Archive for June 2009

MSF speaks out about the deaths–and reveals up to three more training deaths?

June 30, 2009

Yesterday evening, someone who calls himself ET commented on the entry on the bystander injury case, “I asked MSF to comment on student deaths, and they did, here’s what they said:”

I went and looked and will post it along with a different version of the reply I made him last night.  First let me note that ET is a computer programmer named Fergus Nolan of Memphis TN who has been blogging since May 2009–or for just a couple months.  In another entry, he says that he “asked Stacey of MSF about this, she promised me a press release”.  OK, only there is no “Stacey” listed in the latest MSF Contact list. Well, MSF does have a revolving door when it comes to employees so perhaps Stacey is a recent hire. So that’s one issue–who is “Stacey” with no last name?

The next issue is that I could not find this “press release” anywhere else except his blog.  However, this morning, MSF sent out a letter to state and military administrators that says, “We received an inquiry regarding the subject of the statement below.  Below is our response which we are providing to you as a courtesy copy and as an update to what we provided to you at the conference last August.”  So it is legitimate and I suppose one man’s press release is another man’s letter–or MSF looks upon state and military administrators as members of the meda? Who knows.

With those issues already raised, let’s see what allegedly MSF has to say about the deaths:

“Riding, especially learning to ride, has inherent risks. MSF is concerned about any crash that occurs, whether it’s on the road or during training. We take safety seriously in creating the best environment to pursue one’s dream to ride. A primary goal of the MSF is to ensure a low risk, positive learning environment for beginning students so that they can make the best choices while learning and riding.
Since its founding in 1973, more than 4.6 million students have been trained using Motorcycle Safety Foundation curricula, including approximately 2.5 million since 2002. MSF prides itself on making the highest quality research-based and field-tested motorcycle training curricula available to riders and prospective riders throughout the United States and the world.
MSF is unable to disclose details related to fatalities because of privacy considerations. However, since there has been some misreporting on this subject, MSF welcomes the opportunity to provide factual information.
Since 2002, out of the roughly 2.5 million students trained, there have been six crashes that resulted in the death of students, including one that was caused by a serious medical condition. In the past year, three additional students died from medical conditions while not riding. Every fatality has been thoroughly investigated by law enforcement, insurance investigators, or others. The curricula, and the delivery of the curricula by RiderCoaches, have never been determined to be a factor. MSF employs a stringent quality assurance program as part of its ongoing effort to review and refine policies and practices to minimize the inherent risks associated with training.”

For something that’s supposed to be “factual information” there’s precious little facts or information to be had. So let’s look at what MSF has to say–and what it doesn’t say in this “release”:

First of all, though it mentions that 4.6 million have been trained, which means  2.1 million were trained prior to 2002, it does not mention that in those 28 years only one rider died from injuries sustained during training (in 1998 in Valley Forge, PA) and in those 28 years, only one student, apparently, died of a heart attack sustained while at training (though not while riding).

Yet it does mention that with 2.5 million supposedly trained six have been killed from 2002–though it doesn’t mention the case of paraplegia, nor the other life-threatening incidents–and further reveals that three more have died in the past year and implies these weren’t training crashes. This represents a dramatic change in the number of deaths anyway you look at it.

But note how it describes the three deaths in the past year: “from medical conditions while not riding”.  However, if a student ran into a wall, for example, and suffered head and/or thoraic trauma that could be truthfully described as a “medical condition.” Of course in a crash, a rider ejects from the bike. If the rider did not literally die on the bike but succumbed after ejecting from the bike–and even weeks later in the hospital–it could be legitimately though misleadingly described as the rider dying “while not riding.”  Iow, MSF’s tricksy and bizarre way of presenting this “factual” information does not exlude fatal training crashes. Absent complete disclosure of the incidents–of which no one has heard that I know of–we can not rule out 3 more fatal rider training crashes precisely because of the lack of “factual information” in the press release.

It appears to try to fob the Uke’s Harley-Davidson Rider’s Edge death off again by citing the heart attack–even though the student was not treated for a heart attack at the scene but was treated for head trauma. He did suffer a heart attack after five days in intensive care and that was the final medical straw.

MSF says, “Every fatality has been thoroughly investigated by law enforcement, insurance investigators, or others. The curricula, and the delivery of the curricula by RiderCoaches, have never been determined to be a factor.” However, neither law enforcement nor insurance agents are qualified–nor concerned with–whether the curriculum nor delivery is to blame. Their interests are concerned with other things. Note that MSF doesn’t say rider educators investigated this–nor even that its own QAV specialists did or what their conclusions were.

