**** Elections thread *****

[Poopadoop]

It would be one thing if you were just ignorant. But I know you know a thing or two about stats, so the only explanation for your conclusion is a deeply rooted partisan bias and an outright refusal to entertain a thought that challenges your chosen ideology.

But you've repeatedly proven that you don't know a thing, never mind a thing or two, about stats. Which makes it hard to understand why you think you can criticize the facts. I mean I know you have no qualms with doing it, you do it all the time. But it's just amusing you keep trying to do it when you're arguing with someone who actually has a clue about these things and can step on you like an ant.

First of all, a 1% swing in Montana isn't nearly the same thing as a 1% swing in New York.

1% is still 1%. It doesn't matter if it's in MT or NY or on Jupiter. wtf are you even talking about here?

Secondly, a 17.7% average swing (as if that number means anything) in a handful of purple states is simply not comparable to a 5.5% swing in a handful of DEEPLY BLUE states.

You're finally right about something. 17.7% is roughly 3x as much as 5.5%

If I climbed Everest in a month, and you hiked Mt WhoGivesAFuck in a day, would you say you're a better climber than I am?

What has this to do with anything? Please explain your argument by reference to facts, not imaginary reductio ad bananums.

That's kinda the crux of your argument here. You're saying that Democrats enjoyed more success, but you fail to acknowledge the decidedly lesser amount of challenge involved in their endeavor compared to Republicans in 2009.

Oh is that it? That when you win a congressional race by 45% it's a lot more meaningful to lose 5% of your support than when you win a race by 20%, and lose 15% (or some such shit)?

Tell me, which of those differences is more likely to end up in a change ownership of a seat from one party to the other? The one that changes a gigantic lead into a slightly less gigantic lead, or the one that changes a moderate lead into a toss-up?

[BananaStand]

Which makes it hard to understand why you think you can criticize the facts.

Cherry picked results used to form a biased presentation does not constitute a fact

But it's just amusing you keep trying to do it when you're arguing with someone who actually has a clue about these things and can step on you like an ant.

This ant is still alive and well. I'm coming for your picnic!!

1% is still 1%. It doesn't matter if it's in MT or NY or on Jupiter. wtf are you even talking about here?

Do you really think anyone is fooled by this? You know exactly what I'm talking about here. Would you rather have 1% of a dollar, or 1% of a truckload of dollars?

You're finally right about something. 17.7% is roughly 3x as much as 5.5%

And one apple has 100x the mass of one blueberry. How does that affect the apple's ability to function as an apple, or the blueberry's ability to function as a blueberry?

What has this to do with anything? Please explain your argument by reference to facts, not imaginary reductio ad bananums.

FACT: It's harder for a republican to win in California and NY, than it is for a Democrat to win in South Carolina and Kansas.

Oh is that it? That when you win a congressional race by 45% it's a lot more meaningful to lose 5% of your support than when you win a race by 20%, and lose 15% (or some such shit)?

It depends on what state/district we're talking about.

Tell me, which of those differences is more likely to end up in a change ownership of a seat from one party to the other? The one that changes a gigantic lead into a slightly less gigantic lead, or the one that changes a moderate lead into a toss-up?

It's kind of a moot question since you're comparing congressional elections with presidential elections.

Scott Brown pulled out a stunning senate victory in MA in 2010. Does that mean Obama should have been worried about losing the state in the 2012 presidential election? Not even a little bit. Fuck, the guy ran against MA's former governor and still won the state easily in 2012

[Poopadoop]

Cherry picked results used to form a biased presentation does not constitute a fact

I reported all the congressional special elections in the one year following each election you mentioned. You're the one who set the parameters here.

This ant is still alive and well. I'm coming for your picnic!!

You may soon find the picnic has moved on without you.

Do you really think anyone is fooled by this? You know exactly what I'm talking about here. Would you rather have 1% of a dollar, or 1% of a truckload of dollars?

Not a valid comparison by any means.

And one apple has 100x the mass of one blueberry. How does that affect the apple's ability to function as an apple, or the blueberry's ability to function as a blueberry?

It doesn't.

FACT: It's harder for a republican to win in California and NY, than it is for a Democrat to win in South Carolina and Kansas.

That's generally correct, yes.

It depends on what state/district we're talking about.

Only inasmuch as the swing makes a meaningful difference. An election is winner-takes-all. The impact of a swing that changes a safe district into a toss up is meaningful. The impact of a swing that has no potential outcome on the election is not.

It's kind of a moot question since you're comparing congressional elections with presidential elections.

Nope, I'm comparing congressional elections with congressional special elections that took place a year after a general election.

Scott Brown pulled out a stunning senate victory in MA in 2010. Does that mean Obama should have been worried about losing the state in the 2012 presidential election? Not even a little bit. Fuck, the guy ran against MA's former governor and still won the state easily in 2012

That argument would be a lot more convincing if Trump had smashed PA in 2016. He didn't.

[BananaStand]

Not a valid comparison by any means.

