And again, it's changed in the initial round of counting, but if the RCV projections from today hold, Malia Cohen still holds on to win a fairly convincing victory.
BeyondChron has a good analysis of how a candidate can move from fifth place in the raw numbers up to first after ranked-choice voting.
The numbers as of 11/14...
| 11/04 | 11/05 | 11/06 | 11/07 | 11/08 | 11/09 | 11/10 | 11/11 | 11/14 | |
| Total Votes | 10764 | 11973 | 13666 | 15272 | 15368 | 15520 | 15661 | 16177 | 16284 |
| Tran | 1166 (10.83%) | 1309 (10.93%) | 1595 (11.67%) | 1834 (12.01%) | 1855 (12.07%) | 1857 (12.04%) | 1897 (12.11%) | 1947 (12.04%) | 1964 (12.06%) |
| Kelly | 1375 (12.77%) | 1504 (12.56%) | 1644 (12.03%) | 1860 (12.18%) | 1870 (12.17%) | 1889 (12.24%) | 1907 (12.18%) | 1951 (12.06%) | 1962 (12.05%) |
| Sweet | 1300 (12.08%) | 1451 (12.12%) | 1629 (11.92%) | 1823 (11.94%) | 1829 (11.90%) | 1837 (11.91%) | 1867 (11.92%) | 1931 (11.94%) | 1939 (11.91%) |
| Cohen | 1254 (11.65%) | 1377 (11.50%) | 1557 (11.39%) | 1732 (11.34%) | 1737 (11.30%) | 1743 (11.30%) | 1771 (11.31%) | 1856 (11.47%) | 1862 (11.43%) |
| Moss | 1198 (11.13%) | 1364 (11.39%) | 1527 (11.17%) | 1726 (11.30%) | 1738 (11.31%) | 1743 (11.30%) | 1772 (11.32%) | 1821 (11.26%) | 1834 (11.26%) |
| Duque | 778 (7.23%) | 904 (7.55%) | 1129 (8.26%) | 1246 (8.16%) | 1258 (8.19%) | 1258 (8.15%) | 1279 (8.15%) | 1330 (8.22%) | 1347 (8.27%) |
| Lacy | 793 (7.37%) | 867 (7.24%) | 1007 (7.37%) | 1125 (7.37%) | 1128 (7.34%) | 1131 (7.33%) | 1149 (7.34%) | 1184 (7.32%) | 1195 (7.34%) |
| Jackson | 645 (5.99%) | 718 (6.00%) | 834 (6.10%) | 929 (6.08%) | 936 (6.09%) | 944 (6.12%) | 961 (6.14%) | 991 (6.13%) | 996 (6.12%) |
| Enea | 324 (3.01%) | 365 (3.05%) | 424 (3.10%) | 469 (3.07%) | 470 (3.06%) | 473 (3.07%) | 484 (3.09%) | 500 (3.09%) | 502 (3.08%) |
| Smith | 295 (2.74%) | 319 (2.66%) | 352 (2.58%) | 396 (2.59%) | 397 (2.58%) | 399 (2.59%) | 402 (2.57%) | 414 (2.56%) | 415 (2.55%) |
| Calloway | 272 (2.53%) | 308 (2.57%) | 338 (2.47%) | 354 (2.32%) | 355 (2.31%) | 355 (2.30%) | 356 (2.27%) | 368 (2.27%) | 368 (2.26%) |
| Wesley-Smith | 239 (2.22%) | 268 (2.24%) | 291 (2.13%) | 324 (2.12%) | 328 (2.13%) | 328 (2.13%) | 330 (2.11%) | 342 (2.11%) | 344 (2.11%) |
| Weber | 221 (2.05%) | 240 (2.00%) | 254 (1.86%) | 266 (1.74%) | 267 (1.74%) | 270 (1.74%) | 271 (1.73%) | 278 (1.72%) | 281 (1.73%) |
| Morris | 202 (1.88%) | 210 (1.75%) | 240 (1.76%) | 259 (1.70%) | 260 (1.69%) | 260 (1.69%) | 260 (1.66%) | 268 (1.66%) | 272 (1.67%) |
| Rhodes | 173 (1.61%) | 185 (1.55%) | 204 (1.49%) | 214 (1.40%) | 218 (1.42%) | 218 (1.41%) | 221 (1.41%) | 227 (1.40%) | 229 (1.41%) |
| Hampton | 134 (1.24%) | 149 (1.24%) | 159 (1.16%) | 181 (1.19%) | 184 (1.20%) | 184 (1.19%) | 189 (1.21%) | 194 (1.20%) | 196 (1.20%) |
| Donaldson | 129 (1.20%) | 143 (1.19%) | 159 (1.16%) | 176 (1.15%) | 177 (1.15%) | 177 (1.15%) | 178 (1.14%) | 186 (1.15%) | 188 (1.15%) |
| Norman | 103 (0.96%) | 113 (0.94%) | 125 (0.91%) | 139 (0.91%) | 139 (0.90%) | 140 (0.91%) | 142 (0.91%) | 151 (0.93%) | 152 (0.93%) |
