Tom Van Dyke

Tom Van Dyke, businessman, musician, bon vivant and game-show champ (The Joker's Wild, and Win Ben Stein's Money), knows lots of stuff, although not quite everything yet. A past contributor to The American Spectator Online, the late great Reform Club blog, and currently on religion and the American Founding at American Creation, TVD continues to write on matters of both great and small importance from his ranch type style tract house high on a hill above Los Angeles.


  1. Yahn. Why don’t we hear about evangelical blacks?
    Wonder how much of the “defecting” of Republicans is caused by them changing registration to Democrat???
    I mean, if you changed to vote Obama over Hillary, you mgiht still plausibly call yourself a Republican.

    But four years later? cmon…

  2. The thing is, of course, as you head down the chart, you get to proportionally smaller and smaller segments of the 2008 ‘Obama Coalition’. So large percentage defections don’t really translate to that many votes in the big picture.

    • and iirc the ‘women’ defection needed to be around 20% for Romney to cover the gender gap.

    • Except the topline number points toward nearly 7% of the electorate shifting to Romney in the coming election from 2008 numbers — or, in the less Obama-friendly way of looking at it, his percentage plummeting to 44%. Those numbers point toward a Romney landslide — so I’m more than a little skeptical, given the rest of the polling.

      As for what the breakdown tells us — which groups are, in fact, more likely than others to “divorce” Obama — I’m willing to believe the rankings. That the “die-hards” category includes Hispanic voters points, I think, toward one of the GOP’s future problems — that number can’t be something they want to see.

      • Those numbers point toward a Romney landslide — so I’m more than a little skeptical, given the rest of the polling.

        Agreed. That’s why this really isn’t an interesting topic to pursue, if you know what I mean.

      • Also, it doesn’t show the reverse — how many people who voted for McCain are voting for Obama?

        Useless, really.

    • 13% of 69 million votes = ~8 million votes

      Interesting number crunching from reader John Bono:

      This looks like a good math experiment to try and predict the popular vote for ’12.

      In ’08, Obama had 52.9% of the vote(69,456,897), and McCain had 45.7% of the vote(59,934,814).

      According to the Washington Post, only 87% are committing to voting for Obama, and 13% are going to vote for Romney. 3% are either unsure, or voting for someone else. Let’s be generous, and assume 1.5 points of that 3% are going to Obama, 1 point is going to Romney, and .5 point to third parties. . Assuming this poll is halfway accurate(a big giant if), that means Obama can reliably count on 88.5% of his ’08 electorate.

      That means Obama’s vote total of the ’08 vote will be 61.5 million give or take(88.5% of ’08 voters).

      It also means that Romney’s ’08 vote total will be (69,456,897*.135) + 59,934,814, or 69.3 million, give or take . That’s the 13.5% of disaffected Obama voters plus McCain voters. Note this does not take into account the wild differences between GOP and Dem motivation of ’08 vs. now.

      I’m sure there’s the Elvis Factor, the McCain voter who’s going Obama in 2012, but the Obama defector is not an insignificance.

      • You can’t spot the flaw, can you? I’ll give you a few to consider:

        1) First time voters.
        2) Irregular voters.
        3) Third parties.
        4) Turnout changes.
        5) Demographic changes (minority demographics, party affiliation demographics, etc).

        That…”math experiment”…is hilarious. It’s undoubtably what innumerate people think pollsters or the evil Nate Silver do.

        Anyways, the five simple examples listed above should be enough to invalidate any conclusions from that…(hehe) “math experiment”.

        • Hmm. That was probably unnecessarily combative. Anyways, just stick to the high points: There are a number of variable factors that that ‘math experiment’ simply doesn’t address which renders the numbers utterly useless.

          They’d be worth something IF AND ONLY IF the 2012 electorate was identical to the 2008 in every respect. Same number of people, exact same demographics, same party identfication (down to third party support).

          Those are rather critical questions that pollsters routinely work with, because if you ignore them your numbers are basically useless meanderings.

