Download: PresidentialPrimaries.xls

My predictions for the North Carolina and Indiana primaries on Tuesday proved to be pretty accurate. While Barack Obama did about 2% better in each state than predicted, the delegates broke slightly in Hillary Clinton’s favor. Obama’s win in North Carolina translated into a 65-50 delegate victory, a net gain of four over my prediction, and Clinton’s win in Indiana gave her a delegate win of 38-34, exactly what the model forecast.

So, at the end of the night, Obama picked up a net gain of 11 delegates over Clinton, 4 more than predicted. Funnily enough, today’s updates to the opinion polls for the upcoming races (courtesy of Pollster.com) give Clinton back those 4 net delegates. Which means that our prediction of the total number of pledged delegates each candidate will end up with hasn’t changed since the beginning of the week.

What has changed is the share of the delegates each candidate needs to reach their “magic number” (which I’ve defined as a lead of 1 delegate for Clinton and a lead of 150 delegates for Obama). Clinton now needs 90% of the remaining delegates to win the pledged delegates race while Obama needs 46% of the delegates to keep his lead above 150.

And what that means is that, while the situation in the pledged delegate race is essentially exactly where it was before Indiana and North Carolina, the perception of the race has changed. Clinton is now seen as defeated, and Obama is now picking up unpledged delegates at an increased pace — he’s up 16 since the primaries on Tuesday, while Clinton is up a net 6, according to CNN’s tally. (I’ve seen other lists that say that Obama has now taken the lead in unpledged delegates.)

So our trend graph shows Obama accelerating towards the nomination again.

If this trend holds, then Obama will effectively wrap up the nomination on May 21, the day after the Kentucky and Oregon primaries. If I was a gambling man, I would bet that’s the day Clinton will concede defeat, although there’s a chance she’ll wait until June 4, once the final primaries in Montana and South Dakota are over.

Download: PresidentialPrimaries.xls (Excel spreadsheet)

Today sees the last of the big primaries in the Democractic presidential nomination race: Indiana and North Carolina. Right now, the momentum in the race is with Senator Clinton, who has gained 9 extra pledged delegates in our prediction since the Pennsylvania primary thanks to better scores in the opinion polls for upcoming states. However, Senator Obama’s continued courting of unpledged delegates has kept him inching closer to the nomination. More on that in a second. First, my predictions for today’s primary states.

Indiana
Clinton: 38 delegates (52% of vote)
Obama: 34 delegates (48% of vote)

North Carolina
Clinton: 52 delegates (46% of vote)
Obama: 63 delegates (54% of vote)

Overall, Obama should gain 7 net delegates today. That’s what is factored in to the projection above. As usual, in order for the dynamics of the race to change, one candidate will need to do considerably better or worse than these projections.

Speaking of the overall dynamics of the race, I though it would be interesting to look at how each candidate’s “Unpledged Delegates Needed” total in my chart has changed since I started projecting the final results (click on the image for the full-sized version).

As the chart shows, Obama had a lot of momentum coming out of the Super Tuesday primaries, and pretty much every result and new opinion poll showed him closing in on the nomination. The big pause before Pennsylvania gave Clinton a chance to regroup, and since then she’s started to move into a more competitive situation.

However, time is running out fast for Clinton. She’s still 200 delegates away from winning, according to our projection, while Obama’s goal is down to 79. So he just needs to convince one unpledged delegate to support him for every two that Clinton persuades to support her. If Obama does better than expected tonight, he could pull his target down closer to 70, and the ratio would be practically 3 to 1… surely too much for Clinton to overcome.

Download: PresidentialPrimaries.xls (Excel spreadsheet)

Opinion poll averages for Pennsylvania:

Clinton: 49%
Obama: 43%
Undecided: 8%

Predicted delegates:

Clinton: 84
Obama: 74

My last couple of updates on the Democratic primary race have shown a slight shift towards Clinton, although nothing like the kind of momentum that she needs to win the eventual nomination.  Right now, it looks as though Obama will finish the primaries with a delegate lead of around 153 — above the “magic number” lead of 150 delegates.

