Pre-dawn crashes
A note from the ABEA Program Committee: the American Bicycling Education Association’s programs are data-driven. Sometimes the data points to caution, and a warning. We send along a warning by CyclingSavvy co-founder Mighk Wilson in all good faith. While this research is sobering, we teach situational awareness and we need to be aware of risks.
For some years Iโve responded to critics of CyclingSavvy who say itโs not perfect with, โOf course CyclingSavvy strategies arenโt perfect; theyโre just the best.โ I always feel comfortable using our strategies wherever I go.
But over the years, some serious crashes have given me pause because they involved CyclingSavvy graduates or even instructors. Most notable are the pre-dawn crashes.
I work as a safety planner for MetroPlan Orlando, a regional transportation planning agency. In that capacity, I routinely review police crash reports involving cyclists. In my 30-plus years working for the agency, Iโve reviewed, classified and catalogued more than 10,000 cyclist crashes. I see information and data that very few people do. My work helped in the development of the CyclingSavvy curriculum, giving it some of the scientific heft it needed to support its strategies.
Bicycle commuters canโt typically choose their riding time, but recreational cyclists have more flexibility. Pre-dawn recreational cycling is very popular in Florida, to beat the summer heat, and because traffic tends to be very light. But these hours hold a special risk that we need to understand.
Reports of pre-dawn crashes in the Orlando area
In July of 2021, Stephen McKinney-Steck, a CyclingSavvy graduate and notable retired leader from Orlandoโs public television station, was hit from behind and killed during his regular early morning ride. Stephen was one of those people who does everything to extreme excellence. We knew he would have had much-more-than-adequate lights and reflectors on his bicycle. He would have been using lane control on the arterials and collector roads he rode for his daily early-morning rides. His death hit us hard both because he was a friend and because we knew he was doing everything rightโฆand using the strategies we teach. Stephenโs crash occurred about 20 minutes before sunrise.
Only six days after Stephenโs fatal crash, two other Orlando metro area cyclists were hit from behind while riding side-by-side. Fortunately, neither was seriously injured. They were at the rear of a large recreational group ride. All the members were using good, bright taillights.
The officer who wrote the report on the crash said that the driver โappeared confused and fatigued,โ and said he โnever saw the cyclists.โ Clearly, the driver fell asleep at the wheel. He could have just as easily rear-ended another car at the traffic signal just ahead. The crash occurred 15 minutes before sunrise on a 4-lane residential collector posted at 30 mph, and the cyclists were likely riding at 20 mph.
About 12 years ago, CyclingSavvy Instructor Charles Badger was hit from behind around 4:30 a.m. less than a mile from where Stephen was hit. Charles was an avid brevet rider on a training ride. The college student who hit him also fell asleep. Charles suffered a significant and life-altering brain injury.
And elsewhere…
A couple years after that, CyclingSavvy Instructor candidate David Spranger was hit from behind while commuting to work around 5:00 a.m. in Charlotte, NC. The motorist who hit him was intoxicated, but drowsiness likely aggravated the intoxication. David similarly suffered a life-altering brain injury. Like Stephen, both Charles and David had excellent lighting and reflectors and wore helmets.
Massachusetts CyclingSavvy graduate Ian Whiting was riding in the pre-dawn darkness when he was struck from behind. His helmet did its job and he recovered fully, despite a broken neck. He has this to say about the incident:
โThis happened on November 30, 2018, around 5:30AM, so it was full dark. The state trooper who wrote the report cited the driver with a marked lanes violation, which didn’t make sense to me because I was squarely in the right lane. It seemed to me that he thought I must have been on the shoulder and that that’s where she hit me, but he wouldn’t have known because he never interviewed me.
โMy big problem was that on this ride, I had no rear view mirror; otherwise I would have stood a better chance of seeing that someone was behind me. As it happened, I had no clue that anybody was approaching me until I was flying through the air.
โI had tried a couple of mirrors that didn’t work well for me so I went without for a time. I would eventually settle on the drop bar end mirror that I still use today and highly recommend.โ
Ian has prepared and posted video of the incident. In the video, he makes that point that vehicles had been overtaking him safely in large numbers, but a solitary vehicle struck him following a long gap in traffic. That may reflect another factor: drivers are likely to pay less attention to the road when driving conditions appear uncomplicated.