Now let’s deal with the “low-risk” environment that MSF says it’s so concerned about creating: the death of an instructor (of which ET is  erroneously informed) in Valencia, CA, the death of a student in Honesdale, PA, the paraplegia case at the range in Sugar Notch, PA and the near-fatal crash in West Virginia were all in programs administrated by MSF on ranges certified directly by MSF employees. So one wonders exactly how MSF defines “low-risk”.

The release does say that it has a “stringent quality assurance program” however, in all but the cases mentioned above, MSF was not in the position to exert QAV on those sites nor instructors–or even the programs. Nor, of course, did it’s QAV program make an iota of difference in California, Pennsylvania or West Virginia where MSF employees were directly supervising training.

It says that it “unable to disclose details” because of privacy reasons–though in previous years it’s claimed that it couldn’t because of pending litigation. Neither excuse holds up–the personal information on the student (and instructor(s)) could be redacted and all the relevant information needed by the true experts be viewed.

What the real news is–and thank you very much Stacey–is that there have been three more deaths in one year–deaths that cannot be assumed to be not related to training crashes.

So now let’s return to this “press release” that Stacey gave Nolan aka ET. I find it interesting that after years of MSF stonewalling every rider education expert and magazine writer and refusing to say anything at all about any of the deaths that they should suddenly become so forthcoming with someone who has been blogging since May of this year. Especially when Nolan says on his About Us page on the site that he has “no qualifications to run this site” and has almost no knowledge of motorcycle safety–let alone training.  He also describes himself at another point as just “a simple-minded bike rider who has some time on his hands, access to a web server.”

Especially since Tim Buche impressed upon Dave Searle and Fred Rau an I during a personal tour of MSF/MIC/SVIA how very carefully they vet those the communications department will talk to–let alone come up with their very own press release on something as controversial as the deaths.

After years of stonewalling and a very rigid, paranoid relationship with the media, MSF choses to come out of the closet to a newly-minted, self-proclaimed uninformed blogger. And if that don’t just beat all… Not to mention that had MSF examined his site–as Buche claimed the communications department does before treating with the media–they would’ve found numerous factual errors in what he writes. So why this blogger? I have no idea.

Nolan says in another blog entry he’s sure he’s going to be accused of being part of some conspiracy “pretty soon”.  I hadn’t heard of him until he himself drew his blog and this “press release” to my attention. He says he contacted MSF and Stacey of no last name nor listing as an employee sent him a press release. As this is very much in the open, it hardly seems to be a conspiracy–though perhaps Nolan has confused the word “conspiracy” with “self-promotion” or perhaps “meglomania”.

But just who is this ET, Fergus Nolan? On the About Us page on his site it says he’s employed by FNSK Company INC.  However, according the Tennessee Secretary of State site, that corporation was dissolved in 1995 and he was  the principal agent–so even though that corporation doesn’t legally exist anymore, it appears in search engines with Nolan’s home address (the one he incorporated under).  I suppose one could say they are employed by a company that they themselves own–just like one could say a rider that ran into a wall or off a slope died from a medical condiion but not while he was riding.

Nor is BikeSafer is registered as a dba in Tennessee–or rather not one I could find at any rate. It is owned by FNSK Company, which is owned by Nolan. Iow, “ET” creates the illusion that there is an “About Us” when there really is an “About Him”.

He is a programmer who works out of his home as a consultant–in fact, he’s listed as Fergus Nolan Corp–but there is no business entity listing for that either on the Secretary of State site. I presume the laws must be looser in TN and anyone can claim they are something they have not legally registered to be. Sortof like MSF claimng it’s curriculum is both research-based and field-tested and of the “highest quality”.


What can we learn from looking at all four states?

June 17, 2009

Now that we’ve looked more closely at Pennsylvania, Illinois, Idaho and Oregon, let’s do a little reflection and examine some things that emerged by looking at four states:

In all four states, training expanded enormously: In Oregon, it increased over 220%. In Idaho, training increased over 660%. Illinois increased almost 82% and in Pennsylvania it went up 86%.

The untoward increase in Idaho was directly related to the state program’s inception—but not entirely. Illinois, for example, trained only 200 the first year and 4,500 the next as it was established statewide—but between 1977-1987 didn’t increase 660%.

