The twelve closures, graded by evidence

What this note measures

This is a data note, not a second comparison. It records how the twelve closures in the statute-versus-enforcement dossier are evidenced: what kind of instrument exists, whether a date is recorded, how many of the twenty documented lists still name the state and how confident a reader can be in the record.

The unit is a state closure. The sample is the ten statutory closures and the two enforcement closures in the site's ledger. The list data comes from the twenty documented state lists and their read dates. The grades are a summary of those three inputs.

The note does not grade the state's policy, the strength of the law or the seriousness of the regulator. A low grade means the public record is thin, not that the closure is weak. A high grade means the record is easy to verify, not that the closure is permanent.

The note is also a replication sheet. A reader who wants to check a row can follow the source links, read the instrument and compare the list count. The table is a map to the evidence, not a substitute for it.

The twelve closures

The table has six columns. The first is the state. The second is the mechanism: statute or enforcement. The third is the instrument. The fourth is the date or notice record. The fifth is the list lag: how many of twenty documented lists still name the state. The sixth is the evidence grade.

The grades are deliberately coarse. A means a public statute text with a clear instrument and date. B means a public enforcement summary or order with a dated record but no full text. C means a single-source or incomplete record. No row in this sample is C, but the scale is part of the method.

The lag column is a snapshot. The read date is the list date, not the publication date. A row can change after this note is published. The grade does not change with the lag; it describes the instrument, not the commercial response.

The table is sorted by mechanism and then by the order used in the dossier: the ten statutes first, then the two enforcement records. The order is not a ranking of strength.

Instrument quality

The ten statutes all have a public instrument. A reader can open the bill or chapter, read the definition and find the penalty. That is why they receive an A on instrument quality, even when the definition is broad or the penalty is civil.

The two enforcement records have a public summary or order, but not always the full letter. Illinois publishes a page that describes the campaign, the legal boundary and the demand to block residents; it does not publish a dated recipient list in the text reviewed here. Michigan has orders, but the public record is thinner than a statute.

The grade is about the instrument, not the legal theory. An enforcement record can be legally strong and evidentially thin. A statute can be legally narrow and evidentially clear. The two dimensions are separate.

A reader who wants to verify a statute can quote the text. A reader who wants to verify an enforcement record has to use the regulator's summary, the reporting and the operator's list. The second path has more gaps, which is why the grade is lower.

Date quality

The ten statutes have a date. Some are effective dates; some are signature dates with immediate effect. The table uses the date the record gives, not a normalized date. That is why New York's row says 8 December 2025 and Tennessee's says signed 22 May 2026.

The enforcement rows do not have an effective date. Illinois has a campaign described as May 2026; Michigan has orders in February and July 2025. The table records the notice window and says “no effective date.” It does not invent one.

The date grade is part of the overall grade. A statute with a clear effective date is easier to verify than an enforcement campaign with a month but no day. A reader who needs a planning date should use the statute date; an enforcement reader should use the letter or the list.

The absence of a date is not a data error. It is a property of the instrument. The table preserves the absence instead of filling it.

List quality

The list lag is the most volatile column. It depends on the twenty documented lists and the date they were read. A list update can change a row without changing the law. A brand can leave a state before a statute takes effect or after an enforcement letter.

The highest lag is Illinois at 19 of 20. The lowest is Michigan at 0 of 20. The statutory range runs from 1 of 20 in New York to 13 of 20 in Tennessee. The spread is the finding: the mechanism does not determine the commercial response.

The list count is not a quality score for the state. A high lag can mean the list is old, the operator is waiting for a court decision or the brand has a different commercial position. The table records the count; the coverage matrix records the per-brand evidence.

The reader should treat the lag as a date-stamped observation. It is useful for the brand question, not the legal question. The legal question is answered by the instrument.

The confidence grades

A is a public statute text with a clear instrument and date. The reader can verify the definition, the penalty and the effective date. The grade does not mean the statute is clear in every application; it means the record is public.

B is a public enforcement summary or order with a dated record but no full text. The reader can verify the regulator, the demand and the list, but not every letter or recipient. The grade is lower because the record has more gaps, not because the closure is less real.

C would be a single-source or incomplete record. No row in this sample is C. If a future row has only a news report and no regulator page, instrument or list, it should receive a C and be treated as provisional.

The grade is a communication tool. It tells the reader how much of the record they can check. It is not a legal opinion and not a prediction.

What the data cannot say

The data cannot say whether a statute will be enforced, whether an enforcement letter will survive a court challenge or whether a list will update next week. It records what exists at the read date.

The data cannot compare the severity of the laws. A felony and a civil violation are different, but the table does not rank them. The penalty is described in the dossier, not scored in this note.

The data cannot prove that the twenty lists are representative. They are the lists the site could read, not a random sample. The five brands without a readable list are named on the coverage page. The sample bias is part of the limitation.

The data cannot resolve a reader's individual account. A state closure, a list entry and a settlement are not a redemption decision. The terms and the account record are the final layer.

