November 17 - Insurance Institute for Highway Safety

November 17, 2008
David Kelly
Acting Administrator
National Highway Traffic Safety Administration
1200 New Jersey Avenue, SE, West Building
Washington, DC 20590
Notice of Proposed Rulemaking; 49 CFR Part 571 Federal Motor Vehicle Safety Standards,
Motorcycle Brake Systems; Docket No. NHTSA-2008-0150
Dear Mr. Kelly:
On September 17, 2008, the National Highway Traffic Safety Administration (NHTSA) announced a
proposal to amend Federal Motor Vehicle Safety Standard (FMVSS) No. 122, Motorcycle Brake Systems.
The amendments would strengthen the requirements and test procedures that cover many aspects of
motorcycle brake systems, including antilock braking systems (ABS) voluntarily installed by
manufacturers. The Insurance Institute for Highway Safety (IIHS) supports these changes and submits
with this comment our recent research showing that ABS is improving motorcycle safety.
The proposed changes to FMVSS 122 will create minimum performance standards that have been tested
successfully elsewhere in the world and will help keep motorcycles with unsafe brakes from being sold in
the United States. The proposed ABS tests in particular will help ensure that if ABS is installed on
motorcycles, the system will provide operators with adequate stopping distances and stability.
IIHS and its affiliated Highway Loss Data Institute (HLDI) are studying the safety effects of motorcycle
ABS as its availability increases. Results from our initial analysis of a variety of models show that ABS is
having a positive effect (IIHS, 2008). Both the rate of fatal motorcycle crashes and the frequency of
crashes for which insurance collision claims are filed are lower among motorcycles with ABS compared
with the same motorcycles without ABS. The attached IIHS study (Teoh, 2008) shows 6.7 fatal crashes
per 10,000 registered vehicle years among motorcycles not equipped with ABS during 2005-06. The
corresponding rate for the same models equipped with optional ABS was 4.1 fatal crashes per 10,000
registered vehicle years — 38 percent lower. The attached HLDI (2008) study shows that the estimated
effect of ABS was a 21 percent decrease in overall collision losses, primarily because collision claim
frequencies for motorcycles with ABS were 19 percent lower than for motorcycles without ABS.
The importance of equipping motorcycles with ABS increases as motorcycling continues to grow in
popularity. Motorcycle sales more than tripled between 1997 and 2005, and motorcyclist deaths have
more than doubled since 1997. Due largely to safety features that are increasingly available as standard
equipment on passenger vehicles, there were fewer passenger vehicle occupant deaths in 2007 than in
any year since NHTSA began collecting these data in 1975. On the other hand, more motorcyclists died
in crashes in 2007 than in any other year during the same time period.
David Kelly
November 17, 2008
Page 2
The Institute supports the proposed strengthening of FMVSS 122 to establish stronger minimum
performance standards for motorcycle brake systems. Given the beneficial effects of ABS, as
documented in our research, we urge NHTSA to consider further changes to FMVSS 122 to encourage or
require ABS on all motorcycles.
Sincerely,
Joseph M. Nolan, M.S.
Senior Vice President, VRC Operations
cc: Docket Clerk, Docket No. NHTSA-2008-0150
References and Attachments
Highway Loss Data Institute. 2008. Motorcycle antilock braking system (ABS). HLDI Bulletin 25(1).
Arlington, VA.
Insurance Institute for Highway Safety. 2008. Antilock brakes on motorcycles reduce both crash
frequencies and deaths. Status Report 42(9):1-3. Arlington, VA.
Teoh, E.R. 2008. Effectiveness of antilock braking systems in reducing fatal motorcycle crashes.
Arlington, VA: Insurance Institute for Highway Safety.
SR43-9 TO PDF:SR 43-9
10/16/08
11:37 AM
Page 1
Vol. 43, No. 9, Oct. 22, 2008
IF THEY NEED TO
STOP
on a dime, these riders will enjoy
an advantage most others don’t
because of the antilock brakes on
their motorcycle. Two new studies indicate crash reductions associated
with anti-
SR43-9 TO PDF:SR 43-9
2
10/16/08
11:37 AM
Page 2
Status Report, Vol. 43, No. 9, Oct. 22, 2008
locks. Both the frequency of crashes for
which insurance claims are filed and the
rate of fatal motorcycle crashes go down
among bikes with antilock brakes.
The importance of equipping bikes with
antilocks increases as motorcycling proliferates. Motorcycle sales more than tripled
from 1997 to 2005. Deaths of motorcyclists
have more than doubled since 1997, with
some kinds of bikes having much higher
death rates than others (see Status Report,
Sept. 11, 2007; on the web at iihs.org). About
5,000 motorcyclists died in crashes last year.
The new study of fatal motorcycle crashes was conducted by Institute researchers,
while the analysis of insurance claims is by
researchers at the affiliated Highway Loss
Data Institute (HLDI). Adrian Lund is president of both organizations.
“Even though adding antilocks won’t
make motorcycling as safe as going by car,
it’s something manufacturers can do to reduce the risk of traveling on 2 wheels instead
of 4,” Lund says. “It’s a way to reduce the
chances of overturning a bike and crashing,
so it can save lives among people who choose
motorcycles for their basic transportation,
to save on gasoline, or just for fun.”
When antilocks are needed: Stopping a
motorcycle is trickier than stopping a car.
For one thing, front and rear wheels typically
have separate brake controls. Both underbraking and overbraking the front and rear
wheels contribute to crashes (see Status
Report, June 21, 1979). In an emergency, a
rider faces a split-second choice to brake
hard, which can lock the wheels and cause
a motorcycle to overturn, or to hold back
on the brakes and risk running headlong
into the emergency.
This is when antilocks can help. They
reduce brake pressure when they
detect impending lockup and
increase the pressure again when traction is
restored. Brake pressure is evaluated multiple times per second, so riders may fully
brake without fear of locking the wheels.
Antilocks won’t prevent every motorcycle crash. They won’t help a rider who’s
Regression analysis revealed 21 percent
lower insurance losses for motorcycles with
antilocks, primarily because the claim frequency was 19 percent lower than for bikes
without antilocks. These findings are based
on a dataset of 72,000 insured years of 2003-
FATAL CRASHES
INSURANCE CLAIMS
PER 10,000 MOTORCYCLE REGISTRATIONS,
2001-06 MODELS DURING 2005-06
PERCENT CHANGE IN COLLISION LOSSES FOR
2003-07 MODEL MOTORCYCLES WITH ANTILOCKS
6
0
4
-10
2
without
antilocks
with
antilocks
about to be struck from behind, for example.
But the new studies indicate that antilocks
reduce crashes overall and save lives.
Crash reduction benefit: The HLDI study
compares insurance losses under collision
coverage for 12 motorcycle models with
optional antilock brakes versus the same
models without this option. The researchers evaluated the effects of antilock brakes
on both the frequency of insurance claims
that are filed for crash damage and the
average cost of the damage, after accounting for rider age and gender,
motorcycle age, and other factors
that influence the likelihood of a crash.
claim
frequency
claim
severity
SR43-9 TO PDF:SR 43-9
10/16/08
11:37 AM
Page 3
Status Report, Vol. 43, No. 9, Oct. 22, 2008
07 model Honda, Suzuki, Triumph, and Yamaha bikes (an insured year is 1 motorcycle
insured for 1 year or 2 insured for 6 months
each, etc.). BMW models aren’t included
because it’s impossible to determine from
vehicle identification numbers which ones
have optional antilocks and which don’t.
Harley-Davidsons aren’t included because
antilocks were added after the study years.
Antilock brakes “appear to reduce collision claims,” says Matthew Moore, HLDI
vice president and lead author of the study,
“but they don’t affect the severity of the
crashes for which claims are filed. The cost
of these claims doesn’t go down.”
Lives are being saved: In a complementary study, Institute researchers examined
rates of fatal crashes of motorcycle models
with and without antilocks. Eight models
were studied, a subset of the 12 included in
the HLDI analysis. The other 4 models were
excluded because of sample size limitations.
