How fault actually gets decided

Collision Bureau is not a law firm and this is not legal or medical advice. It is general information about what happens after a crash. For advice on your situation, talk to an attorney licensed in your state or a treating clinician.

The sentence everyone misreads about fault

Ask most people how fault gets decided after a crash and the answer is some version of the police decide. The sentence is a shortcut, and it is wrong in a way that matters, because the shortcut shapes decisions in the first week that cost cases in the last. Police officers at the scene produce a record. They do not produce a verdict. The verdict, for insurance purposes, is a conclusion the insurers reach inside their own files after reading what the officer wrote and what else arrives in the days that follow.

Put a different sentence in. Fault gets decided by whoever is going to pay for it, inside rules the state wrote. Most of the time, that is a pair of insurance companies reading the same file. Some of the time it is one insurer, when there is no claim against the other. Rarely, when nothing resolves, it is a court. The piece people most expect to be dispositive, the officer at the scene, is a strong early witness. The piece people often forget to prepare for, the paper that arrives over the next weeks, is the file.

None of this means officers are wrong or that reports do not matter. They matter a lot, and in many crashes the report alone settles the fault question because the facts are plain and the insurers follow the officer's probable cause call. The point is that the decision does not freeze at the moment the officer finishes writing. It opens a short window in which other evidence can arrive, and the window closes at a specific moment inside each insurer's internal process. Knowing the window exists is the whole move.

Here is the practical implication for the first week. Everything you do at the scene and in the first few days contributes to a file somebody will read later. Nobody announces the file. The file assembles itself around what you did or did not do. A fault outcome you are satisfied with later is almost always a reflection of a file that was built thoughtfully in that first week, and a fault outcome that disappoints usually traces back to pieces that were available early and did not get collected.

The police report is not a fault finding

Police reports do several specific things, and understanding them in order helps you read your own report when it arrives. The report records the officer's observations at the scene, the statements each driver and each witness gave on the record, the position of the vehicles when the officer arrived, any citations issued, and a short narrative in the officer's own words. In some jurisdictions, the officer completes a diagram and codes a probable contributing cause from a list that the agency maintains.

What the report is in the law varies by state. In many states, parts of the report are inadmissible as hearsay in a civil trial, and other parts come in through specific exceptions. Insurers do not operate under the trial rules of evidence inside their claim files; they read the report in full and weigh it against everything else they collect. Many of the lines in the report that would never reach a jury still shape the determination the adjuster writes. The report is strong evidence for insurance purposes even where it has limits in court.

Report accuracy is a real variable. Officers write reports quickly, after bad calls, from the facts they can gather at the scene. Direction of travel can get recorded wrong when both vehicles came to rest in confusing positions. Street names can swap on cross-streets. The identity of who was driving a given vehicle can rest on who happened to be standing nearest to it when the officer arrived. None of this is a complaint about police work. It is a reminder that the report is a human document that benefits from a careful read.

Most agencies allow a short window during which a driver can request a correction to the report, based on specific evidence. The window is usually days, not months. The correction request has to be concrete, naming the exact error, and ideally attaching the piece of evidence that supports the correction. A corrected report is a different piece of evidence than an uncorrected one, and insurers do pay attention to corrections, especially when a conflicting original report would otherwise support the opposite finding. If something in the report looks wrong, do the correction request early.

Report element	What it is	Weight insurers give it
Officer narrative	Short, dated prose version of what happened	High, read for sequence and tone
Diagram	Scaled or rough sketch of vehicle positions and roadway	Medium to high, especially when matched to damage patterns
Probable cause code	Pre-defined contributing factor selected from a list	High when consistent with narrative and damage
Citations	Formal traffic citations issued at scene	Very high, though not legally determinative on their own
Statements captured	Each driver and witness, often as a short quoted sentence	Variable; strongest when consistent across sources

THE PIECES INSIDE A TYPICAL REPORT AND THE WEIGHT INSURERS TEND TO GIVE EACH. YOUR AGENCY'S REPORT FORMAT MAY DIFFER, AND YOUR STATE'S ADMISSIBILITY RULES CONSTRAIN HOW A COURT TREATS EACH PIECE LATER.

One quiet detail about citations. A citation is a specific allegation that the driver violated a specific rule. Insurers read citations as strong evidence of fault for the facts the citation covers, and some courts allow a conviction on the citation to be used later as proof of the underlying facts. That has two implications. First, taking a citation to traffic court is sometimes worth doing even when the fine is small, because a dismissal or amendment removes a piece of evidence from the civil claim side. Second, pleading guilty to a citation for convenience can later be used against you in the civil claim. The attorney conversation addresses this one, carefully.