Comparing pre-dawn and post-dawn crash counts
In Orlando metro-area crash data for the years 2015 through 2025, I found 12 fatal pre-dawn overtaking crashes (between 5:00 and 7:00 AM and before sunrise). (Some of these may have involved cyclists with no or poor lighting; crash reports are often incomplete on this point.) By comparison, daylight hours between 7:00 and 9:00 a.m. saw only four fatal overtaking crashes. A total of 49 fatal overtakings occurred during those years, so 24% of them were between 5:00 and 7:00 AM; only 8% between 7:00 and 9:00 a.m.
Crash counts can reflect not only an increased risk for drowsy driving, but also an increased number of cyclists on the roads. But we donโt have good data on how many cyclists are riding in the pre-dawn hours.
The circadian trough
โCircadian troughโ is the name for a time of day when alertness suffers. (Thereโs a pre-dawn one and also a mid-afternoon one, as anyone whoโs tried to stay awake through an after-lunch business meeting can attest.)
Daylight may make the difference. Thereโs some evidence that daylight can help to mitigate sleep deprivation. I experienced this effect in my early 20s (young and stupid) during a solo overnight drive from Orlando to Cleveland, Ohio. I recall struggling to stay awake from around 4:00 a.m. until sunrise around 6:00 a.m., but as soon as the sun came up, that struggle stopped.
It can be tempting to approach pre-dawn crashes as yet another โgotta fix those awful motoristsโ problem. Sure, it needs solving, and not only for cyclists: all road users are at risk. But we can also develop strategies for ourselves. Clearly, being highly conspicuous is not enough. Using bike lanes and paved shoulders also appears to be of little help. And this is certainly not me putting in a pitch for separated bike lanes, which add a host of new problems for bicyclists.
Instead, we can rethink how we treat those pre-dawn hours. Risk management is often about trade-offs. The trade-off for cycling has always been about balancing the relatively lower risk of overtaking crashes against the higher risks of turning and crossing conflicts. The circadian trough shifts the overall risk towards the overtaking motorist, but by how much would be very hard to quantify.
Seeking solutions
Are you riding recreationally? Perhaps it would be worthwhile to wait until sunrise (though of course staying aware of the risks of the low rising sun). Or you might ride trails in independent rights-of-way (though many are technically closed from dusk to dawn). Are you riding to work? Look at your route. Consider using neighborhood side streets rather than collectors and arterials. If thatโs not an option, you could consider using a sidepath or sidewalk. On sidepaths and sidewalks, youโll need to ride significantly slower to reduce your risk of right-hooks, left crosses and drive-outs.
As Ian Whiting suggests, a rear-view mirror can alert you to a vehicle approaching from behind. If you are controlling the lane, it will also show you whether a driver is reacting to you by slowing or changing lanes.
Someday, most motor vehicles will have systems that, even if they arenโt fully autonomous, will at least track the eyes and face of the driver and determine if they are drowsy or otherwise impaired. My Tesla already nags me if Iโm not paying adequate attention to the road while using the Full Self-Driving program. But until then, we should all be aware of the circadian trough and plan accordingly.
Mighk stated:
“I routinely review police crash reports involving cyclists. In my 30-plus years working for the agency, Iโve reviewed, classified and catalogued more than 10,000 cyclist crashes. I see information and data that very few people do.”
Crash reports that are accessible to the public, such as those on SWITRS and CCRS in California, provide enough information to identify fault and vehicle code violations. But the original report includes so much more, especially for a fatality.
Three of us on a Bike Committee in Oceanside were given access to original, unredacted reports. We cataloged them using an Excel spreadsheet for 192 crashes from 2004 to 2009. We have used a circle chart based on the results in LAB Traffic Skills 101/Smart Cycling classes in Oceanside and other Southern California locations.
When I read that over 10,000 original reports have been reviewed, it dwarfs what we did. Thank you for sharing this.
Thank you Pete.
One of my retirement hobbies/gigs will (hopefully) be to train planners and engineers (and interested advocates) how to do the crash typing I’ve been doing for all these years. Nothing challenges your assumptions like the reality of seeing how people get into serious trouble.
I think we could even create a big public database on crash data. It would be stripped of any personal data, but contain all the key factors and crash types.