A main finding is that there were 6.6 fatal
crashes per 10,000 registered motorcycles
without antilocks during 2005-06. The corresponding rate for the same bike models
equipped with optional antilocks is 4.1, or
38 percent lower. Institute statistician Eric
Teoh, author of the study, says the findings
are statistically significant at the 90 percent
confidence level.
Antilocks on cars versus motorcycles:
Passenger cars began to be equipped with
antilock brakes during the 1970s, after studies conducted on the test track indicated
they reduce stopping distances. However,
this promise didn’t pan out in real-world
crashes (see Status Report, Jan. 29, 1994).
Antilocks didn’t reduce relevant collisions.
“It isn’t surprising that antilock brakes
are more beneficial on motorcycles than
they are on cars because the 2-wheelers are
so much less stable, and it’s this instability
that contributes to so many crashes,” Lund
points out. “By reducing wheel lockup during braking, antilocks keep a lot of motorcycles from overturning.”
Antilock brakes are recent additions to
motorcycles. They’re available almost exclusively as optional equipment (see list), which
means shoppers have to find models on
which the option is offered and then pay
extra for it. Antilocks were on only 18 percent of the motorcycles included in the new
studies of effectiveness.
For a copy of “Antilock braking systems
for motorcycles and insurance collision losses” by M. Moore and Y. Yan or “Effectiveness
of antilock braking systems in reducing fatal
motorcycle crashes” by E. Teoh, write: Publications, Insurance Institute for Highway
Safety, 1005 N. Glebe Rd., Arlington, VA 22201,
or email [email protected].
2008 MODELS WITH ANTILOCKS
Antilocks are optional except as noted; models
in bold are included in one or both of the new
studies of the effectiveness of antilock brakes
BMW
BMW
BMW
BMW
BMW
BMW
BMW
BMW
BMW
BMW
BMW
BMW
Can-Am
Harley-Davidson
Harley-Davidson
Harley-Davidson
Harley-Davidson
Harley-Davidson
Harley-Davidson
Harley-Davidson
Harley-Davidson
Harley-Davidson
Harley-Davidson
Harley-Davidson
Honda
Honda
Honda
Honda
Honda
Kawasaki
Moto Guzzi
Suzuki
Suzuki
Suzuki
Suzuki
Suzuki
Triumph
Triumph
Yamaha
Yamaha
K1200GT (std)
K1200LT (std)
R1200RT (std)
F800S/F800ST
G650 Xchallenge
G650 XCountry
G650 XMoto
HP2 Megamoto/HP2
K1200R/K1200R Sport
K1200S
R1200R/R1200S
R900RT
Spyder
Electra Glide Classic
Electra Glide Standard
Night Rod
Night Rod Special
Road Glide/Road King
Road King Classic
Screaming Eagle Electra Glide
Screaming Eagle Road King
Street Glide
Ultra Classic Electra Glide
V-Rod
Gold Wing
Interceptor 800
Reflex
Silver Wing
ST1300
Concours 14
Norge 1200
Bandit 1250S
B-King
Burgman 650 Executive
SV650/SV650S/SV650SAF
V-Strom 650
Sprint ST
Tiger
FJR1300 (std)
FJR1300 Electric Shift (std)
3
SR43-9 TO PDF:SR 43-9
10/16/08
11:37 AM
Page 4
4
Status Report, Vol. 43, No. 9, Oct. 22, 2008
MORE STATES BAN DRIVERS’ TEXTING
California and Alaska are the latest US states to ban text messaging by drivers of all ages,
not just teenagers. The two states join Connecticut, the District of Columbia, Louisiana,
Minnesota, New Jersey, and Washington in banning texting by all drivers amid concern
that such distractions increase crash risk. Alaska’s law took effect Sept. 1. California’s ban
begins Jan. 1, 2009. Both states make the use of an electronic device to write, send, or
read text messages a primary offense, meaning that police officers can pull over drivers
solely for violating the bans. Texting while driving also is a primary offense
in Connecticut, the District of Columbia, Minnesota, and New Jersey.
Bans in Louisiana and Washington are secondary, so motorists must be violating another traffic law in order to be
stopped by police for texting. Nine states have text
messaging bans that apply only to novice drivers.
For details on state bans on text messaging by drivers, go to www.iihs.org/laws/cellphonelaws.aspx.
There’s lots of anecdotal evidence tying texting
while driving to crashes, but not much data from
real-world collisions. Studies have linked cellphone
use with crash risk (see Status Report, July 16, 2005,
and March 22, 1997; on the web at iihs.org).
In one of the first published studies on texting
and driving, the Transport Research Laboratory in the United Kingdom (on the web at
trl.co.uk) found that texting degrades performance in a driving simulator. Researchers found that composing a text
message affected driving more than
reading one. The 17 drivers in the
study — all were 17-24 years old —
had slower reaction times, were
more likely to drift out
of their virtual
lanes, and were
more likely to
reduce their
speeds while
they were
texting.
CHILD SEAT USE
AMONG KIDS IN
CRASHES GOES UP
Use of child safety seats has surged since
1999 among restrained children younger
than 9 riding in insured vehicles. Restraint
types also have changed. These are the main
findings of new research from the decadelong Partners for Child Passenger Safety
study of the Children’s Hospital of Philadelphia (CHOP) and State Farm, with support
from the Association of International Automobile Manufacturers. It’s based on 19982007 insurance claims and phone survey
data on more than 875,000 kids in crashes.
Overall safety seat use among restrained
children 8 and younger rose to 80 percent in
2007 from 51 percent 8 years earlier.
Virtually 100 percent of restrained children 3 and younger in crashes have been in
safety seats since 1999. Safety seat use is
much lower among older children. Progress
has been made, but there’s room to improve.
In 1999 only 15 percent of restrained 4-8
year-olds in the CHOP study were in an
appropriate restraint — a harness restraint
or booster. By 2007 appropriate restraint
use in this group had quadrupled to 63 percent. The rest of restrained 4-8 year-olds rode
in adult belts alone. Typically, such belts
don’t begin to fit properly until kids grow to
about 4 feet, 9 inches tall.
“Along with the increase in the number of
kids riding in child safety seats, we can also
see changes in the types of restraints they are
using now versus 10 years ago,” says Kristy
Arbogast, director of engineering at CHOP’s
Center for Injury Research and Prevention,
where the study was conducted. She says
more restrained 4 and 5 year-olds ride in
boosters now instead of harness restraints.
Only 31 percent of appropriately restrained 4-5 year-olds rode in harness restraints during 2007. Highback boosters are
slightly more popular now than backless
ones among restrained 4-5 year-olds. But
backless boosters are used nearly 3 times as
often as highbacks for 6-8 year-olds.
SR43-9 TO PDF:SR 43-9
10/16/08
11:37 AM
Page 5
Status Report, Vol. 43, No. 9, Oct. 22, 2008
5
Previous CHOP research shows boosters
lower crash injury risk by 59 percent for 4-7
year-olds compared with belts alone. Boosters elevate children so lap and shoulder
belts are properly positioned. Earlier this
month the Institute released evaluations of
41 booster models, finding that several
Of the states in the study, booster seat
use among 4-8 year-olds was lowest in Ohio
(18 percent) and Texas (20 percent). Not
surprisingly, these states don’t have booster
seat laws. On the other hand, 2 of the 5
states with the highest use of boosters,
Pennsylvania (72 percent) and Illinois (62
Among the study’s other findings are
that 60 percent of crashes involving children occur within 10 minutes of home, and
84 percent take place within 20 minutes of
home. Only 14 percent of crashes are on
roads where posted speed limits are 55 mph
or higher, but these crashes result in the
don’t improve belt fit (see Status Report,
Oct. 1, 2008; on the web at iihs.org).
Arbogast attributes the increase in booster use among older kids to education of parents and caregivers plus state laws requiring
older kids to ride in safety seats. Laws in 43
states and the District of Columbia include
booster provisions (on the web at iihs.org/
laws/restraintoverview.aspx).
“More parents than ever now realize that
kids need the help of a booster seat to make
sure the belt fits properly across the bony
parts of their lap and shoulder rather than
across the soft belly or the neck, which are
more prone to injury,” Arbogast says.
percent), do require child restraints or
boosters for children through age 7.