What a liability determination actually is

Every open claim produces, at some point in its life, a liability determination. The term is specific. A liability determination is the insurer's internal conclusion about who was at fault for the crash and in what percentage, based on the evidence the file holds at the moment the determination is written. The determination lives in the adjuster's notes and in letters to the parties. It is a working conclusion, not a court finding. It changes when new evidence arrives, and it can be challenged inside the insurer and then outside it.

Insurers reach the determination differently inside each company, but the shape is consistent. The adjuster assembles the file. A supervisor often reviews it on a defined threshold. The determination is written down with a short rationale that references the evidence. The adjuster then communicates the determination to the parties, often in a letter that states the insurer's position on who was at fault, and sometimes stating a percentage split in states where splits apply.

Not every claim gets a formal liability determination letter. On straightforward files where no money is in dispute, the determination lives inside the file notes and informs the offer directly. On contested files, the determination becomes a visible step and often arrives in writing. The visibility of the determination is often the first signal that the file is heading toward dispute, which is why reading the letter carefully matters when one shows up.

What a liability letter typically looks like. A short paragraph introducing the claim and the parties. A statement of the insurer's position on fault, often naming a percentage or a straightforward attribution. A short rationale that references the evidence the insurer is relying on. A closing paragraph that invites a response if the recipient disagrees, with a window for the response that can be short. Reading the letter line by line is the move. The sentences that look like boilerplate are often the sentences that carry the position the insurer wants to lock in, and a response that answers them specifically is more useful than a response that argues the whole thing from the top.

The letter also carries procedural implications. Some policies and some state rules require claimants to object to a liability determination inside a specific window if they want to preserve certain challenges. Missing the window can narrow the available remedies, which is one more reason the attorney conversation is worth timing right after the letter arrives. A short review on the day the letter lands is cheaper than a scramble months later when the window has quietly closed.

A quiet detail about these letters. The adjuster writing the letter is reading the file through the scoreboard the sibling article on the adjuster described. Cycle time pressures the letter to arrive sooner rather than later, so determinations sometimes get written before every piece of available evidence is in the file. Severity pressures the allocation toward the claimant where there is room to do so honestly, within what the file supports. Litigation rate pressures the specific language toward positions the insurer thinks it can defend if a dispute follows, which is one reason the sentences in the letter often sound carefully written. None of that is sinister. All of it is useful to know, because the letter is a product of institutional forces as much as it is a reading of your particular file. Reading it that way keeps the response calibrated.

Rules of thumb insurers use to write the determination. The driver who violated a traffic law relevant to the crash is usually at fault. The driver who struck the other vehicle in a scenario where the first driver had the right of way is usually at fault. In rear-end crashes, the following driver is usually at fault, with meaningful exceptions. In left-turn crashes, the turning driver is usually at fault across oncoming traffic, with meaningful exceptions. Fault findings cluster around these presumptions because real-world files cluster around them, and exceptions require specific evidence to overcome the presumption.

Where determinations get interesting is on the exceptions. Yes, the following driver is usually at fault, and no, that is not the end of the inquiry. If the leading driver stopped unexpectedly to turn without signaling, if a mechanical defect created the stop, if the leading driver was driving without working brake lights, the file can shift the determination in whole or in part. Every presumption is a starting point. Evidence can move it, and the attorney conversation often knows exactly which pieces of evidence move which presumptions, because the pattern is well traveled.

The evidence that moves the finding

Not all evidence is equal, and insurers rank it in roughly the same order whether or not they say so out loud. The ranking matters because the file you can affect in the first week is almost entirely the top of the ranking. The report and the ticket are already in motion once the officer leaves. The pieces that most reliably change a close fault call are pieces you can preserve or lose in the first days.

Physical evidence sits at the top. Vehicle damage patterns, debris field on the roadway, skid marks or lack of them, paint transfer, deformation geometry. These are inputs that do not change their minds and that can be interpreted later even if nobody is looking at them now. Photographs taken close in time preserve this evidence effectively. The vehicles themselves preserve it until they are repaired or scrapped.

Camera footage sits close behind. Doorbell cameras, business cameras, intersection cameras in some cities, dash cameras in some vehicles. The quality of camera evidence depends entirely on whether anyone retrieved it before it was overwritten. Business and doorbell cameras often overwrite inside days or weeks, and even city-run traffic cameras may not retain footage for long. If a camera could have caught the crash, the window for pulling footage is short. Canvassing cameras is often done by an attorney's investigator the week after a crash on files where fault is contested, and the pattern of success and failure on that work is one of the most consistent reasons for differences in fault outcomes.

Independent witness statements sit next. A witness is anyone who saw the crash and has no financial interest in it. Not a passenger in your own car. Not a family member driving behind you. A stranger at a bus stop, a store clerk looking out a window, a driver stopped at a light. Their statement is weighted heavily because they have no reason to shape it. The practical issue is that strangers do not stay in touch, so the time to collect their name and number is at the scene, and the time to follow up is the next day while memory is still fresh.

