Inertia, Discontinutiy and Voter Preferences

Social scientists talk about "persistence," "inertia," and "long run equilibrium" in their analyses and descriptions of individual, group social and market phenomena. For example, we find that brands generally revert to their mean i.e. average (sometimes, also called equilibrium) market share levels even after the firms inject some perturbations (such as aggressive advertising and promotion of the brand.) Surely, there are variations -- ups and downs -- created by the perturbations but eventually the level appears to revert to the average measure. Only a very discrete and definitive discontinuity changes this level.

Look at the current Democratic party presidential contest between Senators Clinton and Obama. In a national preference match-up, Senators Clinton and Obama were supported by about (average) 45 percent and 25 percent of the voters respectively in a national preference match-up. These numbers persisted in spite of many events including surprisingly strong fund raising reports by Obama and tentative debate performances by Clinton in October and November. None of those events provided enough discontinuity for voters to change their preference structure.

And then came Senator Obama's convincing victories in Iowa and South Carolina, and close placings in New Hampshire and Nevada in the month of January. Since these results were unexpected events (of course, not to the political class) to the Democratic party voters, the preference structure changed.

Since then Senators Clinton and Obama have both earning about 45 percent of support from the Democratic party supporters in the part contest for nomination, and both have been running about even with Senator John McCain (the presumptive Republican nominee) in the general elections match-up. Obama has been doing slightly better on average but not by much. None of the events -- Bosnia error by Clinton and Wright controversy for Obama -- has yet changed the preference structure.

Look at the perceptions of the three candidates - Clinton, McCain and Obama. As Gallup organization reports that over the course of the presidential campaign (when millions of dollars have been spent) basic perceptions have not changed much. Americans viewed McCain older and likable in January and the same perception dominates now in April. Clinton was perceived as experienced and not trustworthy then and that perception has not changed either. Obama is much better known today than before the campaign got underway, but the dominant perceptions of him (as being young and inexperienced and a fresh face with new ideas) have changed little.

It does not appear that there are likely to be any foreseen event that would shift the voter preferences substantially.

That's why this is such a dogged race in the Democratic party presidential nomination contest.

Clinton-Obama Democratic Party Presidential Contest: Are we seeing Bradley/Wilder Effect?

I have written about the Bradley effect. Sometimes, it is also called Bradley-Wilder effect. I wrote about this in 2003, and that piece is available on this blog.

Simply put, Bradley-Wilder effect asserts that some white voters when they reveal their political preferences state the black candidate as their preference (lest they be misconstrued to be closed-minded) in a white-black candidates race though the real preference of the said voters may be the white candidate.

So we have the situation of the publicly expressed polls overstating the support for a black candidate in a black-white candidates political race. The two political choices that are cited as examples of this effect are the 1981 California gubernatorial race between Dukemajian (white candidate) and Bradley (black candidate), and the 1989 Virgina gubernatorial race between Wilder (black candidate) and Coleman (white candidate.)

The question now is: is the Bradley-Wilder effect is still alive? And is it showing up in Obama-Clinton Democratic party presidential contest?

The most compelling data to assert that Bradley-Wilder effect may be operational comes from the Pennsylvania Democratic primary. About one in five Pennsylvania voters said the race of the candidates was among the top factors in deciding how to vote, according to exit polls, and white voters who cited race supported Clinton over Obama by a 3-to-1 margin.

The polls, conducted by Edison/Mitofsky for the five television networks and The Associated Press, asked voters if the race of the candidate was important: 19 percent said yes, while 80 percent said no. Of those who said yes, 59 percent voted for Mrs. Clinton and 41 percent voted for Senator Barack Obama. Of those who said no, 53 percent voted for Mrs. Clinton and 47 percent voted for Mr. Obama. Broken down by race, 13 percent of whites said race was important to them, and 75 percent of those voters sided with Mrs. Clinton. Of the 66 percent of whites who said race was not important to them, 58 percent voted for her.

Further, a recent Associated Press-Yahoo News poll found that about 8 percent of whites would be uncomfortable voting for a black president. (The actual percentage is probably higher because voters are shy about admitting a racial prejudice to pollsters.)