CHOP researchers found that parents
aren’t widely using lower anchors and tethers for children, or LATCH, which are supposed to make it easier to attach infant and
child restraints securely to vehicle seats
(see Status Report, Jan. 16, 1999; on the web
at iihs.org). LATCH has been required in
new vehicles and on child restraints since
2002. However, only 43 percent of all children buckled into restraints in vehicles
equipped with LATCH in 2007 were riding in
seats attached to the lower anchors, the
CHOP study reports.
highest rates of injury. Nearly half of all
crashes involving children occur on roads
with posted speed limits of 25 to 44 mph.
Although the American Academy of Pediatrics recommends that children younger
than 13 ride in the back seats of vehicles,
about 30 percent of all 8-12 year-olds ride in
the front. Positioning children in back seats
reduces the risk of fatal injuries in crashes
by about one-third among kids 12 and
younger (see Status Report, June 27, 1997; on
the web at iihs.org).
For a copy of the September 2008 “Partners for Child Passenger Safety: fact and
trend report” go to www.chop.edu/carseat.
SR43-9 TO PDF:SR 43-9
10/16/08
11:37 AM
Page 6
6
Status Report, Vol. 43, No. 9, Oct. 22, 2008
FASTER, HEAVIER GOLF CARTS
GET THUMBS DOWN FROM
FEDERAL REGULATORS
The federal government says it isn’t willing to trade highway safety for fuel economy in denying 4 petitions that sought to increase the maximum gross vehicle
weight for low-speed vehicles and also launch a class of medium-speed vehicles. The September decisions by the National Highway Traffic
Safety Administration (NHTSA) mean that the weight and
top speed of these golf cart-like vehicles, which don’t have
to meet all the safety rules that apply to cars, remain
capped at 3,000 pounds and 25 mph.
“While NHTSA agrees with the importance of environmental issues, the agency believes that it is neither necessary nor appropriate to significantly increase the risk of deaths and serious injuries to
save fuel,” the agency said in denying petitions from Environmental
Motors, Porteon Electric Vehicles Inc., and Mirox Corporation. The companies asked the agency to create a class of medium-speed vehicles with
speed capabilities of up to 35 mph, arguing they’d fill a need for fuel-efficient vehicles for use in fast urban traffic.
They’re known as neighborhood electric vehicles, street-legal golf
carts, and minitrucks, among other names (see Status Report,
April 6, 2002; on the web at iihs.org). These electric or gasoline-powered low-speed vehicles are designed to haul people
and cargo on private land, such as retirement communities,
farms, amusement parks, and construction sites, but they’re
often driven on public streets.
Forty-six states regulate their use, with most limiting their
speed to no greater than 25 mph (on the web at iihs.org/
laws) on public roadways with speed limits of no more than
35 mph. They’re exempt from most federal safety standards
that apply to cars, and they aren’t required to meet any criteria for vehicle crashworthiness, so they’d be out of their
league in crashes with other vehicles going 35 mph.
Electronic Transportation Applications had sought to
increase the maximum allowable gross vehicle weight for
electric-powered low-speed vehicles to 4,000 pounds.
In denying Electronic Transportation Applications’ petition,
NHTSA said, “We believe that vehicles over 3,000 pounds are
capable of complying with the full requirements” of the federal motor
vehicle safety standards. Increasing the allowable gross vehicle weight, NHTSA said,
“would encourage the use of [low-speed vehicles] in circumstances where it could pose an
unreasonable risk to safety.” The agency noted that some of the smallest passenger cars —
Honda Insight and Toyota Echo, for example — have gross vehicle weights of about 3,000
pounds or less and still comply with safety standards. (Read both decisions at http://edocket.access.gpo.gov/2008/pdf/E8-22736.pdf and 22737.pdf.)
Federal crash databases don’t include a specific category for low-speed vehicles so it’s
hard to track their crashes. News reports frequently chronicle deaths and injuries that result
when these vehicles collide with larger passenger vehicles.
SR43-9 TO PDF:SR 43-9
10/16/08
11:37 AM
Page 7
NEW ‘UNDERSTANDING CRASHES’ VIDEO
Why do some car crashes produce only minor injuries? How can a single
crash of a car into a wall involve 3 separate collisions? Award-winning
science educator Griff Jones visits the Institute’s Vehicle Research Center
to answer these and other questions in a 24-minute video that’s a follow-up
to a previous Institute production, “Understanding car crashes: it’s basic
physics” (2000). In the new video, Jones examines the laws of nature that
determine what happens to the human body in a crash. Order “Understanding
car crashes: when physics meets biology” ($35) online at iihs.org/videos.
SR43-9 TO PDF:SR 43-9
10/16/08
11:37 AM
Page 8
NON-PROFIT ORG.
U.S. POSTAGE
PAID
PERMIT NO. 252
ARLINGTON, VA
1005 N. Glebe Rd., Arlington, VA 22201
Phone 703/247-1500 Fax 247-1588
Internet: www.iihs.org
Vol. 43, No. 9, Oct. 22, 2008
Antilock brakes on motorcycles reduce
both crash frequencies and deaths ...........1
2008 motorcycles with antilocks .....................3
Text messaging bans now cover all drivers in
7 states and the District of Columbia .............4
Child safety seat use in crashes has increased
during the past decade ....................................4
Souped-up golf carts will remain subject to
limitations on weight and speed, the federal
government decides ........................................6
New Institute video explores what happens
to the human body in a car crash ...................7
Contents may be republished with attribution.
This publication is printed on recycled paper.
The Insurance Institute for Highway Safety is a
nonprofit scientific and educational organization
dedicated to reducing deaths, injuries, and property
damage from crashes on the nation’s highways.
The Institute is wholly supported by auto insurers:
21st Century Insurance
AAA Mid-Atlantic Insurance Group
Affirmative Insurance
AIG Agency Auto
AIG Direct
Alfa Insurance
Alfa Alliance Insurance Corporation
Allstate Insurance Group
American Family Mutual Insurance
American National Property and Casualty
Ameriprise Auto & Home
Amerisure Insurance
Amica Mutual Insurance Company
Auto Club Group
Auto Club South Insurance Company
Bituminous Insurance Companies
Bristol West Insurance
Brotherhood Mutual
California State Automobile Association
Capital Insurance Group
Chubb Group of Insurance Companies
Concord Group Insurance Companies
Cotton States Insurance
COUNTRY Financial
Countrywide Insurance Group
Erie Insurance Group
Esurance
Farm Bureau Financial Services
Farm Bureau Mutual Insurance Company of Idaho
Farmers Insurance Group of Companies
Farmers Mutual of Nebraska
First Acceptance Corporation
Florida Farm Bureau Insurance Companies
Frankenmuth Insurance
The GEICO Group
Gainsco
General Casualty Insurance Companies
Georgia Farm Bureau Insurance
GMAC Insurance
Grange Insurance
The Hartford
Hanover Insurance Group
High Point Insurance Group
Homeowners of America
ICW Group
Indiana Farm Bureau Insurance
Kemper, a Unitrin Business
Kentucky Farm Bureau Insurance
Liberty Mutual
The Main Street America Group
Markel Corporation
Mercury Insurance Group
MetLife Auto & Home
Michigan Insurance Company
MiddleOak
MMG Insurance
Mutual of Enumclaw Insurance Company
Nationwide Insurance
N.C. Farm Bureau Mutual Insurance Company
Nodak Mutual Insurance
Norfolk & Dedham Group
Ohio Casualty Group
Oklahoma Farm Bureau Mutual Insurance Company
Oregon Mutual Insurance
Palisades Insurance
Pekin Insurance
PEMCO Insurance
The Progressive Corporation
Response Insurance
Rockingham Group
Safeco Insurance
Samsung Fire & Marine Insurance Company
S.C. Farm Bureau Mutual Insurance Company
SECURA Insurance
Shelter Insurance
Sompo Japan Insurance Company of America
State Auto Insurance Companies
State Farm
Tennessee Farmers Mutual Insurance Company
Tokio Marine Nichido
The Travelers Companies
Unitrin
USAA
Virginia Farm Bureau Mutual Insurance
West Bend Mutual Insurance Company
Zurich North America
FUNDING ASSOCIATIONS
American Insurance Association
National Association of Mutual Insurance Companies
Property Casualty Insurers Association of America
Highway Loss Data Institute Bulletin
Motorcycle Antilock Braking System (ABS)
VOL. 25, NO. 1
Antilock braking systems (ABS) on motorcycles are designed to
allow riders to make urgent, yet controlled stops. Motorcycles
are fundamentally unbalanced; they are kept stable at very low
speeds by a rider holding the handlebar and maintaining balance. At higher speeds, stability comes almost exclusively from
the gyroscopic effect of the wheels. While at speed, if one of
the wheels stops rotating for a fraction of a second, the result is
immediate instability. The effect is more pronounced if it happens to the front wheel, where a fall is almost inevitable —
especially while cornering or leaning the motorcycle. ABS has
independent braking sensors for each wheel. If the system
detects a locked wheel, it releases the brake to allow that tire
to retain grip before reapplying the brake. ABS then modulates
braking pressure to achieve optimum braking.