Driver statements sit lower on the ranking than most people expect, because each driver has an interest in a particular version. That does not mean statements do not matter; they shape the file heavily, and inconsistent statements across time can be used against the driver who gave them. It does mean that a plain fact in a damage pattern or on a camera outranks a plain fact in a driver's recollection in nearly every determination. Being right at the scene without preserving physical evidence is a weak position in paper months later.

Evidence type	Why it is weighted this way	Window to preserve
Vehicle damage patterns	Physical, replicable, interpretable later by experts	Until the car is repaired or scrapped; days to weeks
Camera footage	Direct observation, hard to argue with	Short, often days; business and doorbell cameras overwrite fast
Independent witnesses	No financial interest in the outcome	At the scene; memories soften within days
Police report	Officer's contemporaneous record of statements and observations	Issued within days; corrections possible for a short window
Driver statements	Both drivers have interests in the version they tell	Shift over time; later statements get checked against earlier ones

A WORKING RANKING. YOUR SPECIFIC FILE MAY WEIGHT ITEMS DIFFERENTLY, AND SOME STATES CONSTRAIN HOW CERTAIN PIECES ARE ADMITTED LATER.

Vehicle damage, the stubbornest witness

Vehicle damage is the piece of evidence that gets least argued over inside insurance files, because physics is not open to interpretation in the usual sense. A rear bumper crushed in with no corresponding front-end damage on the other vehicle tells a specific story about impact and direction. A side crumple consistent with a ninety-degree impact is not consistent with a glancing lane change. Patterns of damage, when read against the damage of the other vehicle, support or weaken each driver's narrative without the vehicles saying a word.

Insurers employ or hire crash reconstruction specialists on larger cases, and their reports are routinely read in detail by both sides. On smaller cases the adjusters themselves read damage patterns and compare to the drivers' accounts, often informally. The quality of this analysis depends on the quality of the photographs and, in some cases, access to the vehicle itself. Photos taken from several angles close in time are the usable record. Photos from one angle, taken days later after the vehicle has been moved and partly cleaned, are much weaker.

What to photograph before the vehicle leaves the scene or the tow lot. Every corner, from a standing distance. Each piece of damage, close up, with a sense of scale. The license plate and VIN plate. The debris field around the vehicle if debris is present. The other vehicle, if accessible, with the same thoroughness. The pavement, including skid marks or their absence. Any street signs, signals, or lane markings that figure in the sequence. The habit takes 10 minutes at the scene, less in daylight, and preserves the record for months.

A real caveat on damage interpretation. Damage patterns can be ambiguous when vehicles have been pushed, repositioned, or have multiple collisions in the sequence. A car that was rear-ended and then pushed forward into another car has three pieces of damage that have to be read in the right order. Reconstruction specialists are trained to sort these cases, and most adjusters know when to call for the specialist. The main thing from your side is to preserve the record so the specialist has material to read.

One quiet value in damage preservation. When fault is close, the party who did the better job of preserving the vehicle damage record often has the stronger position in negotiation, because the record is harder to argue away. Not because the damage changes the facts, but because the available evidence constrains how aggressively the other side can push an alternate narrative. The 10 minutes of photographs is doing exactly this work, quietly, from day one.

Statements and the record they create

Statements made at the scene become some of the earliest and most quoted pieces of a fault file. The officer records them in short, sometimes paraphrased, sometimes direct quotation form. Insurers reference the officer's recorded statements through the whole life of the claim, and inconsistent statements made later get compared to them. The things said in the first 10 minutes carry weight for months.

Common patterns at the scene that create problems later. Apologetic phrases offered to be civil. Minimizing statements about injuries offered because people say they are fine when they are shaken. Guesses about speed or distance offered because the officer asked. Admissions of partial responsibility offered because partial responsibility felt true in the moment. Each of these is understandable and each of them can be quoted later in a context that reads very differently than the context in which it was said.

The honest alternative at the scene is a short, factual version of what happened, kept to things you observed and know. The sentence I was driving north on this street at the speed limit when the other vehicle entered the intersection from my left is a factual sentence. The sentence I'm sorry, I didn't see them at all is a sentence that can be read as an admission on the duty to look. Both sentences describe the same mental state. Only one of them helps a file in month three. Nothing in this section is a script. It is simply a reminder that scene statements are evidence, and brevity in evidence is almost always safer than warmth.

Statements to adjusters that come later are a different category. The sibling article on the recorded statement covers them in detail. The short version: your own policy may require a statement to your own insurer under its cooperation clause, and the other driver's insurer is a different situation that often does not require anything. The free attorney call sorts the sequence for your state and your facts, and the sorting is the piece most worth taking from this article into practice.