Instrument quality in detail

The ten statutes have a public text. That is the strongest instrument evidence in the sample. A reader can open the chapter, read the definition and check the penalty. The grade is A because the record is public, not because the law is easy to apply.

The two enforcement records have a public summary or order but not the full letter. The Illinois page describes the campaign and the demand; it does not publish a dated recipient list in the text reviewed here. Michigan has orders, but the public record is thinner than a statute.

The difference matters for replication. A reader can quote a statute; an enforcement reader has to cite the regulator summary, the reporting and the operator list. The second path is not weaker, but it has more gaps.

The grade does not measure the regulator authority. An enforcement action can be legally strong and evidentially thin. A statute can be legally narrow and evidentially clear. The note records the evidence, not the merits.

Date quality in detail

The statutes have dates. Some are effective dates, some are signature dates with immediate effect. The table uses the date the record gives, without normalizing it. That is why New York says 8 December 2025 and Tennessee says signed 22 May 2026.

The enforcement rows do not have an effective date. Illinois has a campaign described as May 2026; Michigan has orders in February and July 2025. The table records the notice window and says no effective date. It does not invent one.

A date is a planning tool. A statute gives a reader a date to check against a list. An enforcement record gives a reader a period to investigate. The absence of a date is a property of the instrument, not a data error.

The grade is lower for the enforcement rows because the date is less precise. A reader who needs a legal date should use a statute; a reader who needs a commercial date should use the list.

List quality in detail

The list lag is the most volatile column. It depends on the twenty documented lists and their read dates. A brand can update a list without changing the law. A list can stay stale for months after a statute.

The highest lag is Illinois at 19 of 20. The lowest is Michigan at 0 of 20. The statutory range is 1 of 20 in New York to 13 of 20 in Tennessee. The spread is the finding: the mechanism does not determine the commercial response.

The count is not a quality score for the state. A high lag can mean the list is old, the operator is waiting for a court decision or the brand has a different commercial position. The table records the count; the coverage matrix records the evidence.

A reader should treat the lag as a date-stamped observation. It is useful for the brand question, not the legal question. The legal question is answered by the instrument.

The grades in detail

A is a public statute text with a clear instrument and date. The reader can verify the definition, the penalty and the effective date. The grade does not mean the statute is clear in every application; it means the record is public.

B is a public enforcement summary or order with a dated record but no full text. The reader can verify the regulator, the demand and the list, but not every letter or recipient. The grade is lower because the record has more gaps, not because the closure is less real.

C is a single-source or incomplete record. No row in this sample is C. If a future row has only a news report and no regulator page, instrument or list, it should receive a C and be treated as provisional.

The grade is a communication tool. It tells the reader how much of the record they can check. It is not a legal opinion and not a prediction.

The twenty-list sample

The denominator is twenty documented state lists. It is not every brand in the market. The site could read twenty lists at the check date; five brands without a readable list are named on the coverage page. The sample is large enough to compare states and small enough to audit.

A list is a commercial document. It can be narrower or wider than the law. A brand can exclude an open state for payment reasons; a brand can name a closed state because the list is stale. The lag measures the document, not the brand intent.

The sample is also dated. A list read on 4 November is evidence about 4 November. A reader who uses the table after a list update should recheck the matrix. The grade describes the instrument; the lag describes the document at the read date.

The table does not weight the brands by size, traffic or market share. A small brand and a large brand count the same. That is a limitation of a simple count, and it is why the matrix shows the per-brand cells.

The grade caveats

A grade is not a legal conclusion. An A does not mean the statute is constitutional, enforceable or clear in every application. It means the text and the date are public enough to verify.

A B does not mean the enforcement action is weak. It means the public record is thinner than a statute. The regulator may have a strong legal theory and a private letter that the public cannot read.

A C would mean the record depends on a single source or an incomplete document. The grade is a warning to the reader, not a judgment about the state.

The grades are coarse on purpose. A finer scale would imply precision that the sample does not support. The note is a map, not a measurement of legal quality.

What would change a grade

A statute could move from A to B if the text becomes unavailable or the effective date is amended without a public record. An enforcement record could move from B to A if the regulator publishes the letters, the recipient list and the dates.

A list update does not change the grade. The grade is about the instrument; the lag is about the list. A state can have an A instrument and a high lag, or a B instrument and a low lag.

A court decision could change the grade if it invalidates the instrument or narrows its scope. A settlement could change the record but not the grade unless it produces a public document.

A reader should update the grade when the record changes, not when the headline changes. The note is a living table, but the method is stable.

How to read the table

Start at the state you care about. Read the mechanism and the instrument. If the mechanism is statute, open the bill or chapter; if it is enforcement, open the regulator page and the list.

Read the date column as a record, not a promise. “No effective date” is a finding. It tells you the instrument does not have the same planning clock as a statute.

Read the lag as a snapshot. A high number means the list is behind; a low number means the list is current. Neither number tells you whether the brand will accept your account.