This Highway Loss Data Institute (HLDI) bulletin compared the
collision losses of 12 motorcycle models available with optional ABS. Model years of the motorcycles studied ranged from
2003 to 2007. Significant reductions in collision claim frequencies and overall losses were found for motorcycles
equipped with ABS. No significant reductions were found for
claim severity.
For motorcycles to be included in the study, their vehicle identification numbers (VINs) had to have an ABS indicator. This
allowed for very tight control over the study population. Twelve
motorcycles met this criterion. It should be noted that there
were motorcycles available with ABS that were not included in
the study because their VINs did not have an ABS indicator.
Data were collected by make and series, rated driver age and
gender, and vehicle age and density. Vehicle age was defined
as the difference between calendar year and model year measured in years. Previous HLDI studies have shown that all of
these factors have a significant impact on insurance losses. For
insurance purposes, a rated driver is assigned to each vehicle
on a policy. Information on the actual driver at the time of a
loss is not available in the HLDI database. For the present study
data were stratified by rated driver age group (<25, 25-39, 4064, 65+, or unknown) and gender (male, female, or unknown).
The dataset also was stratified by make/series and vehicle density (<100, 100-499, and 500+ vehicles per square mile). For
example, a 1-year-old Honda Gold Wing, ABS equipped, with
a 40-64 year old male as the rated driver, and garaged in an
area with a vehicle density of 100-499 vehicles per square mile
constituted one unit of observation. The distribution of motorcycle collision exposure for the six independent variables is
listed in Appendix A. It is important to note that rated driver
factors and vehicle density were included to control for their
potential impact on losses and not to produce estimates for
APRIL 2008
those variables. The estimated parameters for those variables
may not generalize from this subset to the much larger motorcycle population.
Regression analysis was used to quantify the effect of ABS on
motorcycle collision losses while controlling for other covariates. Claim frequency was modeled using a Poisson distribution, whereas claim severity was modeled using a Gamma distribution. Both models used a logarithmic link function.
Estimates for collision overall losses were derived from the
claim frequency and severity models. Reference categories for
the categorical independent variables were assigned to the values with the highest exposure. The reference categories were as
follows: make/series = Honda Gold Wing, ABS = without ABS,
rated driver age range = 25-39, vehicle density = 100-499 vehicles per square mile, and rated driver gender = male. Losses for
each unit of observation were weighted by the exposure in the
linear regression. The key independent variable in the model,
ABS, was treated as categorical. Models were constructed that
examined the interaction of the rated driver factors and vehicle
density with the presences or absence of ABS. None of those
interactions were found to be significant.
Summary results of the regression analysis of motorcycle collision claim frequencies using the Poisson distribution are listed
in Table 1. Results for all independent variables in the model,
including ABS, had p-values less then 0.05, indicating their
effects on claim frequencies were statistically significant.
Detailed results of the regression analysis using claim frequency as the dependent variable are listed in Table 2. The table
shows estimates and significance levels for the individual values of the categorical variables. To make results more illustrative, a column was added that contains the exponents of the
estimates. The exponent of the intercept equals 0.0000826
claims per day, or about 3 claims per 100 insured vehicle
years. The intercept outlines losses for the reference (baseline)
categories: the estimate corresponds to the claim frequency for
a new Honda Gold Wing without ABS, garaged in a medium
vehicle density area, and driven by a male age 25-39. The
remaining estimates are in the form of multiples, or ratios relative to the reference categories. For example, the estimate corresponding to female gender equals 0.81, so female rated drivers had estimated claim frequencies 19 percent lower than
those for male rated drivers.
The estimate corresponding to motorcycle ABS (-0.22) was
highly significant (p=0.002). The estimate corresponded to a 19
percent reduction in claim frequencies for motorcycles
equipped with ABS.
TABLE 1 SUMMARY RESULTS OF LINEAR REGRESSION ANALYSIS
OF COLLISION CLAIM FREQUENCIES
DEGREES
OF FREEDOM
Vehicle Age
1
CHI-SQUARE
P-VALUE
196.270
< 0.0001
< 0.0001
Rated Driver Age
4
57.040
Vehicle Density
2
18.150
0.0001
Vehicle Make/Series
11
292.300
< 0.0001
Rated Driver Gender
2
8.060
0.0178
ABS
1
9.580
0.0020
TABLE 2 DETAILED RESULTS OF LINEAR REGRESSION ANALYSIS
OF COLLISION CLAIM FREQUENCIES
PARAMETER
ESTIMATE
EXPONENT
(ESTIMATE)
STANDARD
ERROR
CHISQUARE
INTERCEPT
VEHICLE AGE
RATED DRIVER AGE
Unknown
14-24
25-39
40-64
65+
VEHICLE DENSITY
0-99
100-499
500+
RATED DRIVER GENDER
Female
Male
Unknown
VEHICLE MAKE/SERIES
Honda Gold Wing
Honda Interceptor 800
Honda Reflex
Honda Silver Wing
Honda ST1300
Suzuki Bandit 1250
Suzuki Burgman 650
Suzuki SV650
Suzuki V-Strom 650
Triumph Sprint ST
Triumph Tiger
Yamaha FJR1300
ABS
ABS Model
Non-ABS Model
-9.4014
-0.2628
3.015
0.769
0.0826
0.0192
12948.8
187.54
<0.0001
<0.0001
0.1791
0.5839
0
-0.1433
-0.0747
1.196
1.793
1.000
0.866
0.928
0.0925
0.094
0
0.0649
0.1026
3.75
38.58
0.0529
<0.0001
4.88
0.53
0.0272
0.4663
-0.1332
0
0.122
0.875
1.000
1.130
0.0582
0
0.0514
5.23
0.0222
5.62
0.0177
-0.205
0
0.0431
0.815
1.000
1.044
0.0845
0
0.0527
5.88
0.0153
0.67
0.4135
0
0.8988
0.5596
0.7658
0.1745
1.4525
0.7284
1.0251
-0.084
1.077
0.3632
0.3276
1.000
2.457
1.750
2.151
1.191
4.274
2.072
2.787
0.919
2.936
1.438
1.388
0
0.1077
0.1093
0.1025
0.1128
0.2295
0.0954
0.0747
0.1352
0.1357
0.2177
0.0949
69.6
26.2
55.81
2.39
40.06
58.29
188.54
0.39
63
2.78
11.92
<0.0001
<0.0001
<0.0001
0.1218
<0.0001
<0.0001
<0.0001
0.5345
<0.0001
0.0953
0.0006
-0.2151
0
0.806
1.000
0.0709
0
9.19
0.0024
P-VALUE
Motorcycle collision claim frequencies increased with increases in vehicle density. Claim frequencies in high vehicle
density areas were estimated to be 13 percent higher (p=0.02) than those in medium vehicle density areas, whereas claim
frequencies in low vehicle density areas were estimated to be 13 percent lower (p=0.02). Claim frequencies were estimated to decrease 23 percent (p<0.0001) for each 1-year increase in vehicle age.