So are we watching Bradley-Wilder effect? May be but there are some serious confounding elements. Here is one. Unlike pre-election polls, the exit polls do not involve a "face to face" interview. Rather, the exit poll interviewer's task is to randomly select and recruit respondents, hand them a paper questionnaire, a pencil and a clipboard and allow the respondents to privately fill out the questionnaire and deposit it into a large "ballot box." (Note that the "Bradley/Wilder effect" pertained less to exit polls but to pre-election telephone surveys. The underlying theory was that white respondents were sometimes unwilling to reveal their preference for the white candidate in a bi-racial contest when they felt some "social discomfort" in doing so. That is, respondents would be less likely to reveal their true preference in a telephone interview if they believed the interviewer supported a different candidate.)

The Divide in the support of Clinton and Obama

There certainly has been very discernible divide in the electoral support of Senators Clinton and Obama in the Democratic party presidential nominating contests -- primaries and caucuses. The divide has been along many lines -- gender, race, age, level of education, level of income, and rural versus urban locale -- and some of these divides have been persistent and clear. The enclosed Decision Tree (Source: The New York Times) makes it all evident (you can click on the Decision Tree to enlarge it.)

Democratic Party Presidential Nomination Contest: A Review of the Delegate and Popular Vote Counts

After the Pennsylvania vote, it is time to update the estimated number of delegates (total of pledged delegates i.e., delegates won through elections -- primaries and caucuses, and super-delegates i.e. party officials and activists) and popular vote acquired by Senators Clinton and Obama in the Democratic party presidential nominating contest.

These -- delegates and populate votes -- counts do not include the outcomes of the primaries held in Michigan and Florida. Both Michigan and Florida violated the explicit rules set by the Democratic Party National Committee (DNC) that no state -- other than Iowa, New Hampshire, Nevada and South Carolina -- was to hold its party presidential nomination primary or caucus before so-called Tuesday when any state could have set its elections. (Super Tuesday was semi-national primary election day. The four states -- Iowa, New Hampshire, Nevada (large Hispanic population) and South Carolina (large African-American population) -- together are considered representative of the American electorate and small enough for campaign without the need for extra-ordinary amount of resources.)

As a result of their defiance of the DNC's ruling, DNC mandated that any election would be considered null and void and asked all the presidential candidates (which included Clinton and Obama but it also included Senators Biden and Dodd, and John Edwards) not to campaign in those two states. All the candidates -- including Clinton and Obama -- consented to this. There was absolutely no campaigning. Obama and Edwards even removed their names from the ballot in Michigan (for some reason, Clinton did not.) However, Obama and Edwards (and Clinton) could not remove their names from the Florida ballot -- Florida party would not allow that.

Senator Clinton now argues that the Michigan and Florida votes should be counted but Obama says no -- so far, the Democratic Party is holding on to its rules and punishment of Michigan and Florida.

If Michigan and Florida were excluded, the the number of delegates required to win the Democratic Party's nomination is 2024. (If Michigan and Florida were included, then the majority number would be approximately 2208.)

The following estimates of delegates vary some because the estimates from the caucus results and the the commitments of super-delegates are just estimates with opportunity for slightly different interpretations and counts.

Delegate Counts without Florida and Michigan
NBC: Obama 1,727, Clinton 1,594
ABC: Obama 1,721, Clinton 1,586
CNN:Obama 1,719, Clinton 1,586
CBS:Obama 1,715, Clinton 1,585
AP: Obama 1,714, Clinton 1,589
So Obama leads by about 130 delegates.

Delegate counts with Florida and Michigan (approximation)
Obama 1960-65, Clinton 1925-1930
If Florida and Michigan primaries were honored as they were conducted with no adjustments, Obama would only lead by about 30-35 delegates (Clinton would lead by about 95-100 delegates when the approximately 380 Michigan and Florida delegates are counted, i.e. the approximate delegate counts would be about 1960 for Obama and about 1925-30 for Clinton.)

And now to the estimates of popular votes under various scenarios.
Popular Vote Total without MI and FL and without caucus popular votes
Obama 14,417,134(49.2%), Clinton 13,916,781(47.5%), Obama +500,353(+1.7%)
Estimate without MI and FL but with w/IA, NV, ME, WA caucus popular vote
Obama 14,751,218(49.3%), Clinton 14,140,643(47.2%), Obama +610,575(+2.1%)
Popular Vote (w/FL) but no caucus popular votes
Obama 14,993,348(48.3%), Clinton 14,787,767(47.6%), Obama +205,581(+0.7%)
Estimate with FL and w/IA, NV, ME, WA caucus popular votes