| Joshua | 77 (0.72%) | 84 (0.78%) | 91 (0.67%) | 102 (0.67%) | 103 (0.67%) | 103 (0.67%) | 104 (0.66%) | 112 (0.69%) | 12 (0.69%) |
| Franklin | 46 (0.43%) | 53 (0.44%) | 59 (0.43%) | 66 (0.43%) | 67 (0.44%) | 67 (0.43%) | 67 (0.43%) | 69 (0.43%) | 69 (0.42%) |
| Jennison | 40 (0.37%) | 42 (0.35%) | 48 (0.35%) | 51 (0.33%) | 52 (0.34%) | 52 (0.34%) | 53 (0.34%) | 57 (0.35%) | 57 (0.35%) |
I believe the only possible winners are Kelly, Sweet, or Cohen.
ReplyDeleteThe top 5 took up 58.62% of the vote. That means it would be very unlikely that even when the 2nd/3rd votes from the remaining 16 candidates are added up, that any one of the five will have a majority. Even if any one of the top five candidates got a full 2/3 of the remaining votes, they would still be in the 38-42% range.
I think it is a fair assumption that most people tend to vote for the folks in their neighborhood that they know better. Birds of a feather, etc. (I would like to assume that we voters don't simply vote based on race and neighborhood, that we can look beyond that to the issues and character qualities, but history suggests otherwise.) When I look at the list of bottom 16 candidates, most come from BVHP. I would assume that most of those votes would then tend to shift to Sweet and Cohen.
After that, things may tip in Kelly's favor. Next off the list is Tran. Her endorsement of Kelly as her 2nd choice candidate could swing a large chunk of her 9.91% his way. Next is Moss, who, like Tony, is a Potrero neighborhood candidate. One could assume that sizable chunk of his votes shifts to Tony. Will it be enough to put him over 50%? Very unlikely. Even with all of Tran's and Moss' votes, plus 50% of the lower 16, he ends up at around 45%.
Unfortunately for Cohen (and Kelly), she is the next cut if she doesn't have a majority at this point, which is very possible if we assume that she and Sweet are splitting some of the same demographics and neighborhoods... which also suggests much of her count will shift to Sweet. I'm not a fan of Sweet's, and things could go differently than I've gamed it out here, but I believe that Sweet, being in the number 2 slot, and the likelihood that she will pick up a majority of the 2nd and 3rd choice votes from other BVHP and AA candidates, has the advantage over Kelly.
I only see this changing if Kelly or Cohen had major success garnering votes outside of their home turf and with broad demographic support.
Another factor will be the number of ballots which are "exhausted" in early 'runoffs' (ie., a ballot that chose three lower candidates and does not get applied to any of the top candidates). With as many candidates as we have, anyone who didn't vote for any of the top 3 candidates somewhere in their 3 choices will probably exhaust their ballot before a candidate has a majority. With no clear frontrunner candidate, it is very possible that many voters could have chosen candidates who fell below the top 3, which could have a major influence on who wins.
How?
1.) Certain demographic or neighborhood chunks of voters could simply disappear, rather than add to the vote totals of the top candidates.
