Interesting analysis shows that even if Hillary wins in TX; Obama might end up with more delegates

[TeamJordan23]

Clinton Up 49-41 in Texas Poll; Obama May Win More Delegates
by: Karl-Thomas Musselman
Fri Feb 15, 2008

First, a hat-tip to R.G. Ratcliffe at the Houston Chronicle who has the entire data set posted from which this post draws. Read the original post here. The poll was commissioned by the Texas Credit Union League, conducted Feb 11-13, with a MOE of +/- 4.9%.

DEMOCRATIC PRIMARY (Latest IVR Poll 1/31)

Hillary Clinton: 49 (48)
Barack Obama: 41 (38)
Undecided: 8 (10)

And now for the interesting sub-groups and my analysis. It's here that we find something very surprising!

Even though Clinton leads by 8 points in polling statewide, based upon the following sub-samples, Obama would still come out with a delegate lead.

And that's just among delegates allocated by the primary, not our additional caucus process which Obama has proven deft at winning delegates through.

I'm not kidding, follow me below...

Clinton also enjoys majority support in the South (57%) and Western (61%) regions of the state, and edges ahead in the Eastern part of the state 46% to 40%. Obama is beating Clinton 53% to 32% in the Central region and leads 49% to 44% in the Houston area. The Dallas Fort-Worth region is tied within margin of error (Clinton 42%, Obama 41%).

This section is the most critical when talking about the allocation of national delegates. (For more background as to why, read our two part guide.) While the regions are not defined by Senate District, I'm going to do my best to match them and show what these margins might result in delegate allocation wise. Please refer to this district map and this delegate spreadsheet if you want to follow along.

Region (net delegate gain)

South (57% Clinton, +1 net Clinton)

In order for Clinton to break any 4 delegate border districts, she needs over 62.5% of the vote. Absent that, all three (or four) of these districts are a wash. Right now, all she'd pick up is the odd 3rd delegate in SD-27 (Lucio).

West (61% Clinton, +3 net Clinton)

Again, Clinton faces the same 62.5% issue but the districts are sparse out here. SD-31 only has 2 delegates which will split no matter what. SD-19 has 4 and won't break unless she gets higher and might have even been included in the 'southern' sample because of its anchor San Antonio which was worse for Clinton. That leaves El Paso and the Panhandle with the three 3 delegate districts which only require 51% to get the odd delegate. Racking up the vote in this region doesn't benefit Clinton much once she passes a simple majority, and in this poll, it's her best region.

Eastern (46% Clinton-40% Obama, +/- 0 net)

Now, while this isn't showing anyone over 50%, I believe the delegate match works based upon the split of the viable vote, I'll have to check. Regardless, it doesn't make a lick of difference in this case for East Texas. SD's 1, 2, 3, & 4 are ALL 4 delegate districts that require the 62.5% supermajority to break the tied allocation. Parts of SD-5 could be in this pool, but guess what- it's an even 4 delegate district, too. Hillary sending Bill Clinton through this region needs to ramp up her vote totals by about 15 points before she squeezes any juice out of east Texas. If not, all for nothing.

Central (53% Obama - 32% Clinton, +6 Obama)

The margins here are critical. Looking at these numbers, there is about 15% undecided floating around. If they split evenly along existing proportions, that would put Obama right on the 62.5% line to break 4 delegate districts into 3-1 advantages. Anchored by SD-14 in Austin, Obama can conservatively expect a 5-3 split if he gets over 56.25% and I have no doubt this poll undersamples Obama's college turnout which is highly concentrated here. He'll be able to pick up the odd 3rd delegates in SD-22 & SD-24 with even a simple majority lead. Williamson County north of Austin anchors SD-5 so it's possible Obama could force that 3-1, but I'll leave it tied for now. SD-18 goes east and south so I'm going to leave that 4 delegate district in the split category. SD-25 runs down to San Antonio and has 6 delegates, and the threshold is only 58.3% to break it to 4-2. Keep in mind that an effective college operation at Texas State could help ensure this breaks 4-2 for Obama.

