What can we figure out about the motorcyclist fatality?

Someone posted a comment and I thought deserved more attention:

Dennis wrote: “Curious thing about statistics. You can make things look the way you want, depending on what gets factored in”

Me: In my experience, one of the curious thing about statistics is that the only people who say things like “You can make things look the way you want, depending on what gets factored in” are those who don’t like what the statistics show. Having no proof to counter it–which is the intellectually responsible thing to do, they dismiss it with a statement like that. However, had you read the other entries on the motorcycle safety puzzle, you’d know that is one of the main problems with the motorcycle safety puzzle–The Things They Don’t Ask–or don’t bother to find out–or tell us if they know. But it also speaks to the difficulty (or at least expense) of finding out those things. All of which I’ve discussed ad nauseum.

But what I find amusing is that once you belittle the statistics, you ask not only for more of them but me to find more of them to prove your implied point (that the 5 safety messages actually work). But since I’m a very nice person, I’ll do it–but only for one year. You are welcome to do the work on as many years as you want:

Dennis: “The article said things like 56.7% were helmeted. But how many of those helmeted were not sober or had not taken a safety class. Of the fatalities, how many had followed all 5 of you [sic] safe riding items? Statistically you have only listed them out one item at a time.”

Me: I’m going to come back to the “how many had followed all 5” part. First of all, they aren’t my 5 safe riding items. They are the Motorcycle Safety Foundation safety messages–but they are ones that most safety experts would agree with–and it seems that you do as well. And had you read the other motorcycle safety puzzle piece items, you wouldn’t ask some of these questions–or you could use the data there (or read the material cited) and figure it out for yourself.

In 2008, 5,290 riders were killed. Of that total, 1,322.5 riders were not validly licensed–meaning 3,967.5 dead riders were. So, the vast majority of riders were following at least one of the safety messages. When it comes to training, depending on the state, up to 100 percent of riders are licensed through the waiver earned by passing a state (or state-approved) motorcycle training course. In Florida, for example, in 2009 all riders had to pass a training course to obtain a license.  Iow, in the majority of cases, practicing one to two of the safety messages did not prevent a fatal crash.

Of those 5,290 dead riders 1,587 had a BAC equal to or over the legal limit meaning up to 3,703 were following the safety message to ride sober (though the probability is likely to be high that some of them had a legal BAC at the time they crashed–even so, they were riding legally).

If the improperly licensed and illegally intoxicated cohorts perfectly overlapped (meaning that all improperly licensed riders also drank to excess), then then 264.5 of the dead drunk were legally licensed–meaning that they followed at least one safety message but not the ride sober one.

Of those 1,587 drunk and dead riders, 730.02 were both drunk and unhelmeted. If the overlap was perfect, that means at the very least 856.8 drunk riders were wearing helmets, once again, it reveals that some riders pick and choose among the safety messages–at least on occasion.

However, the above assumes a perfect world where those who don’t follow one safety rule follow none of them–something highly unlikely–particularly in universal helmet law states.

And if you go through media accounts (usually based on police reports) of motorcycle fatalities (as I have extensively) you’d realize that the par for the course isn’t as nice and neat as you’d like to believe. Good riders who make safe choices all too often–it’s unrighteous (in the old-fashioned biker sense), unfair and downright dangerous to try to reduce the carnage on the roads to such simplistic terms.

Dennis: Also, how many of the atgatt riders survived due to having on the safety equipment?

Me: NHTSA/FARS does not calculate the number of ATGATT riders in terms of injuries or deaths. Nor do any studies examine the difference between fatalities with helmet AND gear, gear but no helmet and helmet but no gear in terms of survival. But I would caution you not to confuse wearing a helmet with ATGATT as you seem to have done–as even the most cursory observation on a hot (or even cold day) will show, riders wear helmets and no gear and (at least some) gear and no helmets.

But the problem is this: While MSF’s safety message is ATGATT, in fact not even MSF nor NHTSA claims that gear (jacket, pants, boots, gloves)  prevents fatal injuries. Gear cannot prevent the kind of injuries (head or chest trauma, severed limbs or arteries or internal bleeding) that kill riders. What gear does do is reduce abrasions and (minor) lacerations and may minimize some sprains. It doesn’t prevent (most) fractures.

If you mean the helmet part, NHTSA estimates helmets would save 37%–however, going to your point about statistics–there’s many questions about how they determined that figure that many want answered.

If MAIDS findings can be generalized to American populations (big IF there), then we could guess that 23-50 percent of the time jackets, pants, gloves or footwear reduced injuries.And, in terms of the helmet, in 33.2 percent of the cases a helmet reduced injury and in 35.5 percent of the cases, it prevented injury.

