After Fol-de-Rol, Predictions

Talkin' Baseball

The impetus for this post at this moment is the imminent start of baseball's World Series.  It's the Yankees and the Dodgers, what I've been fearing and dreading.  For years, and decades, really, but this year this particular rivalry for the championship has raised its head once more.  

This is the matchup that has occurred most often in World Series history, which is basically to say that the Dodgers are the team the Yankees have faced most often.  The Yankees have a large edge in their Series showdowns, but the Dodgers have had a couple of memorable wins (1963, 1955) 

The Yankees' rise goes back to acquiring Babe Ruth in 1919; the Dodgers' to breaking the color line with Jackie Robinson, and then Roy Campanella in the early Fifties, . Neither team has spent much time away from the top of the standings since then.  

In my career as a Reds fan, they are both among the teams which I hated, and envied, most often. After my initial negative impressions of them formed, it was their habit of buying the most expensive players which irritated me, as it made my small-market team's chances ever more difficult. 

With the Yankees, their 4-1 defeat of the Reds in the '61 Series imprinted resistance to their inevitability, without my even being aware of it; it was something I became aware of and resented soon after its occurrence as I first experienced the outside world. We did get our revenge against them in 1976, a 4-0 Series sweep, the Big Red Machine's finest achievement. 

The Dodgers had their history with the Yankees in the Sixties, with my reluctant support--the greatness of Koufax and Drysdale was undeniable, and they proved it against the perennials, the Bronx Bombers in that '63 Series. But their re-emergence in the late Seventies as rivals to the Reds was what turned me against them for all time.  Particularly as long as they had Tommy Lasorda as manager. 

It surprised me to discover that this, the most frequent World Series rivalry, has not come up since 1981. (Don't get me started on that year's improvised playoffs, something which burns me still....), not even by accident.  More like it's been an accident it didn't occur sooner.  It's been a welcome respite.  My aversion has even faded a bit in its intensity, though I still root against both teams in all games, on principle. .

Beyond doubt, this is the ideal World Series matchup.  The #1 team in each league, the two top TV markets, Ohtani vs. Judge, as well as Mookie Betts vs. Juan Soto, Freddie Freeman vs. Giancarlo Stanton.  Six probable future Hall of Fame hitters. The network "got lucky" when the Mets won game 5, ensuring that the Series would start on Friday, instead of Tuesday (if both teams in the Series won in 5 games or less).  

Starting pitching is, somewhat surprisingly, a weak spot for both, though the Yankees do boast a top one in Gerrit Cole and a hot one in Clarke Schmidt.  If Ohtani, Betts, and Freeman (followed by Teoscar Hernandez and Max Muncy) are clicking, though, it will not matter who the Yankees send out there.  On the other side, though, the depleted Dodger rotation will give way quickly to their deep and strong bullpen, as they show the Yankee slugging stars frequent changes of pitchers ("different looks"). I expect a couple of slugfests, a couple of pitching duels (maybe not the ones expected), and a couple of blowouts.    

Election Commentary

I have no need to wait for more data to form opinion of my expectations; more poll data will serve for nothing.  The Biden Weakened scenario gave way to Biden Out, but that morphed almost immediately, due to the expert salvage operation by Kamala Harris and the Democratic Party, to get back to Status Quo by the convention.  There was an Inflection Point, indeed, as Harris' improvement trend flattened out.  And there it has stayed, despite everything (Harris destroying Trump in the debate, Trump's meandering but threatening nonsense, and more).  The 48-46 national polling advantage I cited for Harris at the beginning of the month has not really moved, despite a number of polls all over the place. 

The largest uncertainty at this point is about the mysterious uncounted Trump voter from the 2016 and 2020 polls, compared to results in elections, if he is still out there, uncounted.  Or, perhaps overly accounted for. Nate Silver had a piece in his paid Substack, and then one in the Times, showing the frequency table of  different combinations of winner, by state, resulting from his simulations, using correlations and certain other elements of his own secret sauce (but only showing the 20 or so most frequent of the 128 permutations).  His conclusion was like one that James Carville has mentioned, that the most likely outcomes are ones where one candidate or the other wins all, or almost all, of the swing states.  What he has discovered is that the error term, that 1%, plus or minus, on the Trump voter error estimate (which comes in the adjustment from registered voter to likely voter, usually done by respondents' past history of voting, when available) makes all the difference in the outcomes, along with the correlations between swing states, seemingly very high. I take this to mean that he has no clue on the big picture, but his detailed analysis proved useful to me: 

That projection sequence, from poll preference to popular vote to state outcomes, remains very central to any possible prediction based on data.  If we look at the intangibles, and the tangible adjustments we need to make in order to come up with a final estimate, we find those to be largely in the favor of the Harris-Walz campaign.  Things like ground game, money, increasingly caustic media criticism of Trump's performance, along with grudging acceptance that, yes, Harris can answer questions--all day long, if she has to--though she is not going to deviate from her victory plan of retaining all Biden's supporters and adding what she can of disaffected Republicans.  

Both campaigns are running what I would call limited tactical upside strategies. Harris is not going to articulate anything dramatic that could change the dynamics--it's too late anyway, now.  Trump is not going to appeal to moderates and independents not already leaning his way--he can't.