I became skeptical of police reports after my son was run over by a driver as he rode home from school. He was legally riding on the sidewalk against traffic across a driveway–this is legal in CA. Both stopped. My son continued as he thought the driver acknowledged him with her eyes. Apparently, she looked at him but he did not register and she ran him over. The police said my son was partially responsible since the driveway ramp where he was hit is not part of the sidewalk and it does not constitute a crosswalk. They said he was not allowed to ride on the ramp even as they conceded he could ride on the sidewalk. The driveway ramp in a sidewalk is clearly part of the sidewalk. There is such a bias and ignorance of bicycle law amongst law enforcement that you cannot take their reports seriously. This caused massive delay and difficulty in settling with the insurance company. I was thoroughly disgusted. I get worked up recalling it and that was 15 yrs ago.
I’m well aware of all the ways officers apply bias and make errors. Having read so many reports I approach each one with skepticism. As in the crash you described, their bias tends to come from their interpretation of the law (or from what they believe constitutes safe behavior), not so much in the reporting of what occurred. So they still give us plenty of valuable information.
For an individual involved in a crash, the determination of fault is of course of high importance. But for a safety planner, legal fault is of little value. We’re focused instead on the factors that contributed to the crash. Sometimes that information is also erroneous, but a thorough analysis can usually provide a good picture of what happened.
It seems that the factors that contributed to this collision were that the driver looked at my son but did not see him. He interpreted her look as if she saw him, so he proceeded. She did the same shortly after. I do think that was captured in the police report if you read it carefully.
I just hear victim blaming all too often and I am pretty sensitive given my experience with my son.
This article demonstrates the danger of riding in front of cars, especially in darkness. I feel less and less comfortable riding in front of cars as I age and lose speed. I have many videos of angry drivers who tried to intimidate me with close passes because I had the audacity to use the road by “taking the lane”. It wears on you over time. Where I live, the CHP will send warning/education letters for 3′ violations. I keep a log and have issued >100 letters.
This article reeks of victim blaming, and excuses unacceptable behavior by motorists that injure and kill bicyclists because they are incapable of operating their motor vehicle safely – and apparently are not expected to, either. “Suggesting” that bicyclists not ride at certain times (irregardless of the reason, that’s not my concern here) of day because of the unacceptable behavior of some motorists is not only unacceptable, but disrespectful to those bicyclists that have been serously injured (permanently disabled) or killed by said dangerous motorists, because saying that implies that they were at fault for simply being where the were when they were seriously injured or killed after being hit by said dangerous motorist’s motor vehicle, because of said dangerous motorist not driving their motor vehicle safely (or being expected to).
Not only that, but it also reflects poorly upon CyclingSavvy/American Bicycling Education Association(ABEA) because that means that you all know that what you are teaching (CyclingSavvy and variants) doesn’t guarantee absolute safety (despite implying that it should, despite evidence to the contrary) in the dangerous operating conditions that is riding on any major roadway (in the United States of America, where we both are – I live in Shreveport, Louisiana by the way, at the time of this comment) with high speed traffic (anything above 35-45 mph, irreagardles of what is posted on the speed limit sign of said roadway in question).
CyclingSavvy provides a false sense of safety to its clients/students by failing to acknowledge the dangers of the environment that you expect “certified” (CyclingSavvy “graduate”) bicyclists to ride in despite being warned by others with said expertise (such as myself, but I’m sure there have been others, maybe, only y’all know that), that what ABEA/CyclingSavvy is teaching is inadequate to do what it advertises, and this article shows that.
I’ve been riding safely for 10 years by simply following the rules of the road and using mirrors and proper lighting. I’ve only recently (2025) added FMVSS/SAE/DOT reflectors to the rear and sides of my bicycle, along with side lights (amber front/red rear), as well as amber warning lights on a horizontally mounted stick with red FMVSS/SAE/DOT reflectors on the facing back/amber FMVSS/SAE/DOT reflectors facing forward in addition to two orange flags that I made with red FMVSS/SAE/DOT strip reflectors facing back and amber FMVSS/SAE/DOT strip reflectors facing front that identify me as a slow moving vehicle and also mark the width of my bicycle from the back from afar, day or night – no different than the back of a tractor trailer. In other words if you can’t see me, you would likely also crash into the back of a tractor trailer or other car because they are equally as visible.