          I mean, think about it — if 13% of Obama’s 2008 voters defected, but that combined demographic grew 15%, Obama’s still up 2%. (Numbers invented from whole cloth). Since things like party identification and other demographics have changed since 2008 (as well as the size of the voting population) you can’t run numbers like that and have them mean anything.

          Another example: A lot of Americans turned 18 after 2008. Their turnout percentage is low, but Romney ain’t lighting them on fire either. (They break for Obama pretty heavily).

          That math experiment is a waste of someone’s time, but I can certainly imagine those useless numbers will get pulled out to prove something dumb, sooner or later.

          The image at the top is kinda useless without scaling, either. How big is “2% of liberal democrats” compared to 50% of Republicans? What if you’re a liberal democrat and black? Did you get counted once or twice? What about Protestant women with degrees? Are they counted like four times? Or just once?

          *ugh*. Just crappy infographic work there.

        • Nothing prejudiced against the back-of-the-envelope math. Even a BJuicer could do it.

          As for the diehards, we have the irreligious, and some of the rich. Otherwise, the question arises how the Dems will do next time around when their nominee similarly lacks a certain demographic appeal.

      • If Romney wins the EC and takes the popular vote by 8 million, I’ll write a $100 check to the charity of your choice, TVD. No need for you to even reciprocate, just name your charity before the end of the weekend.

        • Young Republicans!

          But really, TVD’s defense of this travesty is that it’s back of the envelope stuff that a Balloon Juicer would do … so it’s totes legit!

          {{Recall the earlier comment about propaganda??? Yeah??? That’s all I’m gonna say about that.}}

      • “3% are either unsure, or voting for someone else.”

        That’s 2 to 3 points lower than the typical ‘undecided/not sure’ response in polling, and if anything I would expect more fence sitters when specifically measuring for defectors and loyalists.

        And I’m not sure that tracking polls (which is the source of the data) are the best way to measure this – you’re actually looking for a snapshot not a trend.

        • Voters divorcing Obama is a fact. Sorry that upsets people. If he squeaks through, he will be the first president to win a 2nd term less convincingly than the first.

          Medved:Among the 24 elected presidents who sought second terms, all 15 who earned back-to-back victories drew more support in bids for re-election than they did in their previous campaigns.

          In the past century, this base-broadening for re-elected presidents hasn’t been modest or subtle. When Woodrow Wilson campaigned for re-election in 1916 (without Teddy Roosevelt as a third party competitor), his percentage of the popular vote soared by 7 points. Franklin Roosevelt in 1936 enhanced his already formidable popularity by 4 percentage points, and Dwight Eisenhower’s landslide re-election in 1956 saw his share of the electorate rise from 55% to 57%. Richard Nixon’s improvement amounted to a staggering 17 points in 1972, while Ronald Reagan’s re-election percentage went up by 8 points.

          More recently, Bill Clinton faced Ross Perot’s “Reform Party” challenge in both his presidential contests but nonetheless raised his popular vote percentage from 43% in 1992 to 49% in his 1996 re-election campaign against Bob Dole. Even George W. Bush, whose disputed victory in 2000 and tumultuous first term produced toxic levels of partisan rancor, substantially improved his standing with the public, drawing an impressive 11.6 million more votes in his 2004 re-election campaign than in his contest with Al Gore, improving from 48% of the popular vote to a slight majority.

          And again, if folks don’t like Medved, or don’t like his facts, that’s a shame but it doesn’t change anything. It is what it is. This also holds some questions for 2016, since I submit Barack enjoys a certain support for the historic-ness of his becoming president that no other candidate is likely to enjoy in our lifetimes.

          • Ah, he thinks folks are upset that some ’08 Obama voters are not voting for him this time. Isn’t it cute how he manages to deceive himself about what folks are criticizing.

            I’m one of those Obama divorcees, TVD. Do you really think I’m upset that you’re pointing out I’m not alone? Congrats, you’ve managed to become a parody of your own self.

