Recommend
 
 Thumb up
 Hide
7 Posts

BoardGameGeek» Forums » Everything Else » Religion, Sex, and Politics

Subject: The Five Types Of Swing States rss

Your Tags: Add tags
Popular Tags: [View All]
linoleum blownaparte
United States
Unspecified
Unspecified
flag msg tools
mbmb
Relative Democratic Lean is a number showing how a state voted relative to the nation overall.

For example, if the nation voted 55% Democratic, a state that voted 60% Democratic would have a 5% Relative Democratic Lean, while a state that only voted 45% Democratic would have a -10% lean.

Why is RDL important?

Even though the national popular vote goes back and forth, the relative margin between a state and the nation is what really matters. You can think of this phenomenon like boats rising and falling on the same tide.

For example, Obama nearly won Missouri (49.9%) in 2008 - but he didn't need to. When you rank the states by relative lean, Obama had already won the electoral college with Colorado (54.6%), a state that voted closer to the overall national margin (53.6%).

Looking at the four elections since 2000, we can start to see some trends in relative lean.

The following states colored blue or red had a RDL of above 3% or below -3% in all four elections. These are the true, solid red and blue states.



Notice Indiana is colored red - even though Obama barely won it in 2008. That's because it still had a RDL of -3.1% in that year.

The above map adds up to 176 Republican and 163 Democratic electoral votes.

That leaves 18 states which at one time were within 3% of the national margin.

Here's how they've voted in the four elections since 2000:



The four states Colorado, Iowa, Nevada and New Mexico used to be close to the national average margin or somewhat Republican-leaning. Over the past 12 years they have gradually shifted to being more blue than average. These states would be worth 26 Democratic EV if the Democratic nominee wins them all again.

Two states in the upper South, Missouri and West Virginia used to be close to the national average but they have practically zoomed over to the Republican side. Missouri used to be a national bellwether, but now it's increasingly a solid red state. These states would be worth 15 EV to the Republican nominee.

Five northern states Washington, Oregon, Michigan, Wisconsin and Maine used to be close to the national average; over the past 12 years they have pulled away to become lean-Dem states (Michigan and Wisconsin fell back a bit in 2012 because they were the home states of the GOP nominees). These states are worth 45 EV.

Just as dramatic has been the movement in two other states in the upper South, Virginia and North Carolina. Urbanization and immigration have transformed these once-solid red states into a true swing state in Virginia's case, and a soon-to-be-swing state in North Carolina's case. These two states are worth 28 EV.

Finally come five states which have stayed close to the national margin for the four past elections, with no significant momentum either way. These are Pennsylvania, Florida, Ohio, New Hampshire, and Minnesota. MN and PA may be surprising inclusions but they've consistently voted only 1-3% more Democratic than the national average; a concerted campaign might be enough to shift their position. All told, these five states are worth a whopping 81 EV.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
non sequitur
United States
Elk Point
South Dakota (SD)
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Linoleumblownaparte wrote:
Even though the national popular vote goes back and forth, the relative margin between a state and the nation is what really matters. You can think of this phenomenon like boats rising and falling on the same tide.

For example, Obama nearly won Missouri (49.9%) in 2008 - but he didn't need to. When you rank the states by relative lean, Obama had already won the electoral college with Colorado (54.6%), a state that voted closer to the overall national margin (53.6%).


I'm not following you -- the latter paragraph is simple proof that the former paragraph isn't what really matters. It's merely, are you going to win the state or not.

I'm not seeing the benefit vs. just looking at a table of percentage changes in voting behavior -- all this does is move the midpoint to the mean rather than looking at percentage change directly, which is everything you're doing later anyway.

All that happens is changing the intercept.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Hello Goodbye
msg tools
mbmbmbmbmb
Linoleumblownaparte wrote:
Relative Democratic Lean is a number showing how a state voted relative to the nation overall.

For example, if the nation voted 55% Democratic, a state that voted 60% Democratic would have a 5% Relative Democratic Lean, while a state that only voted 45% Democratic would have a -10% lean.

Why is RDL important?

Even though the national popular vote goes back and forth, the relative margin between a state and the nation is what really matters. You can think of this phenomenon like boats rising and falling on the same tide.

For example, Obama nearly won Missouri (49.9%) in 2008 - but he didn't need to. When you rank the states by relative lean, Obama had already won the electoral college with Colorado (54.6%), a state that voted closer to the overall national margin (53.6%).

Looking at the four elections since 2000, we can start to see some trends in relative lean.

The following states colored blue or red had a RDL of above 3% or below -3% in all four elections. These are the true, solid red and blue states.



Notice Indiana is colored red - even though Obama barely won it in 2008. That's because it still had a RDL of -3.1% in that year.

The above map adds up to 176 Republican and 163 Democratic electoral votes.

That leaves 18 states which at one time were within 3% of the national margin.

Here's how they've voted in the four elections since 2000:



The four states Colorado, Iowa, Nevada and New Mexico used to be close to the national average margin or somewhat Republican-leaning. Over the past 12 years they have gradually shifted to being more blue than average. These states would be worth 26 Democratic EV if the Democratic nominee wins them all again.

Two states in the upper South, Missouri and West Virginia used to be close to the national average but they have practically zoomed over to the Republican side. Missouri used to be a national bellwether, but now it's increasingly a solid red state. These states would be worth 15 EV to the Republican nominee.

Five northern states Washington, Oregon, Michigan, Wisconsin and Maine used to be close to the national average; over the past 12 years they have pulled away to become lean-Dem states (Michigan and Wisconsin fell back a bit in 2012 because they were the home states of the GOP nominees). These states are worth 45 EV.

Just as dramatic has been the movement in two other states in the upper South, Virginia and North Carolina. Urbanization and immigration have transformed these once-solid red states into a true swing state in Virginia's case, and a soon-to-be-swing state in North Carolina's case. These two states are worth 28 EV.

Finally come five states which have stayed close to the national margin for the four past elections, with no significant momentum either way. These are Pennsylvania, Florida, Ohio, New Hampshire, and Minnesota. MN and PA may be surprising inclusions but they've consistently voted only 1-3% more Democratic than the national average; a concerted campaign might be enough to shift their position. All told, these five states are worth a whopping 81 EV.


NC will not be a swing state, like Indiana, Obama won here in 2008 by the skin of his teeth, since then Republicans have been doing better than ever.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
linoleum blownaparte
United States
Unspecified
Unspecified
flag msg tools
mbmb
Terwox wrote:

I'm not seeing the benefit vs. just looking at a table of percentage changes in voting behavior -- all this does is move the midpoint to the mean rather than looking at percentage change directly, which is everything you're doing later anyway. All that happens is changing the intercept.


The answer to that is that changes in voting behavior don't add up to different electoral outcomes, UNLESS the state is ALREADY CLOSE to the national mean.

For example Mississippi could swing 15% from election to election but who cares. It's not going to affect the outcome. Whereas a 15% swing in Colorado would be earth shaking.

Karl Rove famously said all you need is "fifty percent plus one" - to win you need to invest your resources correctly for a close election.

As a candidate you don't really get to control the national mood (just ask 2008 McCain) all you CAN control is your relative emphasis on the states. So the rational strategy is to invest in states that you predict will be close to the national average, and then hope that the overall "national tide" is enough to make you win.

With such a strategy you can win even if the national mood is overall slightly against you.

DISinvesting in close states to invest in "reach" states is a bad strategy PRECISELY because any election where you win a "reach" state is one you've already won with closer states.

The Romney 2012 campaign didn't really understand this, that's why they misinvested millions in trying to get Wisconsin. They reasoned "Bush had 47% in 2000 and 49% in 2004 so it must be within reach." But they should have looked at relative lean rankings where Wisconsin was leapfrogging left over other states in every subsequent election. Any election where they won Wisconsin would have been an election where Wisconsin was just a cherry on top of 270 EV from FL/OH/VA and some fourth state like NH or CO. The correct investment should have been in THOSE states.

That's why lean relative to the nation matters strategically.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
linoleum blownaparte
United States
Unspecified
Unspecified
flag msg tools
mbmb
Hi_Regis wrote:


NC will not be a swing state, like Indiana, Obama won here in 2008 by the skin of his teeth, since then Republicans have been doing better than ever.


No, North Carolina keeps inching closer to the center

2008 Obama: 52.9% nationwide - 49.7% in NC = NC was 3.2% to the right of the nation
2012 Obama: 51.1% nationwide - 48.4% in NC = NC was 2.7% to the right of the nation

It could be 2.2% in 2016. That means a Democrat who wins the popular vote by 2.2%, all else being equal, would expect to win NC.

Of course there are a lot of other states that are even closer than NC. If you're saying NC will not be the tipping point state - the one that gets the winner over 270 EV - then I agree.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
non sequitur
United States
Elk Point
South Dakota (SD)
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Linoleumblownaparte wrote:
Terwox wrote:

I'm not seeing the benefit vs. just looking at a table of percentage changes in voting behavior -- all this does is move the midpoint to the mean rather than looking at percentage change directly, which is everything you're doing later anyway. All that happens is changing the intercept.


The answer to that is that changes in voting behavior don't add up to different electoral outcomes, UNLESS the state is ALREADY CLOSE to the national mean.

For example Mississippi could swing 15% from election to election but who cares. It's not going to affect the outcome. Whereas a 15% swing in Colorado would be earth shaking.

Karl Rove famously said all you need is "fifty percent plus one" - to win you need to invest your resources correctly for a close election.


Why is the national mean more important than the 50th percentile mark? Like Rove said, it's all the matters.

I'm not following how using the mean leads to better strategic decisions than the actual cutoff point of 50%.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.