Comparative and contributory, kept general

States run two broad families of fault rules, and the family that applies in your state shapes how the dollars come out at the end even on identical facts. The first family is comparative fault, in which both parties can share blame and recoveries get reduced by each party's share. The second family is contributory fault, in which any share of blame by the claimant can bar recovery entirely. Comparative is the common system across most states; contributory survives in a small number of jurisdictions. The specifics are state law, and the article will not try to name which state falls where.

Inside the comparative family, there are two subtypes. Pure comparative, in which recovery is reduced by the claimant's share no matter how large. Modified comparative, in which recovery is reduced by the claimant's share up to a threshold such as fifty or 51 percent, past which recovery is barred. The threshold is the piece that matters most, because a case allocated at 50 percent and a case allocated at 51 percent are the same case in pure comparative and are opposite outcomes in modified comparative.

In contributory states, the system is cleaner and much tougher on claimants. Even one percent of fault by the claimant can bar the entire claim, which produces negotiation that focuses heavily on moving fault off the claimant entirely rather than splitting it. Most contributory states have narrow doctrines that soften the rule in specific situations, but the general shape is strict. Working in a contributory state changes the arithmetic of early statements and settlements in ways that only a licensed attorney in that state can walk accurately.

Family	What it does	Practical effect on your file
Pure comparative	Reduces recovery by claimant's share, no cap	Even a majority-at-fault claimant can recover something
Modified comparative	Reduces recovery by claimant's share up to a threshold	Crossing the threshold bars recovery entirely
Contributory	Any claimant fault bars recovery	Negotiation focuses on moving all fault off the claimant

GENERAL FAMILIES, NOT STATE-SPECIFIC RULES. YOUR STATE'S CURRENT RULE AND ITS NARROW DOCTRINES BELONG TO A LICENSED ATTORNEY READING YOUR FILE.

Why this matters for the first week. A statement that would be a minor dent in a pure comparative state can be the whole case in a contributory state. A scene apology, a scene admission of partial responsibility, a sentence about not seeing the other vehicle, each lands differently against the state system that applies to your facts. The practical move is identical in every state: keep statements factual, preserve physical evidence, and ask a licensed attorney early. The reason is state-specific, and the attorney conversation is where the reason gets specific.

The percentages conversation

Percentages appear on comparative fault files when the parties and the insurers try to express how much each driver contributed. Common splits include numbers like fifty and fifty, seventy and thirty, eighty and twenty, and occasionally cleaner numbers that reflect specific findings. The percentages feel precise. They are rarely as precise as they look, and understanding how they get written is useful for reading them without reading them wrong.

What a percentage represents on an insurer's file is a prediction. It is the insurer's read of how a factfinder, meaning a jury or an arbitrator, would allocate fault based on the available evidence. The number is discounted by the risk of being wrong, which is why percentages often get negotiated rather than calculated. Two careful adjusters reading the same file can honestly land on different numbers, and the negotiation that moves them toward one another is the ordinary shape of a fault conversation inside a claim.

Where the percentages become real money is at the point of settlement or award. A 30 percent allocation of fault to a claimant with a loss of one hundred thousand dollars means the recoverable amount is seventy thousand, in pure comparative. In modified comparative with a 50 percent bar, a 30 percent allocation still allows recovery and a 51 percent allocation bars it. In contributory, the 30 percent would bar it. The arithmetic changes before the facts do, which is the central oddity of the fault system across states.

How to read a percentage proposal from an insurer. Treat it as a position, not a fact. Ask what evidence the insurer is relying on for that percentage, and ask what pieces of evidence would move it. In many cases there are specific pieces the file is missing, and completing them changes the percentage. In other cases the number is a hard anchor the insurer has set for internal reasons. The attorney can usually tell which category you are in after a short review.

A worked example that uses editable numbers. Pretend the gross claim is a hundred thousand dollars. In pure comparative, a seventy and thirty split costs the claimant thirty thousand off the top, leaving seventy. In modified comparative at a 50 percent bar, the same split is still a recovery; a fifty one and forty nine split is zero. In contributory, any share by the claimant is zero, so a ninety and ten split is still a bar. The arithmetic reads very differently once state is applied. Nothing here is a promise about any claim, and the hundred thousand is a placeholder meant only to make the arithmetic legible.

A quiet pattern inside percentage negotiations. Insurers often open with an allocation that puts more fault on the claimant than the evidence supports, because the opening position carries the usual purpose of opening positions. Rounds move the percentage in increments, and the attorney reads the increments against what the file supports. A move from forty to twenty five is a meaningful shift; a move from forty to thirty eight is a performative one. The size of the moves is often the signal, not the specific number at any one moment.

Fault in two-car intersections

Intersections produce a disproportionate share of fault disputes because they involve multiple drivers making simultaneous decisions, often under time pressure, with limited sightlines. Insurers approach intersection crashes with a well-developed set of presumptions, and reading those presumptions helps you understand how your file will be read before you make any decisions about documentation or statements.