Dennis: “Another factor that hasn’t been considered is rider experience. Less experienced riders are more likely to have an accident. What percentage of the helmeted riders had been riding less than a year?”

Me: While some studies show that riders with 6 mos or less riding experience are more likely to be in a crash, others have found that the  “more experience, less crashing” thing doesn’t hold up when the study takes into account variables such as mileage and how much of the year they ride and age. In fact, more experienced riders tend to ride more often, farther and in more riding conditions–and hence any safety experience gives is mitigated by increased exposure.

However, to my knowledge no studies have attempted to determine how much experience those who died had.

Dennis: “Riding does present more risk. But think about how many more injuries and fatalities there would be if all riders ignored the 5 messages that you list.”

“All riders”, “ignored” all 5? That’s a  meaningless thing to say as it’s impossible to prove or disprove. Nor was I, for one, suggesting that anyone let alone everyone should “ignore” the safety messages. How silly. What I am suggesting that riders shouldn’t take on more risk because they follow them or think they’re safe or even safer on the roads because they follow them.

I want to return to your statement, “how many had followed all 5 of you [sic] safe riding items?”  There is no magic formula for safe riding. Please note that MSF does not make a direct or even implied claim that following those 5 “key messages” is “safe riding” or will reduce anyone’s chances of being in a crash. If you–as others have–taken them to mean that, it’s on you.

Explore posts in the same categories: Motorcycle crashes, Motorcycle fatalities, Motorcycle injuries, Motorcycle licensing, Motorcycle Safety, Uncategorized

19 Comments on “What can we figure out about the motorcyclist fatality?”

  1. DataDan Says:

    FYI, I queried FARS ’08 for riders killed who: 1) had a valid driver’s license with motorcycle endorsement; 2) had BAC .00-.07; and 3) wore a helmet. There were 1155 of them. But that’s out of 3574 (not 5290) because many do not have BAC reported in FARS.

    So 30% of riders killed in 2008 for whom we know license status, helmet use, and BAC were following 3 of the 5 MSF recommendations: proper license, helmet, BAC < .08.

    Moon's excellent post #3 on the training puzzle inescapably leads one to conclude that training does NOT reduce crash rate. And since that's from international data as well as the US, MSF can hardly be kicked around for that failure.

    I'm not sure what to make of the MSF recommendation to ride within one's limits. Riding beyond one's limits can be defined only retrospectively: If you're sliding along the pavement on your leathered fanny, watching your scoot tumble over a cliff, you've exceeded your limits. Prospectively, I don't know of a useful definition.

  2. wmoon Says:

    DataDan: Interesting way to approach the problem–and, if correct, scary. I’d say that they were following 4 of the messages (at least) because they were riding within the legal limits–which can you very validly argue they could reasonably interpret as “within their limits”. Oullet, et. al., found that at 0.04 is reached, a rider performs as badly as a drivers–but that’s beside the point here.

    As for “from international data”, almost all those studies examine MSF curriculum. The curriculum that has shown a positive effect was the British Columbia curriculum, which is excellent. MSF can and should be kicked around for that failure. They knew about these studies (and, according to sources, tried to prevent some of those studies from being released–and their response (which I documented long ago on Journalspace but it’s gone now) is to remove any language that claimed training would reduce crashes or make the rider safer. That was a liability issue–and and FTC problem. The only place it still does that is in the driver’s license examiner’s guides–that I know of.

    I agree with you about the “ride within one’s limits” and I love the way you described how you know you exceeded them. But the message sounds nice and helps fill out the list–“4 key messages” sounds a little skimpy.

    I was just thinking of you last evening and here you comment today! That’s a coincidence!

  3. vstromer Says:

    Just to add to the “numbers” issue… As Wendy knows, but some readers and posters may not know, the numbers reported in FARS data are only as good as the data collected at the site of the crash. Having served on a taskforce that studied motorcycle fatalities, in talking with those in law enforcement, responsible for investigating crashes, there is agreement that critical data may not be collected at the time, and at the site, of the crash. Asking for collection of more information, like rider training level, what gear was worn, even BAC, goes way beyond what is sometimes captured by the investigating LEO. So, take the FARS data with a grain of salt. FARS is what we have, so it is what we use, but it is far from perfect (or accurate).

  4. wmoon Says:

    VStromer: Good points to bring up and especially with your experience on the task force. The same can be said of most motorcycle studies–they don’t ask the questions we really need to know since what they do ask doesn’t lead to a solution.
    Which is why we need a comprehensive accident causation study that asks about those things.

