You really don't understand statistics and how to use them.... I think you've hit on one of the biggest misunderstandings in discussions about Thai road safety. People tend to quote one statistic—deaths per 100,000 population—and then assume it tells the whole story. It doesn't. It's only one of many indicators used by road safety professionals, and by itself it tells us remarkably little. "Thailand has one of the world's deadliest roads." Does it? Whenever Thailand's roads are discussed, someone inevitably quotes the WHO figure of road deaths per 100,000 population as though it settles the debate. It doesn't. The problem isn't the statistic itself—it's treating one measure as though it tells the whole story. Road safety is an evidence-based discipline. It relies on engineering, epidemiology, collision investigation and statistical analysis. No serious road safety researcher would judge a country's roads using a single number in isolation. The statistics commonly used include: Fatalities per 100,000 population Fatalities per billion (or 10 billion) vehicle-kilometres travelled Fatalities per 100,000 registered vehicles Crash rates Serious injury rates Minor injury rates Severity of crashes Traffic density Vehicle ownership per capita Vehicle mix (cars, motorcycles, trucks, buses etc.) Road type and engineering Driver behaviour and enforcement Emergency medical response times Each tells us something different. Why population alone can be misleading Suppose Country A and Country B both record 30 road deaths per 100,000 population. That doesn't necessarily mean the roads are equally dangerous. Country A may have twice as many vehicles. Country B may drive twice as many kilometres. Country A may have 80% of its traffic on motorcycles. Country B may be almost entirely cars. Those are completely different risk environments. Exposure matters One of the most useful measures is: Fatalities per billion vehicle-kilometres travelled. This measures risk according to how much traffic is actually using the roads, not simply how many people live in the country. Similarly, Fatalities per 100,000 registered vehicles helps compensate for countries with very different levels of vehicle ownership. Thailand is a special case Thailand's statistics are heavily influenced by one fact that is often ignored. Around 80% of road deaths involve motorcycles. That immediately tells us the problem isn't simply "Thai roads." It reflects: the huge number of motorcycles low helmet compliance poor licensing weak enforcement vulnerable road users. For someone travelling in a properly maintained car, the risk profile is completely different. What about traffic density? Another useful statistic is traffic density. Countries with very high traffic volumes naturally experience more conflicts between vehicles. Researchers therefore examine: vehicles per kilometre of road traffic flow congestion levels vehicle mix These all affect collision risk. The bottom line Road safety isn't measured with one statistic. It is assessed using a whole range of measures, each describing a different aspect of the transport system. The problem is that many discussions about Thailand stop at one headline: "Thailand has one of the world's highest death rates per 100,000 population." True. But that's only one piece of a much larger picture. Good road safety analysis asks why those deaths occur, who is dying, what they're driving, how far they're travelling, where the crashes occur and what can be done to reduce the risk. Simply quoting one statistic without understanding what it actually measures isn't road safety analysis—it's confirmation bias dressed up as evidence. One final point I'd add, because it's often overlooked, is that deaths per 100,000 population was originally designed as a public health indicator, not as a measure of engineering risk on a road network. It's useful for comparing the health burden of road trauma between countries, but it's a relatively blunt instrument if you're trying to assess how safe a transport system actually is. That's why road safety researchers routinely use multiple exposure-based measures rather than relying on population alone.
