LaneSplitting

What the research says

There isn’t much of it, and none of it is a controlled experiment. Here is what the main studies found, what they can’t tell us, and the strongest objections we’ve seen from people who oppose lane filtering.

The short version

In California, riders who crashed while splitting lanes in slower traffic, without going much faster than the cars, were injured less often than other riders who crashed. Western Australia saw no change in rear-end or sideswipe crashes after it allowed filtering at 30 km/h. France saw motorcycle crashes rise 12% on the roads in its trial, and its evaluators found that the crashes they examined mostly involved riders who weren’t following the trial’s rules on speed and position.

None of these studies can say how often a rider who filters crashes compared with one who waits, because nobody has counted how much riding of each kind happens. The evidence fits a law with low speed limits much better than one without them.

The studies

Study California May 29, 2015

Motorcycle Lane-splitting and Safety in California

Thomas Rice, Lara Troszak and Taryn Erhardt. Safe Transportation Research & Education Center (SafeTREC), University of California, Berkeley, May 29, 2015. Full report (PDF, hosted by the California Office of Traffic Safety), checked 2026-10-26.

The researchers looked at 5,969 motorcyclists involved in collisions that the California Highway Patrol investigated between June 2012 and August 2013. 997 of them (17%) were lane splitting when they crashed.

What it found

  • Riders who were lane splitting were injured less often in their crashes: head injury 9% vs 17%, torso injury 19% vs 29%, extremity injury 60% vs 66%, fatal injury 1.2% vs 3.0%. Neck injury rates were about the same.
  • Lane-splitting riders were rear-ended less often (2.6% vs 4.6%), but they were more likely to rear-end another vehicle (38% vs 16%).
  • Injury rates did not rise in a meaningful way until traffic was moving faster than about 50 mph. The speed difference between the bike and traffic mattered more: up to 15 mph faster than traffic showed no change in injury, and above that the likelihood of every injury type went up.
  • The authors concluded that lane splitting "appears to be a relatively safe motorcycle riding strategy" in traffic at 50 mph or less when the rider is no more than 15 mph faster than other vehicles. 14% of the lane-splitting riders in the data were 25 mph or more faster than traffic.

Its limits

  • The study only covers riders who crashed. It has no exposure data, so it cannot say whether lane splitting makes a crash more or less likely. The authors say this is the main limitation.
  • Lane-splitting riders were different from other riders: more weekday and commute riding, better helmets, lower speeds, less alcohol, fewer passengers. Some of the injury gap may come from those differences. The authors adjusted for age, sex and helmet type.
  • Injury was recorded as yes or no, so a minor injury and a critical one count the same.
  • The authors postponed analysis of rear-end crashes by road type until they had roadway data.

Review United States and Europe June 2010

Motorcycle Lane-Sharing: Literature Review

Myra Sperley and Amanda Joy Pietz. Oregon Department of Transportation, Research Section (report OR-RD-10-20), June 2010. Full report (PDF, National Transportation Library), checked 2026-10-26.

Oregon's transportation department reviewed what was known about lane sharing in the US and abroad before Oregon debated it. It is a literature review, not new crash data.

What it found

  • It found that research on the topic was limited and that no studies at the time focused mainly on the benefits or safety of lane sharing.
  • In the large crash studies it reviewed, lane sharing was a factor in somewhere between under 1% and 5% of motorcycle crashes. The European MAIDS study (921 crashes in five countries) found filtering as the pre-crash motion in 4 cases (0.4%). A UK study (Clarke et al., 2004) found filtering was a factor in 5% of the crashes studied, and the other driver was more than twice as likely to be at fault.
  • It describes the 1981 Hurt Report, often cited in favor of lane splitting, as having "only hinted" that lane splitting was not a significant crash factor, quoting a 1999 interview with Harry Hurt.
  • The main hazard it identified is driver expectation: drivers in stopped or slow traffic do not expect a motorcycle to pass between lanes, which leads to sideswipe and turn-into-path crashes. A possible benefit is that riders can see further ahead.

Its limits

  • The review says its range of under 1% to 5% "should be considered with caution" because so few studies looked at lane sharing as a cause.
  • Most of the evidence comes from Europe, where traffic, roads and vehicles differ from US conditions, and from studies where lane sharing was not the main focus.

Study France January 2021

Evaluation of France's experiment with motorcycle lane splitting (in French)

Expérimentation de la circulation inter-files (CIF) des deux-roues motorisés: rapport d'évaluation
Mathis Beltrami and Benoit Hiron (Cerema), with Université Gustave Eiffel and Ergo-Centre. Cerema, for France's road safety agency (ONISR / Sécurité routière), January 2021. Evaluation report (PDF, in French, ONISR), checked 2026-10-26.

From February 2016 to January 2021 France allowed motorcycles to ride between the two leftmost lanes of slow traffic on motorways and similar roads in eleven departments, with a 50 km/h cap. This is the government's evaluation of that trial, and its results are less encouraging than California's.