  5. Young Dai Says:

    Could it be that training, helmet use and sobriety merely have the greatest immediate affect upon a rider’s ability to avoid single vehicle crashes,or to prevent the rider from being sucked into potentially dangerous situations within the traffic flow, and to survive the injuries that may result if it does go wrong ? If you like they are Darwin’s low hanging fruit

    In the UK we have training requirements that are very different to your own, (including restrictions on engine output to 33bhp for up to two years having passed your test). Helmet use is mandatory (unless you are a Sikh), and DUI laws are vigorously enforced. Yet we find too ingenious ways of killing ourselves over here as well.

    The majority of UK deaths in this report http://www.dft.gov.uk/pgr/roadsafety/research/rsrr/theme5/indepthstudyofmotorcycleacc.pdf

    are from accidents with other vehicles, which although perhaps of comfort to some in the motorcycling community, is little help to those left behind.

    I am sorry if you have expanded this idea before: But may it be that if: the greatest danger to a motorcyclist comes from other road users and where operating your bike is not an automatic skill,i.e. you have to think about how to do what you intend to do, then you have nothing left to scan the traffic around for potential risk.

    Therefore learnt machine skills allow better hazard recognition as the brain has the capacity to handle the extra information and to react as necessary. But because the greatest source of hazard is external to the rider, absolute safety perhaps can never be achieved ?

    Even then some occupations or genders may show unusual spikes.

    For example looking that the roll of honour for UK Policemen,the period when they are most likely to die is not during shift time, but during the commute to or from work. Privately ‘street officers’ will admit this is due to shift work so upsetting their sleep pattern they they are constantly in sleep deficit, especially for officers travelling to early turn, or coming off night shift. Therefore they literally crash and die because they are too tired to assess and react to the risk.

    In London at the moment with the use of push bikes being promoted, there is an issue around cyclists being caught up beneath and being killed by left turning (right turning in US terms) lorries at road junctions. But of those who have died over 18 month the majority have been women. This has only recently begun being picke up in the Cyclist’s forums so not work has been done.

  6. wmoon Says:

    Young Dai, what you suggest is a very popular notion and I like how you put it–Darwin’s low-hanging fruit–and/or the inexperienced gets plucked while those of us who follow the safety messages are higher on the tree. However, if riders any where in the world were honest instead of trying to defend a fondly-held belief that makes them *feel* safer and less vulnerable–then they’d realize that they know riders who were good and safe and aware and sober/helmeted, et. al. and yet were killed or injured in a crash. In fact, they may be the one they know (injured, obviously and not dead).

    I don’t think most riders are prepared to self-identify as Darwinian low-handing fruit….

    Nor should many of them.

    The problem is that we cling to beliefs that are not supported even by the little we know–to continue to turn a blind eye to the reality simply dooms more riders to injury and death.

  7. Jeff Brenton Says:

    Some of the missing statistics ARE available. For example, every state-run program I know of has driver’s license information on every rider that took their classes, if a waiver is involved. They know if the rider passed or failed. They know when the rider took the class, and how many times.

    Collating the information to make it useful would require matching those records up against the other databases that NHTSA and others pull from, the crash reports. And every state has a proprietary way of storing it, since they don’t have to distribute it or its derivatives to any other agency.

    About 4 years ago, I made some inquiries into what it would take to get access to the data in a way that would allow matching it with crash data. I was told it wasn’t going to happen, short of a government study, because the information necessary to tie them together is considered too personal to allow out. Privacy concerns and all that.

    Maybe I should ask again…

  8. wmoon Says:

    Jeff, are you saying that state programs have the information on whether the student went on to get the motorcycle endorsement after graduating from class? Because the issue isn’t whether they have a driver’s license but a motorcycle endorsement (as the numbers of unlicensed at all fatalities are very small). That is not true in the state programs I’m familiar with. While the state-run program does have the training data, the DMV has the motorcycle endorsement information and the crash/traffic offense information.

    It’s not like at least one state hasn’t tried–but the money was never there to do it. The problem isn’t scrubbing the information of personal data if state employees (or contractors) are the ones accessing the data–those fields can be deselected. But, yes, you–as a private citizen–wouldn’t be allowed access.

    I’ve been told by more than one state (different agency/departments in each state), that it’s extremely expensive and time-consuming to do that kind of project–and they were unwilling even before these economic times–to spend that kind of money.

    And then it gets more complicated: FARS collects the information from the state and, as Vstromer pointed out, some is incomplete and other data isn’t even asked (such as if they were wearing protective clothing). About 30% of all BAC-involvement is estimated, for example–though NHTSA strongly defends the method used for estimating. Crash reports have their own problems–for one thing, research has found that LEOs check “unsafe speed for the conditions” in lieu of other evidence because, in their thinking, if a crash occurred, obviously it was an unsafe speed. So, if a rider avoided the deer but crashed and died, without witnesses, a LEO might check unsafe speed. And that’s just the beginning of the problems.

