"This shows the chance of winning the general election IF NOMINATED:
[chart showing that P(Newsom wins general election | Newsom wins Dem nomination) is 72%, while P(Biden wins general election | Biden wins Dem nomination) is 42%]
As you can see, bettors say that Democrats should absolutely switch horses, even at this late stage."
Not exactly. These are bets on correlation, not causation — e.g. bettors might be imagining some event E that (a) is outside of anyone in the Dem party's control, (b) makes Newsom much more likely to win the Dem nomination, and (c) makes Newsom much more likely to win the general election.
Yes, I agree, and this is noted on the site itself: "CAVEAT: While these numbers generally show how "electable" a candidate is, an exception is when a party's nomination pick depends on how strong its position is that year. Perhaps a party is more likely to pick Candidate A in a year when the party is popular. In that case, the odds on this page would suggest not that Candidate A is electable, but rather that he or she will be picked in an "electable" year." https://electionbettingodds.com/ElectabilityDEM.html
That being said, I do not think that is a big factor at the moment. The closer you get to the decision point, in general, the less of a factor it becomes.
"This shows the chance of winning the general election IF NOMINATED:
[chart showing that P(Newsom wins general election | Newsom wins Dem nomination) is 72%, while P(Biden wins general election | Biden wins Dem nomination) is 42%]
As you can see, bettors say that Democrats should absolutely switch horses, even at this late stage."
Not exactly. These are bets on correlation, not causation — e.g. bettors might be imagining some event E that (a) is outside of anyone in the Dem party's control, (b) makes Newsom much more likely to win the Dem nomination, and (c) makes Newsom much more likely to win the general election.
DYNOMIGHT has a really good essay on this: https://dynomight.net/prediction-market-causation/
Yes, I agree, and this is noted on the site itself: "CAVEAT: While these numbers generally show how "electable" a candidate is, an exception is when a party's nomination pick depends on how strong its position is that year. Perhaps a party is more likely to pick Candidate A in a year when the party is popular. In that case, the odds on this page would suggest not that Candidate A is electable, but rather that he or she will be picked in an "electable" year." https://electionbettingodds.com/ElectabilityDEM.html
That being said, I do not think that is a big factor at the moment. The closer you get to the decision point, in general, the less of a factor it becomes.