Read the grade last. It summarizes how much of the record you can check. If the grade is B, expect to rely on the regulator summary and the list; if it is A, expect a public text.

The replication log

A replication log should record the date of the check, the state, the instrument, the list read date and the grade. The log is more useful than a screenshot because it shows the sequence.

The log should also record what was missing: a private letter, a list date, a penalty section or a court order. The gap is part of the record, not a failure.

The site's ban tracker and coverage matrix are the two starting points. The state pages and the instruments are the next layer. The log can be as short as five lines per state.

The point is not to produce a perfect dataset. It is to make the comparison reproducible. A reader who can repeat the check can challenge the grade, the lag or the conclusion.

The two enforcement records in detail

Illinois and Michigan are not one model. Illinois is a broad campaign with more than 60 letters, a public page and a 19-of-20 list lag. Michigan is a narrower set of orders with a zero lag. The difference is the operators' response, not the label.

The Illinois record has the stronger public narrative and the weaker public list. The Board's page describes the campaign and the legal boundary; the list data shows that most operators have not updated. The grade is B because the instrument is public but the letters are not.

The Michigan record has a lower public profile and a cleaner commercial response. The orders are dated; every documented list excludes the state. The grade is B for the same reason: an enforcement instrument is not a statute.

A reader should compare the two rows, not average them. The enforcement category contains two different speeds and two different evidence profiles.

The statutory clusters

The ten statutes cluster by drafting choice. Montana and Oklahoma use broad or payment-system definitions with felony penalties. Connecticut, California and New York name the product or the dual-currency structure. New Jersey protects a licensed market. Indiana, Louisiana, Maine and Tennessee add state-specific variations.

The clusters do not determine the lag. New York has the smallest lag; Tennessee has the largest. The difference is the operator response and the age of the list, not the drafting family.

The clusters do help a reader predict the evidence. A product-specific statute gives a reader a definition to quote; a broad statute gives a reader a broader theory; a licensed-market statute gives a reader a market rationale.

The table keeps the rows separate so the reader can see the cluster without losing the state.

The bottom line

The comparison is a record of twelve closures, not a ranking. The ten statutes have the stronger public text; the two enforcement records have the thinner public record. The list lag cuts across both.

A reader who wants a legal answer should use the instrument. A reader who wants a commercial answer should use the list. A reader who wants an account answer should use the terms.

The data note is useful only if it makes those distinctions easier to see. If it collapses them into a grade, it has failed.

The table limits

The table is a snapshot. It cannot show a list update that happened after the read date, a letter that was sent after the campaign summary or a court decision that changed the instrument. The date on each record is part of the limit.

The table is also a sample of twenty lists. It does not cover every brand or every state. A brand outside the sample can have a different list. A state outside the ledger can have a different instrument.

The table is not a legal opinion. The grade is about evidence, not about whether the state will win a case or whether an operator will be prosecuted.

A reader should use the table to ask a better question, not to close the research.

The final check

Before relying on a row, check the state page and the instrument. If the row is a statute, confirm the chapter and the effective date. If the row is enforcement, confirm the regulator and the demand.

Then check the list and the read date. If the list still names the state, the lag is current. If the list has changed, the row is stale.

Then check the account. A state record cannot tell you what the cashier will do. The terms and the support record can.

The data note is finished when the reader can repeat those checks. The grade is only a summary of how easy that is.

The practical use

A reader can use the table in three steps. First, find the state and the mechanism. Second, open the instrument and the list. Third, decide whether the answer is legal, commercial or account-specific. The table is a routing tool.

It is also a way to spot a weak claim. A claim that cites no instrument, no date and no list is not ready. A claim that cites a statute but no effective date is incomplete. A claim that cites a list but no read date is stale.

The point is not to memorise twelve rows. It is to learn the questions that make a row checkable. Once the questions are clear, the table is optional.

The closing note

The value of this data note is not the twelve rows. It is the habit of checking the instrument, the date, the list and the grade before repeating a claim. The habit survives when the rows change.

A reader who can say “this is a statute, this is the effective date, this is the list read date and this is the gap” is already ahead of a reader who can only say “the state is closed.”

The method in one line

Open the instrument; date the record; read the list; grade the evidence. That is the method of this note, and it is the same method a reader can use on the next closure, even before a table exists.

A last caution

Do not read the grade as a prediction. A state can close with an A instrument and a high list lag; a state can close with a B instrument and a zero lag. The record is the evidence, not the outcome, and the next read may change it. The table is a snapshot, not a verdict, and the reader should treat it that way when the next list is read tomorrow.

How to replicate a row

Start with the state page and the ban tracker. Confirm the mechanism, the instrument and the date. Then open the instrument or the regulator page. Then open the coverage matrix and find the state's cells and the read dates. The row should match those three records.

If the row disagrees with the state page, use the state page and the instrument. The table is a summary; the page is the record. If the row disagrees with the list, check the read date before assuming an error.

If a source is missing, mark the row as provisional. The grade should reflect the gap, not the confidence of the headline. The point of the note is to make the gaps visible.