Voting & Representation

seats · None. A charter constant rather than an estimate. Two independent official pages state it identically, and the count report header states the same number as the number of candidates being elected.

years · None as published, but flagged for re-check. The same Commission page records that the City voted in 2025 to adopt a new charter, and does not state when that charter takes effect. The term length is published as current today; the effective-date question is carried as its own gap and is shared with the petitions module rather than restated here.

seats · None. Carried because the School Committee contest sits on the same ballot and has a materially different invalid-ballot rate, so omitting it would let a reader generalise the Council's rate to the whole ballot.

counting system · None. Transcribed from the header the Commission prints on the certified report itself, which is the same block that governs the count the report describes.

votes · None. Integer arithmetic on an integer ballot count. The reproduction is shown as a check on our reading of the source, not as the origin of the number.

votes · None. Carried because the page publishes the Council quota and the rule that produces it, and a reader who applies that rule to the School Committee should be able to check the answer here rather than compute it themselves.

counts · None. The report header names the count, the navigation strip on the summary report lists every count from the first to the final one, and no report exists beyond the final one.

registered voters · No sampling error: this is an administrative count, not an estimate. It is point-in-time, reflects the list as it stood for this election, includes inactive voters, and is not a measure of population.

voters · No sampling error. This counts voters recorded at check-in, which is not the same event as a ballot processed by the count, and the two are deliberately kept in different unit classes so they cannot be added or silently substituted.

ballots · No sampling error. Specific to the City Council contest: the School Committee contest on the same ballot processed a different number, and the charter question recorded a third.

ballots · No sampling error. Named explicitly because a reader who divides turnout by this instead of by registered voters gets a different number that looks equally correct.

ballots · No sampling error. The reason this contest records so few invalid ballots relative to the other contest on the same ballot is not stated by any source held here and is not inferred.

ballots · No sampling error. Carried as its own entry because three figures on this page are derived from it, and a number that other numbers depend on must be traceable to a printed line of its own.

ballots · No sampling error. Given its own entry so that the School Committee validity rate has a sourced numerator rather than a number quoted from another measure's prose.

ballots · No sampling error. This is the clearest published evidence that processed and valid are different quantities, which is why it is carried rather than left to the reader to notice.

percent of ballots processed · Rounding only, because both inputs are exact administrative counts. Contest-specific: the same computation on the other contest gives a materially different answer.

percent of ballots processed · Rounding only. Published beside the Council rate because the contrast between the two is the finding, and neither number means anything without the other in view.

percent of ballots processed · Rounding only. The gap between this share and the equivalent share in the other contest is published without explanation, because no source held here offers one.

percent of ballots processed · Rounding only. The size of this share relative to the Council contest on the same ballot is the single most striking published contrast in this tracker, and it is left unexplained on purpose.

ballot question · None. The wording is its own measure rather than a row of the tally measure, because those rows are checked by adding them to the published total, and a row that cannot be added would either break that check or have to be excused from it. Quoting the question separately keeps the check exact and the quotation intact.

ballots · No sampling error. This is the fourth different total on one ballot and it is carried so the tracker can demonstrate its central claim rather than assert it.

percent of registered voters · Rounding only, because both inputs are administrative counts with no sampling error. The Commission prints the same quotient to the whole percent; both are shown, so the difference is visibly a rounding choice rather than a different measurement.

seats · None beyond the transcription itself, and the transcription is checked against the next-placed first-choice total, which falls below quota, so the boundary between elected and continuing is unambiguous.

seats · None beyond transcription. Worth reading with the note that every councillor finishes on exactly the quota, because everything above quota is transferred away, so a final total is not a measure of support.

seats · None beyond transcription. The per-count transfer amounts are shown as published, and no claim is made about which earlier preference produced them beyond what the report itself records.

count number · None beyond transcription. Every councillor finishes on exactly the quota, so this column records when a seat was settled and says nothing about the size of anyone's support.

first-place votes · None beyond transcription. Two of these figures sit above the quota and the rest below it, which is the whole of what the first count decided.

ballots · No sampling error. Carried because the nine councillors' final totals and this pile together account for every valid ballot, which is the check that the published count balances.

count number · None beyond transcription. All six members finish on exactly the quota, so this column records when a seat was settled and says nothing about the size of anyone's support.

precincts · None. Worth stating because it is a cross-publisher agreement: the City's GIS layer defines the same number of precincts as the Election Commission's turnout table publishes rows for.

registered voters · No sampling error. The published rows add exactly to the printed citywide registration total, which is the check that the transcription is complete and correct.

voters · No sampling error. The published rows add exactly to the printed citywide turnout total, which is the check that the transcription is complete and correct.

percent of registered voters · Rounding only. A small denominator moves a percentage a long way, so the smallest precincts move furthest on the fewest ballots, and none of these rows is a ranking.

percent of registered voters, and the multiple between the two rates · Rounding only, but the selection rule matters more than the rounding: choosing the endpoints from the City's whole-percent column picks the wrong precinct, because two rows print the same whole percent and the higher one is not the one that column suggests. This is explicitly a spread and not a ranking, and no precinct between the endpoints is ordered or characterised.

ballots · The difference itself is exact. Its cause is not stated by any source held here and is not inferred, and no explanation appears anywhere on this page.

ballots · The difference is exact and its cause is not stated by any source held here. Both totals are printed by the same office on the same day for the same election.

ballots · The difference is exact and unexplained. Together with the other two residuals it is the evidence for this tracker's central claim, which is why all three are shown rather than summarised.

published result sets · This is our own observation of the published record on one date, not a figure the Commission states, and it is badged accordingly. A location that stops answering later would change it, which is why the method names the date rather than treating the result as permanent.

published statement · None as to the wording, because it is transcribed rather than summarised. The Commission does not state which figures moved or by how much, and no such detail is inferred here from the difference between the published sets.

report timestamp · None. Carried because a result set is identified by when it was produced, and every figure on this page should be attributable to one identified run of the count rather than to a page that may change.