By La Velle Goodwin Collision Prevention Specialist Founder, Driving Hero Academy Full disclosure: I spent 13 years at Young Drivers of Canada before leaving over 25 years ago. I have no current financial relationship with them, but their approach shaped how I've trained drivers for decades since. I'm launching drivinghero.ca in 2026 and pilot classes based on these principles. That context matters for what follows. Teen drivers aged 16-19 experience crash rates nearly 3 times higher than drivers 20 and older, despite completing standard driver education programs. In 2021, 2,116 drivers aged 15-20 were killed and an estimated 203,256 were injured in motor vehicle crashes in the United States, with similar patterns in Canada where motor vehicle crashes remain the leading cause of death among teenagers aged 15-24. In 2016, the UK's Transport Research Laboratory launched the Driver2020 trial - the largest driver safety study ever conducted with over 28,000 participants. They tested three evidence-based interventions: hazard perception training, parental engagement programs, and telematics feedback systems added to standard licensing programs. The result? No impact was detected on self-reported collisions in the first 12 months of driving amongst any of the 5 intervention groups compared to the control group. But here's what we haven't tested at scale: making comprehensive defensive driving the foundation of in-car training by specially trained instructors, before any other habits form.

My Wake-Up Call

At 17, I thought I was an amazing driver. For two years, I wove through traffic at high speed, confident I was impressing everyone I passed. At 19, I walked into my first Young Drivers of Canada class as a new employee. Within one hour, I realized I wasn't an impressive driver. I was a moving disaster who'd somehow avoided killing anyone. Everything I thought was skill turned out to be luck wrapped in ignorance. That wake-up call changed my life. It's why I've spent nearly 30 years since teaching defensive driving to everyone I care about - because the alternative is unthinkable.

What Driver2020 Really

Proved

Before you dismiss Driver2020 as just another failed study, understand what makes it different. These weren't poorly designed programs thrown together by bureaucrats. They were sophisticated interventions based on decades of road safety research, tested with proper controls and meaningful sample sizes. They failed anyway. Here's the question that's been rolling around in my head since reading this study: What if they succeeded in proving something more important than they realized?

Standard Driver Training

vs. Defensive Driving:

The Critical Difference

Almost every driver safety study I've reviewed, including Driver2020, shares a common design assumption. They all test defensive driving interventions as supplements to standard licensing programs. That seems logical. Standard programs already exist. They're regulated, widely available, and affordable. If we want to improve outcomes, we add better components: hazard perception modules, parental coaching, feedback technology. But there's a distinction buried in this approach that researchers consistently miss: Standard road test training prepares students to pass a driving test. Students learn traffic laws, basic vehicle control, and test procedures, often practised repeatedly in the same quiet neighbourhood until test day. The goal is licensing. Road tests measure rule-following and basic competence, not hazard perception or risk management. Comprehensive defensive driving teaches drivers to identify and reduce risk before it becomes a crash. Students practice in varied traffic environments from day one, not after licensing. The goal is building defensive habits as the foundation of their driving identity, not adding them as an afterthought.

Why This Distinction

Matters

Driving habits form during initial learning. When students spend their first 20-40 hours focused primarily on test procedures - staying in their lane, executing proper turns, parallel parking - those patterns become “what driving is” to them. Their brain codes "driving" as "follow the rules and demonstrate competence to an examiner." The defensive mindset never becomes foundational. It gets bolted on later, fighting against habits that are already hardwired. Driver2020, like virtually every intervention study in this field, tested whether you can meaningfully change driving behaviour after those foundational patterns are set. The results suggest you largely can't. The problem isn't that these skills are useless. The problem is they appear to boost overconfidence more than they improve safety. Students leave feeling more capable of handling dangerous situations, which paradoxically makes them more willing to create dangerous situations. Comprehensive defensive driving works from a different premise entirely.

Calibrated Confidence

vs. False Confidence

Instead of "how to recover from a skid," students learn "how not to skid in the first place." Instead of high- speed vehicle dynamics, they learn the massive differences between controlled racing environments (roll cages, helmets, medical staff on standby, no intersections, no oncoming traffic) and street driving. The approach doesn't eliminate optimism bias in teenagers. That would be impossible - their brains are literally still developing the capacity for risk assessment. What it does is provide reality-based information that makes their existing confidence more calibrated to actual capability. Consider the difference these specific inputs make to decision-making: Generic safety messaging: "Speed kills. Slow down." Calibrated information: "At 160 km/h, you need 205 meters of clear sight distance to a hazard (and at that distance, be able to identify what you’re seeing as a hazard) to stop safely for it. Your car travels 67 meters before your foot even reaches the brake pedal. That curve ahead reduces your sight distance to 150 meters. Here's what happens if you maintain this speed." Teenagers are excellent at detecting when adults oversimplify or hide information from them. When you treat them like they're incapable of understanding physics and consequences, they stop taking the message seriously. When you give them real data and decision-making frameworks that respect their intelligence, that respect becomes the foundation for better choices. Not perfect choices - they're still teenagers. But materially better choices than "speed kills" produces.

Young Drivers of

Canada: The Data That

Demands Investigation

Young Drivers of Canada has trained over 1.4 million Canadians since 1970 using comprehensive defensive driving as foundational training, not an add-on. Critical methodological context before I share their numbers: What follows is self-reported data without randomized controls, independent peer review, or third-party verification. Selection bias almost certainly plays a substantial role. Families who choose comprehensive training programs that cost significantly more and require greater commitment differ from the general population in ways that likely affect outcomes independent of the training itself. These aren't minor concerns. They're fundamental limitations that prevent us from drawing causal conclusions.

The Graduate Survey

Data

With those caveats clearly stated, here's what their data shows: Young Drivers of Canada released their comprehensive Graduate Survey Report in September 2025, analysing outcomes for 881 graduates from 2023 through August 2025 - covering participants tracked during their highest-risk driving years. Key findings: • 96.7% of graduates remained crash-free or were involved only in not-at-fault crashes • 92%+ safe driving rate specifically among 16-19 year-olds • 56.8% of graduates credited their YDC training with helping them avoid at least one specific crash

Why These Numbers

Matter

To understand the significance, you need context about what "normal" looks like for newly licensed teenagers. • Drivers aged 16-19 were involved in 4.8 fatal crashes per 100 million travel miles, compared to 1.4 for drivers aged 30-59—making teen drivers more than 3 times as likely to be involved in fatal crashes. • More 19-year-olds die or are seriously injured than any other age group in Canada. The first six months after licensing represent the absolute peak of crash risk. These are not small differences. These are order-of-magnitude differences in risk exposure. The 3.3% at-fault crash rate among YDC graduates, occurring during these peak-risk years and likely inflated by selection bias in their favour, still represents a substantial deviation from national patterns. Is this proof that comprehensive defensive training as foundation reduces crashes? No. Absolutely not. The selection bias alone prevents any causal claim. Is this signal strong enough to warrant properly controlled investigation? I'd argue yes, and here's why the signal matters despite the noise.

Why This Approach

Hasn't Been Studied

(And Why That's the

Real Story)

Here's where the research gap becomes impossible to ignore: Young Drivers of Canada appears to be the only large-scale organization teaching comprehensive defensive driving as foundational beginner training rather than post-license supplementation. They operate in only a handful of Canadian provinces. Their program costs significantly more than standard training. Most "defensive driving" in North America exists as: • Brief classroom-only modules embedded in standard licensing programs • Short post-license courses for ticket dismissal or insurance discounts • Advanced programs marketed to already-experienced drivers who want skill enhancement

The Theory-Practice Gap

In the United States, many states mandate teaching concepts like SIPDE (Scan, Identify, Predict, Decide, Execute) and Smith System principles. But these typically remain classroom theory. The vast majority of driving instructors haven't been trained to teach these concepts during behind- the-wheel practice in real traffic conditions. Students learn the ideas in a classroom, then spend their actual driving hours focused on executing turns properly and passing the road test. The concepts stay theoretical. The habits get built around test procedures. The crash rates don't budge. When Driver2020 researchers noted that "many interventions currently administered" have "little to no evidence supporting their effectiveness", they were absolutely correct. But they, and almost every other research team, have only tested supplements to a foundational system that, I would argue, are inadequate. We've run dozens of studies asking: "Can we improve outcomes by adding defensive training after standard licensing?" We haven't run the study asking: "What happens when comprehensive defensive training IS the foundation of in-car training by specially trained instructors, before any other habits form?" This isn't a small methodological oversight. This is a fundamental gap in our evidence base.