Common patterns and presumptions. A driver who ran a red or a stop sign is almost always at fault, if the running can be shown. A driver turning left across oncoming traffic is almost always at fault for a collision with that oncoming traffic, with narrow exceptions. A driver who entered an intersection on a yellow light after the other driver entered on a green is almost always at fault, unless the yellow was still active across both entrances. A driver who had the right of way, exercised it, and was hit by a driver who did not yield is almost always not at fault.

Where intersection cases get hard. When each driver claims a green light and no camera is available, when a pedestrian or a cyclist is involved, when visibility was impaired by sun or weather, when a signal was malfunctioning, when speed is contested. These cases often turn on independent witnesses or on footage that a canvass would reveal. The investigation stage of intersection claims is where most of the determination gets made, and the files that spend money on canvassing early tend to produce more defensible outcomes than the files that rely on the officer's probable cause code alone.

Diagramming an intersection crash is one of the most useful exercises any claimant can do early. On a plain piece of paper, draw the intersection with north pointing up. Mark the direction each vehicle was traveling. Mark the point of first contact. Mark where each vehicle came to rest. Mark the signals that applied to each entrance. The exercise clarifies the sequence in your own head and produces a document that your attorney can use as a starting point. The diagram does not have to be artistic. It has to be labeled.

Right-of-way presumptions have a specific shape at four-way stops and at uncontrolled intersections, which are two scenarios that create a quiet share of fault disputes. At a four-way stop, the driver who arrived first has the right of way, and when two arrive simultaneously the driver on the right usually does. At an uncontrolled intersection, the yield rules depend on roadway classification, posted signage, and in many states a baseline duty for the vehicle on the left to yield to the vehicle on the right. These rules are rarely tested in daily driving and surface precisely when a crash happens. Reading the state's specific rule for an uncontrolled intersection is often part of the investigation stage, and the finding depends on the specific signage present.

A specific pattern worth naming. Many intersection crashes involve one driver who clearly violated a rule, a second driver who had a reasonable opportunity to see or respond, and an outcome shaped partly by both. The clean fault read against the rule-violating driver can get split by evidence that the second driver was speeding, was distracted, or had an obvious opportunity to brake. The split is often modest, maybe eighty and twenty, and the attorney conversation sorts whether a split like that is likely on your facts given the available evidence.

Fault in rear-end collisions, with surprises

Rear-end collisions are the archetype of a clean fault case, and the presumption is heavy against the following driver. The rule insurers apply is roughly that the following driver was responsible for maintaining a safe distance and failed to do so. In most rear-end crashes, the determination writes itself and the file resolves on that side quickly. The surprises are worth knowing about anyway, because the exceptions, when they apply, can shift fault partly or wholly to the leading driver.

Common exceptions. Brake lights not working on the leading vehicle. An unexpected and unnecessary sudden stop by the leading driver, such as reversing into a parking spot without warning. Mechanical defects that caused the following driver to lose brake capability through no fault of their own. A third vehicle that caused the stop, where the chain of causation can be reconstructed. A leading driver who was partially in a travel lane while attempting to park. Each of these shifts the determination in a specific direction, and each requires evidence to overcome the presumption.

Multi-vehicle rear-end collisions are their own category. A chain of three or more vehicles presents an evidentiary puzzle, because the sequence of impacts has to be reconstructed to allocate fault. Insurers often work backward from the last impact, allocating fault to each following driver for the gap maintained with the driver ahead, with specific attention to whether some drivers were pushed into the vehicles ahead of them by the force of a later collision. Damage pattern analysis is often decisive in these cases.

Rear-ended drivers who are considering accepting the clean fault read without question should still take care with early statements and settlements, because injuries from rear-end crashes often develop over days and weeks, and the clean fault read on the property side can quietly set up pressure to close the injury claim before the injury picture has resolved. Being right about who hit whom does not make the first offer the right offer. Those are two different questions, and the first attorney conversation separates them cleanly.

A specific rear-end wrinkle inside parking lots. Lot crashes are not governed by the same roadway rules as public streets, and insurers often read lot crashes with a lighter presumption. A driver backing out of a parking space is usually at fault for colliding with a vehicle in the main lane, but exceptions exist when the vehicle in the lane was speeding, driving in a wrong direction, or passing in a prohibited area. Lot crashes often go to split allocations more often than road crashes with similar impact geometry, which is one reason an attorney conversation on these can produce a different result than people expect.

Commercial truck rear-end cases carry additional considerations because of federal regulations on commercial driver conduct, including hours of service rules that may be at issue. Those cases often involve document discovery of driver logs and telematics data, which can corroborate or undercut the trucking company's account of events. Specialized attorneys handle these cases because the regulatory layer adds evidence that routine investigations rarely touch.