2.) As these ballots are exhausted, they are eliminated from the vote totals. What this means is that the percent of the vote the remaining candidates have goes up as ballots are exhausted. Said another way, it takes less votes to reach 50%.
So, Kelly could pull this off if he was VERY successful in picking up significant 2nd/3rd choice votes from other neighborhoods and demographics, and/or he gets pushed over the 50% mark before Cohen is eliminated. Cohen could pull this off if, similarly, she picks up a broad enough swath of voters to put her in the overall lead, AND at the same time enough ballots are eliminated to make that total cross over the 50% line. Quite a few if's and and's there. I still think Sweet has the advantage.
A big question for me is who did the Lacey, Duque, and Jackson voters put in their 2 and 3 spots? Those three represent about 20% of the vote (and probably much more once the bottom 16 are eliminated). I suspect what happens when those three are eliminated will set the stage for the eventual winner.
I'd be curious to see if someone has a different read on this.
I don't think all the "not present at the polling station" votes are in and counted. In addition, I suspect that few people who voted for Moss would have voted for Kelly as the #2. My bet is the eventual winner will have a majority but not 50%. There are probably a ton of vacant votes. Fitting - vacant votes in a vacant district. Sad.
ReplyDeleteDale, I'm glad to see I'm not the only one playing these complicated parlor games!
ReplyDeleteI see many of the progressive's (Jackson and Lacy) #2/#3 votes going to Kelly. I see Duque's votes going to Tran and Sweet, with some going to Moss. I see a lot of the bottom 11 candidates' votes going to Sweet and to a much lesser degree, Cohen. That's roughly 1400 votes up for grabs, and could push Sweet over the top of everyone. Smith's votes likely go to Cohen, while Enea's will be split.
I think the top five will remain the top five, but it's a matter of who's ahead and by how much when there are only five left that matters.
North Slope Voter here.
ReplyDelete1. Cohen 2. Enea 3. Kelly.
Heard that many did the same that I know and that there was opposition to Steve Moss as second and third choice based on some questionable issues that appeared not to be answered.
Think it will be Kelly or Cohen, believe they are honorable people that will make the right choices.
I think for the most part, leaving out the race and neighborhood factor is wishful thinking. Agreed that the lower 1/3 votes are BVHP clustered and if they voted for a #2 and #3 these votes will likely go to Sweet or Cohen based upon a shared identity. My guess is that these votes will lean to Cohen because Sweet is much more polarizing. Weber’s #2 and #3 are a toss up, but I think they will be dispersed and not change the top tier positions, but Cohen will pick up more then the others because unlike Moss and Sweet she more of a populist. E. Smith’s supporters will help Cohen because of the co-branding effort. Kelly and Cohen will benefit from Enea’s #2 and #3 votes. I don’t see Enea’s supporters as Sweet or Moss supporters. Chris Jackson’s supporters will be split by Cohen and Sweet with Cohen getting the biggest push because race matters. There is major beef between Duque and Tran so I can’t see many of Duque’s voters going over to Tran..there’s more to this story then we westerners will ever know. Duque only appeal is to a certain segment of the Chinese community. She came out of nowhere and might be the biggest ‘spoiler’ in this race because race matters. Lacy’s supporters will help Sweet and Cohen, again race matters. Tran’s supporters will clearly help Kelly because of their co-branding. As you can see from my purely unscientific, one sided, and wishful thinking analysis, at the end of the day I concur that the top tier will remain largely unchanged. Kelly will drop to #2. Sweet and Cohen will be neck and neck for #1 until the bitter end. If it goes down to Tran getting eliminated, Kelly could squeak back up to #1 with Moss dropping to a strong #4.
ReplyDeleteDon't count Moss out just yet. His IE and labor supporters spent tons of cash and hours on the phone calling Chinese speaking supporters.
ReplyDeleteI have to say it's all a crapshoot but to answer one of your points I don't see Enea supporters supporting Kelly at all. In fact, I expect they voted against him as much as they voted for Enea.
I for one can't wait to see the certified vote results showing precinct by precinct.
Appreciate the numbers and the updating. Can you put a total at the bottom? Thanks
ReplyDeleteYou want me to add, too!?!
