in each version?

- **Data note, Nov. 11, 2026: no rule change, rating program version 1.9 unchanged.**
- **Patch to v1.9 (Nov. 12, 2026; rating program version 1.9 unchanged), no formula change.** Costs (reimbursement of an investigation, hearing or inspection cost) are money, not a sanction type: they stay on the row, labelled, and no longer make a matter Tier 2 on their own (Tier 2 definition above; a program, a second program, a second program).
- **Version 1.9, Nov. 11, 2026: corrections.** Restitution, reimbursement or disgorgement stated only in the sanction-details text is read into restitution amount in dollars: 76 matters in 34 firms, 7 overalls move.
- **Version 1.8, Nov. 11, 2026: identity rule.** Effect: 0 overall, publication, confidence or rank changes (the lead was never scored).
- **Version 1.7, Nov. 10, 2026: independent re-implementation findings.** The independent re-implementation, run against version 1.6, found 7 factor and 6 overall scores that differed from the rating program; each difference was traced to the rating program and fixed in this version.
- **Version 1.6, Nov. 10, 2026: one rounding rule for every type.**
- **Version 1.5, Nov. 10, 2026.**
- **Version 1.4, Nov. 10, 2026: second review of the absolute floor.**
- **Version 1.3, Nov. 10, 2026.** What changed: coverage rule; absolute floor on the regulatory deduction; duplicate matters; Somerset Securities. Coverage rule (shared): share of the rubric's weight that was scored, the factors left unscored, each with its reason, note on confidence.
- **Version 1.2, Nov. 10, 2026.** What changed: matter classification; confidence; outside record; version labels.
- **Version 1.1, Nov. 9, 2026.** High confidence now requires the reviewed outside-record sweep (SEC name hits and primary-domain web search leads); otherwise the cap is Medium.
- **Version 1.0, Nov. 9, 2026.** First version: five factors, 20-firm trial.

This page is the written rubric the rating program follows, with file names and internal codes put in words. Individuals are never rated, pending matters and allegations are shown but never scored, and missing data is never counted as good or bad. Information, not investment advice.