Individual make/series motorcycles were included in the model, and estimates of their effect on collision claim frequencies were found to be significant. As previously mentioned, the reference category for the make/series variable was the
Honda Gold Wing. Significant predictions for make/series ranged from 1.39 for the Yamaha FJR 1300 to 4.27 for the
Suzuki Bandit 1250. All make/series estimates were significant at the p=0.001 level except for the Honda ST1300, Suzuki
V-Strom 650, and Triumph Tiger. The lack of significance for these individual values of the make/series variable indicate
that estimated claim frequencies for these motorcycles were similar to the Honda Gold Wing (reference category).
Driver age was highly significant in predicting motorcycle collision claim frequency. Estimated claim frequencies for rated
drivers 24 and younger were 79 percent higher (p<0.0001) than those for rated drivers ages 25-39 (reference category),
whereas estimated claim frequencies for rated drivers ages 40-64 were 13 percent lower (p=0.03). The estimated 7 percent decrease in claim frequency for rated drivers 65 and older was not significant. Driver gender also significantly predicted collision claim frequencies. Estimated claim frequencies for female rated riders were 19 percent lower (p=0.02)
than those for male rated riders.
Summary results of the regression analysis of motorcycle collision claim severities using the Gamma distribution are listed in Table 3. Of the six variables included in the analysis, only vehicle age and make/series had p-values less than 0.05.
Neither the rated driver nor the driving environment affects the claim size.
TABLE 3 SUMMARY RESULTS OF LINEAR REGRESSION ANALYSIS
OF COLLISION CLAIM SEVERITIES
DEGREES
OF FREEDOM
CHI-SQUARE
P-VALUE
Vehicle Age
1
12.400
0.0004
Rated Driver Age
4
3.420
0.4907
2
2.490
0.2880
Vehicle Make/Series
Vehicle Density
11
316.960
< 0.0001
Rated Driver Gender
2
0.190
0.9075
ABS
1
0.070
0.7953
Detailed results of the regression analysis using motorcycle collision claim severity as the dependent variable are listed
in Table 4. The structure of the table, as well as the variables and reference categories, are the same as those used for
claim frequency in Table 2. The variables and reference categories that were used for claim frequency were used for claim
severity. The exponent of the intercept equals $9,089. The intercept outlines losses for the reference (baseline) categories:
the estimate corresponds to the claim severity for a new Honda Gold Wing without ABS, garaged in a medium vehicle
density area, and driven by a male age 25-39.
The estimate corresponding to the ABS effect, a 2 percent decrease in claim severity, was highly nonsignificant (p=0.8),
indicating ABS does not affect claim severity. As previously mentioned, vehicle age and make/series were significant predictors of claim severity. Not surprisingly, as motorcycles age their claim severities decrease. The model estimated a 6 percent decrease (p=0.0004) in claim severity per 1-year increase in vehicle age. Estimated claim severities for the 11
make/series motorcycles, compared with those for the Honda Gold Wing (reference category), ranged from 21 percent
lower for the Honda ST1300 to 75 percent lower for the Honda Reflex. All of the make/series estimates were significant
at the p<0.05 level.
TABLE 4 DETAILED RESULTS OF LINEAR REGRESSION ANALYSIS
OF COLLISION CLAIM SEVERITIES
EXPONENT
(ESTIMATE)
STANDARD
ERROR
CHISQUARE
PARAMETER
ESTIMATE
P-VALUE
INTERCEPT
VEHICLE AGE
RATED DRIVER AGE
Unknown
14-24
25-39
40-64
65+
VEHICLE DENSITY
0-99
100-499
500+
RATED DRIVER GENDER
Female
Male
Unknown
VEHICLE MAKE/SERIES
Honda Gold Wing
Honda Interceptor 800
Honda Reflex
Honda Silver Wing
Honda ST1300
Suzuki Bandit 1250
Suzuki Burgman 650
Suzuki SV650
Suzuki V-Strom 650
Triumph Sprint ST
Triumph Tiger
Yamaha FJR1300
ABS
ABS Model
Non-ABS Model
9.1148
-0.0608
9088.817
0.941
0.0849
0.0172
11515.1
12.57
<0.0001
0.0004
0.0819
0.1743
0
0.0624
0.0415
1.085
1.190
1.000
1.064
1.042
0.0838
0.1029
0
0.0586
0.0866
0.95
2.87
0.3286
0.0905
1.14
0.23
0.2866
0.6317
-0.0342
0
0.0553
0.966
1.000
1.057
0.0563
0
0.0514
0.37
0.5432
1.16
0.2822
0.0319
0
0.0078
1.032
1.000
1.008
0.0728
0
0.0501
0.19
0.6613
0.02
0.8766
0
-0.5373
-1.3808
-1.1479
-0.2339
-0.8037
-0.9265
-0.9031
-0.989
-0.5166
-0.5712
-0.4631
1.000
0.584
0.251
0.317
0.791
0.448
0.396
0.405
0.372
0.597
0.565
0.629
0
0.0928
0.0916
0.0922
0.0989
0.1774
0.0908
0.0801
0.1179
0.1154
0.1755
0.0863
33.54
227.27
154.93
5.59
20.53
104.19
127.02
70.33
20.03
10.59
28.81
<0.0001
<0.0001
<0.0001
0.018
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.0011
<0.0001
-0.0155
0
0.985
1.000
0.0597
0
0.07
0.795
Table 5 summarizes the effects of the independent variables on motorcycle collision overall losses, derived from the claim
frequency and severity models. Overall losses can be calculated by simple multiplication because the estimates for the
effect of ABS on claim frequency and severity were in the form of ratios relative to the reference (baseline) categories. The
standard error for overall losses can be calculated by taking the square root of the sum of the squared standard errors for
claim frequency and severity. Based on the value of the estimate and the associated standard error, the level of statistical
significance (p-value) can be obtained from a probability distribution table.
The estimated effect of ABS was a 21 percent decrease in collision overall losses that was significant (p=0.01). This is a
strong indication that ABS is effective in reducing collision overall losses for motorcycles.
Motorcycle collision overall losses were predicted to increase with increased vehicle density. Estimated overall losses in
high vehicle density areas were 19 percent higher (p=0.01) than those medium vehicle density in areas (reference category), whereas estimated overall losses in low vehicle density areas were 15 percent lower (p=0.04). Estimated overall
losses for the 11 make/series motorcycles, compared with those for the Honda Gold Wing (reference category), ranged
from 66 percent lower for the Suzuki V-Strom 650 to 91 percent higher for the Suzuki Bandit 1250. Only about half of
the make/series estimates were statistically different from the reference category.
Driver age was a significant predictor of motorcycle collision overall losses. Estimated overall losses for rated drivers 24
and younger were 113 percent higher (p<0.0001) than those for rated drivers ages 25-39 (reference category). Rated drivers ages 40-64 and 65 and older had slightly lower estimated overall losses, but these estimates did not reach statistical
significance. Vehicle age also was significantly predictive of collision overall losses. Overall losses decreased by an estimated 28 percent (p<0.0001) for each 1-year increase in vehicle age.