And now we provide numbers from Michigan primary too. But this is seriously problematic because Clinton is given about 328,000 votes (because her name was on the ballot) and Obama is given zero (because the rest of votes went to "uncommitted" as Obama's name was not even on the ballot.)
Obama 15,327,432(48.4%), Clinton 15,011,629(47.4%), Obama +315,803(+1.0%)
Popular Vote (w/FL & MI) but no caucus popular votes
Obama 14,993,348(47.4%), Clinton 15,116,076(47.8%), Clinton +122,728(+0.4%)
Estimate with FL and MI, and w/IA, NV, ME, WA , and caucus popular votes
Obama 15,327,432(47.5%), Clinton 15,339,938(47.5%), Clinton +12,506 (+0.04%)

In summary, even with the disputed Florida popular votes included Obama leads Clinton. However, only when Michigan is included (where Obama is given zero votes) Clinton leads Obama narrowly.

An update on Pennsylvania Democratic Party Primary

Here is my update on Pennsylvania Democratic primary.

The latest and final round of reliable public polls seem to affirm that Senator Clinton is leading by about 6-7 points. Rasmussen and Survey USA polls both of which use automated calls, show 5-6 points lead for Senator Clinton. Mason-Dixon and Zogby (tracking) polls, both which use live interviews, also show5-6 points for Senator Clinton. There are other polls (American Research Group, Quinnipiac, Strategic Vision and SuffolkUniversity) that show a range 7-13 points lead for Senator Clinton, and one of them -- just one of them i.e. Public Policy Polling (PPP) -- shows Senator Obama leading by about 3 points, 49-46 (PPP has been quite reliable, and it was both in direction and magnitude correct in its final polled forecasts in Ohio and Texas.)

There are reports that suggest that Senator Clinton's (own) internal polling suggests a 11-points lead for her. Further, Senator Clinton has just released a last-minute political advertisement that is dark and wonders on the imponderables, -- it even has an image of Osama bin Laden -- and asks the question, "Who do you think what it takes?" This is a sterner version of Clinton's (really, Mondale's 1984) celebrated 3 am advertisement that saturated the Ohio and Texas markets and that was considered responsible for Obama's unexpected loss in Texas and a larger magnitude of loss in Ohio.

So a substantial (7-9 points) Clinton victory appears likely. However, there are tow huge caveats. First is a comparative (empirical) observation about the publicly measured preference of votes and the movement of this preference in Ohio and Pennsylvania. Examining just the averages of the polls in the days leading up to the primaries, Senator Clinton led in Ohio by about 5 percent (47.8 to 42.4) with three days to go. And this average lead of about 5 points persisted at least for about 4 days before we started seeing an uptick for Senator Clinton. With three days to go, Clinton's average support grew from 47.8 to 50.1 -- a statistically significant 2.3 percent increase. In the same time-period, Obama's support remained flat -- the final number being 43.0 which was statistically insignificant. So there was a clear, statistical uptick in support for Clinton in the last three days and the upward trend carried over to the election day with the final tally providing Clinton with a 10-points victory margin.

However, in Pennsylvania the numbers just don't show that trend. For the last 4-5 days, the average lead for Senator Clinton has been about 5 points (average support of about 47.5 to 42.2.) However, today -- with one day to go -- Senator Clinton's support has edged from 47.5 to 49.0 -- a significant gain of 1.5 percent. However, Obama's support too has edged upward today from 42.2 to 43.6 -- about 1.4 percent increase, the same magnitude of increase as observed for Clinton. So the differential appears to be stable, and we don't quite yet see the electorate breaking in favor of Clinton.

Second, the registration of substantial number of new voters and their distribution in the various regions of Pennsylvania tilts the field in favor of Obama. This is best analyzed by Jeanne Cummings of Politico as abstracted below.

"It's clear that a disproportionate share of the 300,000 new Democratic voters this year are young, and they're underrepresented in opinion surveys because pollsters don't have their cell-phone numbers. People under 34 represent 25 percent of the state's Democratic electorate, but they're only 7 percent of the likely voters in the Daily News/Franklin & Marshall poll that showed Clinton up by 6 percentage points.

According to the Secretary of State’s office, since January about 217,000 new voters have registered for the April 22 primary, the vast majority of whom signed up as Democrats. In Philadelphia, by far the state’s largest city, more than 12,000 new Democrats were added to the rolls in the final week before the March registration deadline, compared to just 509 Republicans.