These states are not unusual; across the nation—except in, notably, the MSF-administered state of West Virginia—training numbers have increased enormously. This is nothing new—however, that training has increased so greatly in specifically Oregon and Idaho, is of special note:

Does length or rigor of the course affect numbers trained?

The Motorcycle Safety Foundation has consistently claimed—and a great many rider educators believed—people didn’t take training because the course was too long. Each new iteration of the MSF curriculum has been shorter. When it sold the rider training community on the Basic RiderCourse, MSF then claimed that people didn’t take training—and many failed or dropped training—because it wasn’t fun and relaxed enough.

As a result, many in the rider ed community have believed that training numbers have increased because the course was both shorter and more relaxed/fun. However, even with only looking at four states, we find that assumption is unfounded:

Oregon and Idaho taught the longer RSS course then switched to the BRT, which is shorter than the RSS but still longer than the BRC. And Illinois’ BRC is significantly longer than MSF’s standard BRC.

So with longer and less “student-friendly” courses, training increased far more in Oregon and Idaho than in Pennsylvania and Illinois. Iow, neither the length or the pedagogical basis of the course seem to drive student participation.

In fact, it appears to be the opposite. MSF claimed the MRC:RSS was too rigid and demanding and TEAM Oregon’s course, the BRT was too much like the RSS—both reasons MSF had sold the motorcycling community on as reasons why more riders weren’t trained. Yet the  states that used longer, more “rigid” courses than the BRC saw far more significant increases in enrollment year after year. In fact, enrollment in the BRC-only Pennsylvania dropped as it was still increasing in Oregon and Idaho (although training in Illinois increased every year as well).

Only in Pennsylvania, a state where the pass and fail rate changed dramatically with the BRC so that more students passed did training rise then fall then rise then fall.

In fact, Illinois had dropped it’s fail rate and raised it’s pass rate soon after the switch to the BRC and training numbers continued to increase. But, when the state lowered the percentage of those who passed and raised the fail rate to what they were under the RSS student numbers kept increasing.

Length of course and how fun/relaxed or student-centered/adult-learning does not appear to influence student participation in a negative way—if anything, the longer course in Illinois and the less adult-centered courses in Oregon and Illinois (and in one region in Illinois) may have a positive influence on student participation.

Is it true that “something’s better than nothing”?

MSF has convinced rider educators that poor students should be allowed to continue, counseling out should be rare and students should self-select to opt out if at all possible.

Rider educators overwhelmingly tell me that if they didn’t keep these poor riders in class “they would go out and ride anyways” and “something is better than nothing”. The assumption behind the second is that a poor but trained rider is a safer rider.

And rider educators at least in MSF-controlled states overwhelmingly report the supposed objectivity of the test forces them to pass students who should not pass, are not competent and should not ride in traffic.

Otoh, in complete contradiction to this common thinking among rider educators, Billheimer’s study said that one of the greatest benefits of rider training is that it convinces those who should not ride not to—and therefore lowers the fatality rate.

Yet here, in three out of four states (and we didn’t do these graphs for Idaho), we find this odd little correlation that may be just that: When the percentage of “Did not graduate” goes up fatalities go down. And we see in states that teach the BRC there’s at least a visual correlation between trained and passed numbers and fatalities.

Every study has found that trained riders were less crash-involved, no less traffic offense involved than the untrained at best. At worst, they had a higher probability of crashing. The one study—one of the Dr. Mortimer ones done a very long time ago—found that training only made those who rode small bikes and rarely rode, didn’t ride far when they did and were women were safer.

While this is hardly conclusive it suggests that MSF is wrong and Billheimer is right: failing a student or counseling them may save their lives by convincing them they aren’t cut out to ride.

Iow, nothing is better than something and they don’t go out and ride anyways—and yet MSF—whose manufacturer members benefit from selling more motorcycles if riders take training—and whose manufacturer members get product liability from riders taking training—keeps pushing this notion that instructors shouldn’t be gatekeepers and something is better than nothing.

And that should be a problem for good-hearted men and women who sincerely believe that training should be saving lives.

And it’s also something that should be studied further. One of the ways that it could be examined further is if the comprehensive accident study was actually done—but, once again, it’s the motorcycle manufacturer members behind MSF and MSF itself that has made that happening less and less likely.