          • Dr. Hanley, your $100 donation to the USO will be appreciated, per and regardless of the terms you set in this comment


            terms I do not propose as a bet or a prediction.

            If Romney wins the EC and takes the popular vote by 8 million, I’ll write a $100 check to the charity of your choice, TVD. No need for you to even reciprocate, just name your charity before the end of the weekend.

            Just send it, a good and worthy cause. I want nothing to do with you, James, not even in LoOG comments sections—but in the off chance you are obliged to pay up if Romney wins by the high bar you’ve set, I accept—you just bought my giving you the time of day.

            In future, if you want me to give you the time of day, it’ll cost you $100 to the USO.

            Until then, good day, Dr. Hanley. I wish us both good luck, most of all, the USO.

          • Lovely, Tom, you act as though I’m setting too high a bar when I’m using the very numbers that you quote approvingly. Run from them, you coward, even though I didn’t even ask you to put anything on the line. Of course you don’t want anything to do with me, you don’t want anyone calling you on your intellectual cowardice and dishonesty. Run, coward, run.

          • Colin Powell, Michael Bloomberg, Gov. Christie.

            Nice divorce you’ve got there.

          • Well, Mr. K, as you note there’s a first time for everything. If he does squeak through, I’ll even allow for the possibility that his 2nd term will be better [not as wretched?] as his first. With his electoral future no longer in question and the GOP House there to thwart his progressive agenda, he might just learn the art of triangulation and exercise leadership instead of what has so far been Trumanesque divisiveness.

            [Truman left office with approval in the 30% range, ala Dubya. It’s only later media/academic revisionism that’s made him an icon.]

          • fwiw, I don’t think Truman has been rehabilitated all that much. He gets credit for being ahead of the curve on Civil rights and ensuring the military recognized who is boss in the civ-mil relationship, but the left blames him for allowing Taft-Hartley regime to come into being (and Truman could break strikes as good as any Pinkerton), the right doesn’t like him because he was after all a Democrat, and everyone questions whether or not Korea could have been done better (though nobody questions whether or not it should have been done).

            Somebody who’s stock has risen over the last few years? It’s Ike’s. As much as JFK’s has fallen. (and Wilson’s continues to fall).

          • Voters divorcing Obama is a fact. Sorry that upsets people. If he squeaks through, he will be the first president to win a 2nd term less convincingly than the first.

            The other way to interpret the situation: if Obama wins despite defections and economic conditions that have defeated previous incumbents, then the Republican candidate must really suck.

  3. So, out of curiosity, was this useless set of numbers meant to be informative, or entertaining, because it fails at the first and few enough people take recreational mathematics that I doubt there is much success at the second.

    Also, I don’t see how your point about reelection numbers holds much water when you consider that Bush in 2000 failed to win the popular vote while Bush in 2004 managed to only improve his numbers by 3% (and so landed in the low 50% range). Are you trying to point out that close elections are close, or are you simply wanting us to imagine that an Obama reelection would be less legitimate than the previous two termer even if he manages to beat Bush’s numbers by a substantial margin?

    • Re-elections are referendums: Historically, a president has built on his original totals, or been denied a second term. I confess why the simplest of truths garner so much static is beyond me.

      In Obama’s case, squeaking out re-election will come from his diehards who care not a whit about his actual record–they dig his demographic profile or are every bit as leftist as he is and cannot abide anything less left.

      As for the former, the next Dem nominee will likely be as white as Mitt; as for the latter, I have no problem with someone as ideological as my blogbrother Tod Kelly going all-in on gay issues and socialized healthcare

      and whathaveyou. I respect that. I’m pretty much that ideological meself on the other side, although I do hope I would consider voting against a Republican with a record as wretched as Barack Obama’s.

      Off the Medved grid, I have seen Obama ’12 as a repeat of Truman’s squeaker over Dewey in 1948, running against the GOP “Do-Nothing Congress” and sliming the other side from head to toe.