That prediction assumes Clinton wins by 6 or 7 points today, and gains a net 10 delegate advantage.  So, that’s her baseline…  I think she needs to win by at least 10 points to retain any sort of a hope in the overall race — and even then, she will still need a complete collapse by Obama in the remaining states.

Download: PresidentialPrimaries.xls (spreadsheet)

It’s been a little while since I last updated the delegate counts for the Democratic Presidential Primaries, but, even though Mississippi is the only state to have voted since my last article, there’s actually been some movement in my predictions.

I was just a single delegate out on my prediction for Mississippi, with Hillary Clinton doing slightly worse in the overall vote, but slightly better in the delegate battle than predicted. That obviously didn’t change the shape of the race much.

What did have an impact was the Iowa County Conventions, which determined the final delegates for the state. This proved to be a major boost for Barack Obama, who gained many of the delegates formerly pledged to John Edwards, and thereby had a net gain of 10 delegates. Finally, I got some updated delegate counts from the Democrats Abroad.

Looking forward to future races, Hillary Clinton has gained somewhat from my last survey, thanks in part to improved polling in Pennsylvania, and also because Pollster.com dug up a poll from West Virginia that heavily favors Clinton. She also gained 5 superdelegates to Obama’s 3, according to CNN’s count.

All of which adds up to Hillary Clinton getting 24 delegates closer to the nomination while Barack Obama picked up 13 delegates.

That all sounds like pretty good news for Clinton, but what’s interesting is that even with these good numbers, she has actually fallen further behind on my “magic number” analysis. If you’ve been reading the blog for a while, you’ll know that I believe that Clinton will win the nomination if she reaches the convention with the majority of the pledged delegates, while Obama will win the nomination if he gets to the convention with a lead of 150 delegates. If the margin is less than 150 delegates, Clinton could still win by virtue of picking up enough superdelegates (although her chances get increasingly slim as Obama’s lead gets closer to 150).

Because we’re now coming down to the last few states, and Obama’s current lead in the pledged delegate race is 168, Clinton needs to pull off some really big victories in order to make the situation competitive. In fact, Obama now just needs 49% of the delegates across all the remaining states to secure his 150 delegate advantage — the first time his goal has fallen to less than 50%. Clinton, meanwhile would need to secure 66% of the delegates to go to the convention in the lead. That’s pretty much impossible, unless Obama suffers a catastrophic collapse.

In short, Clinton needs to win Pennsylvania by a big margin to keep the race alive. If she doesn’t, I think the remaining superdelegates will start breaking decisively for Obama, as it becomes apparent that he can’t be beaten.

A follow up to my earlier post about the state of the Democratic race…

My prediction for the Mississippi Primary today, based on polls from Pollster.com, is that Barack Obama will gain 20 extra delegates to Hillary Clinton’s 13.  That’s factored in to the overall prediction below.  Based on current polling, Obama should score something like 59% of the vote in Mississippi, which would be another comfortable win, but not enough to get a really big bump in his delegate lead.

Download: PresidentialPrimaries.xls

The past week has been an interesting one for the Democratic primaries. The conventional wisdom is that Hillary Clinton kept the race alive with her performance in last Tuesday’s elections, and that now the race will head to the convention without a clear winner. In reality, everything turned out pretty much as predicted. Based on the best information I can find, it looks as though Clinton won last Tuesday 188-182, which compares to my prediction of 187-183… a net difference of 2 delegates. Clinton also picked up the last 2 delegates from Maryland, according to CNN. Meanwhile, based on this careful research, Obama did a little better in California than previously estimated (although my count was already pretty close to the numbers calculated by David Deyan).

The overall result is that Clinton gained a couple of pledged delegates and Obama lost three over the last week, based on my model. That leaves Obama leading by 182 delegates going in to the convention.

Meanwhile, on the unpledged delegate side, CNN reports Obama’s total as having increased by 21 since my last review, and Clinton has gained two. So, overall, Obama is 18 delegates closer to the nomination, Clinton is 4 delegates closer, and Obama looks a more clear favorite than he did a week ago, if one looks solely at likely delegate scores.

Of course, perception counts for a lot in politics, and Obama could yet lose ground if Clinton does well in the remaining states and picks up some more endorsements. But, right now, he still looks to have a healthy lead.

