3 ms·

Yes, a thing can be most likely, and not likely. It is talking about the relative likelihood of the possible origins.

TeMPOraL 4y ago
> It is talking about the relative likelihood of the possible origins. That's how you read it. GP is reading "most likely" as statement of absolute likelihood, making it false here, because the most likely thing is "any of the possibilities not under consideration", which takes most of the probability mass. IMO, GP's interpretation is the one that most people take for "most likely" statements, which in our opinion makes this headline a blatant lie.
- wizofaus 4y ago
  If you told most people you had 10 rolls of a dice to land a 6 and then suggested the most likely number of rolls you'll get it on is 1 I wonder whether they'd believe you...
  - TeMPOraL 4y ago
    I think most people would have the same reaction as me, that is "er, I'm having trouble parsing this, could you rephrase?". Anyway, a different way of putting my point is: "most likely" modifies meaning of the thing it attaches to. "Lab leak [is the] most likely origin of" strongly suggests the absolute take. Relative take would be phrased "Lab leak [is the] most likely of hypotheses proposed for the origin of", or something to that effect - that is, attaching "most likely" to the body of proposed explanations (which, by definition, excludes "anything else not on this list" as an option), not to the origin itself. A yet different way: NYT is doing the equivalent of me putting some object under a color detector, and announcing "Blue most likely color of Object", without mentioning that my color detector consists of separate R, G, and B color sensors, and that I'm just reporting which sensor gave the highest reading. Most people would read my announcement as "Object is some shade of blue, unless their machine is broken or something", whereas the object itself could be purple, or white, or yellow, or near-black, or any other color that just has "B" component stronger than "G" and "R".

> It is talking about the relative likelihood of the possible origins. That's how you read it. GP is reading "most likely" as statement of absolute likelihood, making it false here, because the most likely thing is "any of the possibilities not under consideration", which takes most of the probability mass. IMO, GP's interpretation is the one that most people take for "most likely" statements, which in our opinion makes this headline a blatant lie.

wizofaus 4y ago
If you told most people you had 10 rolls of a dice to land a 6 and then suggested the most likely number of rolls you'll get it on is 1 I wonder whether they'd believe you...
- TeMPOraL 4y ago
  I think most people would have the same reaction as me, that is "er, I'm having trouble parsing this, could you rephrase?". Anyway, a different way of putting my point is: "most likely" modifies meaning of the thing it attaches to. "Lab leak [is the] most likely origin of" strongly suggests the absolute take. Relative take would be phrased "Lab leak [is the] most likely of hypotheses proposed for the origin of", or something to that effect - that is, attaching "most likely" to the body of proposed explanations (which, by definition, excludes "anything else not on this list" as an option), not to the origin itself. A yet different way: NYT is doing the equivalent of me putting some object under a color detector, and announcing "Blue most likely color of Object", without mentioning that my color detector consists of separate R, G, and B color sensors, and that I'm just reporting which sensor gave the highest reading. Most people would read my announcement as "Object is some shade of blue, unless their machine is broken or something", whereas the object itself could be purple, or white, or yellow, or near-black, or any other color that just has "B" component stronger than "G" and "R".