The Vegas Odds put the probability of a Hillary Presidency at .95 against a Trump .2
The reasons are simple:
The phony scandals are only believed by the far right - they are not potential Dem voters
The Republican governance history is abysmal - Bush did not keep us safe, militarily or financially
The Obama administration has achieved steady, substantial progress for seven years
Clinton has the experience and skill to govern
The odds makers are correct - It’s going to be Mrs. President!
Crane
Out of curiosity, can you link to your source of “Vegas Odds”? Most places I’ve seen have Clinton favored, but not that degree.
That doesn’t have Clinton anywhere near 0.95.
I can’t see the link, but that .95 sounded like odds to win the nomination, not the general election (contrary to the post that first mentioned it). So my guess is a misunderstanding of what the number is for.
She’s actually at 58% right now. 81% chance of winning the nomination.
Oddly I’d put her chance at the nomination at considerably higher (close to 95%), with her chance at the general at about 55% right now.
What’s hard to predict is how Sanders wins in Iowa and NH would change things. Nate Silver seems to think that he can win those two and then lose everywhere else, but history tells us that winning IA and NH means a huge bump in support.
That’s hard to predict, but his chances at Iowa seem pretty low based on recent polling.
Three things.
I’m not sure I trust PredictIt yet. It’s still in beta and its smaller volume makes it susceptible to more noise and more it’s more easily influenced by a handful of big bettors. I prefer PredictWise, but it uses a combination of polls and real money betting data.
You are computing your probabilities incorrectly by failing to account for a small vig at PredictIt. The difference is small, but it is a difference.
Possible losses in IA and NH are already baked in to the probabilities derived from prediction markets. That’s how prediction markets work.
The odds vary and perhaps I calculated the probability incorrectly:
5/6 to win the Presidency, I calculated as a probability of .83
-245 to win the Presidency I calculated as a probability of .95
In any case Clinton is the strong favorite with the odds makers.
The current chaos in the Republican Party does not bode well for their candidates.
Crane
If I’m understanding this right, this gives an implied probability of 55% without adjusting for vig. (6/(5+6)
The implied probability here is 71% (also no vig). (245/(100+245))
I’d still like to see where these numbers came from. In particular the 71% is higher than I’ve seen anywhere.
Lance,
Couldn’t find the -245.
Here is a similar link:
Using your equation puts Fiorina (1000) at .90.
Crane
-x gives the implied probability x/(100 + x).
+x gives the implied probabilty 100/(100 + x).
I’ll work this out below for the dem nomination for clarity. For those who don’t care feel free to skip this post.
From the linked page at present we have:
Hillary Clinton -1200
Bernie Sanders +600
These yield implied probabilities:
Hillary Clinton: 1200/(100 + 1200) ~ 0.923076923
Bernie Sanders: 100/(100 + 600) ~ 0.142857143
Note that the sum of these is greater than one so we normalize to correct for the vig:
Hillary Clinton: 0.923076923/(0.923076923 + 0.142857143) ~ 87%
Bernie Sanders: 0.142857143/(0.923076923 + 0.142857143) ~ 13%
And since I’m bored, here’s what we get for president from the linked OddsShark page (with intermediate steps omitted).
Hillary Clinton -110 48.2%
Donald Trump +500 15.3%
Marcio Rubio +600 13.1%
Bernie Sanders +1000 8.4%
Jeb Bush +1200 7.1%
Ted Cruz +3300 2.7%
Carly Florina +4000 2.2%
Chris Christie +5000 1.8%
John Kasich +7500 1.2%
Lance,
Thanks.
Using your equations:
Clinton .43
Bush, Trump, Rubio .14
Fiorina .05
That makes sense.
Thanks again,
Crane
Lance,
I used older numbers.
Your equation converges on .5.
Crane
Lance,
-110 means that the bet is on the favorite and will return $1.10 for a $1.00 bet.
That’s odds of 10:1 or a probability of .9.
Crane
Crane,
-110 means you have to bet $110 to win $100 (plus getting your original $110 back).
See here. Although many sites will tell you the same thing. Search for the phrase, “Converting ‘minus’ moneyline odds into implied probability formula.”
Lance,
Oh, I see. The + values converge on 50% and the - values diverge.
Thanks,
Crane