As an addenum to what I just said, no state in the United States currently requires that a bicyclist equip their bicycle to FMVSS/DOT visibility specifications for nighttime riding like mine currently is, as all states require a white headlight visible from a specified distance (depending on state), a red rear reflector visible from a specified distance (depending on state) that may also have to meet FMVSS/DOT requirements (such as Texas, for example – there are other states that require the same), in addition to mandating useless CPSC reflectors (California), or where I’m at in Louisiana, ambigously, convolutedly worded description of such reflectors (“perpendicular to the bicycle frame”, doesn’t specify pedals like California does, or the color) – meaning that a bicyclist riding on a roadway at night could meet the legal lighting equipment requirements of their respective state and still be killed because they do not meet FMVSS/DOT visibility/warning requirements, which is the minimum any bicyclist should be going by if they are riding on any high speed roadway (which is most any major roadway in any major city in the United States, not including freeways masquerading/”converted” to roads like in Ian Whiting’s video link showing his crash) and want to do so safely – in addtion to using rearview mirrors and following the normal rules of the road as if you were driving a car – but CyclingSavvy/ABEA is not saying/promoting that (which actually works -ask me how I know) or acknowledging the aforementioned hazards of riding in high speed traffic, before including the additonal hazard of doing said activity at night while being hard to see (due to defective equipment despite being “legal”).
Also, how do you expect to tell motorists (not including the ones who are incapacitated/incompetent/malicious) that you are stopping/slowing or turning if they can’t see your hand signals/you don’t have turn signals and a brake light at night on from/on your bicycle, so that they don’t run into you from behind? I wear color coded amber and red lights on my arms so that my turn and stop hand signals respectively are clearly visible at night.
Invisible hand signals (at night) are not safe either (in addition to defective, non-FMVSS/DOT compliant/equivalent lighting/reflectors and no rearview mirrors on a bicycle ridden at night in high speed traffic) and doesn’t help to prevent crashes at night – so victim blaming dead or severely injured bicyclists doesn’t help either, no matter how much CyclingSavvy/ABEA try to sugar coat it, and it’s also just plain wrong.
I’m not the NHTSA nor the CPSC, nor do I work for either. I have nothing to prove (or gain) by telling it like it is other than preventing further needless loss of life/loss of quality of life from preventable causes with known solutions – including safer riding conditions so that such crashes don’t happen in the first place. I use vehicular cycling/bicycle driving too, but let’s not act like that it’s not being done in dangerous conditions that the average person is not willing to put up with – and should not be expected to. Why should I or anyone else have to equip a bicycle to FMVSS/DOT spec with lights, reflectors, and rearview mirrors (in addition to gears and brakes) just to ride safely, day or night. Just because I can doen’t mean others should have to, especially when such equupment has to be self manufactured/modified for use on bicycles outside of mirrors/brakes/gears. Safety should not be a privelege solely based on the ability to make the equipment to be able to do so – nor should it be required.
Giovannte Sims
Saturday, February 21st 2026
2:52 P.M. Central Standard Time (CST)
I agree there are limitations to the Cycling Savvy techniques and have been similarly frustrated with their lack of acknowledgement. They seem anti-bike lane but few people are willing to ride in front of cars, whether or not it works. Separation is key to getting more people on bikes–the more separation the better. “Driving your bike” does not always work, especially at night or with hostile drivers (which I encounter regularly). I was therefore surprised that this article seems to acknowledge some limitations. At least that is a start.
I am anti bike lane for the sole reason that they violate the normal rules of the road in ways that kill, in addition to authorizing dangerously close passes at high speed (that would be normally be illegal in most states/provinces with minimum safe passing laws if said bike lane did not exist), accumulating surface hazards that can cause deadly crashes (like sand, gravel, wet leaves/mud, trash, and broken glass), the door zone/obstruction from turning (assumedly motor) traffic by parked motor vehicles until an intersection/driveway (also causes deadly crashes from dooring/being run over at blind intersections), and gives justification to the idea that bicyclists do not deserve equal respect and access to the roadway (and by extension access to any destinations on that roadway, which is reflected by motorist, including police, harassment (including death threats with their motor vehicle being the weapon by trying to deliberately hit you or run you off the road) for refusing to use suicide gutter bike lanes/dirty and or narrow or overgrown shoulders designed only for motorist convience.
I do support properly designed paths – with clear sightlines, adequate space for passing other users, clearly marked hazards (with reflectors – or lights if necessary) and grade separation (like speed tables, or over/underpasses, depending on topography and water table), along with medians, and curb extensions narrowing the roadway to force slower speeds upon appoach to said path crossing a roadway, if feasible – but in no circumstances should that mean forfeiture of the right to use the roadway outside of a neighborhood street solely because such a facility exists – like it is in some European countries currently. Nor should it absolve a motorist of their responsibility to operate their vehicle safely in the first place, irregardless of the existence of such a facility just because a bicyclist isn’t using it and is using the roadway instead.