The question of turnout is central now; the drive to get that 1% that could make the difference in these close state races.  It also bears on the close Senate races, many of which in the same swing states closely  contested for the Presidency, and indirectly bears on the critical outcome of the contest for control of the House. In this regard, the effort by the Harris campaign to do a big event in Houston is relevant:  while the GOTV effort may not be fully realized in NY and CA (apart from the districts with the close races), the underdog effort to try and break through statewide in TX (and in FL, wait and see) will pad the popular vote margin in a similar way to 2020.  

Predictions

I start with an estimate for the national popular vote:  Harris-Walz 82 million vs. Trump-Vance 77 million. This out of 165 million total votes.* The percentages corresponding are Harris 49.7%, Trump 46.7%.  So, a bit more improvement for the Democrats than Republicans from the present poll averages as we scale up to the full 100% of voters (with 3.6% others).  By applying the adjustments to the national average by state that I reviewed in my post "Inflection Point", this then plays out to the following outcomes by state (mean estimates of Democrats' margin, with 80% MOE), with state-level adjustments and brief comments; 

 North Carolina  (-2)  +/- 2%:  1 pt to Democrats for the reverse coattails because of a miserable R Governor candidate;

Georgia (-0.5) +/- 2%:   1 pt to Republicans for Democrats' turnout being not quite as amazing this time;

Nevada (0) +/- 3%:  The largest uncertainty, but the least important for the EC;

Arizona (+0.5) +/- 2%:  I'm going against the consensus with an adjustment for the strong Senate candidate and the abortion referendum. 

Pennsylvania (+1) +/- 2% : 1 pt Adjustment for the focused turnout I expect from the Democrats. 

Wisconsin (+1.5) +/- 2% : 1 pt Adjustment for Walz-adjacent effect!

Michigan (+2) +/- 1% : Democrats should never lose Michigan, except by neglect, and that is not the case here. 

If we start with the standard 226D - 219R, the count without the swing states and with the 1 CD vote from NE and ME switching from their statewide winners, these point estimates result as 281 Harris, 251 Trump, with the 6 EV in Nevada TBD.  So that makes the final margin (prior to some potential actual fol-de-rol in the event) either 36 or 24 electoral votes. Slightly reduced from the 2020 Biden margin, more like that of Jimmy Carter in 1976 vs. Gerald Ford, which I think is appropriate in multiple ways. 

Senate

We have to accept as starting point that the Democrats will lose the seat in WV and that Tester's chances in MT against a 15-20 point gap in the Presidential race are slim.  That takes us to 51R, 49D, prior to any other seats changing hands.  The most likely ones after that are WI (Tammy Baldwin) and OH (Sherrod Brown), though I would give each better than 50% chance of surviving.  A little lower than that are others like the slim Democratic pickup chances (FL, NE, but especially Allred defeating Cruz in TX) and those for which the Republicans had hopes (NV--Rosen, MI--open seat, PA--Casey). My mean estimate would be a little worse than 49-51, but as we are limited to integers I will stick with that. Hopes will then turn to Lisa Murkowski of Alaska (Independent, now caucusing with the Republicans), especially if Harris wins.

House

I give credit to those pundits who do the research at the district level and report back, though I do not necessarily agree with their broad approach in their ratings.  Beyond the 13 or so "Toss-up" seats, all of which are truly likely to flip, there are the "Tilt' seats, which are shown siding with the incumbent in almost all cases.  This is the pundits' conservatism in calling for defeat of an incumbent or a change of party control.  When we consider those just a little tilted to one side as also highly vulnerable, there are 32 seats at high risk, 18 Republican and 14 Democrat, with the Republicans leading 204-199 in seats rated Safe, Likely, or "Lean" (going by Nathan Gonzales here, though his is very similar to those such as Cook Political Report.  Lean means the pundit has a clear idea who is ahead, though they are a little insecure about it.)

 That means the Democrats would need to win 19 of these 32 jump balls, or 60% of them, to take control of the House.  The means to take them presents itself in just a few states, and the Democrats are focused on them, almost as much as they are on the Presidential swing states (some--in AZ, for example, and PA--are the same, some are in less-urban areas of NY and CA, and a couple are in Iowa).

I believe the Democrats will succeed in gaining control of the House, picking up 5-8 seats.  A gain of 8 would bring the Democrats back to 222-213, the inverse of the Republicans' in 2022, and the same as the Democrats won in 2020.  The reversal of control of both the House and the Senate in the same election, in different directions, such as I am predicting, would be a first in US history.  Perhaps that is why the pundits are so reluctant to go there. 

Baseball

I think this will be regarded as one of the great World Series of days (post-DH).  One reason it might not be is because it may be just the first of a series of such Series, which may surpass this one (as with Dodgers-Yankees in the Fifties). Another is if one team dominates, but I don't expect that to happen.  Instead I expect it to be seven games, or if not that, six.  I reluctantly predict Yankees (or the Crank-biters, as we colloquially call them) to win in 7.  

As for the NBA, which began its regular season this week, the big question is whether the Boston Celtics can repeat.  There is a lot of parity in both conferences, if we exclude the fact that the Celtics seem a notch better, a bit more complete, as long as their quadrilateral (in order of importance, Tatum, Brown, Holiday, and Porzingis) remains intact.  We'll see, around Christmas, if that is the case.

*The comparable numbers from 2020:  Biden 81-74 million over Trump with total votes of 160 million, a margin of 51-47%.