TABLE 5 RESULTS FOR COLLISION OVERALL LOSSES DERIVED FROM CLAIM FREQUENCY AND SEVERITY MODELS
PARAMETER
INTERCEPT
VEHICLE AGE
RATED DRIVER AGE
Unknown
1. 14-24
25-39
40-64
65+
DENSITY
0-99
100-499
500+
RATED DRIVER GENDER
Female
Male
Unknown
VEHICLE MAKE/SERIES
Honda Gold Wing
Honda Interceptor 800
Honda Reflex
Honda Silver Wing
Honda ST1300
Suzuki Bandit 1250
Suzuki Burgman 650
Suzuki SV650
Suzuki V-Strom 650
Triumph Sprint ST
Triumph Tiger
Yamaha FJR1300
ABS
ABS Model
Non-ABS Model
FREQUENCY
STANDARD
ESTIMATE ERROR
SEVERITY
STANDARD
ESTIMATE
ERROR
ESTIMATE
OVERALL LOSSES
STANDARD EXPONENT
ERROR
(ESTIMATE)
P-VALUE
-9.4014
-0.2628
0.0826
0.0192
9.1148
-0.0608
0.0849
0.0172
-0.2866
-0.3236
0.1185
0.0258
0.7508
0.7235
0.0155
<0.0001
0.1791
0.5839
0
-0.1433
-0.0747
0.0925
0.094
0
0.0649
0.1026
0.0819
0.1743
0
0.0624
0.0415
0.0838
0.1029
0
0.0586
0.0866
0.2610
0.7582
0
-0.0809
-0.0332
0.1248
0.1394
0
0.0874
0.1343
1.2982
2.1344
1.0000
0.9223
0.9673
0.0365
<0.0001
-0.1332
0
0.122
0.0582
0
0.0514
-0.0342
0
0.0553
0.0563
0
0.0514
-0.1674
0
0.1773
0.0810
0
0.0727
0.8459
1.0000
1.1940
0.0387
-0.205
0
0.0431
0.0845
0
0.0527
0.0319
0
0.0078
0.0728
0
0.0501
-0.1731
0
0.0509
0.1115
0
0.0727
0.8411
1.0000
1.0522
0.1207
0.4839
0
0.8988
0.5596
0.7658
0.1745
1.4525
0.7284
1.0251
-0.084
1.077
0.3632
0.3276
0
0.1077
0.1093
0.1025
0.1128
0.2295
0.0954
0.0747
0.1352
0.1357
0.2177
0.0949
0
-0.5373
-1.3808
-1.1479
-0.2339
-0.8037
-0.9265
-0.9031
-0.989
-0.5166
-0.5712
-0.4631
0
0.0928
0.0916
0.0922
0.0989
0.1774
0.0908
0.0801
0.1179
0.1154
0.1755
0.0863
0
0.3615
-0.8212
-0.3821
-0.0594
0.6488
-0.1981
0.1220
-1.0730
0.5604
-0.2080
-0.1355
0
0.1422
0.1426
0.1379
0.1500
0.2901
0.1317
0.1095
0.1794
0.1781
0.2796
0.1283
1.0000
1.4355
0.4399
0.6824
0.9423
1.9132
0.8203
1.1298
0.3420
1.7514
0.8122
0.8733
0.0110
<0.0001
0.0056
0.6921
0.0253
0.1325
0.2653
<0.0001
0.0017
0.4570
0.2908
-0.2151
0
0.0709
0
-0.0155
0
0.0597
0
-0.2306
0
0.0927
0
0.7941
1.0000
0.3549
0.8047
0.0147
0.0128
APPENDIX A DISTRIBUTION OF EXPOSURE FOR INDEPENDENT VARIABLES
VEHICLE AGE
0
1
2
3
4
5
RATED DRIVER AGE
Unknown
14-24
25-39
40-64
65+
VEHICLE DENSITY
0-99
100-499
500+
RATED DRIVER GENDER
Female
Male
Unknown
VEHICLE MAKE/SERIES
EXPOSURE
PERCENT OF TOTAL
291
12,177
20,388
18,147
12,748
8,724
0%
17%
28%
25%
18%
12%
4,886
1,696
11,274
47,687
6,933
7%
2%
16%
66%
10%
21,281
29,331
21,863
29%
40%
30%
7,034
43,147
22,294
10%
59%
31%
EXPOSURE
ABS
WITHOUT
Honda Gold Wing (Touring)
28,904
Honda Interceptor 800 (Sport)
1,925
Honda Reflex (Scooter)
3,036
Honda Silver Wing (Scooter)
2,882
Honda ST1300 (Sport)
3,118
Suzuki Bandit 1250 (Standard)
136
Suzuki Burgman 650 (Scooter)
3,219
Suzuki SV650 (Unclad Sport)
8,468
Suzuki V-Strom 650 (Dual Purpose)
2,986
Triumph Sprint ST (Sport)
908
Triumph Tiger (Dual Purpose)
695
Yamaha FJR1300 (Sport)
3,610
Total
59,886
PERCENT
OF SERIES
82%
76%
87%
85%
68%
80%
92%
100%
97%
82%
94%
59%
83%
EXPOSURE
ABS
WITH
6,366
623
443
506
1,468
33
288
11
90
198
42
2,520
12,589
PERCENT
OF SERIES
18%
24%
13%
15%
32%
20%
8%
0%
3%
18%
6%
41%
17%
The Highway Loss Data Institute is a nonprofit public service organization that gathers, processes, and publishes insurance data on the human and economic losses
associated with owning and operating motor vehicles.
COPYRIGHTED DOCUMENT, DISTRIBUTION RESTRICTED © 2008 by the Highway Loss Data Institute, 1005 N. Glebe Road, Arlington, VA 22201. All rights
reserved. Distribution of this report is restricted. No part of this publication may be reproduced, or stored in a retrieval system, or transmitted, in any form or by
any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the copyright owner. Possession of this publication
does not confer the right to print, reprint, publish, copy, sell, file, or use this material in any manner without the written permission of the copyright owner.
Permission is hereby granted to companies that are supporters of the Highway Loss Data Institute to reprint, copy, or otherwise use this material for their own business
purposes, provided that the copyright notice is clearly visible on the material.
HIGHWAY LOSS
DATA INSTITUTE
1005 North Glebe Road
Arlington, VA 22201
Effectiveness of Antilock Braking Systems
in Reducing Fatal Motorcycle Crashes
Eric R. Teoh
October 2008
ABSTRACT
The effect of antilock braking systems (ABS) on motorcyclist fatal crash risk in 2005-06 was
studied by comparing fatal crash rates per registrations of motorcycles with and without ABS. Study
motorcycles included those for which ABS was optional equipment and could be identified as present by
the model name. Fatal motorcycle crashes per 10,000 registered vehicle years were 38 percent lower for
ABS models than for their non-ABS versions.
INTRODUCTION
Annual motorcyclist deaths in the United States have more than doubled, from 2,077 in 1997 to
4,697 in 2006 (Insurance Institute for Highway Safety, 2008), and motorcycle registrations have increased
by about two-thirds, from 5,167,693 in 2000 (earliest year for which data are available) to 8,642,243 in
2006, according to data from R.L. Polk and Company. Many factors contribute to motorcycle crashes,
but improper braking was identified as a major pre-impact factor in a study of motorcycle crash causation
(Hurt et al., 1981) and again, 20 years later, in the Motorcycle Accident In-Depth Study (MAIDS)
(Association of European Motorcycle Manufacturers, 2004).
Operating the brakes on most motorcycles is much more complicated than on four-wheel
vehicles. Most motorcycles have separate controls for the front and rear brakes, with the front brake
usually controlled by a lever on the right handlebar and the rear brake controlled by a pedal operated by
the rider’s right foot. During braking, a rider must decide how much force to apply to each control. As
with other types of vehicles, much more deceleration can be obtained from braking the front wheel than
from braking the rear wheel.
Motorcycles are inherently less stable than four-wheel vehicles and rely on riders’ skills to remain
upright during extreme maneuvers such as hard braking. Braking too hard and locking a wheel creates an
unstable situation. Locking the front wheel is particularly dangerous, with loss of control being almost
certain. A locked rear wheel is more controllable but still can lead to loss of control if the rider
simultaneously tries to steer the motorcycle, as in an emergency avoidance maneuver. However, in an
emergency requiring full stopping power, riders concerned about wheel lock may be reluctant to apply
full force to the brakes, particularly to the front brake. Both Hurt et al. (1981) and MAIDS (Association
of European Motorcycle Manufacturers, 2004) had examples of both loss of control due to wheel lock and
failure to adequately brake.
Although proper braking practices can be taught, rider training courses have not been shown to be
effective in reducing motorcycle crashes (Mayhew and Simpson, 1996) or have provided mixed results at
best (Billheimer, 1998). In an effort to address the issue of under-braking (especially the front wheel),
manufacturers have developed braking systems that essentially link the front and rear brake controls.
These systems, collectively referred to here as combined braking systems (CBS), apply braking force to
1
both wheels when either control is engaged. The degree to which braking force is applied to the front
wheel, for example, when the pedal for the rear brake is pressed varies by design, but the concept is the
same. CBS has been shown to reduce stopping distances of experienced riders on closed test tracks
(Green, 2006) and would be expected to be beneficial in situations in which a rider under-brakes (or does
not brake) the front wheel to avoid locking it or causing the motorcycle to pitch forward. Even with CBS,
however, it still is possible to lock a wheel during hard braking, often with catastrophic consequences.