    That’s why Hurt spearheaded the method of accident causation that had investigators trained to accurately determine crash causation and have them out there right away while there was the best (and most accurate) chance of determining it. At that point, it would be possible to check if they had been trained–but only if there was a national database of all riders ever trained since riders can be trained in one state and die in another–and if they do, if only a state was doing it, it would appear they were not trained.

    It’s complicated and expensive and time-consuming–and rife with missing data–in the best situations and at the end, all you know about is dead and injured riders rather than much about the bulk of riders who aren’t crash-involved. Still, I wish something would be done.

  9. Young Dai Says:


    I am sorry, I was posting late at night and went off on a bend in my arguement. The point I was trying to make is that UK riders undergo a different training process to US yet still there seems to be a level below which M/C deaths do not fall

    Maybe education can never deliver total safety. Ultimately are motorcyle deaths the real problem or a just a symptom of wider issue ?

    As is frequently pointed out motorcyclists exist as a minority amongst all types of road users

    Perhaps the reason better training of riders (however defined ), struggles to deliver safer riders, is because of our inter-actions with those other road users, and the roads themselves.

    When we ride the reasonable assumption is that those on the road around us are sober, aware, able to handle their vehicle to the standard required by the law. And that test itself is of a sufficiently high standard to be credible. But gut feeling is that a group do not reach this ideal. . Once you remove the deaths of those rider who were riding impared for any reason, and those single vehicle accidents due to lost of control etc, do the actions of sub-standard drivers have a disproportionate effect upon other road users and M/C’s in particular ?

    If the education message is targeted at only one sub-set of all road users, will any advantage that should accrue to that group, be lost within the general poor driving standards of the other road user groups ? No matter how you position yourself, you could still be part of someone elses’s accident.

    Is the idea of a License for life (as is the case in UK) still valid ? Perhaps road users need to be retested to the current driving standard of the day say every five or ten years That includes an eyesight test as well. At least that would bring a modicum of regular upskilling to the ‘bog standard’ driving masses.

    Advances in car safety seem to have brought unexected consequences elsewhere. Cars that can now survive an impact with an elk, or being rolled at autobahn speeds, need huge A and B pillars directly in the drivers sight line. Similarly cars now have heavily raked windscreens both for fashion, ‘economy and the requirements to protect pedestrians. again in the drivers’ sight-lines.

    It is now very easy to loose sight of a rider or pedestrian behind a B pillar as we sit waiting at a junction. And we are looking for them, never mind the guy with a moble glued to his ear, or the look-but-do-not-see group.

    Is the reported increase in M/C deaths as a proportion of the deaths overall, because the roads are dangerous than 10 years ago, or because driver deaths decrease as the survivability of the modern car improves ?

    Are they in fact crashing just as often as always, it is just that people are now able to walk away from a crash that would have killed their parents 30 years ago ? Whereas you are just as exposed on a M/C as you were 50 years ago.

    If the ability of soft education programmes to control road deaths has its limits, should society instead be looking at hard $ engineering changes.

    If the majority of deaths are caused by ROWV on cross lane left turns, can they be designed out ? Should speed limits be decreased on these roads to reduce closing speed of impact. Do we look to more automated enforcement (average speed camera’s, signal camera’s and the like)

    Euducation is cheap , engineering and enforcemetn are expensive. Perhap society will have to find a compromise. Accept that as a system what we have is imperfect, but to have an honest debate the level of deaths and injuries it will find acceptable for the amount it is willing or able to spend on the roads.

  10. wmoon Says:

    Young Dai (oh, I am so curious as to how you got/chose that name!):
    I don’t think that all crashing can ever be eliminated. However, injury and fatal crashes have gone up–in relation to their previous numbers–125% since 1998. Motorcycle registrations have gone up 84%–so there’s clearly a huge disparity there (41% over what could be expected if simply more motorcycles on the road was driving it). While the casualty rate in Britain has also gone up, it’s nowhere near what’s happened in the USA. For example, motorcyclists are about 3% of the total road users in both the USA and Britain. In 2007, out of 1.3 million motorcycles in Britain, there were 588 fatalities or 1 fatality to 2,210.88 riders. In the USA in 2007, there were 5,174 fatalities and 7,138,476 registered motorcyclists–which translates to 1 fatality to 1,379.68. That’s an enormous difference. When it comes to riding impaired, 1.4 % Britain’s riders failed a breathalyzer test after a crash in 2008. In 2008, 28% of USA rider fatalities were over the legal limit. Forget about the new EU license test–the old UK one was so hard that MSF instructors would fail it when they moved to Britain. So it’s a very different situation here in the USA–and, I firmly believe, because our motorcycle training scheme is so very inferior to what is done in Britain.

