Translate

Tuesday, March 03, 2020

Super Tuesday Special Edition

So much has changed so quickly that I have to make a few comments prior to the election returns this evening (and beyond).

There is an unusually high degree of uncertainty, due to the mix of three new elements:

  • The arrival (finally!) of Mike Bloomberg on voters' ballots--the great, pure test of the effect of saturation TV advertising;  
  • The effect of another real-world 21st-century experiment: using multiple endorsements and perceived momentum in achieving real electoral results without traditional grassroots organization (the Biden campaign); 
  • The changing fluid dynamics from the rapid, simultaneous withdrawal of significant candidates in a race that remains multi-lateral.  
Apart from having real significance for the outcome of this momentous general election coming this November, today's results will provide fabulous grist for political scientists.   There is little experience on true multi-sided contests.  Much as the big picture, along with the tendency of US first-past-the-post election systems, would suggest we should expect convergence to a two-person showdown, right now there are still four contenders who can draw significant numbers of delegates today, and the variations of support evidenced for each lead to some very interesting contests to watch tonight. 

A Note on Methods
Somewhat amazingly, two polling outfits have been able to file results in each of the 14 Super Tuesday states, conducted entirely after Saturday's surprisingly-strong Biden victory there. So these reflect his sudden resurgence.  Even more impressively, in one of the two (Data for Progress), many of the states' polling seems to have been done on Monday, with some knowledge that Buttigieg and Klobuchar were dropping out. 

So, for my own analysis, I have allocated the percentages shown for Buttigieg and Klobuchar (and Steyer, when applicable) as:  40% to Biden, 20% to Sanders, 20% to Bloomberg, and 20% to Warren.  There are reasonable arguments as to how much of their previous supporters would move in the direction of their endorsement and how much to their previous second- or third-choice, if they were not Biden; this is supposed to be a simple compromise approach to estimate that unknown.  

The other adjustment is the "gross-up" to 100% (actually 98 or 99%, to give Tulsi her due) of these results.  The polls vary in the extent to which the sum of all the candidates' shares are less than 100%--the ones with 95-98% accounted for clearly used more pressing methods to get respondents to choose than the ones with 85-90% allocated.  I personally prefer the latter, as many of those "borderline don't know" people don't actually vote. 

Five Major Outcomes to Watch For Tonight
0. Delegates Won 
This is the most important one, but I expect the results will not be clear enough to report with confidence tonight, despite what will be very earnest efforts by the news networks.  The main reason will be the uncertainty about which candidates will achieve the binary 0/1 outcomes on the 15% vote threshold, for the statewide and congressional-district level delegates, and the leverage those will have on the calculations.  We can see (from prior coverage efforts) that the networks will try to use partial returns toward the objective of being able to call (to something like 95% likelihood) at least some of those outcomes, but there will be many, many delegate decisions remaining outstanding by breakfast tomorrow. 

1. How Strong the Biden Surge?  
 This primary campaign has had a "Flavor of the Month" characteristic for a long time.  Remember the surges for Kamala, Beto, Julian, Pete, Amy, Warren (last fall), Gabbard, Booker?  (OK, I guess not Booker)  Generally they have resulted from spotlighted media exposure driven by news stories or attacks they made.  Most of them didn't last, and the attackers usually suffered in the end.   Is Biden the flavor of this month, or will his momentum fade, due to our fickleness or some unusually huge gaffe (there will still be many small ones, I can guarantee). 
There are two aspects to the force of the surge we will be able to measure:  1) the rate of increase for Biden in states where he was not even viable prior to SC (Ex. CA, MN, CO); 2) The degree to which his support consolidates to produce big wins in states where he was already strong (NC, OK, AL, VA)--will these match up to the degree of his wipeout win in SC? 
Key test:  1)  Does Biden's % in CA exceed 25%; 2) Does his % in NC exceed 40%?
As for whether the Biden Surge is merely this month's flavor (great as this month/week was to have as his), or something that will either endure or grow stronger, that will have to wait. 