What Proper

Investigation Would

Require

I'm not advocating we accept anecdotal evidence or self-reported data as proof. I'm advocating we finally test what remains untested. Here's what rigorous research might look like: 1. Partner with Existing Implementation Young Drivers of Canada exists. Their curriculum is documented. Their instructors are trained in specific methodologies. Whatever its other limitations, it's a real-world implementation that's been refined over 54 years and 1.4 million students. 2. Use Random Assignment Recruit participants before they begin any driver training. Randomly assign them to comprehensive defensive training versus standard licensing programs. This controls for selection bias that plagues observational data. 3. Match for Confounding Variables Control for socioeconomic status, parental involvement, baseline risk- taking tendency, academic performance, geographic location. We know these factors influence crash risk. Proper matching or statistical controls can account for them. 4. Track Long-Term Outcomes Measure crash rates, near-miss incidents, traffic violations, and hazard identification performance. Follow participants for minimum 2-3 years post-license to capture patterns beyond the initial novelty period. 5. Use Established Safety Research Protocols This doesn't require inventing new methodologies. Road safety research has well-developed standards for crash tracking, incident reporting, and outcome measurement pioneered by organizations like NHTSA. 6. Conduct It at Scale Driver2020 tested 28,000 participants. We don't need that many for initial investigation, but we need sufficient sample size to detect meaningful differences in relatively rare events like serious crashes. The curriculum exists. The research protocols exist. One organization has implementation data suggesting this warrants investigation. Multiple research institutions have the expertise to design such studies. What's missing is the decision to test the foundation rather than testing another add-on.

What We Already Know

Works (And Why It's Not

Enough)

To be clear: I'm not suggesting we abandon proven interventions. Graduated Driver Licensing Works. Graduated Driver Licensing (GDL) restrictions demonstrably reduce crashes by limiting exposure during highest-risk conditions. The most restrictive GDL programs are associated with a 38% reduction in fatal crashes and a 40% reduction in injury crashes among 16-year-old drivers. That intervention doesn’t work by making them better drivers, but by eliminating the highest risk factors for them. Components that work: • Limiting nighttime driving • Restricting teenage passengers • Extending supervised practice periods • Minimum learner permit holding periods These are valuable tools. They're evidence-based. They should remain part of comprehensive safety strategies. But GDL Has Limitations. Research shows that GDL policies reduce the number of accidents by limiting the amount of teenage driving rather than by improving teenage driving. The same study found that teen driving quality does not improve ex-post GDL exposure. GDL restrictions reduce exposure to risk. They don't fundamentally change how drivers think about risk when they eventually face it without restrictions. Hazard perception training improves identification. It doesn't necessarily change whether drivers act on what they've identified, especially when that action requires overriding habits formed during initial training. We've optimized the supplements. Maybe it's time to question the foundation.