ReplyDeleteCan you merge the two posts together? (First day results and second day). And, put any updated numbers on it, too. That way the comments are all together and chronological.
ReplyDeleteOne issue not factored yet: uncounted votes. As I understand it, as much as one third of the votes are not yet counted, and that could dramatically shift the positions on the list. I'm not quite sure, but I've heard a large proportion of those uncounted votes are mail-in and provisional ballots from Potrero Hill. We know who benefits from that.
ReplyDeleteIn addition, here's a question: do candidates shift in the rankings after each new count, such that someone who was in, let's say 17th place right now, could shift to 20th place after the second count, thus being next in line for elimination? Or, is it as I understand, that the 1st choice votes set the elimination positions, regardless of any additional votes a candidate may get?
Depending on the answer to that question, things could play out very differently.
Whups - just got an answer to my question, and an update from candidate, Steve Moss:
ReplyDelete"The rankings shift after each rcv round. In other words, a second place vote looks exactly like a first place vote after its reallocated.
So far today they've counted 800 additional d10 votes, mostly from vis valley. New results should be posted in a couple of hours."
It is a little concerning that the department of elections would report this type of information to a specific candidate.
ReplyDeleteHey Chris. Thanks for the changes. Big improvement. If you're using Excel you can easily total at the bottom. That would be cool.
ReplyDeleteJust an observation: Had Duque not run, Tran might have been in the top three and C. Jackson would have done much better.....
ReplyDeleteAgreed. And, Tran's just a little too early to the party. In another 10 years the Bayview and Viz Valley will look like the Sunset and the Richmond. We have better weather, though! As we speak, Asians are buying up foreclosed houses in the 10 and sending their kids to school. The tram takes you directly downtown, to the CalTrans station, and to the UCSF complex. The times are a changin.
ReplyDeleteFirst thing, we need an image re-do. Let's just call it "The 10." I'm open to any other ideas. Perhaps Chris can open a section in the blog called "Make it a 10: Improving the District." Or, something like that.
Doubt we'll get them today.
ReplyDeleteUseless city network is down.
To Anonymous of 11/4, 2:57pm
ReplyDeleteNot really. Moss has a representative monitoring the count, just like any candidate can. Counting votes is an open process precisely so that no sneaky s#!t is going on behind closed doors.
Thanks Dale.
ReplyDeleteReally useful and interesting comments.
Chris/
ReplyDeleteOn a lighter note and votes cast:
In California Meg Whitman received 3.14 million votes and lost. However, Prop 19, legalizing marijuana yes votes received 3.4 million, nearly 300K more than Whitman, interesting.
Do we have today's counts?
ReplyDeleteOff topic:
I finally got a real security 'word': "ration".
Yesterday I got "knisses". I've started thinking up definitions for these weird non-words. "knisses - kissing with one's knees."
@Dale, we do, but I'm off to an event, so can't post right now. In the meantime, check out http://d10canwatch.blogspot.com/2010/11/preliminary-rcv.html
ReplyDeleteWhat, you aren't sitting idly by the keyboard 24/7 waiting to do our bidding? The nerve!
ReplyDeleteAny sense how many more counts are left, or how many votes uncounted, and which neighborhoods those votes come from?
ReplyDeleteToday's had almost none from d10. Not sure how many more there are, but it's kind of funny how we feel that these results are taking so long. I guess everything in the future won't be faster after all!!
ReplyDeleteI know a couple of folks were at the Dept. of elections earlier, expect a D10 update in the AM to where they stand in the count.
ReplyDeleteI have to say it all makes my head explode. Mama mia!
ReplyDeleteLooks like BeyondChron has called it for Cohen (http://tinyurl.com/2ce35fq). I think they have it right. With the number of remaining ballots left, and the distribution of where they're from, it seems unlikely that there will be any shuffling of the RCV results. The best shot for Kelly (or anyone else) would be to have Cohen drop before Sweet in the RCV, but it seems no matter which order the middle candidates go out, Cohen is the key beneficiary most of the time. The only other outside chance is that the remaining ballots to be counted show a surprising variance from the results already counted for each neighborhood. That's pretty unusual, but not unheard of. We'll know soon, I think.
ReplyDelete