A look at Oregon’s training, registration and fatality stats

June 12, 2009

Oregon and Idaho are the only two states that have never used the BRC curriculum (beyond a limited field-test). Looking at Oregon, then, offers a comparison to Idaho just as Illinois was a comparison to Pennsylvannia.

However, it’s not apples to apples: Oregon has a universal helmet law and the number of unhelmeted fatalities is miniscule. In this way it differs from Idaho—but does offer an additional data point to compare the years before Pennsylvania changed its helmet law.

So some basic stats:

  • Training increased by 221.1% with an total of 53,565 students trained from 1997-2007.
  • Oregon’s motorcycle registration increased by 33,620 total (54.2%).
  • Using MIC’s formula, that translates to 63,763 owners.
  • There was a 112.5% increase in the fatality rate (but see below).

Oregon trained 19,945 more riders than there were increased motorcycle registrations—and that was just the basic trained riders.

Unlike Pennsylvania and Idaho, Oregon does not track “dropped” students.


MC fatalities


Passed %

% Failed/Did not finish course
























































In Oregon the Pass rate was higher and then dropped. This is in direct contrast to the experience in Pennsylvania and Illinois. However, Oregon’s pass rate is higher than Pennsylvania’s.

Motorcycle Registration v. fatalities

Fatalities did more than double over the 11 years however the number of fatalities per year look more like a bouncing ball going up and down year by year—though rising overall. In this way, it’s more like Idaho where fatalities also showed more variance year by year while Illinois and Pennsylvania’s fatality numbers seem to be more similar. There’s many reasons why this may be and is yet another thing that deserves study.

Motorcycle registrations show a very different pattern than in any of the other three states. This much is similar though: in none of the states was there a unilateral increase from year to year. While we know that nationally motorcycle sales went up year by year and it was assumed that motorcycle registrations did as well, on the individual state level it’s more dynamic.

The rise and fall of both fatalities and registrations seen in these four states change that popular notion that there’s been an unrelenting rise of both over the past 11 years—while that may be true as a nation, it breaks down on the state level. It also emphasizes the need to look at longer term trends than a year or two—or even three.

However, as in the other three states, there appears to be no correlation between increases in motorcycle registrations and fatalities:Fatalities v mc regs

Instead, when motorcycle fatalities jumped, motorcycle registrations went down and they only drew together again during the last two years of available data.

What’s perhaps most remarkable is that this same pattern now can be seen in all four states to a lesser degree: not only isn’t there a positive correlation between increased registrations and fatalities but in all four states to one degree or another the exact same pattern can be seen—registrations go down and fatalities go up.

In the motorcycling community, there’s the belief that registration and fatalities went together like bread and butter, horse and carriage: more motorcycles naturally mean more deaths. Four states in, this particular correlation doesn’t appear to be true on the states level—and NHTSA has been saying for years now that it isn’t true on the national level. However, a correlation no one’s talked about might be true on the state level: less motorcycles, more deaths. There is, actually, research that does link the lack of motorcycles with more deaths—but we’ll get to that in another entry.

Training v. registration

As in the other three states, training has dramatically increased over the 11 years and have consistently increased year after year.

OR increases mc reg trained

However there’s even less correlation between training in and increased registrations than there was in the other three states.

What is most significant, perhaps, is potentially how many of Oregon’s riders have taken training: If absolutely no other motorcycle owners in Oregon had been trained prior to 1997 or in another state and if no students took the course more than once and there was no attrition—all of which we know is false—then up to 84% of Oregon’s owners are trained. And it means they have been trained in the past 11 years.

And that makes that the fourth state where there’s a profound correspondence between numbers trained and registrations:

In Illinois, we found that using MIC’s owner’s formula, more riders were trained than there were owners—and that Illinois trained just 30% less riders than the increases in motorcycle registrations. In Pennsylvania, the numbers trained were just 214 less than the increase in motorcycle registration. And, in Idaho, the data suggests that just 19% of riders registered a motorcycle but didn’t take training.

Training v. fatalities

Like Pennsylvania and Illinois, there appears to be some correspondence or correlation between training and fatalities—though not as pronounced as in Illinois or Pennsylvania. In Idaho, we know exactly how many riders trained in that state have died.