That statewide Democratic surge has been accompanied by a flood of party-switching. More than 178,000 voters have changed their party status since January — and the Democrats have captured 92 percent of those voters. In Delaware County, a Philadelphia suburb once home to a storied Republican machine, nearly 14,000 voters have switched their party affiliation to Democratic since January compared to just 768 who became Republicans.

A poll of those switchers and new registrants released by Madonna last week found that Obama was the preferred candidate for 62 percent of them. Clinton insiders said they are also bracing for the same 60-40 split among newly registered Democrats. Depending on turnout, Madonna (of Franklin and Marshall) said, those newcomers could help Obama cut a Clinton victory margin by 2 to 3 percentage points and keep her below a double-digit win that would breath new life into the hard-fought race.

For instance, about 143,400 Democratic newcomers – including newly registered and party switchers — are in Philadelphia and its suburbs. Those numbers could help Obama rack up big margins in what is considered his strongest turf. About 28,400 of them are in or around Pittsburgh, an urban area Clinton needs to counter Obama’s Philly support. Another 30,000 of them hail from the generally smaller, conservative counties in the state’s northwest and southwest, a region that Clinton is hoping to draw Reagan Democrats back to the party and to her cause.

Finally, the Clinton-friendly sections of central Pennsylvania are now home to more than 70,000 of the Democrats’ new recruits, including more than 6,000 in Centre County which is home to Penn State University. An area where Obama and Clinton are likely to battle for voters is the state’s northeast corridor. Those ten counties, ranging from Carbon to Wyoming, have recorded more than 40,000 newly registered Democrats and party switchers. In Lehigh County, for instance, Clinton is expected to have an edge in working-class Allentown. But Obama could tap a vein of votes from the host of small universities and liberal arts colleges based in the county. "

Update on Tuesday, April 22 morning (9 am). Based on a flurry of additional polls (new Insider Advantage and Zogby polls among others), I can now report that there is now a more discernible uptick in the average measure of the preferences for Senator Clinton than for Senator Obama since yesterday. Clinton's average preference has grown from 47.6 to 49.3, and Obama's from 42.3 to 43.3 -- a statistically significant difference of 0.7 percent in the upward tick. Further, the trend estimator from the polls of last two days now shows a lead of about 6.6 points for Senator Clinton as against a lead of about 5.3 points on Sunday. The possibilities from the new registrations and the voter turnout still remain.

Pennsylvania Democratic Primary: What might be the outcome?

What might be the outcome of the Pennsylvania Democratic party contest between Senators Clinton and Obama? Here is a look at some data.

Basic demographics (larger percentage of electorate that is older, catholic and without any college-education, and in general similar to Ohio) provides a structural advantage to Senator Clinton. In the democratic primary contests this year, demographics have been substantial explanatory variables.

Apart from the basic structural demographic advantage, there are at least two other significant events that strengthen Clinton's appeal. First, the offensive comments by Obama's pastor Reverend Wright and Obama's own less-than artful articulation of the angst of the small-town American citizens. Second, amplification and illumination of these issues in the debate among the candidates on April 16th.

Based on all the public polling data available, it appears that Clinton is leading Obama about 48-43 with about 10 percent of the electorate remains undecided. (And in almost all the polls Obama has so far earned about 40-44 percent of the support but Clinton has consistently earned in the high 40s and lows 50s.) If the undecideds break 60-40 in favor of Clinton, she could win the primary with about 7-8 points margin. Since, so far in the democratic primaries (in 20 out of 27 contests) the late-deciders have broken in favor of Clinton (the structural and events-driven issues are stacked in favor of Clinton in Pennsylvania), it would be reasonable to assume that Clinton will secure at least 60% of the late-breakers, may be even more.

There has been a down turn in Obama's support (both nationally, see Gallup tracking poll and in Pennsylvania, see Zogby tracking poll) since the debate on April 16th. The debate appears to have crystallized some of the allegations (fairly or unfairly) against Obama.

In sum, Clinton should win the Pennsylvania primary and by at least about 7-8 points. Most likely by about 10-12 points unless Clinton's voters -- Catholics, blue-collar workers and older voters do not turn up at the polls and conversely, Obama's voters -- African-Americans, voters with college education, and very liberal voters turn up in very large numbers. A final point -- the Pennsylvania Democratic party primary is a closed primary i.e. the independents cannot vote in the primary unlike some other state primaries, and one of Obama's strength has been independent voters. However, slice the Pennsylvania pie it just does not compute well for Obama.