Frequently Asked

Questions

What is comprehensive defensive driving? Comprehensive defensive driving is an approach to driver education that focuses on hazard recognition, risk assessment, and crash risk mitigation from the very first lesson. Meaningful emergency maneuvers are included (well beyond parallel parking). Unlike standard driver training that primarily prepares students to pass a road test, comprehensive defensive driving builds the defensive mindset as the foundation of a driver's skill set as their habits form. Key elements include: • Training in varied traffic conditions from day one (not just quiet test routes) • Specialized instructor training in teaching the practical application of defensive methodology • Typically 40+ hours of behind-the- wheel practice • Focus on identifying and reducing risk before it becomes a crash • Decision-making frameworks based on physics and real-world consequences At what age should teens start defensive driving training? Most comprehensive defensive driving programs, including Young Drivers of Canada, begin instruction when students are eligible for their learner's permit (typically 14-16 years old depending on jurisdiction). Starting early is crucial because driving habits form during initial learning. The goal is to build defensive patterns from day one rather than trying to retrofit them after test-focused habits are already established. How much does comprehensive defensive driving cost? Comprehensive defensive driving programs typically cost $2,500 + compared to $500-800 for standard driver education. The higher cost reflects: • 14+ hours of practical defensive mindset instruction vs. 10+ hours of test preparation in standard programs • Specialized instructor training in teaching defensive methodology in car • Practice in all traffic environments rather than repeated test routes • Smaller class sizes for more personalized instruction While the upfront cost is higher, many families consider it an investment in safety, especially given that motor vehicle crashes were the leading cause of unintentional death for the 15-24 age group in 2020. Does driver's education actually work? The Driver2020 study - the largest real-world trial of driver safety interventions with 28,000 participants - found that adding hazard perception training, parental engagement programs, and telematics feedback to standard licensing programs produced zero reduction in crashes. However, this doesn't mean all driver education is ineffective. It suggests that: • Supplements added after standard training does not work • The foundation of standard licensing programs may be inadequate • We haven't properly tested making comprehensive defensive driving the foundation Graduated Driver Licensing programs, which limit exposure to high-risk situations, have shown 38-40% reductions in crashes, but these work by reducing exposure rather than improving driving quality. Is defensive driving worth it for teens? From a safety perspective, the theoretical case is strong: teens face crash rates up to 3 times higher than experienced drivers, and 19-year-olds show the highest death and serious injury rates of any age group in Canada. Young Drivers of Canada reports a 96.7% crash-free rate among their graduates, but this data suffers from selection bias and lacks randomized controls. The honest answer: We need properly controlled research to determine effectiveness definitively. Until then, families must choose based on: • Observational data suggesting benefit • The logic of building defensive habits from the start • Their own risk tolerance and resources In my experience, the self reported data from Young Drivers of Canada tracks with the outcomes I have observed in the people I have trained over the decades. What's the difference between defensive driving and driver's ed? Standard Driver's Ed: • Primary goal: Pass the road test • Typical duration: 10-20 hours • Practice environment: Quiet neighbourhoods, repeated test routes • Focus: Rules, procedures, vehicle control • Outcome measure: Test passage rate Comprehensive Defensive Driving: • Primary goal: Avoid crashes through hazard recognition • Typical duration: 30+ hours • Practice environment: All traffic types trained prior to road test • Focus: Risk identification and reduction • Outcome measure: Crash avoidance through proactive methods The fundamental difference is when the defensive mindset is introduced - as a foundation versus as an afterthought. How can I choose a good driving school? Look for programs that: ✓ Train instructors specifically in an established defensive driving system for incar training (not just basic certification) ✓ Begin hazard recognition training on day one (not after test prep) ✓ Practice in varied traffic conditions (not just quiet test routes) ✓ Teach risk calibration and decision- making frameworks (not just rules) ✓ Have a quality control system in place to ensure all instructors are teaching all of the curriculum ✓ Focus on crash avoidance (not just test passage) ✓ Include real-world scenario discussion ✓ Have measurable safety outcomes (even if self-reported) Ask specific questions: • "Are your instructors trained differently than standard certification?" • “Do you have a quality control system in place to ensure all of your instructors follow your curriculum? • "What percentage of driving time is spent in varied traffic vs. test routes?" • "When do you introduce hazard perception training?” (assuming they do at all). • "Do you track crash rates among graduates?"

What Happens Next

I don't have funding for multi-year randomized trials. I don't have academic institutional backing or access to national crash databases. What I have is decades of experience seeing the difference between students trained to pass tests and students trained to avoid crashes. I have enough conviction in that difference to build a business around it.

Driving Hero Launch

Driving Hero Academy is running beta classes now with full launch in 2027, based on comprehensive defensive training as foundation, not supplement. That's not a research project. It's a commercial venture based on my belief that this approach works better than the alternatives currently available. But belief isn't evidence, and commercial implementation isn't research.

