2020 Presidential Election

[PackMan97]

FWIW, I see that three most inaccurate years from the sample pool were from 1980 or before. Presumably, polling has gotten better (and as the article states, the most accurate oddly enough was the Clinton - Trump match in 2016 which ended up only being off by about 3%).

I imagine the more well known the candidates, the more Predictive early polling might be. Clinton in 2016 was one of the most well known candidate in my lifetime (excluding incumbents like Obama in 2012). Trump was also a very popular due to celebrity status even if an unknown politically. His fame went back at 30 years.

[OldPhiKap]

I probably won't be active in this thread until after the first few debates, but I also don't want to be accused of hindsight, so just putting this here. I concur with PM97 that Biden will crash and burn. (Not at all an out-on-the-limb prediction, of course -- prior to the 2016 cycle, it was routine for early polling leaders to crash and burn. Think Herman Cain in the fall of 2011 being the polling frontrunner, for example.) And if the crash and burn occurs as the result of a 1-on-1 confrontation at a debate (think Rubio getting crushed by Christie), watch out for Buttigieg (a very talented politician, imo) to be the destroyer of Biden's presidential dreams.

And President Michelle Bachmann says hello too!

I read an article this morning, noting that Joe's support in Iowa was broad but not terribly enthusiastic. He has been out of the game for awhile. It will be interesting to see if he still has his chops or not.

Unlike some of those front-runners who crashed, though, Joe has a ton of endorsements and a ton of money raised. Even if his candidacy is mortally wounded, he can zombie on for a long time. But I agree that Mayor Pete or Kamala Harris could well claim the center(ish) lane from Joe before all is said and done.

[dudog84]

I probably won't be active in this thread until after the first few debates, but I also don't want to be accused of hindsight, so just putting this here. I concur with PM97 that Biden will crash and burn. (Not at all an out-on-the-limb prediction, of course -- prior to the 2016 cycle, it was routine for early polling leaders to crash and burn. Think Herman Cain in the fall of 2011 being the polling frontrunner, for example.) And if the crash and burn occurs as the result of a 1-on-1 confrontation at a debate (think Rubio getting crushed by Christie), watch out for Buttigieg (a very talented politician, imo) to be the destroyer of Biden's presidential dreams.

Are you seriously comparing Herman Cain to Joe Biden? Or anyone, for that matter?

[Troublemaker]

FWIW, I see that three most inaccurate years from the sample pool were from 1980 or before. Presumably, polling has gotten better (and as the article states, the most accurate oddly enough was the Clinton - Trump match in 2016 which ended up only being off by about 3%).

While one certainly cannot take these polls as predicative gospel, the article notes that it does give us present-time horse race information:

Should be noted that the "No" on Trump winning the Presidency in 2020 is trading on PredictIt at a relatively low 59 cents (relative to the polling he receives, that is).

Basically if these polls are to be taken seriously at all, we should all be rushing to buy "No" shares at 59 cents in order to receive a 69% profit if I'm doing the math right.

The betting markets are basically saying that the polls today are near meaningless and that the polls in the future will inevitably show a closer horse race.

[OldPhiKap]

Should be noted that the "No" on Trump winning the Presidency in 2020 is trading on PredictIt at a relatively low 59 cents (relative to the polling he receives, that is).

Basically if these polls are to be taken seriously at all, we should all be rushing to buy "No" shares at 59 cents in order to receive a 69% profit if I'm doing the math right.

The betting markets are basically saying that the polls today are near meaningless and that the polls in the future will inevitably show a closer horse race.

But what if Trump played Alabama?

[OldPhiKap]

Democratic Debate line-up is out:

https://www.cnn.com/2019/06/14/polit...020/index.html

Biden, Sanders, Buttigieg, Harris all on Day Two. Gillibrand as well.

[luvdahops]

But what if Trump played Alabama?

Or Mini Ditka?

[dudog84]

Jason, I'm really quite surprised you didn't catch my wink at Herman Cain's famous nine-nine-nine plan in saying no no no. I wasn't calling anyone a Nazi. Maybe I'm just too clever. Or need to use more emojis.

[dudog84]

Democratic Debate line-up is out:

https://www.cnn.com/2019/06/14/polit...020/index.html

Biden, Sanders, Buttigieg, Harris all on Day Two. Gillibrand as well.

Time for conspiracy theories. Resident statisticians, what are the chances that 4 of the top 5 candidates are on the same day? Especially since there's a pretty big drop-off after those 5.

