Hurt on a flight? Start a free case evaluation
RESQ.com. When your future is at stake. (800) 300-0001 Free case evaluation Call Now
Plane Crashes by Model: How to Compare Counts, Rates, and Accident Records

Home / Blog / plane crashes by model

Plane Crashes by Model: How to Compare Counts, Rates, and Accident Records

Published by The Ledger Law Firm · · general information, not legal advice

Were you in a plane crash, or did you lose someone in one? See the numbers and what to do next.

What aircraft-model crash statistics can—and cannot—show

Aircraft-model crash statistics can describe patterns in a defined group, but they cannot answer every question about safety or a particular accident. The first question is what you are counting: an FAA-reportable accident, an incident, a fatal accident, or another category. The next questions are where the records come from, which aircraft are included, what dates they cover, and what exposure measure makes the comparison fair.

A raw total is not enough to compare models without a defined scope. It does not, by itself, show how often those accidents occurred relative to departures, how many aircraft or flights were exposed, or why a particular event happened. A rate can make groups with different numbers of departures more comparable, but it still describes a population rather than deciding cause or responsibility in one event.

The most useful comparison keeps three measures separate:

Measure What it can tell you What it cannot tell you
Raw accident count How many defined accident records fall within the chosen model, period, geography, and operation group How frequently the group experienced an accident per departure or what caused any event
Accidents per million departures How the accident total relates to a stated number of departures or flight cycles Whether an individual aircraft, operator, crew member, or manufacturer was responsible for a specific event
Case-specific record What the documented event involved, including its classification and known investigative facts Whether a broad model-level pattern establishes a legal claim or a general ranking

The FAA commercial-jet summary defines accident rates as accidents per million departures and explains why departures, or flight cycles, are used for that calculation. That measure is not interchangeable with a raw count. If you connect a model statistic to an actual injury or accident, treat the statistic as context and turn to the specific record for the event itself.

Define the event before comparing models

An aircraft accident has a specific federal reporting meaning. Under the FAA reporting definition, it is an occurrence associated with aircraft operation that takes place from the time a person boards with the intention of flight until everyone has disembarked, and in which someone suffers death or serious injury or the aircraft receives substantial damage. That definition is narrower and more precise than every everyday use of the word "crash."

An incident is different. An incident is a safety-related occurrence associated with aircraft operation that does not meet the definition of an accident but affects or could affect operational safety. The NTSB online aviation database classifies investigations as either incidents or accidents. The NTSB also investigates only selected incidents and records those selected incidents alongside accidents, while noting that incidents typically do not involve the level of injury or damage characteristic of an accident.

Those distinctions change the denominator of any model comparison. If one count includes incidents and another includes only accidents, the figures do not describe the same event population. A list that combines substantial-damage accidents, near misses, minor operational occurrences, and uninvestigated reports may look comprehensive while answering no precise question.

Use a short definition statement before you read or build a model comparison:

  • Event category: State whether the count includes accidents, incidents, fatal accidents, hull losses, or another defined category.
  • Operational window: State whether the record begins at boarding and ends at disembarkation when you are applying the federal accident definition.
  • Record source: Identify the agency or database that classified the event.
  • Inclusion rule: State whether selected incidents appear alongside accidents and whether the comparison removes them.

A fatal accident is still an accident; the FAA commercial-jet summary defines it as an accident that results in fatal injury. If you want to study fatal events, label that category instead of using "crashes" as a substitute. A reader who sees a model with more records should be able to tell whether the difference comes from the type of event counted rather than from a difference in aircraft performance.

Set the aircraft, time, geography, and operation scope

A model comparison is meaningful only after you define the population on both sides of the comparison. Start with the aircraft category: commercial jet, regional aircraft, turboprop, helicopter, private aircraft, or another group. Then decide whether a model name includes every variant or only a particular series. A broad label can combine aircraft used in different operations and periods, so record the exact category you intend to compare.

The time period must be identical for every model. A count covering several decades cannot be placed beside a count covering a few recent years and treated as a current safety measure. State the beginning and ending dates, then identify whether the data represents historical records, a particular edition of a statistical summary, or a database search conducted on a stated date.

Geography and operation also need separate labels. A United States-focused comparison may use records involving U.S. locations, U.S. aircraft, or both, but those are not the same scope. The FAA accident-reporting rule covers U.S. civil aircraft wherever they operate, specified public aircraft wherever they operate, and foreign civil aircraft when an event occurs in the United States, its territories, or its possessions. The NTSB database uses location codes for U.S. states, some bodies of water, U.S. territories and possessions, foreign countries, other foreign countries, and unknown locations. It includes only a limited number of foreign accidents.

