Quote Originally Posted by 4eyedbuzzard View Post
Actually, I didn't. If you'll note, I did say that there were some 90,000 thru-hiker attempts out of the roughly 20,000+ completions, and that the presence of section and other hikers reduces the effective rate: "There are probably 10 times as many section hikers on the trail as thru-hikers, all potential targets as well, which would reduce the AT rate substantially. But not by a factor of 10 as they typically spend a week or less on the trail. And it's probably fair to assume that the murderers probably don't know if you're a thru-hiker or section hiker. But even so, the presence of other hikers/potential victims obviously would reduce the overall any-hiker murder rate. The section hiker factor probably cuts the actual any-hiker murder rate in half or more."

Are these VERY rough estimates? Yes. There just isn't enough data to get really accurate rates. But it was just to point out that the effective murder rate isn't the almost alarmingly high one of 7 out of 20,000 - or even 90,000. I would hesitate to use the 2-3 million number as most are day hikers, and almost all the murders have been related to hikers staying overnight camped or at shelters. That's also why day hiker murders like Meredith Emerson weren't considered, even though a strong case could be made to include at least 4 other murders of day and other hikers related to the AT, but not on the AT proper. The best way would be to come up with a ratio of hiker miles or time vs murder rate, but this would be most difficult given available data.
"Based on this, the AT thru-hiker murder rate is slightly higher than the US average." This was your conclusion. It's very wrong, based on the reasons you stated.