Another question...is that good or bad for Warren (the 5th of those 5). Is she upset she's at the kiddie table, or does that give her a better chance to shine? However it plays out, I bet she feels right now like the teenager stuck with all the rambunctious toddler cousins at Thanksgiving.

[OldPhiKap]

Time for conspiracy theories. Resident statisticians, what are the chances that 4 of the top 5 candidates are on the same day? Especially since there's a pretty big drop-off after those 5.

Another question...is that good or bad for Warren (the 5th of those 5). Is she upset she's at the kiddie table, or does that give her a better chance to shine? However it plays out, I bet she feels right now like the teenager stuck with all the rambunctious toddler cousins at Thanksgiving.

Don't know the odds, but my understanding was that every candidate would have a rep present when the random drawing was done. After complaints of putting their fingers on the scale last time around, I am sure the DNC bent over backwards to at least have the appearance of fairness. After all, that's why we have this 20-car pile-up mess to start.

I think it helps Warren, she has a chance to shine in her debate while the rest slug it out in theirs. Put another way, if Warren can't stand out against the crowd she drew, that would not be a good sign.

Realistically, each candidate is going to get about 8-10 minutes if the time is equally shared. Some no doubt will gobble up more time than others, either due to the moderators or because they are able to grab it (that's a major skill it seems). A few may blip up, a major faux pas may ding someone, but this is not likely to change much it seems to me.

[Wander]

Time for conspiracy theories. Resident statisticians, what are the chances that 4 of the top 5 candidates are on the same day? Especially since there's a pretty big drop-off after those 5.

Well, it can only be split 5-0, 4-1, or 3-2. The chances are

5-0: 1/16, or 6%
4-1: 5/16, or 31%
3-2: 10/16 , or 63%

So, no need for conspiracy theories.

[sagegrouse]

Well, it can only be split 5-0, 4-1, or 3-2. The chances are

5-0: 1/16, or 6%
4-1: 5/16, or 31%
3-2: 10/16 , or 63%

So, no need for conspiracy theories.

Wander, you are brilliant. Here is the post I was about to make:

Not that improbable. I'll let someone else model it more precisely. But if you were tossing a coin five times, there are 32 possible outcomes. The probability of either zero or one heads is 2/32 (all candidates on one day or another). The probability of one or four heads is 10/32. The probability of either two or three heads is 20/32. So, 12 times out of 32 one would have a result with either four or five of the top five candidates on one day or the other -- 0.375.

(If you did the sampling "without replacement" -- there are only 20 candidates, not an infinite number -- the probability would be somewhat lower -- but someone else can figure that out.)

Cheers,
Sage Grouse

[pfrduke]

And I thought I had heard that they were guaranteeing that the top 8 would be divided 4 and 4, so the 5-0 option is out, making the odds even more likely.

[sagegrouse]

Wander, you are brilliant. Here is the post I was about to make:


Cheers,
Sage Grouse

Oops. Meant probability of zero or FIVE heads...

[RPS]

Well, it can only be split 5-0, 4-1, or 3-2. The chances are

5-0: 1/16, or 6%
4-1: 5/16, or 31%
3-2: 10/16 , or 63%

So, no need for conspiracy theories.

Agreed.

But if Mayor Pete ends up between Biden and Bernie, I might rethink things.

[budwom]

They should go shirts and skins, IMO.

[CrazyNotCrazie]

They should go shirts and skins, IMO.

Trump in public without a shirt on might be the only thing that could destroy his “shoot someone on 5th Avenue” theory...

[Bostondevil]

Time for conspiracy theories. Resident statisticians, what are the chances that 4 of the top 5 candidates are on the same day? Especially since there's a pretty big drop-off after those 5.

Another question...is that good or bad for Warren (the 5th of those 5). Is she upset she's at the kiddie table, or does that give her a better chance to shine? However it plays out, I bet she feels right now like the teenager stuck with all the rambunctious toddler cousins at Thanksgiving.

Warren is on the first night. I suspect she's happy about that. IMHO going first is an advantage.

[Bostondevil]

I notice the female candidates are equally balanced at 3 per night. The non-white candidates are also equally balanced, I think. Is Tulsi Gabbard a person of color? If she is, it's 3 and 2. If she isn't, it's 2 and 2.

[OldPhiKap]

They should go shirts and skins, IMO.

Not quite, but:

https://politics.theonion.com/jimmy-...rcu-1835383770