      He won, but his term was a disaster, so much so that although eligible to run again in 1952, he was so unpopular that didn’t even bother. Obama has been a lame duck since he lost the House in 2010 anyway.

      But I will say here that history need not always repeat itself. If Barack Obama wins a historic/anti-historic 2nd term with a slimmer margin than he won his first, perhaps he’ll surprise us. Another miserable term like this past one and history will have to adjudge him the worst 2-term president ever. Except for, you know. ;-/

      • ” [A] Republican with a record as wretched as Barack Obama’s.”

        Mitt Romney?

  4. Medved:Among the 24 elected presidents who sought second terms, all 15 who earned back-to-back victories drew more support in bids for re-election than they did in their previous campaigns.

    So what’s our standard here? Electoral votes or popular votes? Let’s go check the records.

    Popular Vote
    Andrew Jackson: 1828–55.93% of the vote; 1832–54.74.

    FDR: 1932–54.71%, 1936–60.8%; 1940–54.74%; 1940–53.93%.

    Electoral College
    Woodrow Wilson: 1912–435 electoral votes; 1916: 277 electoral votes.

    FDRL 1932–472 electoral votes; 1936–523 electoral votes; 1940–449 electoral votes; 1944–432 electoral votes.

    Whichever measure we use, Medved is wrong on the facts.

    • Medved did saysecond terms, and you know very well why 1912 was a special case. Good catch on Andrew Jackson, though.

      Washington didn’t get more support in his second run — it was unanimous both times.

      I’m not sure to do with Grover Cleveland:

      1884: 48.5%, 216 EV (victory)
      1888: 48.6%, 168 EV (loss)
      1892: 46.0 %, 277 EV (victory)

      It looks like Medved put “back-to-back” in because he knew there was an exception otherwise.

      But, holy cow, look at Madison:

      1808: 64.7% PV, 72% EV
      1812: 50.4% PV, 59% EV

      Not even close. Medved is full of it.

      To a different point, this all brings to mind Buck and McCarver and their idiotic collection of baseball playoff records (“He’s only the fourth player ever to go from first to third on an infield hit three time in an NLCS!”)

      • Easy there, Mike. We have enough full diapers already. 😉

        Medved qualifies the Madison thing elsewhere. I notice that only 9 of the 18 states of that time used the popular vote to decide their electoral votes, so it’s probably along those lines, that the two-party system as we know it wasn’t evolved yet. It’s rather pointless arguing over the phrasing in a single article but it seems to amuse some people.

        “Of the 42 men who served as president before the current incumbent, only 15 won two consecutive elections.

        Among the others, 5 died during their first terms, 7 incumbents declined to run, 5 tried but failed to win their party’s nomination, and 10 won the nomination but lost their bids for re-election. What’s more, three former presidents (Martin Van Buren, Millard Fillmore and Theodore Roosevelt) attempted to make comebacks and roared out of retirement as third party candidates; all three of them failed miserably in November, winning between 10 and 27 percent of the popular vote.

        The numbers look even worse for second terms if you remove the early “cocked hat” presidents (George Washington, Thomas Jefferson, James Madison, and James Monroe) who easily won re-election before the emergence of the modern two-party system. Washington and Monroe, for instance, both eased into second terms without campaigning and without facing even token opposition. With these early chief executives withdrawn from the equation, 70 percent of those who have served as president since 1825 (26 of 37) failed to win two consecutive terms.”

        So like whatever. The point is that ALMOST ALL presidents receive either a mandate or the boot.

        • The point is that ALMOST ALL presidents receive either a mandate or the boot.

          And that is true. But it seems not to be what Medved actually claimed. One of the things pounded into us in grad school was to not overstate your case. If you try to make your case look better than it is, you’ll end up looking bad.

          Also, when we’re talking about presidential elections, there’s just not really that many of them. Even after 200 years we’re dealing with a fairly small-n sample. This is especially true when we realize that not all those 200 years are relevant to today. As you accurately note, in the early days not all states allocated their electoral votes through popular election. In addition, the mid-late 19th century was an era of mostly weak presidents, because of how nominees were selected. The modern presidency began arguably, with McKinley, but the nature of the job changed again after WWII. But we didn’t begin selecting nominees solely through the primary system until 1972, which changed the nature of the electoral game, so arguably we can only really compare within that era.