Exactly

It doesn't.

Right

That's generally correct, yes.

Good. Now try considering that in your analysis

Only inasmuch as the swing makes a meaningful difference. An election is winner-takes-all. The impact of a swing that changes a safe district into a toss up is meaningful. The impact of a swing that has no potential outcome on the election is not.

So....why is the Lamb victory meaningful? There's going to be a whole new election in November for ALL of PA's congressional districts in November. The district that Lamb just won won't even exist by then. This special election doesn't have any effect on the outcome of anything. So why are you so emboldened by this 26% swing?

Nope, I'm comparing congressional elections with congressional special elections that took place a year after a general election.

Maybe that's what you're doing now. But that was definitely NOT the aim of the Politico article you linked originally. Maybe I should coin a phrase "reductivio al poopicus". I doubt it will catch on because this already has a name. It's called "moving the goalpost".

That argument would be a lot more convincing if Trump had smashed PA in 2016. He didn't.

You don't think he did? What do you consider "smashing"??

Didn't you just get through saying how impressive it is when a candidate makes gains in an opposing party's stronghold? How would you describe Trump's victory in PA then?

[Poopadoop]

Exactly


Right


Good. Now try considering that in your analysis

You said nothing of note nor provided any argument about anything related to my reporting of the results of congressional special elections. And now you advise me to apply these non-sequiturs to my analysis. Tell you what - they're your statements, why don't you apply them and make a logical case for how they alter the conclusions?

So....why is the Lamb victory meaningful? There's going to be a whole new election in November for ALL of PA's congressional districts in November. The district that Lamb just won won't even exist by then. This special election doesn't have any effect on the outcome of anything. So why are you so emboldened by this 26% swing?

I never said I was emboldened by it. Your response to me posting the 26% figure for PA was:

Jesus tapdancing Christ!!

It's this kind of group identity mentality that is moving the democratic party further towards the radical left.

Do you not see how fucking dumb this game you're playing is??

which I took to imply that you thought such a figure had little significance. I then went on to argue that it represents a general trend against Trump, for which I provided evidence from all the special congressional elections that have taken place since he became president. Your response to that was:

Bad science. Go do the same analysis of special elections during the first years of the Reagan, Clinton, Bush 2, and Obama administrations then tell me if anything in your previous two posts amounts to a hill of dogshit

Which I did for the previous two presidents, showing again it was something other than a hill of dogshit.

I won the argument through the numbers. You lost.

Maybe that's what you're doing now. But that was definitely NOT the aim of the Politico article you linked originally. Maybe I should coin a phrase "reductivio al poopicus". I doubt it will catch on because this already has a name. It's called "moving the goalpost".

Actually this is another reductio ad bananum. I never said I agreed with everything written in the Politico article, those are your assumptions. My argument has been and continues to be that Trump is facing a severe drop in popularity since he became president, and this is evident in the results of these special elections.

If you want to argue with what's in the Politico article, go write a letter to Politico. I was only quoting it as a source for the numbers I was using.

You don't think he did?

He won by .7% in 2016. No I don't consider that 'smashing'.

What do you consider "smashing"??

Winning by a comfortable margin at the very least. Barely scraping a victory by < 1 % doesn't qualify. You see, < 1% is a very small margin of victory. 10% is larger and 20% is very large. See how easy it is to think with numbers?

Didn't you just get through saying how impressive it is when a candidate makes gains in an opposing party's stronghold? How would you describe Trump's victory in PA then?

Romney lost in PA in 2012 by ~6%. So, Trump's gain in PA relative to Romney in 2012 was ~6%. Now his nominal popularity has gone down in one PA county by 20%. Draw whatever conclusions you like.

You see, if Trump had won PA by 40% then a swing of 20% would be a lot less threatening to him, because there would still be a cushion. If you win by a small amount like .7% then even a tiny swing can change ownership of the state.

[BananaStand]

You said nothing of note nor provided any argument about anything related to my reporting of the results of congressional special elections. And now you advise me to apply these non-sequiturs to my analysis. Tell you what - they're your statements, why don't you apply them and make a logical case for how they alter the conclusions?

I'm not sure what you're missing. Moving election results in states with larger populations, or where opposing political views are more entrenched is HARDER. Your analysis doesn't account for that at all. You just say 17 >5....as if that's indicative of anything relevant.

I never said I was emboldened by it. Your response to me posting the 26% figure for PA was:

Wrong again. I posted that response in response to your implication that all R voters are homogeneous

which I took to imply ....

Doesn't matter. You don't even remember what you said five posts ago

I won the argument through the numbers. You lost.

By cherry picking data and moving the goalpost. It started with comparing Trump's results to Saccone's. Then you changed your mind and decided you were actually comparing Saccone to his predecessor. Then you decided that Scott Brown didn't count because it wasn't within a year from the election. I don't know where you got that idea. I assume shit like that just pops in your head when you're making up the rules as you go along. The passage you quoted challenged you to review elections within the first year of an adminstration. I never declared that the election was when the stopwatch starts. Scott Brown won Kennedy's senate seat 364 days after Obama's inauguration. So go back and put that little nugget into your analysis. But you won't, because for some insane reason you've also decided that Senate races don't count. WTF??

My argument has been and continues to be that Trump is facing a severe drop in popularity since he became president, and this is evident in the results of these special elections.

Except the guy that won has policies that most align with Trump. Again, this is part of the dangerous group-identity mentality that you have on the left. You think just cause a guy has a "D" next to his name, that means he's categorically opposed to everything Trump.

If you want to argue with what's in the Politico article, go write a letter to Politico. I was only quoting it as a source for the numbers I was using.

You can't even decide which numbers you're using.

Winning by a comfortable margin at the very least. Barely scraping a victory by < 1 % doesn't qualify. You see, < 1% is a very small margin of victory. 10% is larger and 20% is very large. See how easy it is to think with numbers?

See how idiotic it is when you use objective numbers to represent subjective ideas. I'd say that Trump winning a democratic stronghold is a "smashing" victory. If Trump wins California, even by one vote, I'd call that "smashing" as well.

You see, if Trump had won PA by 40% then a swing of 20% would be a lot less threatening to him, because there would still be a cushion. If you win by a small amount like .7% then even a tiny swing can change ownership of the state.

Well this particular congressional district represents 1 electoral vote. Trump won by 77. Draw whatever conclusions you like.

[Poopadoop]

...

Wuf is the only one of you two trying to have a serious discussion about this. You're just screaming and throwing your toys out of the pram, while pulling any old excuse out of your ass for why the numbers don't mean what you don't want them to mean, and claiming that I'm trying to trick you with simple math.

[Poopadoop]

Moving election results in states with larger populations, is HARDER

This is the wrong way to think about % data. You use % data for the very reason that different places have different populations and the only value that matters in winning is who gets a higher % of votes. It doesn't matter if there are 50k voters or 500k voters, it's the % value that matters. So, when comparing elections across different years it makes sense to look at % data.

Thought exercise for you: the next time you look at a poll, try to figure out why they report the values in terms of % favouring X or Y rather than raw number of respondents favouring X or Y. If you can come up with a cogent answer, you're making progress.

or where opposing political views are more entrenched is HARDER.

To analyze this, you need to provide evidence of stability ("entrenchment") in long-term election patterns (in % of victory by one party, not just what party tends to win), and be able to correlate it with the outcomes of the 2017 S.E. So, provide the data and let's see if it works.

My intuition is the idea of entrenchment is confounded with base support. That is, D are more likely to win CA in any given election because D popularity is on average, very high in CA, and it is thus less affected by swings than other states where base D and R popularity are more comparable. It doesn't mean significant swings in opinion don't occur in CA, just that they are much less likely to affect the outcome in CA given what the base %s of R and D voters are.

Your analysis doesn't account for that at all. You just say 17 >5....as if that's indicative of anything relevant.

It also doesn't account for a lot of things I don't find plausible, like whether there was a full moon on the night of the election or not. If you DO find these alternate explanations plausible, then provide the data you think will support your position and I'll be happy to analyse it for you.

By cherry picking data and moving the goalpost. It started with comparing Trump's results to Saccone's. Then you changed your mind and decided you were actually comparing Saccone to his predecessor.

Nope. You're the one who compared Trump to Saccone. I was comparing 2016 congressional candidate R v D % to 2017 figures in the special election. It just so happened 20% was the same margin of victory for both Trump and whoever ran for R in 2016 congressional election in that district. I tried to explain that to you but you seem to prefer to ignore the idea that you were confused and try to blame it on me doing something nefarious instead.

Then you decided that Scott Brown didn't count because it wasn't within a year from the election.

I didn't include any senate S.E. in my analysis because I didn't think of it at the time, also it's a lot of work to go and find all these numbers. But you keep harping on about this ONE case as if it makes all the difference in the world that it's not in the set of data I analysed. So, if you're so convinced that including the senate S.E.s will change the overall conclusion, by all means provide me with the data.

for some insane reason you've also decided that Senate races don't count. WTF??

I'm tired of explaining myself to you over and over and over. So either collect and post all the senate S.E. data or just keep railing about my unfair analysis that doesn't include the one S.E. you cherry-picked out of all the senate S.E.'s that it also doesn't include. Either way is fine with me, but I'm done arguing about it. K, thx.

I'd say that Trump winning a democratic stronghold is a "smashing" victory. If Trump wins California, even by one vote, I'd call that "smashing" as well.

You consider a 6% swing from Obama to Trump a 'smashing victory' in a D stronghold, but you're not impressed by an average 17.7% swing in the other direction a year later. Hard not to see that as biased.

Well this particular congressional district represents 1 electoral vote. Trump won by 77. Draw whatever conclusions you like.

Lol, well if I were you I might conclude that 1 electoral vote is the same as 77 because the views were so entrenched there or some other garbage explanation I have no evidence for.