Whoa, hang on...this survey isn't attempting to determine scientific significance, or to pull from a representative sample of the world's tool buying population. So let's not criticize this humble survey for not spending $50,000 to get a randomized, representative sample.
This poll is a mix of avid users and collectors, but likely more users than collectors. Collectors are even more harsh on markings. I know lots of user/collectors who pass on most owner-marked tools (unless they just can't find a missing item otherwise, or it's a ridiculous bargain). And the data above bears that out. My point is, this particular survey sample is probably more forgiving of markings than are collectors. So if we are talking collectible tools, the possibility exists that the required percentage off becomes greater than 40%, not smaller.
Now, any particular person can dismiss the data collected here and say, "That's not my experience." So what? One person's experience is an anecdote. A lot of people's experience is data. At this point we have data based on 71 responses from a pool of tool fanciers. I'd take the data of these 71 people, over a single person's guesstimate, any day.
What I'm saying is, it's unfair to criticize a survey as "not a representative sample," and then replace it with a single person's guesstimate as being more accurate...talk about an incredibly unrepresentative sample, a sample of one!
You could do a real study that went beyond preference (as in this survey) to actual behavior. Study eBay auctions and compare marked vs unmarked items, and see what they sold for. Then do a comparison. That's a lot of work but the results would be fascinating.
Until there's better data, however, I think the 40% number is useful. It helps me not overpay for marked tools, which are obviously a poorer investment than clean tools.
But until this survey, I had no idea how much worse an investment they really were.