          The upshot is that it’s really hard to make meaningful claims about patterns in prez elections because we just have too few really comparable ones, not just for this election, but for all the ones coming up in the remainder of our lives. It’s essentially the reason all the predictive models designed by political scientists are so sketchy.

          That doesn’t mean it wouldn’t be a noteworthy and interesting outcome. It’s just that it’s unclear if it would really mean anything.

          • Also, when we’re talking about presidential elections, there’s just not really that many of them.

            This will be the 57th presidential election. There have been 43 NLCS’s Did you think I brought championship series states up just because I like to talk about baseball?

          • Actually I meant to explicitly tie into your comment. Somewhere along the way I forgot.

            My favorite type of sports stat goes something like, “The Hummingbirds have never won on a Thursday night with a left-handed pitcher when it’s raining. They lost in 1932, 1959, and 1994. So we’ll see if the can change history tonight.”

          • The other good one is “Smith hits .500 against Jones, He’s got a double and a single in 4 at-bats.”

          • Did you think I brought championship series states up just because I like to talk about baseball?

            Well, … uhhh …

          • I was paraphrasing House:

            “Do you think I made that patient undergo a painful and humiliating procedure just for my own amusement? Well, yes, I do that, but not this time.”

        • Yes, if you remove the exceptions, the rule looks absolute. Funny how that works.

        • So, of the 27 presidents who failed to be elected to an additional term, 5 were dead and another 12 didn’t make to the general election. But unless Obama should happen to die, none of those apply to him. To say that the odds are against Obama’s reelection because a majority of presidents were not elected to a second term, based on figures that include presidents who, unlike Obama, failed to receive their party’s nomination or indeed died in office is questionable at best.

          Now, the column was written in March, not today, so it was perhaps theoretically possible that he would decide to withdraw or otherwise lose the nomination. But likely? Was there a serious chance in March that anyone other than Obama was going to receive the Democratic nomination (unless Obama died, in which case I admit it would be highly unlikely for him to serve a second term)?

          • Obama’s re-election sans mandate would be the historical ? here.

            Off the grid of this particular quibble on Medved is what I think is the closest analogue, Harry Truman winning a squeaker in 1948 by sliming the other party but going on to have such a miserable term that he didn’t even bother to run for re-election in 1952, though he was constitutionally eligible.

            Oy. I would not enjoy a replay of that, nor of the last 4 years, which have sucked.

        • No one has ever been elected to the presidency with a running mate who’s both Roman Catholic and younger than he is. It would be absurd to expect history to change this time.

          • Moreover, back in the days of cock-hat presidents (as opposed to these days of cocked-up logic), the only ones who didn’t get re-elected were guys from Massachusetts that were impossible to like.

      • this all brings to mind Buck and McCarver and their idiotic collection of baseball playoff records

        Personally, I blame Al Michaels for that horrible trend. He’s the bane of all modern announcing. (He’s the first player to have exactly 56 rushing yards in the second quarter in Monday Night football history!).

        Now, it’s the all important Sunday Night Records!

  5. I know Nate Silver and the 538 blog are part of the Liberal Cconspiracy and therefore non-gratis around these parts, but Nate has Obama back up to 85% likely to win the election, which is exactly where he was prior to the first Presidential debate.

    Just throwin that out there.

    • He also says that pretty much all of the 15% Romney chance is “systemic bias in the polls against Romney”.

      So apparently Romney’s chances of winning are down to “all the polls are wrong”. Which will be an interesting result for the pollsters, who are gonna have to figure out who they missed and how not to miss them next time.

      it does make me laugh about those “Romney 315” blowout speculations — insofar as you’re looking at something like a 10 point anti-Romney bias at that point.

Comments are closed.