Before comparing two model groups, write down these boundaries:

  1. Aircraft population: Which make, model, series, or variant qualifies?
  2. Event category: Are you counting accidents, incidents, fatal accidents, or another defined record type?
  3. Dates: What is the first date and last date in the comparison?
  4. Geography: Does the comparison cover U.S. locations, U.S.-registered aircraft worldwide, or a broader set of records?
  5. Operation: Are the groups commercial passenger operations, cargo operations, private flights, public operations, or mixed activity?
  6. Record treatment: Are selected incidents included with accidents, and how are unknown locations handled?

A scope statement prevents a category error before it reaches the conclusion. For example, a commercial-jet accident-rate summary should not be presented as a rate for every airplane, private aircraft, or aviation operation. A U.S. location search should not be described as a complete worldwide record.

Clear boundaries keep unlike aircraft records out of the same comparison.
Matching these boundaries keeps the comparison focused on the same population.

Compare raw crash counts without turning them into rankings

A raw model count answers a limited question: how many defined records were assigned to that model within the selected scope? That information can describe representation in a historical record. It cannot, by itself, establish that the model is inherently unsafe or safe, identify the model with the highest accident rate, or explain the cause of any individual event.

A larger total may reflect a larger fleet, more departures, a longer service period, broader geographic coverage, or a record system that captures more events. A smaller total may reflect fewer exposures or a narrower record set. Those possibilities do not prove a particular explanation; they show why the total must remain attached to its scope and denominator.

The FAA commercial-jet summary also specifies how it counts a collision involving two or more airplanes: the event is counted as a separate accident for each airplane. That rule matters when you interpret totals by aircraft model. A single collision can add more than one aircraft-level accident record, so an event count and a count of distinct occurrences are not automatically the same measure.

Use raw totals as a descriptive starting point rather than a league table. For each model, preserve the same fields:

  • number of defined accident records;
  • event category included in that number;
  • date range;
  • geographic and operational scope;
  • aircraft population and variant rule;
  • whether collisions are counted per airplane or per occurrence; and
  • the exposure measure, if a rate is also shown.

Then phrase the result narrowly. "This model has more recorded accidents in the defined period" describes representation. "This model is more dangerous" goes beyond a raw count. The second statement would require a comparable measure, an appropriate population, and analysis of the underlying events. Even a well-matched rate would not establish the cause, fault, or legal responsibility for a particular accident.

Use accidents per million departures for exposure-adjusted comparison

An exposure-adjusted comparison asks how many defined accidents occurred relative to how much flying the comparison group performed. The FAA commercial-jet statistical summary measures accident rates as accidents per million departures. In that context, a departure is a flight cycle, so the rate relates the accident count to the number of takeoffs or operating cycles rather than simply to the number of airplanes.

The FAA summary explains that departures are used because accidents have a stronger statistical correlation with departures than with flight hours, the number of airplanes in service, passenger miles, or freight miles. That choice does not make every denominator interchangeable. A flight-hour rate, an aircraft-in-service ratio, and a per-departure rate answer different questions and should not be placed in one ranking without clearly labeling the measure.

The rate calculation can be expressed plainly:

Accident rate per million departures = defined accident count ÷ defined departures × 1,000,000.

The formula is useful only when the numerator and denominator describe the same population. If the count covers one set of aircraft while the departure estimate covers another, the resulting number can create false precision. Match the model category, period, geography, operation, and event definition before comparing rates.

A careful rate comparison should show more than a single decimal. Include the accident count, the estimated departures, the stated period, and the source’s definition of the rate. If the source supplies separate fatal and nonfatal categories, keep them separate rather than blending them into one number. If you cannot establish that the exposure estimate matches the accident population, present the raw count with a clear limitation instead of presenting an unsupported rate.

Rates also need context when the groups are small. A modest change in the numerator can produce a large percentage change when the underlying number of events is low. That observation is a reason to show counts and scope alongside the rate, not a basis for declaring one model safer. The comparison should tell you what was measured, how it was normalized, and how much of the aviation population the result represents.

A departure-based rate makes exposure differences visible without replacing the underlying count.
A rate is meaningful only when its numerator, denominator, and scope describe the same population.