ABS has been developed to help riders solve this dilemma. The system monitors wheel speed
and reduces brake pressure when impending wheel-lock is detected. Brake pressure is increased when
traction is restored, and the system evaluates and adjusts brake pressure many times per second. These
systems allow riders to apply brakes fully in an emergency without fear of wheel-lock. ABS was first
developed for commercial aircraft in 1929 (Maslen, 2008) and was first implemented in production
automobiles with the 1971 Chrysler Imperial (Douglas and Schafer, 1971). BMW was the first
manufacturer to implement ABS on a motorcycle with its K100RS Special model in 1988 (Tuttle, 2001).
ABS and CBS are not necessarily related; either or both can be implemented on a motorcycle.
ABS has not significantly reduced crash risk for passenger vehicles (Farmer et al., 1997; Farmer,
2001), but there is reason to expect ABS will be more helpful to motorcycles because of the instability
that occurs with any wheel-lock. Studies conducted on closed test tracks have demonstrated that ABS can
reduce the motorcycle stopping distances (Green, 2006; Vavryn and Winkelbauer, 2004). It is clear that
reducing wheel-lock is crucial in maintaining stability during hard braking. These results suggest that
ABS has the potential to reduce motorcycle crashes in real-world situations. Serious motorcycle crashes
identified from insurance claims were analyzed in a small study to determine, by crash reconstruction,
how certain crashes would be affected by ABS (Allianz Center for Technology, 2005). About half of the
200 crashes studied were deemed to be relevant to ABS, and the majority of those involved another
vehicle violating a motorcyclist’s right-of-way. Crash reconstruction analyses showed that between 8 and
17 percent of these crashes could have been avoided had the motorcycles been equipped with ABS. No
results were presented on how increased stability or stopping power provided by ABS might have
decreased the severities of the crashes that were deemed inevitable.
A study by the Highway Loss Data Institute (HLDI), conducted in conjunction with the present
study, found that motorcycles equipped with optional ABS had 19 percent fewer insurance claims for
collision damage per insured vehicle year than the same motorcycle models without ABS (Moore and
Yan, in process). The goal of the present study was to evaluate the effectiveness of ABS in reducing the
rate of fatal motorcycle crashes on public roads in the United States. Specifically, rates of fatal crash
involvement per registered vehicle were compared for motorcycle models with and without ABS installed
as optional equipment.
2
METHODS
Data on fatal motorcycle crashes were extracted from the Fatality Analysis Reporting System
(FARS), a national census of fatal crashes occurring on public roads that is maintained by the National
Highway Traffic Safety Administration. Exposure data consisted of national motorcycle registration
records obtained from R.L. Polk and Company. Each vehicle record in both databases was indexed by its
vehicle identification number (VIN), which encodes vehicle information, and the first 10 digits of the
VINs were used to determine make, model name, and model year according to records in a motorcycle
features database created and maintained by HLDI. Vehicles with missing or invalid VINs were
excluded.
To be included in the study, a motorcycle model was required to have ABS as an option and the
presence of that option must have been discernable directly from the model name (e.g., Honda Gold Wing
vs. Honda Gold Wing ABS). This eliminated bias due to the comparison of different makes or styles of
motorcycles. Although ABS has been an option on BMW models for much longer than the study period,
the BMW systems cannot be identified from the model name alone. All BMW models were excluded.
The final study population (Table 1) included eight make/model motorcycles, each with both ABS and
non-ABS versions. Some vehicles were excluded due to zero registrations of the ABS model during the
study years in the Polk records. Because none of the study vehicles with ABS were available in model
year 2000 or earlier, the analysis was restricted to 2001 or later model year vehicles. Among the
motorcycles included, all of the Hondas (both ABS and non-ABS) were equipped with standard CBS;
CBS was not available on any of the others.
Table 1
Study motorcycles
Non-ABS motorcycles
ABS motorcycles
Model years
Make/model
Model years
Make/model
2001-06
Honda Gold Wing
2001-06
Honda Gold Wing ABS
2001-06
Honda Interceptor 800
2002-06
Honda Interceptor 800 ABS
2001-06
Honda Reflex
2001-06
Honda Reflex ABS
2003-06
Honda ST1300
2003-06
Honda ST1300 ABS
2002-06
Honda Silver Wing
2003-06
Honda Silver Wing ABS
2003-06
Suzuki Burgman 650
2006
Suzuki Burgman 650 ABS
2001-06
Triumph Sprint ST
2006
Triumph Sprint ST ABS
2003-05
Yamaha FJR1300
2004-06
Yamaha FJR1300 ABS
At the time this study was conducted, registration data were available only for 2000 and 2005-07,
and FARS data were not yet available for 2007. There were no registrations of the ABS versions of these
motorcycles in 2000. Therefore, data were analyzed for years 2005-06. Fatal crash rates per 10,000
registered vehicle years for each motorcycle model, both ABS and non-ABS versions, were calculated by
dividing 10,000 times the number of motorcyclist fatal crash involvements in 2005-06 by the number of
3
motorcycles registered during these years. Because registration counts spanned 2 years, the denominator
was interpreted as registered vehicle years instead of registrations.
Fatal crash rates per vehicle registrations for ABS and non-ABS motorcycle models were
compared by calculating a rate ratio (RR) equal to the crash rate for ABS models divided by the crash rate
for non-ABS models. If ABS has no effect, then the rate ratio should be 1.0. A rate ratio of less than 1.0
would indicate the fatal crash rate for ABS models is lower than the rate for non-ABS models. Similarly,
a rate ratio greater than 1.0 would indicate the fatal crash rate for ABS models is higher than the rate for
non-ABS models. One way to calculate the rate ratio is to calculate the rate for non-ABS motorcycles as
total crash involvements divided by total registered vehicle years, and analogously for ABS motorcycles.
However, to reduce any bias that may have occurred from relative differences in registrations among
motorcycle models, an alternative rate for non-ABS motorcycles was calculated as the weighted average
of fatal crash rates for each vehicle model, where the weights were taken as the number of registered
vehicle years of ABS-equipped motorcycles. Thus, for any given motorcycle model the ABS and nonABS fatal crash rates received the same weight in calculating the overall fatal crash rates for motorcycles
with and without ABS.
Ninety and 95 percent confidence intervals (CIs) for the rate ratios were calculated based on
standard error estimates derived by assuming that the number of fatal crash involvements follows a
Poisson distribution as follows:
Var(ln(RR)) = VarABS + Varnon-ABS
where VarABS =
∑8 1 X ,ABS
and Varnon-ABS =
∑8 1 r X ,non‐ABS
∑8 1 r X
with Xi,ABS as the number of fatal crash involvements of the ith model motorcycle
equipped with ABS, analogously for Xi,non-ABS, and ri as the ratio of registered vehicle
years of the ABS model to registered vehicle years of the non-ABS model for model i.
lower 95% confidence limit = exp[ln(RR) – 1.96(Var(ln(RR)))1/2]
upper 95% confidence limit = exp[ln(RR) + 1.96(Var(ln(RR)))1/2]
RESULTS
Table 2 presents fatal crash involvements, registered vehicle years, and the rate of fatal crash
involvements per 10,000 registered vehicle years for the study motorcycles during 2005-06. Motorcycles
manufactured by Honda, particularly the Gold Wing model, dominated the sample, but the pattern for all
but two of the motorcycles was a lower fatal crash rate for ABS-equipped motorcycles. Across all ABSequipped motorcycles, the rate of fatal crash involvements per 10,000 registered vehicle years was 4.1,
compared with 6.7 for the same motorcycles not equipped with ABS.