Yes, roadways should be designed for slow speed traffic (which includes bicycles) with slow lanes (like on steep bridges), wide shoulders, especially on long bridges, angled curbs to allow exiting the roadway to avoid a rear end collision on a curbed roadway, as well as pedestrian access (continuous wide sidewalks without dropoffs, for ease of use in a wheechair), as well as clear signage in advance/at the pedestrian crossing, speed tables and curb extensions to slow traffic to ensure safety at pedestrian crossings – pretty much the same as the safe bicycle path crossing described earlier.
The problem here in the United States of America is that the majority (who are motorists, inlcuding the vast majority of CyclingSavvy members) can’t even admit that we do not expect motor vehicle drivers to follow the basic rules of the road when operating said vehicle aroound others (including bicyclists) – this article is exhibit A of that, and yet and still, CyclingSavvy/American Bicycling Education Association(ABEA) purports that “acting better” (i.e., following the normal rules of the road/vehicular cycling/bicycle driving) will protect you in a hazadous operating environment where the hazard (dangerous motorists who are not following the normal rules of the road) isn’t even acknowledged.
I can tell you from personal experience that “acting better” only allows you to see hazards before they get to you, and to react accordingly so that you don’t get hit, assuming you can see said hazard in the first place.
Case in point – this article about “[Inevitable] Pre-Dawn Crashes”. So you mean to tell me that the so-called “experts” here at CyclingSavvy/ABEA don’t know that some motorists don’t follow the rules of the road, and that in addition to that, also don’t understand that 99% of bicyclists riding on a high speed roadway at night aren’t visible from a sufficient distance (at least 1000 feet) due to inadequate lighting/reflectorization equipment and requirements for bicycles at both the state and federal levels, despite the fact that FMVSS (Federal Motor Vehicle Safety Standards) 108 (in the United States of America) already covers this base and if a bicyclist were to equip their bicycle to this standard, would exceed meet and exceed the laws of every state for nighttime lighting equipment, as well as the StVZO requirements in Germany, and be clearly visible from far enough away that such crashes would be virtually nonexistent? And if such a crash were about to occur despite being properly lit and reflectorized, simply using rearview mirrors in addition to the aforementioned would alert you of such a situation with enough time to take evasive action to avoid entirely, or minimize any collision that would result from such a scenario, really?
I’ve done exactly that recently (equip my bicycle to FMVSS 108 spec, which includes side lights and reflectors in appropriate color per location, as well as a rear width marker stick with clearance lights, amber warning lights, and reflectors, including reflectorized flags, again in appropriate color per location) in addition to already using rearview mirrors. The result is night and day with how you are treated as a bicyclist at night, as you are clearly visible, as well as the fact that your width is clearly marked from the rear with said lights and reflectors (and from the front with amber reflectors facing forwards from the same stick on each side), making it clear that a lane change is required from afar (1000 feet) at night.
The video that Ian Whiting shows of his crash from three years ago is what happens when you don’t use rearview mirrors and also don’t acknowledge the dangers of the environment you’re riding in. Even if he had FMVSS 108 compliant/equivalent lighting/reflective equipment on his bicycle at the time of the crash, not using mirrors makes it useless as he did not know of the impending crash until it happened. At no point did CyclingSavvy/ABEA mention that using mirrors would have prevented such a crash, nor do they require mirrors for safety training, even though they require helmets, which do not prevent crashes (like mirrors and FMVSS 108 compliant/equivalent lighting/reflective equipment), nor serious injury or death from said crashes – at best it mitigates some severe head injuries.
Rearview mirrors (in addition to proper lighting and reflectorization as per FMVSS 108, with the obvious of already following the normal rules of the road – or vehicular cycling/bicycle driving) are the simple solution to this problem – so telling people to not ride because a few dangerous motorists instead of identifying the hazard for what it is and the already existent solutions is insane when coming from an organization (CyclingSavvy/ABEA) that presents itself as the bicycle equivalent of the MSF (Motorcycle Safety Foundation) when they don’t even tell you something that should be blatantly obvious to any bicyclist with equal or greater experience than what I have in operating a bicycle safely in traffic.
At least I don’t charge membership fees for the truth and lifesaving information. Your safety is payment enough.
Giovannte Sims
Tuesday, February 24th, 2026
8:28 A.M. Central Standard Time (CST)