Recognize the limits hidden inside a model comparison

Fleet size and exposure can change the meaning of a total. A model with more aircraft in service or more departures has more opportunities to appear in an accident record, so its raw total may be larger without answering the per-departure question. The FAA summary specifically distinguishes departures from the number of airplanes in service, flight hours, passenger miles, and freight miles when it explains the basis for its accident-rate measure.

Use that distinction as a practical check. Put fleet size and service years in the descriptive context of the comparison, but do not use either as a substitute for the stated exposure denominator. Use departures or flight cycles when applying the FAA commercial-jet rate definition, and label any other denominator as a different measure rather than calling it the same rate.

Aircraft age, variant, and operation type also deserve separate fields. You can record them to avoid combining unlike groups, but do not infer a direction of effect from the field alone. A model comparison that mixes older and newer variants, cargo and passenger operations, or different operating environments may describe a mixed population rather than a single model characteristic.

Geographic coverage can create another hidden difference. The FAA reporting rule has one scope for U.S. civil aircraft operating anywhere, specified public aircraft operating anywhere, and foreign civil aircraft involved in an event in the United States, its territories, or its possessions. The NTSB database contains location codes for domestic and foreign records but includes only a limited number of foreign accidents. A U.S. database search and a worldwide manufacturer history therefore should not be described as equivalent populations.

Reporting definitions can change the count as well. The same event may be classified as an accident or an incident depending on the applicable definition, and selected incidents may appear in the NTSB database alongside accidents. Keep the classification rule visible when you compare records. Do not silently add categories because they share a model name.

A source edition is also a time-stamped measurement. The FAA summary states that departure data is continually updated and revised as new information and estimating processes become available. It also states that accident rates may vary between editions for that reason. If two editions show different rates, record the edition and data date before treating the difference as a change in aircraft safety. A historical rate should remain labeled as historical.

Use a limitations note beside every comparison, not several paragraphs later. The note should identify the event definition, aircraft population, period, geography, operation, denominator, and source edition. It should also say whether the data was revised or whether the record set includes selected incidents. That short note gives you the boundary needed to interpret the number without turning it into an absolute judgment.

The practical conclusion is modest: a model can appear more or less represented within a defined record set, and a rate can make exposure differences more visible. Neither measure should be read as a complete safety rating or as proof about the cause of a particular accident.

Check the model, variant, and operational record

The aircraft model name is only one part of an aviation safety record. A responsible comparison should preserve the exact aircraft identification used in the record and distinguish a broad model family from a particular series or variant. If the record supplies an engine field, keep that detail instead of assuming that every aircraft carrying the same model label had the same configuration.

FAA aviation safety data includes fields for the engine make and for the engine model or series installed at the time of an event. Those fields can help you describe what was recorded about the aircraft involved. They do not, by themselves, establish that the engine caused the event or that a particular configuration explains a model-level pattern.

The operational record can add another layer of context. FAA aviation safety data includes the pilot in command’s total flight time, time in the make-and-model aircraft involved in the event, and flight time during the last 90 days. Treat those as documented fields about the event record, not as standalone explanations. A comparison should preserve the distinction between a recorded fact and a conclusion about cause.

When you inspect a model-level record, separate these fields:

  • aircraft manufacturer and model;
  • variant, series, or type designation when recorded;
  • engine make and engine model or series;
  • event date and operating location;
  • operation type;
  • pilot-in-command total flight time;
  • pilot-in-command time in the make-and-model aircraft; and
  • pilot-in-command flight time during the last 90 days.

A database field tells you what information the system is designed to record. It does not make every field complete in every event, and it does not turn a field into a causal finding. Keep unknown or unavailable details marked as such rather than filling them with assumptions. That discipline matters when a broad model label is being used to interpret a particular accident.

Separate statistical safety context from cause and legal responsibility

A model statistic does not establish why a specific accident happened. It does not identify a design defect, maintenance failure, pilot error, operator decision, weather condition, air-traffic issue, or other cause unless the specific investigation and documented facts support that conclusion. It also does not determine who is legally responsible or what rights may follow from an injury.

The gap between a broad statistic and a specific event matters because an accident record has its own aircraft, operation, location, date, people, agencies, and investigative history. A model-level rate can provide background, but it cannot replace those event-specific facts. Keep the statistical question—how often defined records appear in a population—separate from the case question—what happened in this event and what legal rules apply.

At ResQ Legal, we believe experienced legal representation is crucial in aviation accidents because federal regulations, flight standards, and multiple parties can make these matters complex. That business guidance describes why a case may require focused review; it does not convert a general model comparison into a finding about your matter.

A raw count or rate is not a substitute for that factual record.

If you are trying to understand an accident connected to a model statistic, keep three questions separate:

  1. What does the defined population show? This is the statistical comparison.
  2. What is documented about the event? This is the accident or incident record and its investigative status.
  3. What legal questions remain? These depend on the specific facts, governing rules, and the parties involved.

Answer the first question with a carefully scoped comparison. Answer the second with the event record. Leave the third open until the documented facts and applicable law have been assessed.

A model statistic provides context, not a finding about cause or responsibility.
Separating these questions prevents a model statistic from becoming a finding about one accident.

What to gather before connecting statistics to a specific accident

Documentation is essential in aviation injury claims, and the useful starting point is a factual file for the specific event. Gather what you already have without treating a general model statistic as proof of what happened. Label each item with its date, source, and connection to the event so that a model comparison remains separate from the underlying record.

Use this checklist:

  • Aircraft identity: Record the manufacturer, model, series or variant, and any identifying information shown in the documents.
  • Operation: Note whether the flight involved passenger, cargo, private, public, training, medical, or another operation when that fact is known.
  • Time and place: Record the event date, location, route or operating context, and the geographic scope used in any comparison.
  • Event classification: Note whether the agency record identifies an accident or incident and whether the event involved injury, death, or aircraft damage under the applicable definition.
  • Investigation record: Keep the agency, report number or identifying reference, report date, and investigation status when those details are available.
  • Aircraft and operating details: Preserve recorded information about the variant, engine, and other model-specific details rather than relying on a broad aircraft name.
  • Injury and treatment records: Organize documents that describe the injury, care, expenses, and effects connected to the event.
  • Communications and observations: Keep factual messages, notices, photographs, or witness information that helps identify what occurred and when.

This file does not need to answer every legal question before you seek help. Its immediate purpose is to keep the event identifiable and the facts distinguishable from general statistics.

When you compare the event with aircraft-model records, write the comparison in a separate section of your notes. State the event definition, model category, period, geography, and denominator. Then write a separate section for the specific accident’s facts and investigative status. This simple separation helps prevent a historical total from being mistaken for a finding about your accident.

If a record uses an unfamiliar model name or classification, preserve the original wording and mark the question for follow-up. Do not replace an uncertain model, variant, or event category with a guess merely to make the comparison fit.

Use the documented facts to decide whether a case evaluation fits

Aircraft-model statistics are useful when they stay in their proper lane. Use them to describe a defined population, compare raw counts with exposure-adjusted measures, and identify questions that require the specific accident record. Do not use them to assign blame, establish cause, or predict legal responsibility from a model label alone.

Preserve the documents and factual details tied to your event, keep the event classification and investigation status clear, and avoid treating a general ranking as an answer about your injury. At ResQ Legal, we invite you to reach out for a free consultation to discuss your options if you are unsure about hiring an aviation lawyer. General information cannot determine whether an individual legal matter fits; a qualified attorney must assess that question from the documented facts and applicable law.


What should I verify before relying on a plane-crash-by-model comparison?

Which plane model has had the most crashes?

No reliable answer can identify a model with the highest crash count. Any answer would depend on the event definition, aircraft population, time period, geography, operating category, and whether collisions are counted per aircraft or per occurrence. A raw total should be treated as a scoped historical count, not as a safety ranking.

What should I check before comparing accident rates for two aircraft models?

Confirm that both figures use the same model or variant categories, dates, geography, operation type, event definition, and exposure denominator. For the FAA commercial-jet measure, the denominator is departures, or flight cycles, and the rate is accidents per million departures. Keep the underlying accident counts and departure estimates visible because small counts and revised data can make rate differences difficult to interpret.

Why can two official aviation databases show different totals for the same aircraft model?

The databases may cover different aircraft populations, locations, dates, operations, or event categories. One record set may include selected incidents alongside accidents, while another may focus on a narrower accident definition. Collision-counting rules, unknown locations, and later revisions to departure data can also affect results. Compare each source’s scope and definitions before treating the totals as inconsistent findings.

Can a model-level crash rate determine what caused a specific aviation accident?

No. A rate describes a defined population and does not establish whether a particular event involved a design issue, maintenance problem, pilot action, operator decision, weather, or another factor. Cause and responsibility require the specific accident record, investigative findings, and case facts. Preserve the aircraft identification, variant, date, location, event classification, and available records separately from general statistics.


Were you hurt on a flight?Tell us what happened. The consultation is free.
Free case evaluation

Keep reading