4
Table 2
Motorcycle fatal crash involvements and registered vehicle years, 2005-06
Non-ABS models
ABS models
Fatal crash
Registered
Rate
Fatal crash
Registered
involvements vehicle years
(x 104)
involvements vehicle years
Honda Gold Wing
63
93,608
6.7
6
19,547
Honda Interceptor 800
8
10,437
7.7
3
2,307
Honda Reflex
6
14,858
4.0
1
2,644
Honda ST1300
2
7,003
2.9
2
3,580
Honda Silver Wing
11
12,273
9.0
1
1,278
Suzuki Burgman 650
8
9,618
8.3
0
309
Triumph Sprint ST
1
4,476
2.2
0
135
Yamaha FJR1300
8
8,734
9.2
2
6,486
Total
107
161,007
6.7*
15
36,286
Rate
(x 104)
3.1
13.0
3.8
5.6
7.8
0.0
0.0
3.1
4.1
*Overall non-ABS rate is weighted by registered vehicle years of ABS-equipped motorcycles.
The effect of ABS on fatal crash involvement is given by the rate ratio estimate for ABSequipped motorcycles against non-ABS motorcycles. This estimate and associated 90 and 95 percent
confidence intervals are provided in Table 3. The rate ratio estimate corresponds to an approximate 38
percent reduction (computed as (RR-1)×100%) in the rate of fatal crash involvements per 10,000
registered vehicle years for the ABS models over the (weighted) non-ABS models.
Table 3
Estimated rate ratios and confidence intervals for those
estimates for comparing ABS and non-ABS fatal crash rates
Rate ratio
0.615
95% confidence interval
(0.352, 1.074)
90% confidence interval
(0.385, 0.982)
DISCUSSION
Results of this analysis provide evidence that ABS is effective in reducing fatal motorcycle
crashes. Study motorcycles with ABS had a fatal crash involvement rate 38 percent lower than that for
their non-ABS versions during the study years.
Although the estimated effect of 38 percent is large, it is not statistically significant at the
customary 0.05 level. ABS is a relatively recent option on motorcycles, and the option was purchased on
only 18 percent of registered study vehicles during 2005-06. More data are required to obtain a more
precise estimate of ABS effectiveness in reducing fatal motorcycle crashes. However, as the estimate
becomes more precise, it is quite likely that it will continue to indicate a benefit of ABS. If there were no
effect of ABS on fatal crash involvement, an estimate as large as the 38 percent reduction in this study
would be expected to occur by chance less than 10 percent of the time. Thus, there is considerable
confidence that ABS is preventing fatal crashes among motorcyclists. This confidence is bolstered by the
fact that a separate analysis of insurance collision coverage losses among crashes of all severities also
shows a reduction in crashes of about 19 percent for motorcycles equipped with ABS (Moore and Yan,
5
2008). These results provide confirmatory evidence of the expected benefit of ABS from engineering
principles, test-track trials, and a crash reconstruction analysis.
The substantial effectiveness estimate observed in this study is not, however, without limitations.
ABS was studied as optional equipment, so the cohort of motorcyclists who choose to purchase ABS may
differ from those who decline to purchase it. In particular, motorcyclists who choose ABS may be more
concerned about safety than those who decline, thus leading to lower fatal crash rates due to safer riding
practices. Because of the small sample of ABS-equipped motorcycles, it was not possible to carefully
examine how rider factors such as helmet use and speeding differ between the two groups. For example,
78 percent of the non-ABS riders were helmeted, compared with 60 percent (9 of 15) of the ABS riders.
Also, the prevalence of these factors is not known for riders of study motorcycles that were not involved
in fatal crashes. Therefore, it was not possible to study how such factors influenced the observed
reduction in fatal crash rate for ABS-equipped motorcycles. Aside from differences in rider factors, it is
also possible that riders who choose ABS accumulate more miles than those who decline, which would
result in an upward bias in the fatal crash rate for the ABS cohort relative to the non-ABS cohort. As may
have occurred in passenger vehicles (Grant and Smiley, 1993), motorcyclists may tend to drive ABS
motorcycles more aggressively than non-ABS motorcycles, also resulting in a higher than expected crash
rate for the ABS group. Without more extensive data, it is not possible to know the magnitude or nature
of any bias of the estimated rate ratio comparing crash rates for ABS and non-ABS motorcycles.
With or without ABS, CBS also may reduce the likelihood of certain types of crashes. However,
due to the small sample of non-CBS motorcycles in this study, the effect of CBS could not be evaluated.
Still, CBS is not expected to bias the results because the braking systems of the ABS and non-ABS
motorcycles differed only by whether or not they were equipped with ABS. In other words, each ABS/nonABS pair either did or did not have CBS. ABS showed a benefit in both the CBS and non-CBS groups,
suggesting the presence of CBS on some of the motorcycles did not confound the observed effect of ABS.
ABS cannot be expected to prevent or mitigate all types of crashes, as demonstrated in the
Allianz Center for Technology (2005) study. For example, ABS would not affect the outcome or
likelihood of a crash involving a motorcycle struck from behind. The small sample of ABS motorcycles
in FARS and the lack of detailed information on precrash events in FARS precluded examination of the
effects of ABS on crashes that would likely be influenced by its presence.
ACKNOWLEDGMENTS
Motorcycle VINs were decoded by Marvin Campbell of the Highway Loss Data Institute, which
also maintains the database used to decode these VINs. Adrian Lund, Anne McCartt, and Charles Farmer
of the Insurance Institute for Highway Safety contributed helpful comments and suggestions. This work
was supported by the Insurance Institute for Highway Safety.
6
REFERENCES
Allianz Center for Technology. (2005) AZT Accident Research on Effectiveness of ABS in Motorbikes.
Presented at 2005 RCAR Conference,
Association of European Motorcycle Manufacturers. 2004. MAIDS in-depth investigations of accidents
involving powered two wheelers. Final Report 1.2. Brussels, Belgium
Billheimer, J.W. 1998. Evaluation of the California motorcyclist safety program. Transportation
Research Record 1640:100-09.
Douglas, J. and Schafer, T. 1971. The Chrysler ‘Sure-Brake:’ The first production four-wheel anti-skid
system. SAE Technical Paper Series 710248. Warrendale, PA: Society of Automotive Engineers.
Farmer, C.M. 2001. New evidence concerning fatal crashes of passenger vehicles before and after adding
antilock braking systems. Accident Analysis and Prevention 33:361-69.
Farmer, C.M.; Lund, A.K.; Trempel, R.E.; and Braver, E.R. 1997. Fatal crashes of passenger vehicles
before and after adding antilock braking systems. Accident Analysis and Prevention 29:745-57.
Grant, B. and Smiley, A. 1993. Driver response to antilock brakes: a demonstration of behavioural
adaptation. Proceedings of the Canadian Multidisciplinary Road Safety Conference VIII. Ottawa, Ontario:
Transport Canada.
Green, D. 2006. A comparison of stopping distance performance for motorcycles equipped with ABS,
CBS and conventional hydraulic brake systems. Presented at the 2006 International Motorcycle Safety
Conference, Long Beach, CA.
Hurt, H.H.; Ouellet, J.V.; and Thom, D.R. 1981. Motorcycle accident cause factors and identification of
countermeasures; Volume 1: technical report. Report no DOT HS-805-862. Washington, DC: National
Highway Traffic Safety Administration.
Insurance Institute for Highway Safety. 2008. Fatality facts, 2006: motorcycles. Arlington, VA.
Available: http://www.iihs.org/research/fatality_facts_2006/motorcycles.html. Accessed: July 21, 2008.
Maslen, J. (Spring 2008) ABS: 30 years of life saving. Roadsafe Magazine. Available:
http://www.roadsafe.com/magazine/2008spring/abs.htm. Accessed: July 23, 2008.
Mayhew, D.R. and Simpson, H.M. 1996. Effectiveness and role of driver education and training in a
graduated licensing system. Ottawa, Ontario: Traffic Injury Research Foundation.
Moore, M. and Yan, Y. In process. Antilock braking systems for motorcycles and insurance collision
losses. Arlington, VA: Highway Loss Data Institute.
Tuttle, M. 2001. One track mind. Rider Magazine July;28(7), p. 8.
Vavryn, K. and Winkelbauer, M. 2004. Braking performance of experienced and novice motorcycle riders
– results of a field study. Presented at the 2004 International Conference on Transport and Traffic
Psychology, Nottingham, United Kingdom.
7