OR trained passed

And here’s the comparison between those who failed and fatalities:

OR Trained failed

In Oregon, as in Pennsylvania and Illinois, we see that odd correspondence—when failure rates go down, fatalities go up and when failure rates go up, fatalities go down. It’s a correspondence that’s not evident in the numbers passing and fatalities. This may be just a odd little coincidence and not mean anything at all—still it would be an odd little coincidence found in three out of four states.

Fatalities v. Helmet Use

Oregon is a universal helmet law state—and we see the number of unhelmeted fatalities is between 1-3 all years but one when 4 unhelmeted riders died. The rise in fatalities in Oregon, then, cannot be blamed on non-helmet use.

OR fatalities helmeted unhelmeted

Quick poll

June 3, 2009

This is an updated version of the original entry.  There’s a whole lot of people reading this entry but very few have filled in the form. Please do so, especially if you live in a state where you have a choice to wear or not wear a helmet. There’s no judgment here: I am no helmet nazi. As silly and small as this straw poll is, it may suggest something important about why you choose to wear or not wear a helmet.

One of my regular readers, Gymnast, speculates that temperature plays a significant factor in whether people wear helmets or not in states without universal helmet laws and how that would affect helmeted v. unhelmeted fatality rates. That’s fairly easy–though tedious–to check by correlating the deaths against weather records over a period of time (preferably years).

But his idea is based on an assumption I’ve heard from others: that riders are flexible in their behavior–they makes choices based on X (in  this case weather) to wear or not wear a helmet if they live in a state where that choice can be made.

My question to you: is that assumption true? Do riders change helmet habits based on the weather? And, if they do, why do they?

So let’s put Gymnast’s supposition to the test–answer via the comment function:

Do you wear a helmet (full, 3/4, half-shell)?

How often do you wear one?

Do you–regardless if it’s legal not to wear a helmet–wear it?

If you do NOT regularly wear a helmet, in what conditions do you choose TO wear a helmet? Why?

If you DO  regularly wear a helmet, in what conditions do you choose NOT to wear it? Why?

Any other comments?

Please send this link far and wide.

Let’s do a straw poll and see what turns up.

What about Illinois fatalities and helmets?

June 2, 2009

Illinois only had a universal helmet law for two short years; passed in 1967, it was repealed in 1969. Illinois checks seat belt and helmet usage for one week each year in June or July. Only DOT/Snell helmets are counted. Limitations of the NOPUS survey (which also measures seatbelt usage) are many—one obvious one is the weather which may lead to more or less helmet usage.

From 1998-2007, helmet usage by percentage:

98 99 00 01 02 03 04 05 06 07
Helmet 29.9% 29.1% 27.7% 22% 26.3% 35.9% 36.1% 37.6% 35.5% 29.5%
Non-Helmet 70.1% 70.9% 72.9% 78% 73.7% 64.1% 63.9% 62.1% 64.5% 70.5%

In Illinois, the median helmet usage is approximately 30% and the non-helmet usage is approximately 70%.

IL Observed Helmet UseHowever, in the five years from 2002-2006, helmet usage rose significantly (≥9.2%). In 2007, helmet use dropped 6% and was essentially what it was in 1999. If this is a one-year blip, helmet usage appears to be increasing though the reason(s) are unknown.

It raises the question: from whence comes this increase in helmet use? Is a result of new riders? Different kinds of motorcycles (as sportbike/tourers are associated with helmet usage)? Or have experienced riders decided to wear a helmet? Since training has increased so much in the past five years, does this indicate training does influence helmet usage—and there is anecdotal evidence that it does.  These questions are discussion-worthy—and research worthy.

Setting that aside and merely based on their proportion of the motorcycling population, it would not be untoward if unhelmeted fatalities were 70% and helmeted fatalities were 30%.

As we see from the table, unhelmeted riders are from 60% to a high of 78% of fatalities:

97 98 99 00 01 02 03 04 05 06 07
Fatalities 82 99 103 128 138 103 140 161 159 133 161
Helmeted 13.3% 2.3% 16.5% 11.1% 12.9% 24% 12.6% 18.5% 18.5% 20.5% 17.8%


66.3% 74.7% 70.9% 76.2% 75% 60% 76.9% 75.8% 75.8% 78% 77.1%
Unknown 20.5% 23.2% 12.6% 12.7% 12.1% 16% 10.5% 5.7% 7% 1.5% 5.1%

In all but three of those years, unhelmeted riders are over-represented for their proportion of the motorcycling population by 5-8% with a median of 75.8%—or 5.8% over their proportion of the population.