Houston (49% Obama - 44% Clinton, +2 Obama)

Again, if the undecided vote is in the same proportion, Obama would take a 52% simply majority. This is hard to allocate simply because the Houston area is so diverse and the general number is certainly highly weighted by local variances. Four of the six districts here have odd amounts. Based on this, Obama would easily win 4-3 in SD-13 (Ellis who supports Obama) which is African American. He'd need 64% to make this 5-2 and with the support of the district's Senator actively working for him, likely will get there by election day. SD 17 has 5 delegates so as long as Obama leads, he wins the odd delegate. SD's 11 & 15 are both even 4 delegate districts so they are a wash either way if it's close. SD-6 (Gallegos) is a 3 delegate district but Hispanic, so we'll assume this is Clinton's support in the Houston sample and break it 2-1 for her. SD-7 (Patrick) is Anglo so it's 2-1 Obama if he has 1 more vote than Clinton.

Dallas/Ft. Worth (42% Clinton - 41% Obama, +2 Obama )

There are 26 delegates at stake here. Problem is, 18 of them are in even numbered districts, three of which are going to split 2-2. The one that is 6 delegates is SD-23 (West who supports Obama) which is the African American district. Obama needs only 58% to make that a 4-2 split and given the size of the metro area, I'm reasonably confident in asserting that portion of this area's sample includes enough support from SD-23 to do that. That leaves SD-10 (Brimer) and SD-9 (Harris) to whomever wins the simple majority. Obama could do well in SD-9 which include the mid-cities and home to lots of independents but I don't feel comfortable assigning either of these. So lets just assume they break 2-1 for either candidate an cancel each other out for now.

Total (49% Clinton - 41% Obama, +6 net Obama delegates!)

http://www.burntorangereport.com/showDiary.do?diaryId=4972

[anamandujano]

1. Mind numbing, but thank you.

I tend to doubt any of your fellow Obama supporters will be able to wade through the information but you'll get a few pats on the back for it anyway. :)

[Spider Jerusalem]

3. Arrogant and condescending, much?

If you can say something about the figures, then please. Do. Snide commentary about a candidate's supporters, rather than actually discussing the topic at hand, doesn't help anyone.

[ErnestoG]

5. As mind numbing as your lack of commentary on the facts?

[NMMatt]

2. The prroblem with this kind of win

I really think we need to win the popular vote as well as the delegates in the end, otherwise camp Hillary can spin this into a "tie" and take it to the convention.

It will help Hillary make the case that delegates don't represent the will of the people, which she will try to do (regardless of the popular vote, which is estimated anyway). What better for her spin than a large state in which she wins the most votes, but does not get the most delegates.

So even though I feel we need to win outright in Texas, it is a good thing that the delegate breakdowns look favorable to Obama.

[ShadowLiberal]

4. Yep, that's the cold hard reality, it's very hard to catch up when your down by 100 delegates

That's the cold hard reality about proportionally awarding the delegates. It's very hard to build up a pledged delegate lead over your opponents, and when you're behind by like 100 delegates and over half the states have voted already it becomes extremely difficult to take back the pledged delegate lead.

In order to catch up to Obama's pledged delegate lead and pass it Hillary needs to start blowing Obama out of the water by like 20% or more in some states, that's how Obama got such a large pledged delegate lead, and it's the only way for Hillary to catch up to it.

[Barack_America]

6. Delegate math hurts my brain, but...I'll take it!

:D

[TexasObserver]

7. The analysis is correct as the numbers are presented. But Obama will win the popular vote, too.

These polls showing Hillary UP in Texas are deficient. They chronically underpoll key elements of Obama's support, underestimating the turnout of under 30s, blacks, indies, and crossovers.

Obama will win the popular and delegate vote.

[Spider Jerusalem]

9. If that happens in TX and OH...

then Hillary needs to wake up to the fact that her candidacy is finished. Although I don't think she will.

[ButterflyBlood]

10. True. If she loses either Texas or Ohio, it's over

The end.

[TexasObserver]

11. The fact that no one told her how bad the $ situation was means she can't hear bad news.

This means no one is going to say "Hillary, it's over, time to do the right thing."

She won't quit until they drag her out, kicking and screaming "BUT I'M HILLARY! HILLARY!!"

[ButterflyBlood]

8. Let's not forget the caucus too

Since Obama seems to overperform by about 10-20 points in caucuses, that should equal a victory for him with Hillary's current lead.