Fault in lane changes and merges

Lane change and merge collisions involve some of the most consistently litigated fault questions, because the physical evidence is often ambiguous and both drivers have reasons to tell the story their way. The presumption usually puts fault on the driver who changed lanes, under rules that require that driver to yield to traffic already in the destination lane. The presumption is not absolute, and the evidence that moves it is often subtle.

Common lane-change scenarios and their evidence. A merging driver who entered a highway lane without yielding, with scrape damage on the right rear and the left front of the vehicles. A driver who changed lanes across a solid white line, where the markings in the roadway become additional evidence. A driver who misjudged the speed of traffic in the destination lane, where the damage pattern is similar but the fault question remains. Each pattern lives at a slightly different intersection of markings, speed, and distance.

Blind-spot cases are their own subset. The adjective blind-spot is a mechanism, not an excuse, and insurers do not treat not seeing the other car as a defense in itself. The duty to look applies even when seeing is hard, and failure to see is often treated as part of the failure to yield. There are exceptions when the other driver was speeding substantially or occupying the destination lane at a position that evaded a reasonable check, but those exceptions require specific evidence rather than general claims about limited visibility.

The documentation habits for lane-change crashes mirror the intersection pattern. Photograph damage from multiple angles, including the lateral geometry that shows which part of each vehicle contacted which part of the other. Record the lane markings in the roadway and the direction of travel. Note the position of each vehicle when it came to rest, which can indicate the relative speeds and the point of first contact. The paper trail is the whole case on many lane-change files.

Highway merge cases carry their own patterns. A driver entering from a merge lane generally has the duty to yield to traffic already on the highway, which places the presumption on the merging driver when a collision occurs during the merge sequence. Exceptions include situations where the highway driver was in the merge lane itself, where the merge lane was closing on the merging driver, or where the highway driver changed lanes into the merging driver without allowing adequate space. The physical evidence that distinguishes these scenarios lives in the damage geometry and the position of the vehicles at rest, both of which the first-week photographs preserve.

Fault in pedestrian and cyclist collisions

Pedestrian and cyclist collisions bring a different fault analysis than vehicle-to-vehicle crashes, because the state rules often impose heightened duties on drivers to anticipate and yield to vulnerable road users. The presumption in many cases is weighted toward the driver, though the rule is not universal and depends on the specific circumstances. The analysis is heavier on duty-of-care considerations and lighter on the physical geometry that dominates vehicle collisions.

Common patterns. A driver who struck a pedestrian in a crosswalk, with signal timing and visibility becoming the dominant questions. A driver who struck a cyclist in a bike lane, where the lane markings and the driver's duty to check become central. A pedestrian who crossed outside a crosswalk, where the state's rules about jaywalking interact with the driver's duty to see and respond. Each pattern lives at a specific intersection of statute, physical layout, and reaction time.

Vulnerable-user cases are also sensitive to timing and visibility factors that vehicle cases often weight less heavily. Weather, lighting at dusk or dawn, reflective clothing on the pedestrian or cyclist, speed of the vehicle, and the driver's actual sightline at the moment of first visibility are all evidence that can shift fault substantially. These cases often need investigation that goes beyond what a routine insurance adjuster conducts, and the attorney conversation often brings an investigator in early.

The evidence habits for these cases include everything from the vehicle collision pattern plus specific attention to the pedestrian or cyclist's path and visibility. Photographs of the roadway at a similar time of day, under similar weather, can be useful later. Statements of witnesses to the pedestrian or cyclist's approach, not just the collision itself, carry weight. The hospital record from the pedestrian or cyclist is a different piece of evidence than a vehicle driver's medical record, with different implications for the fault analysis.

Vulnerable-user cases also tend to involve insurance interactions that look different from vehicle-only crashes. The pedestrian or cyclist's own auto insurance can sometimes come into play through specific coverages that pay regardless of vehicle ownership, and health insurance plays a bigger role when the injured party does not have the usual med-pay or PIP coverage attached to a car. The attorney conversation covers these coverage interactions early, because they can change the sequence of who pays which bills and when, which matters a lot while a pedestrian or cyclist is in treatment.

A detail worth saying out loud. Fault findings against pedestrians and cyclists are not unheard of and can be significant. A pedestrian darting into traffic between parked cars, a cyclist running a red light at night without lights, a pedestrian walking into the roadway while looking at a phone, each can draw a share of fault that reduces the recovery. The duties on vulnerable users are real and well established, and the fact that drivers often carry a heavier duty to anticipate does not mean the vulnerable user has no duty at all. The investigation runs on everyone's conduct, and insurers write determinations that reflect what the facts actually say.