    Much can be done to improve road safety–and you’re right, many of them would be expensive. But much can be done on the soft side rather than the hard technological (expensive) side. Not that the others shouldn’t be done–but no matter how good the hardware the wetware between our ears is our best chance of staying out of trouble.

  11. Dave B Says:

    “Forget about the new EU license test–the old UK one was so hard that MSF instructors would fail it when they moved to Britain”.

    I don’t know if all of us would fail it.

  12. wmoon Says:

    Dave, maybe you wouldn’t, but are you familiar with what it was–and what it is now?

  13. Dave B Says:

    Yes to both questions.

  14. wmoon Says:

    Do you really think that the average graduate could’ve–on their own bike–pass even the old test?

    As for MSF instructors not passing the old test–they didn’t. As to whether you or others would, it’s a moot point. But I’d just love to see Ray Ochs have to take the new EU test…

  15. Dave B Says:

    No to your first question.

    Ray would find some problem or flaw with the new EU test. Remember, his answer was Britain’s training hasn’t been proven to reduce motorcycle accidents. That’s why the US hasn’t adopted it. (And if we did, motorcycle sales in the US would plummet).

    Unfortunately, the training in the US to get a motorcycle license/endorsement will never get tougher.

  16. wmoon Says:

    hmmm…I wonder if it really would?

  17. Young Dai Says:


    Nothing as exciting as a misquote Billy Joel ,(“Only the good die young”). I am sorry to say the truth is rather more proasaic.

    My father was known throught his career as Dai, not because he was Welsh, but because he entered an inter office dart’s competition soon after he joined.

    In the first match he played in he was up against man who’s surname began with the same 4 letters as his: Davidson, as against Davis. The match scorer then marked the senior man up as DA, and my father up as Dai, and so on the whim of the dyslexic with the chalk, he became known as Dai to those he worked with
    for the next 40 odd years and into retirement as well.

    As a boy when he took me to his office, or to the sports matches he played in, I would be greeted by his friends and team mates as ‘Young Dai’.

    Then when I took a bloging identity shortly after his death, it just seemed a nice way to commemorate him to use it again.

    Thank you for saying kind things about the UK training standards. Funny but many of the UK forums bang on about how much worse things have got on the roads ! You will also have picked up on the dispair about the EU test, although I think that is more to do with who it is being administered in the UK than any basic arguement over the skills being tested.

    Is this clip from the California DMV typical for the space allocated to M/C testing across US ?

    How can you test road sense in a car park /service area stuffed behind a group of industrial warehouses ?

    28% riding drunk ! Is that to do with the bad boy, ‘rock ‘n’ roll and brew’ lifestyle projected by some manufacturers, or because folk really are that stupid ?

  18. wmoon Says:

    Young Dai–lovely story. One question–when you’re old, gray and balding will you still be Young Dai?

    Things have gotten worse on the roads on your side of the pond too–just not as bad as it is on this side. And I really do think a lot of it has to do with your training scheme that requires in-traffic training as part of the basic form (either CBT or the accelerated form). I don’t think you realize that in American training riders can never go above 12 mph on a 125-250cc bike and never ride more than 15 mph and ride in a smooth, paved parking lot and never in traffic. After about 8 hrs of which at least one third of the time is spent listening to instructions or waiting in line or taking a break, they take a test to get a waiver so they can avoid taking it at the DMV–though I’ve never seen one like the one in the video clip (others chime in here).

    The test consists of a u-turn in each direction and a swerve to the right. Then they stop in a line and listen to directions wait for their turn, then do a quick stop. Stop, listen to directions and wait to do two gentle curves. At no point do they have to ride above 12 mph (though they’re supposed to get up to 20 mph between the two curves).

    Passing that cream puff test then they can go out and hop on a Ducati or Hyabusa or Road King and ride in extreme congestion or the twistiest road you can imagine. So you can imagine why I think GB’s training that requires real in-traffic riding and the EU test that actually tests the swerve and stop at the speed you travel on slower roads to be exemplary.

    As to why people ride drunk (or drive drunk for that matter)? Wasn’t it you who used the phrase Darwinian low-hanging fruit?

  19. Young Dai Says:

    50 is the new 20 Wendy.

    At least that’s what Cosmo is saying. 🙂

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: