Methodology

How we compute the statistics on this site

Methodology refinement — May 14, 2026. Our dataset is now strictly limited to criminal cases filed January 1 – December 31, 2025. Prior to this date, our public-facing aggregates included some cases visible in Virginia's Case Information System (CIS) that were filed in earlier years but still appeared as active or recently closed at the time of extraction. We now enforce the calendar-year-2025 filter at the data layer, with automated tests that fail if any non-2025 record enters the dataset. Research reports published before May 14 have been refreshed against the corrected dataset; substantive findings are unchanged, specific counts and rates have been updated. The change applies retroactively to every page on this site.

Data source

All statistics on VirginiaCourtFile are derived from public case records maintained by Virginia's General District Courts and Circuit Courts. Under Va. Code §17.1-208(B), circuit court records are open to public inspection. General District Court records are similarly publicly accessible.

The current dataset covers 1,681,767 criminal case records filed during calendar year 2025, across 125 jurisdictions in both court systems.

Unit of analysis

Each row in the dataset is a charge, not a defendant or an arrest. Virginia's court system assigns a separate case number to each charge filed. A defendant arrested for DUI and reckless driving produces two rows — two case numbers, two independent dispositions.

In the 2025 dataset, 97.9% of case numbers appear only once. Multi-charge cases do not materially affect per-charge rates. When we say "cases," we mean charges unless explicitly stated otherwise. This is consistent with how Virginia's courts report case data.

Definitions

Filed. A case with a filing date in the period under analysis.

Resolved. A case that reached a final disposition: dismissed, nolle prosequi, acquitted, guilty plea, or found guilty. Cases that are pending, certified to circuit court, on fugitive status, or in other non-terminal statuses are excluded from the resolved count. All outcome rates on this site use resolved cases as the denominator.

Nolle prosequi. The Commonwealth's Attorney formally declined to proceed with the charge. The charge is dropped by the prosecutor. This is distinct from a judicial dismissal.

Judicial dismissal. The charge was dismissed by the court — on the defendant's motion, at trial, or by the judge's order. This is distinct from nolle prosequi.

Acquittal. The defendant was found not guilty at trial or hearing.

Dismissal rate. (Nolle prosequi + judicial dismissal + acquittal) ÷ resolved. Where these components are reported separately (as in our research reports), we label each one. Where a single "dismissal rate" is shown, it combines all three non-conviction outcomes.

Headline dismissal rate on charge × jurisdiction pages. On a page that argues a single jurisdiction's record for a single charge, we surface the General District Court dismissal rate as the page's headline number when district data exists. Most criminal cases in Virginia start in (and resolve at) General District Court; circuit-court cases involve different procedural posture (felony indictments, jury trials, appeals from district) and are shown separately on the same page. For charges heard only at Circuit Court, the circuit rate is the headline. Combined district + circuit aggregates remain available in the underlying dataset for researchers who need them.

Conviction rate. (Guilty plea + found guilty) ÷ resolved. Includes convictions on amended (lesser) charges.

Charge reduction rate. Cases where the original charge was amended to a lesser offense and the defendant was convicted of that lesser charge ÷ resolved. Cases where the charge was amended but later dismissed are counted as dismissals, not reductions.

Case duration. Days from filing date to disposition date. We report the median (not the mean) and the P25–P75 interquartile range. Cases with disposition dates more than two years after filing or with scheduled future disposition dates are excluded from duration calculations.

Plea rate. Guilty pleas ÷ total convictions (guilty plea + found guilty). Measures what proportion of convictions were negotiated pleas vs. findings of guilt at trial.

Denominators

All outcome rates (dismissal, conviction, acquittal, reduction, plea) use resolved cases as the denominator. Case counts and filing volumes use filed cases. We do not mix denominators within a single statistic. When two numbers in the same surface use different denominators, both are labeled.

Charge categorization

Cases are classified into 26 named charge categories based on the charge description text in the court record (e.g., "DUI / DWI," "Assault & Battery," "Reckless Driving"). The categorization is rule-based: charge text is matched against patterns for each category.

Cases that do not match any named category are classified as "Other." In the 2025 dataset, 641,750 cases fall into the "Other" category across all case types. This includes charges that are substantively criminal — assault on law enforcement, malicious wounding, drug manufacturing/distribution variants, and others — but that our current taxonomy does not yet classify into named categories. The "Other" category also includes probation violations, toll violations, and registration-related offenses classified as misdemeanors or felonies.

Traffic infractions (728,229 cases) and failure-to-appear warrants (32,838 cases) are excluded from criminal charge analysis.

Coverage

The dataset covers Virginia's 125 General District Courts and Circuit Courts in 30 of Virginia's 31 judicial circuits.

Not included: Fairfax County Circuit Court (19th Judicial Circuit) operates an independent case management system not accessible through the statewide public case information system. Circuit Court cases for Fairfax County, Fairfax City, and Falls Church are not represented. General District Court cases for those three jurisdictions are included.

Independent cities that share a judicial circuit with a surrounding county under Va. Code §17.1-506 appear under the parent circuit's records. This is how the Code organizes those circuits, not a coverage gap.

Minimum-case thresholds

Individual charge × jurisdiction pages require at least 20 resolved cases. Individual attorney profiles require at least 20 cases of record in 2025. Individual officer profiles require at least 20 charging incidents in 2025. Research report tables use higher thresholds (50 or 100 resolved cases, as noted per table). Pages below the threshold return a 410 Gone response with an explanatory note rather than a stat-bearing page. Attorney and officer pages with case counts between 20 and 29 display an in-page small-sample caveat to flag the higher statistical uncertainty of low-volume rates.

VCF Sentencing Outcome Index

The VCF Sentencing Outcome Index describes what happened to convicted defendants on each charge × jurisdiction page. It uses cases filed on or after January 1, 2025 (calendar year 2025 onward).

Denominator (conviction). A case enters the denominator only when its final disposition reflects a conviction. For General District Court, that means a FinalDisposition of Guilty or Guilty In Absentia. For Circuit Court, it means a DispositionCode of Guilty. We exclude probation- and sentence-revocation hearings (already-convicted defendants returning to court), prepaid traffic infractions, and dismissed/nolle cases.

Active jail versus suspended sentence. Virginia courts impose a sentence and then often suspend most or all of it. We separate the two. Received Active Jail means the case had at least one day of active jail time after suspension — computed as SentenceTime − SentenceSuspendedTime. Median when imposed is the median of active jail days across only those cases where active jail was imposed. Cases with a fully suspended sentence are reported as not having received active jail.

Median fine. Median of Fine across convicted cases with a non-zero fine. The fine rate is the percentage of convictions where any fine was imposed.

License suspension. Median of OperatorLicenseSuspensionTime across convicted cases with a non-zero license suspension. The license suspension rate is the percentage of convictions where the court imposed a license suspension. License suspension is most commonly attached to driving offenses (DUI/DWI, reckless driving, driving on suspended) and applies only when statutory.

Probation. Median of ProbationTime across convicted cases with a non-zero probation order. The probation rate is the percentage of convictions placed on probation.

VASAP enrollment (DUI/DWI charges only). The percentage of convictions where the VASAP field indicates enrollment in the Virginia Alcohol Safety Action Program. Surfaced only on DUI/DWI charge pages.

Suppression. Per-metric suppression at fewer than 30 cases. If a page has fewer than 30 convictions, the entire Index is suppressed. If a metric subset (e.g., active jail) has fewer than 30 cases in its numerator, the median for that metric is suppressed (the rate is still shown when the overall conviction count is ≥ 30).

What it does not tell you. The Index reports outcomes of cases that ended in conviction. It does not include dismissed, dropped, or acquitted cases, and it does not predict any individual case outcome. Sentences reflect the imposed sentence including any suspended portion; only the active jail component reflects time the defendant was ordered to serve. Probation conditions, restitution, community service, and other terms are not surfaced. The Index is descriptive of the historical record, not predictive of future cases.

Defense attorney pages

Per-attorney pages at /attorneys/{name} aggregate cases by the DefenseAttorney field as recorded in DCIS (district) and CIS (circuit). They are a publication-of-record surface, not a directory: attorneys cannot edit or claim their pages, and there is no opt-out. Factual corrections are accepted via the “Report a correction” link on every page.

  • Threshold: 30 cases of record per attorney to publish a page. Cases like yours rows require 3+ cases per charge × jurisdiction combination.
  • Generic entries excluded: office-level labels (PUBLIC DEFENDER, PD, PUBLIC DEFENDERS OFFICE, etc.), waived-counsel rows, and pro-se entries are dropped before aggregation.
  • Name aliasing: initial-form variants are merged when unambiguous (e.g., PAULDING, K → PAULDING, KRISTIN when only one full-name variant starts with K). A small manual override map handles spelling variants verified to refer to the same person.
  • Outcome definitions: the same Filed / Resolved / Convicted / Dropped / Reduced / Acquitted definitions used elsewhere on the site — convicted requires a guilty disposition on the originally filed charge with no amended charge; reduced requires a guilty disposition with a non-empty amended charge.
  • Plea vs trial: derived from outcome category — guilty_plea = pled guilty; found_guilty = trial conviction; acquitted = trial acquittal.
  • Coverage: attorney pages reflect cases where the defense counsel of record was entered in the court system. Retained matters that resolved before formal counsel entry, federal cases, and the 19th Circuit (Fairfax County / Fairfax City / Falls Church) Circuit Court records are not included.
  • Refresh: attorney aggregates are rebuilt quarterly from the underlying case database.

Demographic data

Race and gender are recorded in the court system for nearly all cases (99.8% have a race field; all have a gender field). However, the ethnicity field is unreliable: only 10 of 278,950 criminal cases in 2025 are recorded as "Hispanic or Latino," in a state where approximately 10% of the population is Hispanic (U.S. Census Bureau, 2024 American Community Survey). Hispanic defendants are almost certainly recorded under other racial categories. Any analysis that compares racial proportions in the court data to Census population figures is unreliable due to this gap.

What we publish, and what we don't

Everything on this site is an aggregate statistic. We do not publish individual case pages or per-defendant detail. We do not publish:

We do publish: jurisdiction-level case counts, dismissal/conviction/reduction rates by charge category, median case durations, and statistics for law enforcement officers acting in their official capacity, with minimum-case thresholds applied.

What this data does and does not tell you

What this data shows:

  • How often charges of a given type are dismissed, convicted, or reduced in a given court
  • How a specific officer's case-outcome distribution compares to the local baseline
  • How long cases of a given type typically take from filing to disposition
  • What lesser charges most often appear when reductions happen
  • How rates vary across jurisdictions, charge types, and court levels

What this data does not show:

  • Why a judge ruled a particular way in a particular case
  • The reasoning behind a prosecutor's decision to nolle prosequi, reduce, or pursue a charge
  • The plea negotiation dynamics that produced a particular outcome
  • The strength or weakness of the evidence in any individual case
  • Whether a specific defendant's case will resolve the way aggregate patterns suggest
  • Prior criminal history, income, or bail status of defendants

The data captures what happened. It does not capture why. Aggregate patterns are useful for understanding the structural behavior of a court, a jurisdiction, or a charge type over time. They are not predictions of individual case outcomes.

Limitations

  • Statistics reflect historical patterns and do not predict any individual outcome
  • Data may include cases under appeal or with pending post-trial motions
  • Officer name matching is approximate within a jurisdiction (name variants are aggregated)
  • General District Court records do not include a judge field for all cases
  • Fairfax County Circuit Court (19th Judicial Circuit) is not included — see Coverage above
  • The "Other" charge category (641,750 cases) contains unclassified charges including substantively criminal matters — see Charge categorization above
  • The ethnicity field is unreliable — see Demographic data above
  • This is the first year of this dataset; year-over-year comparisons will be possible when 2026 data is complete

Corrections and removal requests

If a record is inaccurate, has been expunged or sealed by court order, or you are a public official who believes a statistic reflects a misattribution, contact richard@virginiacourtfile.com. We act on documented requests and update the dataset on the next refresh cycle.

Last updated

Dataset: May 2, 2026. Methodology page last revised: May 2026.