If Ron Paul runs as a third party candidate in the 2012 presidential election, then he will receive greater than 10% of the popular vote.
Created by Jayson Virissimo on 2011-11-10; known on 2012-11-07; judged wrong by ChristianKl on 2012-11-29.
- Jayson Virissimo estimated 70% on 2011-11-10
- Jayson Virissimo said “BTW, what is the convention about judging conditional predictions? Should we consider the prediction true if the antecedent is false like in propositional logic or just withdraw the prediction?” on 2011-11-10
- muflax estimated 60% on 2011-11-10
- chemotaxis101 estimated 40% on 2011-11-10
- moridinamael estimated 60% on 2011-11-10
- RobertLumley estimated 75% and said “I would say that you withdraw it.” on 2011-11-10
- lavalamp estimated 25% on 2011-11-10
- gwern estimated 20% and said “Jayson, I have been treating them as material conditionals and marking correct if the antecedent is false. anything else seems wrong” on 2011-11-10
- faul_sname estimated 33% on 2011-11-10
- fork estimated 15% on 2011-11-12
- NathanMcKnight estimated 20% and said “If the antecedent proves false, then no prediction will have been made, by the statement’s own internal logic. In which case, it should be withdrawn. ” on 2011-11-12
- RobertLumley said “So this is to be withdrawn?” on 2012-11-07
- RandomThinker said “looks like it to me” on 2012-11-07
- Jayson Virissimo said “Marking true per convention (see gwern’s comment).” on 2012-11-13
- Jayson Virissimo judged this prediction right on 2012-11-13.
- RandomThinker said “@jayson @gwern I think these should be withdrawn. Otherwise your stats for calibration will be screwed up because you make a lot of questions that never materialize. I consider these “ifs” more assumptions than logical conditionals.” on 2012-11-13
- RandomThinker said “Otherwise there’s no penalty for wildly overguessing on conditional predictions, E.g. for this question, say Paul has 10% chance of running as third party. Then you can guess up to 90% and be calibrated, which over-estimates his chances.” on 2012-11-13
- RandomThinker judged this prediction unknown on 2012-11-13.
- Jayson Virissimo said “RandomThinker, I am hoping to get some kind of agreed upon convention going here; I don’t really care all that much what it is (although, I slightly lean towards standard formal logic rules).” on 2012-11-13
- Jayson Virissimo said “As long as you keep in mind that the principle of indifference gives conditional statements a probability of 75%, you won’t see these higher estimates as “wildly overguessing”.” on 2012-11-13
- RandomThinker said “In prediction, 90% means something is highly likely to happen. If it turns out it’s just predictions that trigger on very unlikely conditions, that would be very counter intuitive. IMO. What do others think?” on 2012-11-13
- Jayson Virissimo said “Assigning a 90% probability to this statement is (logically) equivalent to assigning a 10% probability to the statement ‘Ron Paul will run as a third party candidate and he will receive equal to or less than 10% of the popular vote’.” on 2012-11-14
- RobbBB said “If ‘→’ is material implication, p → q is truth-functionally equivalent to ¬(p ∧ ¬q). But ‘if p then q’ in colloquial English almost never means precisely ‘p → q’.” on 2012-11-24
- RobbBB said “When I say ‘If it snows tomorrow, school will be cancelled’ and it doesn’t snow, I don’t ordinarily conclude ‘I was right!’. I conclude that I was neither right nor wrong, or that the truth-value remains unknown.” on 2012-11-24
- RobbBB said “To increase expressivity, I suggest just saying ‘¬(p ∧ ¬q)’ when you mean ‘¬(p ∧ ¬q)’. Say ‘if p then q’ when you mean ‘(p → q) ∧ (¬p → X)’. When p fails to obtain, the X blocks evaluation. ” on 2012-11-24
- RobbBB said “Remember the point of PredictionBook. It’s not to propagate the crude semantics of propositional logicians, or to reward people who make disproportionately conditional predictions.” on 2012-11-24
- ChristianKl judged this prediction wrong on 2012-11-29.