U.S. Presidential preference analyses based on level of education

The 2008 November U.S. Presidential elections may be quite different if Barack Obama is the Democratic party nominee. (We now know that John McCain is the presumptive Republican party nominee.) And here is one reason why.

Traditionally, the proportion of support for the Democratic party presidential candidate rises as the level of education decreases. Conversely, the proportion of support for the Republican party presidential candidate increases as the level of education level increases. This pattern has been robust and significant at least over the last 8-10 presidential elections. (Since level of education and income are highly correlated, the patterns of support for income have been similar to education levels.)

However, Gallup poll has discovered an interesting finding in its recent presidential preference survey data. In its April 10th report, Gallup Organization (based on 6,158 interviews with registered voters between March 31-April 6 with a 1-point error margin) found that both Obama and McCain were preferred by 45% of the electorate.

But among a sample of 1,440 voters (with a 3-point error margin) with a high school education or less, McCain had 46 percent to Obama's 40 percent -- a deviation from the traditional model of support. In comparison, Clinton had 48 percent to McCain's 43 percent among this sample -- consistent with the earlier patterns of support in presidential elections.

Further, among a sample of 1,936 registered voters (with a 2-point error margin) with some college, but no four-year degree, Obama got 46 percent to McCain's 45 percent. (In McCain-Clinton choice, the numbers were 49 percent to 44 percent.) Furthermore, in a sample of 1,388 registered voters (with a 3-point error margin) with a four-year college degree but no postgraduate education, Obama and McCain were tied at 46 percent (as against, 51% for McCain and 41% for Clinton.)

And, finally, among a sample of 1,350 registered voters (3-point margin of error)with some postgraduate education, Obama had 52 percent to McCain's 42 percent (as against, 48% for McCain and 45% for Clinton.)

Gallup's April 10th report is a confirmation of other similar findings in surveys measuring presidential preference. Among voters with no college education, voters prefers Clinton to McCain in Clinton-McCain match up, and McCain to Obama in Obama-McCain match-up. Among voters with some college education, McCain is preferred when matched against Clinton but it is Obama when it is McCain-Obama.

Clearly, thus, the pattern of support is predictable (and consistent with historical data)in a match-up between Clinton and McCain but not so in a match-up between Obama and McCain. So the churning and uncertainty will be greater with Obama's candidacy. Of course, who (Democratic or Republican party presidential candidate)will benefit from the new pattern of support if it persists into November is not clear? How will this translate into competition for electors is not clear? But it looks like that change and uncertainty are certain outcomes in a McCain-Obama match-up.

India's modulated approach to economy

Economic-development choices and experiments in the last fifty years (e.g., China, India, Soviet Union) have now fairly conclusively demonstrated that mass economic prosperity is better achieved through a focus on the growth of wealth rather than its distribution.

However, there are many influential Indian intellectuals, policy-makers and elected officials who do not agree with this assessment. In some quarters, these policy-makers may be seen -- and they may be right -- as impediments to even faster economic growth in India. But the caution and wariness expressed by these officials have led to a more stable and predictable economy in India. And, here, are two prime examples.

(1) Caution on foreign capital investment: India has been more careful and sometimes even rigid in approving foreign capital investment -- particularly in some areas such as insurance, media, industries with implications for national security. These regulations -- though quite often a bit too restrictive -- have generally provided stability to India's economic growth. For example, there has never been run-away inflation (e.g., Latin American countries) or serious collapse of confidence in the currency (e.g. Asian crisis.) And now look at the serious challenges that the U.S. economy in regulating the hedge funds and sovereign wealth funds -- India's policy appear prudent in the context of this.

(2) Focus on the poor, and poverty: The unprivileged and those struggling in deep poverty constitute about 800 million, about two-thirds of India's population. Simple computations would establish that India's macro economic growth rate, and GDP would increase more dramatically even if a fraction of the 800 million Indians become part of the economic engine. That requires that they participate in the economy -- consumer investment is the engine -- which, of course, translates into policies that are likely to favor some level of redistribution even if it forsakes some growth. At the margin, the benefits of such policies are monumentally greater than any losses.

Overall, these Indian policy-makers who have not adopted the market-economic philosophy without substantial skepticism have contributed to a modulated approach to economic development and growth.

Tibet: India's Dilemma

Tibet remains one of those difficult diplomatic and political, and moral challenges for the world. This is even more so true for India because India's interests are more closely and interwined with Tibet and China.