Phantom drivers and third-vehicle cases

Phantom drivers are drivers whose vehicle was never found. The sibling article on when the other driver has no insurance covers the coverage side. The fault side has its own shape. A phantom driver is treated as at fault for the collision in nearly every case, because the fact of fleeing is itself evidence of responsibility and because the typical reasons to leave suggest fault. The investigation runs through your own uninsured motorist coverage, and the fault analysis is often more vigorous than it would be against a reachable driver, because the UM payout requires the phantom to be at fault.

Third-vehicle cases, where a car swerved to avoid an action by a third vehicle and ended up crashing without any contact with that third vehicle, are complicated. Fault analysis has to decide whether the third vehicle's action was itself negligent and whether the swerving driver's response was reasonable. In some states these are workable UM claims, often with physical contact requirements that are tricky to meet. The attorney conversation covers both the physical contact rule and the fault analysis together, because they interact.

Chain reactions involving three or more vehicles produce fault questions that are proportionate to the number of vehicles involved. Insurers sort these by working backward from the final impact and reconstructing the chain. Attribution of fault to each driver depends on whether they maintained a safe following distance, whether they were pushed by a later collision, and whether their own driving contributed to the chain. The physical damage evidence and camera footage from vehicles in the chain, if any, become central.

These cases are also where multiple insurers get involved in a single file, and the article's section on insurer disagreements becomes relevant. Three insurers reading the same chain can produce three different allocations, and reconciling them often requires intercompany arbitration or civil court. The file management on chain cases is heavier than on two-car cases, and the attorney who handles them is reading the chain as a sequence rather than a snapshot.

Another third-vehicle pattern that comes up often enough to name. A driver swerves to avoid a child, an animal, or an object in the roadway and strikes another vehicle in the process. The fault analysis has to decide whether the swerving driver had a reasonable opportunity to brake instead, whether the hazard itself was negligently created by someone else, and whether the resulting collision was the kind of harm the duty to react was meant to prevent. These cases almost always need an attorney, because the chain of causation is harder to argue cleanly and the available evidence varies widely.

Chain cases also raise the question of whose insurer should be opening the record retrieval and the scene canvas. Each insurer tends to defer to the others on investigation steps that benefit the shared file, which can mean nobody takes the lead and evidence quietly decays while jurisdictional confusion plays out. A plaintiff-side attorney who takes on a chain case often becomes the de facto coordinator of the investigation, pushing each insurer to pull its share of records and canvass its share of cameras. The file ends up more complete because somebody owns it on your side.

When two insurers disagree

Most two-car crashes resolve with the insurers agreeing on fault. On the files that do not, the insurers have two main mechanisms for resolving the disagreement between themselves. The first is intercompany arbitration, a private process shared across major insurers that resolves specific types of disputes under a defined rule set. The second is civil court, which handles cases that fall outside intercompany scope or that involve amounts above its limits.

Intercompany arbitration is not something most claimants ever see directly. It is a dispute between two insurers over the fault allocation on a file, resolved by arbitrators the insurers share. The claimant may be asked for documents, but typically does not appear. The process is faster than civil court and produces a decision that binds the insurers for that file. The decision is not public, and it does not create binding precedent, which means similar facts can get allocated differently in different panels.

Civil court is the other path, and it has the shape most people imagine when they imagine fault disputes. A lawsuit gets filed, discovery runs, and a jury or judge decides the fault allocation at trial, if the case does not settle first. The vast majority of cases that enter civil court on a fault question settle before trial, because trials are expensive for both sides and the risk of a bad outcome motivates settlement. The possibility of trial is often the only tool that forces the insurer to move its allocation to a figure that reflects the evidence, which is one of the reasons representation changes insurer behavior on hard files.

Where this matters for your file. If two insurers are reading the same facts and reaching opposite answers, your position is caught between them, often with a property claim or an injury claim held up by the dispute. The insurer who agrees fault lies with you may try to settle fast at a low number. The insurer who agrees fault lies with the other driver may delay while the dispute runs. The attorney conversation sorts this out by acting on your side rather than being caught between the two insurers, which is a very different place to be than inside the dispute as one of the drivers.

Challenging a liability determination

Liability determinations are not the final word, and the mechanism for changing them is more routine than people assume. The move is almost always evidence rather than argument. A reconsideration request supported by new information, a corrected police report, witness statements that were not reviewed in the initial determination, expert analysis of damage patterns or scene photographs, each can shift a determination inside the insurer's own file.

The first level of challenge is internal. The adjuster who wrote the determination can revisit it, often with supervisor involvement, when new evidence is presented. The request has to be specific. A general dispute about the outcome rarely moves a determination; a specific request pointing at specific new evidence often does. Writing the request as a short letter that lists the evidence and the exact piece of the determination it addresses is more effective than a long argument about the whole file.

The second level is intercompany, when the challenge is between insurers and both are willing to run it through arbitration. The third is civil court, which handles cases that cannot resolve inside the first two. Each level has different rules, costs, and timelines, and a licensed attorney reading your file picks the right level for the facts. Challenging at the wrong level is a common waste, and skipping the first level often forces a worse outcome at the second.