2. Who Wins More States, Biden or Bernie? 
In terms of clear indications from the current polls, eight states have clear indicated "winners"--plurality of the popular vote.  I'd say they range from 80% probability to 95%, so probably one of these will be wrong.  
For Biden:  NC, VA, OK, AR, AL;   For Bernie:  VT, CO, CA. 
The six others present a variety of close races, which will draw a whole lot of the attention.  They don't matter much for the delegate numbers, because of proportionality, but still: 
Bernie/Biden showdowns: TN (Biden favored--turnout may be adversely affected by tornadoes overnight in Nashville, yet another uncertainty), MN (Bernie favored--see #3 below), and, most significantly, TX.  
Three-way races:  UT (Bloomberg and Biden narrowly ahead of Bernie); ME (probably Bloomberg's best chance for a state today; he's slightly leading Bernie and Biden); MA (Bernie/Biden/Warren).  MA could be a close four-way race, as the poll from the other outfit (Swayable, which conducted their polls Sunday and Monday) actually had Bloomberg ahead of Warren.  A fourth-place finish in her home state might actually convince her to withdraw, which would create another ripple in the force fields later this week.   
My only prediction:  Biden 7 states, Bernie 6, Bloomberg 1. 

3. How transferable is endorsement support? 
Buttigieg's endorsement may be more meaningful to Biden in the long run than Klobuchar's, but the value of Amy's endorsement will have a very clear measure in MN.  Before last weekend, Klobuchar and Bernie were in a close race, with support around 30%, with Biden in the also-rans, well below the 15% threshold.   Though Bernie should be expected to finish first, how close Biden gets to him will make a meaningful difference in delegates.  It might also give an indication if, down the road sometime, Warren bows out and either endorses Sanders directly or implicitly. 
Key test:  Does Biden's number reach 30%?   

4. Is Bloomberg a real threat to Biden's dominance among moderates? 
Put another way, what is Bloomberg's ceiling, when he dominates voter contact through his ads?  It's more complicated than just a Biden/Bloomberg showdown anywhere, as Sanders and Warren also competed in all the states Bloomberg has saturated.  He was planting the seed, which might have flowered if Biden had not rallied, but Joe's still there.  The best ways to measure the relative strength of Biden and Bloomberg will be in states where they have both polled relatively well--AR,  UT, and ME--and those where Biden was weak, but coming on strong--CO, CA. 
Personally, I expect the strength of Biden's surge may falter in weeks to come.  Bloomberg may feel the same way, but if he does not do well enough today, he may feel the urge to fall in behind Joe.  It may also take a week or two to re-calculate his ROI for his first half-billion of expenditure.  
Key test:  In how many of those five states--AR, UT, ME, CO, CA--does Bloomberg edge Biden?  

5. Can Warren survive?   Is she still "viable"? 
Trends are clearly not in her favor, as conventional wisdom settles on the inevitability of a 'B-geezer' battle (Bernie/Biden, or if not that,  Bernie/Bloomberg); she gets no attention and not much respect for her First Four finishes in the First Four--none of which were better than third (IA; one was fifth).  Her long-run strategy of building organizations in many states ran afoul of the whiplash of the early primary/caucus results and associated news stories, with the result that her ceiling for results in Super Tuesday states are uniformly on the low side.   The only positive is that the withdrawal of Buttigieg, Klobuchar, and Steyer will give her a few extra percent, which may make the difference as to whether she reaches 15% or not in a number of states.   This result is important also for the other candidates' ultimate delegate results:  how many ways will the pies be divided? 
Key test:  In how many states will Warren reach 15% statewide?  
Should be safely above:  VT, MA, CO.  Likely below:  AL, NC.  Can't tell--projects right around 15%: VA, UT, TN, OK, MN, ME, AR, and most importantly, TX and CA. 
Her plan is to continue on, regardless, but embarrassment in MA or in her birth state of OK might change her mind.   I would say that if she reaches 15% in seven of these, and including both TX and CA, she should stay in, though I don't know if she will see it that way.  The true shape of the delegate race is not likely to be clear for a couple of weeks yet, and her share of delegates (probably will end up being 5-10% of the total) may be important, or meaningless, in the big picture. 


Watching the results from Texas and Massachusetts will answer many of these questions.  I would also cite--apart from the obviously important, but slow-moving CA results--the extent of Biden's win in NC.  If it is comparable to the one he had in the other Carolina, it suggests a strong argument for his eventual nomination:  if Biden can bring in NC for Democrats, it would be very good news for the Electoral College race.