A Call to Researchers

If you're a researcher interested in designing proper trials, I'll contribute whatever expertise I can to help design them. The complete Driver2020 methodology provides a template for what doesn't work when added to standard training. Now let's test what we haven't tested yet. You can reach out to me through LinkedIn or contact page.

A Note to Parents

If you're a parent exploring this approach for your family, Driving Hero beta classes are running now. But understand you're choosing based on observational data and personal conviction, not peer- reviewed evidence. In my opinion, it is your best bet to ensure the degree your new driver is trained to. The research gap is real. The theoretical case is sound. The observational data is suggestive but flawed. Until we have proper controlled trials, this remains a calculated risk based on incomplete information.

To the Sceptics

If you think this entire argument is flawed, that selection bias explains everything, that standard training is adequate, that I'm overselling unproven methods, drop your critique in the comments. Seriously. This conversation needs sceptical voices as much as it needs advocates. Driver2020 proved the current approach isn't working. Now let's test what we haven't tested yet.

Key Takeaways

Current State: Standard driver education programs, even with evidence-based supplements, show no reduction in teen crashes Teen drivers (16-19) have crash rates 3x higher than experienced drivers Motor vehicle crashes remain the leading cause of death for ages 15-24 The Gap: We've tested adding defensive training after standard programs (doesn't work) We haven't tested making comprehensive defensive training the foundation One organization has 54 years of implementation data suggesting this warrants investigation What Works Now: Graduated Driver Licensing reduces crashes 38-40% by limiting exposure. But GDL reduces exposure, not skill - it limits driving rather than improving it. What's Needed: • Randomized controlled trials of foundation-level comprehensive defensive training • Proper controls for selection bias • Long-term outcome tracking • Willingness to question foundational assumptions about driver education What's your experience with the gap between test-ready and road-ready? Have you seen differences in how foundational training versus add-on training affects actual driving behaviour?

References

National Highway Traffic Safety Administration. (2022). Young Drivers: 2020 Data (Traffic Safety Facts. Report No. DOT HS 813 313). U.S. Department of Transportation. MADD Canada. (2024). Statistics & Links: Youth and Impaired Driving. Weekley, J., & Helman, S. (2024). Driver2020: An Evaluation of Interventions for Novice and Learner Drivers - Report D3: Delivery and Engagement (PPR2011). Transport Research Laboratory. UK Department for Transport. (2025). Introducing a Minimum Learning Period for Learner Drivers (Category B Driving Licence). GOV.UK. Young Drivers of Canada. (2025). Graduate Survey Report 2023-2025. Baker, S. P., Chen, L. H., & Li, G. (2007). Graduated Driver Licensing Programs and Fatal Crashes of 16-Year-Old Drivers: A National Evaluation. National Highway Traffic Safety Administration. Referenced in: NHTSA Countermeasures That Work: Graduated Driver Licensing. Dee, T. S., Grabowski, D. C., & Morrisey, M. A. (2005). Behavioral Impact of Graduated Driver Licensing on Teenage Driving Risk and Exposure. Journal of Health Economics. PMC2824081.

About the Author

La Velle Goodwin has spent nearly three decades thinking about why driver behavior is so hard to change, and how to actually change it. Her path into road safety was not a career plan. Hired into the sales department at Young Drivers of Canada, she was required to take the program as a condition of employment. She went from believing her aggressive driving habits were a sign of skill to recognizing they had been a sustained act of luck. That shift was sharp enough that she immersed herself in crash research, driver psychology, and industry training standards before becoming their on-air expert for media, and eventually completing YDC's instructor certification, a process requiring more than four times the training of a standard driving instructor license with mandatory annual recertification requiring instructors to retrain and meet progressively higher scoring targets on practical in-car exams, advancing through successive certification levels as a condition of continued employment. She delivered YDC's commercial driver training program, Collision Free, working directly with experienced drivers and observing firsthand the attitudes, blind spots, and psychological reluctance that make behavior change so difficult to achieve in professional driver populations. She founded Driving Hero Academy on the same principles she has been teaching privately for decades, long before it became a business, because she has seen firsthand what foundational defensive training produces compared to everything that gets bolted on after the fact.