Again, I’m looking for a prior. All those things come after. When I tried to do my own analysis, I got stuck on a prior, and realized I don’t have a good idea where to put it.
I’ll rephrase the question: what percentage of such offers (counting everything regardless of these other factors) are more or less true? After you have that, you can update up or down based on any other info.
Different people consider different claims “too good to be true”. To produce a specific number you would have to provide a more precise notion of “too good to be true”.
How about take the reference class of “things publically accused of being a scam by non-official entities, but not by official ones (like any government agencies, bbb, media, etc.).” Weigh different offers by publicity (more precisely, the number of people who potentially thought about using whatever it was.)
I don’t think it’s useful to make decisions based on whether or not a random person on the internet calls something a scam or an illegal scam.
When it comes to a financial investment it makes sense to ask: “If they advertise those returns and those returns are real, why doesn’t some smart bank simply invest money into the vehicle?”
“The deal still seem to look good even if they company would reduce the rates of return, so why don’t they reduce the rates to make more profit for themselves?”
For a lot of other products: “What do trustworthy knowledgable people say about the product?”
“Who can I ask?”
I’m not using this as a decision maker end-all, I’m using it to define the reference class to get a prior. All those questions are calculated afterwards. I’m trying to define the “too-good” category, not define something that would discriminate between scams and non-scams.
The core question isn’t whether something is too good to be true, but whether it’s reasonable that an opportunity like that exists without information being hidden from you.
Hm. What I mean is that when I try to weigh up the evidence, it seems pretty balanced (perhaps slightly weighted toward genuine), so the prior will determine it. If the prior was 10%, I would conclude that it was probably real, versus if the prior was 1%, I would conclude it was fake.
If you want me to explain what I mean by prior …
Before doing any investigating, what is the probability of something that I am likely to hear about that fits the intuitional category of “too good to be true” to be more-or-less true? (I’m assuming an implicit weighting for popularity, which seems fair. OTOH it might be hard to estimate popularity for different people.)
I think many people in this subthread are suggesting ways of interpreting the evidence that you may not have (or may have) thought of, or alternately, additional pieces of potential evidence that may not obviously be evidence. So it seems like the question you should really be asking is, “how do I assess this opportunity?” rather than “what should the prior be?”
Tentatively, how much too good to be true is it? Does resemble past scams? Do the people making the offer get angry when they’re asked about details?
Again, I’m looking for a prior. All those things come after. When I tried to do my own analysis, I got stuck on a prior, and realized I don’t have a good idea where to put it.
I’ll rephrase the question: what percentage of such offers (counting everything regardless of these other factors) are more or less true? After you have that, you can update up or down based on any other info.
Different people consider different claims “too good to be true”. To produce a specific number you would have to provide a more precise notion of “too good to be true”.
How about take the reference class of “things publically accused of being a scam by non-official entities, but not by official ones (like any government agencies, bbb, media, etc.).” Weigh different offers by publicity (more precisely, the number of people who potentially thought about using whatever it was.)
Is that well-defined enough?
If you search well enough I think that you can find for most products that somewhere on the internet a person calls it a scam.
How about specifically an illegal scam? The way scam is used doesn’t always imply illegal.
Do you have any ideas about how we can define this in a useful way?
I don’t think it’s useful to make decisions based on whether or not a random person on the internet calls something a scam or an illegal scam.
When it comes to a financial investment it makes sense to ask: “If they advertise those returns and those returns are real, why doesn’t some smart bank simply invest money into the vehicle?” “The deal still seem to look good even if they company would reduce the rates of return, so why don’t they reduce the rates to make more profit for themselves?”
For a lot of other products: “What do trustworthy knowledgable people say about the product?” “Who can I ask?”
I’m not using this as a decision maker end-all, I’m using it to define the reference class to get a prior. All those questions are calculated afterwards. I’m trying to define the “too-good” category, not define something that would discriminate between scams and non-scams.
I think it’s a poor reference class.
The core question isn’t whether something is too good to be true, but whether it’s reasonable that an opportunity like that exists without information being hidden from you.
I may not know what you mean by a prior. Could you give me some examples?
Hm. What I mean is that when I try to weigh up the evidence, it seems pretty balanced (perhaps slightly weighted toward genuine), so the prior will determine it. If the prior was 10%, I would conclude that it was probably real, versus if the prior was 1%, I would conclude it was fake.
If you want me to explain what I mean by prior …
Before doing any investigating, what is the probability of something that I am likely to hear about that fits the intuitional category of “too good to be true” to be more-or-less true? (I’m assuming an implicit weighting for popularity, which seems fair. OTOH it might be hard to estimate popularity for different people.)
I think many people in this subthread are suggesting ways of interpreting the evidence that you may not have (or may have) thought of, or alternately, additional pieces of potential evidence that may not obviously be evidence. So it seems like the question you should really be asking is, “how do I assess this opportunity?” rather than “what should the prior be?”