What a well-prepared challenge looks like. A short cover letter stating the determination and the specific request. A clear exhibit list naming each new piece of evidence. The evidence itself attached, with notes pointing at the specific pieces that address the determination. A request for a written response within a reasonable window. The format is simple, and offices that handle this work keep letters like it on templates for exactly this reason.

The limits of challenge. A determination that rests on strong evidence against a weak counter-narrative will not move no matter how well-written the challenge is. The challenge works when the file was incomplete at the time of the determination, when the officer made a specific error that has now been corrected, or when evidence that moves the presumption has been assembled since the original read. Expectations should match the evidence, and the attorney conversation helps calibrate them.

A short list of specific evidence that tends to move determinations in practice. Doorbell camera footage retrieved from a nearby residence within days of the crash. Store or business camera footage that captured the sequence from an angle the officer did not see. A witness statement from someone who stopped at the scene and left a name, confirmed by a short follow-up call that locks the account in writing. A crash reconstruction expert's short written analysis of the damage patterns and vehicle positions, pointing at a specific inconsistency with the current determination. Any one of these can move a determination; a combination of two or three usually does.

The time dimension on challenges is unforgiving. The sooner the challenge arrives, the more likely it is to succeed, both because evidence that supports it is more available and because the insurer has not yet committed to the determination in writing in ways that are harder to walk back. A challenge raised in week three is almost always easier to resolve than the same challenge raised in month six. The attorney conversation surfaces this calendar early, which is the practical reason the early call is more useful than people assume.

If this is your week

Fault is read from a file. Build the file.

One request covers the attorney, the tow, the repair, and the rental. It costs you nothing, ever.

Get help now How it works

Where the article hands off to a licensed attorney

Everything to this point describes a mechanism that applies across states. The pieces specific to your file are the ones no article should try to answer. Which state's fault rules apply. What the admissibility rules do with the specific pieces of your police report. How the citation your officer wrote is likely to play in civil court. How likely a reconsideration request is to succeed given the current file. Which direction, intercompany arbitration or civil court, is the right next step. Each of these is a licensed attorney reading your pages against your state.

The free attorney conversation is the piece that puts those questions on a surface that can answer them. The call costs nothing, binds you to nothing, and produces an honest read of the fault picture at the moment it happens. The call also sits adjacent to several other questions this article could not answer. The adjuster conversation. The recorded statement question. The uninsured motorist question when the other driver has no coverage. Each has its own article in the library, and each hands back to the attorney call on the fact-specific questions. The sibling pieces are honest attempts to describe the mechanism. They are not substitutes for the lawyer.

Where this article has done its job. The reader now knows that fault is a conclusion inside insurer files, not an officer's declaration. The reader knows the evidence that moves the conclusion and the limits of each piece. The reader knows the general shape of comparative and contributory systems and why state matters so much. The reader knows that disputes get resolved through arbitration or court and that challenges respond to evidence rather than argument. The specifics are for the lawyer, and the article says so plainly, which is the only honest shape a general article on this topic can take.

One last practical note. If a crash is being argued over on your file and you have not made the attorney call, this week is the week. Evidence that moves fault is almost always evidence that decays. Cameras overwrite, witnesses move, vehicles get repaired or scrapped, memories soften, and the officer's report sits inside its correction window for only a short time. The call costs nothing. The call reads your file. The call tells you which decaying pieces are worth chasing before they stop existing. That is a sequence that happens to be free, and it is the piece the general article can genuinely point at.

A small supplementary habit that pays over the life of any contested fault file. Keep every piece of correspondence related to the fault question in one place, dated and ordered. The adjuster letter stating the determination. Your written challenge, with exhibits numbered. The insurer's response to the challenge. Any witness statements collected after the fact. Any expert analysis commissioned. Any revised determination letters. Six or eight documents, kept together, become the file the attorney can read in one sitting rather than reconstruct from phone calls months later. The habit is small. The payoff is large and quiet, the way most of the useful habits in the aftermath turn out to be.

The last paragraph is a short reset. Fault is not a sentence at the scene. It is a conclusion inside two insurer files, reached inside a state's rules, from evidence that mostly decays on its own schedule. The ways to influence that conclusion are to preserve the evidence that would be lost otherwise, to keep statements factual, to read the police report when it arrives and correct what is wrong inside the window, and to make the free attorney call early so somebody licensed can read all of it with you. None of these is heroic. Each is a specific thing, done on a specific day, that compounds across the weeks the file takes to resolve. That is the honest sum of the article. The specifics live with the lawyer.

Questions people actually ask

Legal

Fault is a file. Open it with someone on your side.

One request covers the attorney, the tow, the repair, and the rental. It costs you nothing, ever.

Get help now How it works