What it found

  • Motorcycle crashes fell 10% across the trial departments as a whole, but rose 12% on the specific roads where the experiment applied (comparing 2012 to 2014 with 2016 to 2018). The authors say the rise should be tempered because the numbers stabilized during the trial.
  • Gironde (Bordeaux) was the outlier: crashes involving lane splitting on the trial roads rose from 13 in 2015 to 57 in 2018, at a time of fast-growing traffic and congestion on the Bordeaux ring road.
  • Of 16 fatal crashes involving motorcycles riding between lanes, none followed the trial's rules on speed or position. Most were well above the 50 km/h limit.
  • The typical crash was a commuter at rush hour riding between the leftmost lanes when a driver changed lanes without seeing the motorcycle. Compliance with the 50 km/h cap improved from 23% of riders observed in 2015 to 40% in 2018, but stayed a minority.

Its limits

  • Crash counts on the trial roads were small, and the authors say they should be read with caution.
  • It compares before and after periods, so changes in traffic volume and congestion (as in Gironde) are mixed in with the effect of the rule.
  • French motorways, rules and riding habits differ from US roads.

Review Western Australia November 2023

Motorcycle Lane Filtering Post-implementation Review: Summary of Findings

Road Safety Commission, Government of Western Australia. Government of Western Australia, November 2023. Summary of findings (WA.gov.au), checked 2026-10-26.

Western Australia allowed lane filtering at up to 30 km/h from 29 March 2021. More than 18 months later the state's Road Safety Commission reviewed crash data, offence data and a community survey.

What it found

  • There was no notable change in the number of rear-end and sideswipe crashes involving motorcyclists after the law took effect. The review says the crash data is "somewhat limited".
  • It concluded that the change "has not created any negative outcomes for the community", and noted shorter travel times for riders and less exposure to heat and exhaust.
  • Most riders surveyed felt safer filtering than waiting in front of traffic. Some other respondents thought filtering was unsafe because of sideswipe risk and riders being harder to see, especially when riders ignore the 30 km/h limit.
  • Many people in the wider community did not know the rules, and most respondents said they had seen riders breaking them. The commission planned public education.

Its limits

  • This is a short government summary, not a peer-reviewed paper, and the full crash analysis is not published in it.
  • The review covers a fairly short period after the law changed.

The case against

These are the objections legislators hear most. We’ve summarized them as fairly as we can, and each one links to the original testimony.

AAA Oregon/Idaho

Written testimony opposing Oregon HB 3542 (lane filtering), Joint Committee on Transportation, April 7, 2025. Read the testimony, checked 2026-10-26.

  • AAA said there is not enough data to support a law authorizing lane splitting, and that the evidence from other places is "far from conclusive".
  • Drivers do not expect vehicles to pass between lanes and may change lanes or open doors without seeing a motorcycle, which raises the risk of sideswipes.
  • Officers would have to judge case by case whether conditions were legal, and pursuing a rider through stopped traffic is rarely safe, so enforcement would be hard.
  • AAA said its surveys show most drivers are uncomfortable with lane filtering, and raised concerns about driver liability and insurance costs after a crash.

Cranston (Rhode Island) Police Department

Letter from the chief of police opposing Rhode Island H 7354 (lane filtering), House Judiciary Committee, February 13, 2026. Read the testimony, checked 2026-10-26.

  • A filtering motorcycle often sits in a driver's blind spot, so the driver may not see it until it is alongside.
  • Modern cars are well soundproofed, so drivers may not hear a motorcycle coming at low speed.
  • If a driver swerves for a pothole or hazard, a filtering rider has no margin for error. The chief argued that a 10 mph speed difference that close to other vehicles invites sideswipes.

An Oregon sheriff's sergeant who leads a county crash reconstruction team, testifying as a private citizen

Written testimony opposing Oregon HB 3542, April 7, 2025. Read the testimony, checked 2026-10-26.

  • Oregon lanes are not designed or measured for two vehicles side by side, and lane splitting removes most of the buffer.
  • Some drivers react to a passing motorcycle by closing the gap on purpose. He wrote that he has seen this many times on I-205 at rush hour.
  • Riders are taught to stay visible and out of blind spots. Filtering puts them in a narrow channel next to one blind spot after another, with little room to avoid a crash.

How we read it, as riders

The objections are mostly about the same thing the studies found: the danger comes from speed differences and from drivers who don’t expect a bike between lanes. France’s fatal crashes involved riders going well over the trial’s limit. In California, injuries climbed once riders were more than 15 mph faster than traffic.

That is why we ask for filtering with firm limits: traffic stopped or slow, a low maximum speed for the rider, and no filtering next to the curb or on the shoulder. It is also why driver education matters. Western Australia found many drivers didn’t know the rules, and the opponents above say drivers don’t expect a bike there.

We would like to see better data, especially counts of how much filtering riders do, so a crash rate can be calculated. States that pass filtering laws could collect it.

Ask your legislators See the state laws