Otoh, helmeted riders, being 30% of the riding population, could be expected to contribute that percentage of fatalities—and yet they do not. Instead, they are significantly underrepresented in fatalities by 6-24% in every single year with a median of 16.5%.

Granted, the “unknown” category is large enough in almost all years to make a significant difference depending on how many of those riders were wearing helmets or not. “Unknown” means simply that—in that number of cases, the police and/or hospital report did not include the information about helmet use or non-use. And it varies quite a bit from year-to-year. This does not negate the trends in the known numbers and the percentages they comprise—so we’ll note the unknown category and set it aside.

If all else is equal, the numbers in Illinois seem to support the position that helmets save lives. But that’s the key phrase: is all else equal?

Do both groups ride as often, as far and as long in similar traffic, on similar roads at similar times? No one knows. Or do those who choose to wear helmets because of perceived safety benefit also choose other behaviors associated with safety benefits such as training or not drinking and riding, etc. All those things can make a significant difference in risk factors and affect the number of deaths.

If, however, all else is equal then the consistent under-representation for helmeted fatalities does appear to support the belief helmets save lives. Based on the median over 11 years, 13.5% fewer died than would be expected. As good as that is, it’s nowhere near NHTSA’s claim that helmets are 37% effective. NHTSA’s percentage  was based on doctor’s theorizing which riders could’ve been saved if only they had worn a helmet.

However, if riding without a helmet is as dangerous as we’ve been led to believe, the 5.8% difference between expected and actual contribution for unhelmeted fatalities barely reaches the level of significance.

Otoh, as we saw in Idaho and Pennsylvania, helmeted fatalities are increasing at a faster rate than unhelmeted fatalities:IL H Unh v Overall fatalitiesRemoving helmeted fatalities from the graph gives a very different picture of what is happening in Illinois:

IL Unhelmeted fatalities v Overall Fatalities

While unhelmeted fatalities increased, they did so proportionally in a remarkably consistent way across all the years. That’s not at all what happens if only helmeted fatalities are compared with overall fatalities:

IL Helmeted Fatalities v Overall Fatalities

We noted the pattern of increased helmet usage at the beginning of the entry. So let’s see if that explains it:

IL Observed helmet use v helmeted fatalities

There is a rough correspondence but the explanation that more helmeted riders simply increases the chances that more helmeted riders will die.

Over 11 years, Illinois’ fatalities increased by just over 94%. We already know that it’s not due to increases in motorcycle registrations—i.e., it’s not as easy as “more riders on the road, more deaths”. Pro-helmet law advocates cite deaths increasing when laws are repealed. Illinios and Idaho’s experience, however, suggest that the reverse is not true: helmet use can increase but not have a demonstrated effect on the fatality rate.

In fact, helmet usage increased over the 11 years but the greatest increase was  9.2%. In 2007, it returned to the 1997 level. However, from 1997-2007 helmeted fatalities rose 154% while unhelmeted fatalities only rose 120%. That enormous increase in helmeted fatalities cannot be simply explained by “more riders wearing helmets means it’s reasonable that more helmeted riders die”.

While helmeted riders are significantly underrepresented in fatalities, the enormous increase in helmeted fatalities should cause motorcycle safety experts concern—it deserves both discussion and study.

Training and helmets?

It’s often implied that helmet use increases because of training whether because training convinces people to wear helmets or because more safety-minded people take both training and make the helmet choice. Comparing observed helmet use with those who took and passed training we see:

IL Observed helmet use v passed training

Now let’s look at helmeted fatalities v. those who have passed the course:

IL Passed course v helmeted fatalities

Now let’s compare unhelmeted fatalities and those who passed the course:

IL Passed course v unhelmeted fatalities

I don’t have a clue what all of this means, but this I’m sure of: at the very least, the fatality picture in Illinois—and in the entire USA—is much more complicated than slapping a helmet on everyone’s head or mandating training.

While there may be more than one explanation, it does appear the correlation between trained licensed riders and helmeted fatalities does appear to be stronger than between trained and observed helmet use and trained and unhelmeted fatalities. It raises the possibility that training—as has been found in several studies—increases a rider’s likelihood to be accident-involved—and training and helmet use increases it that much more.

The easy answers are not sufficient to explain what’s happening on our roads. While some may take comfort that the fatality rate is no longer rising, that does nothing to explain why it did and why riders are really dying on our roads.