Since China's defeat of independent Tibetan army in 1949 and the signing of an agreement by the then Tibetan government, the situation has been rife for disagreement and conflict: The government of China recognizes Tibet as autonomous provincial entity of the mainland China but the government of Tibet in Exile seeks independence for Tibet. And there are other disputes which include what constitutes Tibet i.e., the geography of Tibet.

India has always been involved in this dispute. Dalai Lama who is the symbolic head of the Tibet government in exile was welcomed to India in 195o's and since then the Dalai Lama has made Dharma Shala (in India) his home. Of course, the government of China has been very displeased with this arrangement. Furthermore, in China's view Tibet also includes Arunachala Pradesh -- a province in India.

Thus there are many issues of potential dispute between India and China with regard to Tibet. However, over time China and India have let the status quo prevail for the sake of peace. Both countries have been focused on economic growth and development.

However, the forthcoming Olympic games in summer, and Tibet's recent energetic and loud demands and protests all over the world for independence and Dalai Lama's nuanced stance (of both supporting the Olympic games and the aspirations of the Tibetans for independence) in the context of long-standing support of India for Dalai Lama has made it a knotty situation.

Here is the crux of the issue: should India be focused on its immediate national interests based on assessment of geo-political, economic, territorial and security issues? or should India be an agent for fostering religious and cultural freedom and diversity consistent with its traditions (even if it led to serious consequences to immediate political and economic interests)? The answer to these questions will determine India's response to the current Tibetan turmoil including the relay of Olympic torch in New Delhi later in April.

Estimates of number of delegates acquired by Clinton and Obama

The estimated number of delegates acquired by Obama and Clinton through elections (primaries and caucuses, called pledged delegates) and commitments of appointed delegates (called super-delegates.) There is no complete certitude to the numbers because the exact numbers of delegates are not yet allocated in the various caucus elections (those numbers are not likely to be finalized till June), and the commitments of super-delegates are not verified independently.

The estimates by the various news organizations are as follows. The total number of estimated delegates are shown as sum of pledged and super-delegates in the parantheses.
NBC: Obama 1,640 (1,416+224), Clinton 1,507 (1,252+255)
CNN: Obama 1,629 (1,414+215), Clinton 1,486 (1,243+243)
CBS: Obama 1,627 (1,409+218), Clinton 1,492 (1,243+249)

These numbers do not include the disputed elections in Florida and Michigan. Without counting Florida and Michigan, 2024 represents the majority of the total possible delegate acquisition. So 2024 assures nomination as the Democratic party's candidate for the presidential contest against the Republican nominee, John Macain, in November.

Delegate Count -- Democratic Party Presidential Nominating Contest

Here are statistics as they stand today -- April 3rd -- relating to pledged delegates and super delegates in the contest to elect the Democratic Party's presidential nominee.
(Pledged delegates are those elected on behalf a candidate through a primary or a caucus or both as in Texas. Super delegates are those assigned by the Democratic party with a vote -- these are Democratic party elected officials and party activists.)

Current estimates (based on the elections and commitments made thus far) are as follows. For Senator Clinton, the numbers are: Pledged delegates 1252 , Super delegates 251 and the total is 1503. For Senator Obama the corresponding numbers are 1414, 221 and 1635 respectively.

The projected numbers of pledged delegates from the next ten nominating contests are (based on the assumption that Senator Clinton would win 60% of the delegates from Pennsylvania, Kentucky, West Virginia and Puerto Rico, and Senator Obama would North Carolina, Oregon, South Dakota and Montana by about the same margin, and that Indiana would be a near-draw) are 305 and 290.

So at the end of all the nominating contests, the total numbers of delegates would be at about 1925 and 1808 for Senators Obama and Clinton. Of course, all this leaves out Florida and Michigan because nothing is clear at this point.

If 2024 is magic number, then Obama needs another 100 delegates and Clinton needs about 220 delegates from the current pool of about 350 uncommitted super-delegates.

So Obama's path appears clearly more plausible. Based on the last two months rate of commitment of the super delegates, it is more likely that Obama secures the commitment of at least 55 percent of the remaining delegates. If that were the case, that would leave about 75-100 surplus delegates in Obama's corner. At that point, Obama can agree to even seat the Florida and Michigan delegations based on the disputed elections because those states would give only about 50-60 additional delegates to Senator Clinton, and that would not be enough to overturn Obama's majority. That dispute will just melt away. This is a likely scenario.