Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

Ladies and Gentlemen, I give you the Electoral Vote Strength Factor

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Archives » General Discussion: Presidential (Through Nov 2009) Donate to DU
 
berni_mccoy Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 01:19 PM
Original message
Ladies and Gentlemen, I give you the Electoral Vote Strength Factor
Edited on Tue Jul-15-08 01:24 PM by berni_mccoy
I've been watching the EV data on both electoral-vote.com and fivethirtyeight.com. From time-to-time, I've published the data here in my Journal when something significant has happened, for example, a weakening in McCain's position or an increase in Obama's strength.

One thing I've tried to determine is how strong is each candidate's position. For example, Obama could have the number of electoral votes he currently has, but he could be in an extremely weak position, which would not bode well for the general election.

EV strength is a good measure of momentum and how solid a candidate's hold is on his current position. Of course, electoral-vote.com publishes this data in numerical format, but I thought it would be worth-while to put it in an image. To do so required calculating a new metric.

Presently, EV.com has three categories: Strong (> 10%), Weak (>5%, <10%) and Barely (<5%). If we weight each category of EV's a candidate, say 1.0 times for Strong, 0.5 times for Weak and 0.25 times for Barely and then add that up, we get an idea of how many EV's are solid. This would be a raw strength value. For example, today, Obama's raw strength would be

200 X 1.0 + 39 X 0.5 + 81 X 0.25 = 239.75.

But a raw number doesn't tell us much by itself. We'd like to gage that number against a target or a goal. If all of those numbers were in the Strong Category, the total would be his current EV total, which well exceeds the total number needed. But that is an unlikely event (until the General Election actually occurs). We could use 538 as a goal, but that too is likely not achievable. If we use 270 as a target, then we have a reasonable target such that if Obama had 270 Strong, we could say he's doing great. He could do better, but it's a good goal to start with and so we use it as a target for the EV Raw Strength.

Therefore, we could say Obama's EV Strength Factor today is 89%. He's not at 100%, but he's close.

On the contrary, looking at McCain's EV Raw Strength today, we see he's at

77 X 1.0 + 101 X 0.5 + 26 X 0.25 = 134

McCain's EV Strength Factor for today is therefore 50%.

Now we can compare how the candidates are doing with respect to each other's Electoral Vote Strength over time. Below is a graph I made from June 6 through Today using the data from Electoral-Vote.com:



As you can see, Obama's EV Strength is increasing while McCain's is steadily decreasing. In fact, if the trend continues, Obama will be very close to a strength of 100% and could possibly be higher. The polynomial trend line is a power 4 trend line for all series.

And we can see how Obama's strength is compared directly to McCain's by looking at the Ratio of the Strength Factors of Obama to McCain. The following graph shows this relationship:



As you can see, Obama's EV Strength is nearly 2:1 against McCain's. If it stabilizes where it is at, then things look very good for the Fall, given that the EV.com data is based on solid polling data and uses averages from multiple pollsters. Furthermore, there does seem to be a correlation between the EV Strength Ratio and the Win Percentage (computed at fivethirtyeight.com), even though the numbers are computed in a completely different way.

At any rate, I am finally happy I have a means to track candidate strength and I will monitor this value moving forward.



Printer Friendly | Permalink |  | Top
DangerousRhythm Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 01:26 PM
Response to Original message
1. Very cool!
Thanks! I love the calculations you guys do here and this is a great look at the data. :)
Printer Friendly | Permalink |  | Top
 
berni_mccoy Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 09:24 PM
Response to Reply #1
10. Oooo, I LIKE that Button! At first, I didn't realize it said McBush... VERY subliminal
Printer Friendly | Permalink |  | Top
 
DangerousRhythm Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Jul-16-08 05:42 AM
Response to Reply #10
15. Thanks!
I'd hoped to get very close to his logo, enough that one might do a doubletake. ;)
Printer Friendly | Permalink |  | Top
 
damntexdem Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 01:41 PM
Response to Original message
2. But there's a problem: on what basis did you decide on weights?
True, Obama will be ahead whatever the weights assigned; but it seems that the weighting scheme is still arbitrary.
Printer Friendly | Permalink |  | Top
 
berni_mccoy Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 01:53 PM
Response to Reply #2
4. The weights are based on strength of percentage win.
P-win > 10 = 1.0
5 < P-win < 10 = 0.5
0 < P-win < 5 = 0.25

The values are scaled to the ranges of win percentages that are tracked on EV.