Sources: CNN, The Green Papers, Calitics

Download: PresidentialPrimaries.xls (Excel spreadsheet)

I’ve updated my delegate prediction model with the latest poll results.  (Download the spreadsheet for more details.)  I’ve also made another slight change to the model to, hopefully, give a more accurate projection of how many delegates each candidate will have going in to the convention.  I’m using each candidate’s average vote share from previous states as their expected vote share in future states for which there is no polling data.  I’ll be the first to admit that this is a little rough and ready, but it’s probably more accurate than predicting that the delegates will be split down the middle in those states.

As has been the case in each of my last few updates, the momentum is clearly with Senator Obama.  His performance in tomorrow’s primaries has improved across the board, and he’s now predicted to win Texas.  The model still has Senator Clinton winning more delegates (187-183) thanks to her lead in Ohio, but it would be a surprise if she won that state by 7%, as my “poll of polls” method would suggest.

Even if Clinton does win handily in Ohio and Rhode Island tomorrow, the model says Obama will reach the convention with 187 more pledged delegates, which will almost certainly be enough to clinch the nomination.

I just published our final predictions for tonight’s Oscars.  As I mentioned in a previous post, these are based on votes in our annual Predict the Oscars competition.  It’s the eleventh time we’ve run the competition, and our readers generally score about 70%.

So, if No Country for Old Men wins Best Picture tonight, and the Coen brothers win Best Director, don’t say I didn’t warn you…

Bruce

Link: The Numbers’ Oscar Predictions

Download: PresidentialPrimaries.xls (Excel spreadsheet)

I’ve updated current delegate totals based on CNN reports and the results of the Democrats Abroad vote, which gave Obama 2.5 delegates to Clinton’s 2. (And, no, I’ve no idea how someone who’s “half a delegate” votes at the convention.) There’s also new polling data for OH, TX and PA, all of which seem to be moving in Obama’s favor. Pollster.com also finally published some polls for North Carolina, which is looking good for Obama too. Hopefully we’ll get polls for a few more states shortly.

The main reason that the numbers have jumped in my predictions since Wednesday is that I changed the model slightly. Where previously I was ignoring the “undecided” figures in the polls, I am now assuming that the undecided votes will split between the candidates in proportion to their current share in the state. So, for example, if a poll says Candidate A 50%, Candidate B 30%, Undecided 20%, the model assumes that the remaining voters will break 5-3 in favor of Candidate A, producing a final delegate split of 62.5% for Candidate A and 37.5% for Candidate B. This change marginally improves the delegate balance in favor of Senator Clinton, but only by a handful of delegates.

The inclusion of the North Carolina polls makes the task for Clinton look harder than it did already. If she manages to stop the slide in OH, TX and PA, she still needs to score 66% of the vote in the remaining states to come out ahead in delegate count coming in to the convention. Given that she’s only broken that number in Arkansas so far, the question is clearly not whether Obama will lead in delegates going in to the convention, but how big his lead will be.

One other number to watch: Obama needs 68% of the remaining delegates to have an unbeatable lead, assuming that he retains his current unpledged delegates. Given how close the races are in the big states, that’s an improbable outcome, but it’s not an impossible one.

Sources: Pollster.com, CNN Election Center

Download: PresidentialPrimaries.xls (Excel spreadsheet)

Yesterday was a very good day for Barack Obama.  Not only did he do better than expected in the Wisconsin primary and Hawai’i caucus, but a new poll from Ohio has him closing the gap on Hillary Clinton in that state.  Add to that a superdelegate switching sides in his favor (according to CNN), and his target fell by 13 delegates while Clinton’s increased by 8.

Obama’s lead in pledged delegates is now over 150, which means he just needs to secure 50% of the delegates going forwards to reach the convention with a lead that will likely be big enough to win the nomination (the “Magic Number” in the table above).  If one assumes that the polling in OH, RI, TX and PA is accurate, then he needs to average 56% of the delegates in the remaining states to clinch the nomination.  He’s scored better than that in every state election since Super Tuesday.

So it’s looking more and more likely that Obama will be a clear winner before the convention.  If he wins Texas and comes close in Ohio next Tuesday, I think the rest of the contest might be a formality.  I’m sure Senator Clinton has other ideas…