If you consider the top bracket averaging 12.5 and use averages for the other two, you get 12.5, 7.5 and 2.5. Each is 5 points apart and normalizing the differences you get a linear scale. If you then scale by two to emphasize (or reward) larger spreads, you get 0.25, 0.5 and 1.0.

Regardless of the method chosen (they would all be arbitrary), as long as the same calculation is used on both candidates it should not matter.
Printer Friendly | Permalink |  | Top
 
grantcart Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 01:45 PM
Response to Original message
3. Berni can run a spread sheet and graph - who knew lol


useful metric my friend
Printer Friendly | Permalink |  | Top
 
berni_mccoy Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 01:58 PM
Response to Reply #3
5. Yeah, Somehow Excel was able to figure out automatically that Obama should be blue
and McCain red. I was pretty amazed by that.

:rofl:

Thanks Grant.
Printer Friendly | Permalink |  | Top
 
phrigndumass Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 08:01 PM
Response to Original message
6. Now THIS is helpful! :)
I'm very impressed with this. It really doesn't matter what value you weight each category, as long as strong is more than weak, and weak is more than barely, and you use the same for both candidates. It offers an excellent comparison, especially your ratio graph.

Using 270 as a common divider will work, although it may go over 100% for Obama. But that doesn't matter; the ratios will still work if one candidate has over 100%.

Good choice to use polynomial trend lines. Any fluctuation is easier to spot with that method.

Excellent, big K/R Berni! :thumbsup:
Printer Friendly | Permalink |  | Top
 
Political Heretic Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 08:10 PM
Response to Original message
7. AWESOME job!!
K&R
Printer Friendly | Permalink |  | Top
 
ChimpersMcSmirkers Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 08:39 PM
Response to Original message
8. Great work! K&R!
Printer Friendly | Permalink |  | Top
 
madrchsod Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 09:11 PM
Response to Original message
9. interesting....but i`m going with this
Edited on Tue Jul-15-08 09:12 PM by madrchsod
john has`t any more chance against obama than did alan keyes. in fact alan is more intelligent and more in tune with what is happening in the usa and the world than john is.

i`m going out on a limb but once those "debates" start between the two i`m going on record that obama will beat john by of 15% of the popular vote. barack will have enough electoral votes to make it a loss for the republicans that will take 8-12 years to recover.



oh by the way good post and i`m betting those numbers will increase in the next few months
Printer Friendly | Permalink |  | Top
 
crimsonblue Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 11:14 PM
Response to Original message
11. man, the vote riggers have a tough fall ahead of them
i'm sure they can pull through though and deliver and illegal victory for mccain.
Printer Friendly | Permalink |  | Top
 
madura Donating Member (239 posts) Send PM | Profile | Ignore Wed Jul-16-08 12:34 AM
Response to Reply #11
14. cnn is still bellowing dead heat, close race. fucking liars.
Printer Friendly | Permalink |  | Top
 
RNdaSilva Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 11:49 PM
Response to Original message
12. To add fuel to the fire...
Printer Friendly | Permalink |  | Top
 
AtomicKitten Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Jul-15-08 11:58 PM
Response to Original message
13. Now it counts. In the general election.
Funny how the MSM bought into Hillary's BS electoral vote argument that wasn't applicable in the primary, yet nary a word about Obama kicking the crap out of McBush electorally now that we are in the general election.

It seems the MSM traditionally gives the Democratic candidate a hard time.
Printer Friendly | Permalink |  | Top
 
DavidDvorkin Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Jul-16-08 09:01 AM
Response to Original message
16. Nice!
Nifty idea.
Printer Friendly | Permalink |  | Top
 
berni_mccoy Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Jul-16-08 11:38 AM
Response to Reply #16
17. Thanks.
Printer Friendly | Permalink |  | Top
 
DU AdBot (1000+ posts) Click to send private message to this author Click to view 
this author's profile Click to add 
this author to your buddy list Click to add 
this author to your Ignore list Thu Dec 26th 2024, 08:30 AM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Archives » General Discussion: Presidential (Through Nov 2009) Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators


Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC