I mentioned in an earlier post on voting patterns that one precinct, 3626, had an unusually low level of voter turnout compared to the rest of the precincts that voted on Measure S. This is the precinct that covers most of the hilltop area starting where Club Drive and Crestview Drive intersect.
The low level of turnout was particularly odd because the precinct’s turnout for the overall ballot was actually pretty typical. In other words, a lot fewer people cast a ballot for or against Measure S in 3626 than in any other precinct in the District:
| Precinct | Overall Turnout | Measure S Turnout | Relative Turnout | Undervote |
| 3202 | 84.4% | 75.7% | 89.8% | 8.6% |
| 3601 | 81.2% | 76.7% | 94.5% | 4.5% |
| 3602 | 82.6% | 79.1% | 95.8% | 3.5% |
| 3603 | 81.7% | 76.8% | 94.0% | 4.9% |
| 3604 | 84.2% | 79.8% | 94.7% | 4.4% |
| 3605 | 83.8% | 77.1% | 92.1% | 6.6% |
| 3607 | 74.8% | 67.1% | 89.7% | 7.7% |
| 3608 | 74.1% | 67.4% | 90.9% | 6.7% |
| 3609 | 80.2% | 71.3% | 88.9% | 8.9% |
| 3610 | 84.4% | 80.3% | 95.1% | 4.1% |
| 3611 | 77.4% | 71.9% | 92.8% | 5.5% |
| 3612 | 76.8% | 70.8% | 92.2% | 6.0% |
| 3613 | 83.0% | 78.1% | 94.1% | 4.9% |
| 3614 | 85.2% | 82.0% | 96.2% | 3.2% |
| 3615 | 83.4% | 79.1% | 94.9% | 4.3% |
| 3616 | 87.1% | 82.7% | 95.0% | 4.3% |
| 3617 | 85.1% | 80.2% | 94.3% | 4.9% |
| 3618 | 84.7% | 80.0% | 94.5% | 4.7% |
| 3621 | 81.5% | 76.1% | 93.4% | 5.4% |
| 3622 | 83.3% | 76.4% | 91.8% | 6.8% |
| 3623 | 82.4% | 78.5% | 95.3% | 3.9% |
| 3624 | 85.9% | 81.5% | 94.8% | 4.5% |
| 3625 | 79.2% | 72.0% | 90.9% | 7.2% |
| 3626 | 80.6% | 50.0% | 62.1% | 30.6% |
| Overall | 81.7% | 74.8% | 91.6% | 6.8% |
Overall Turnout means the percentage of registered voters in the precinct who cast a vote for or against something on the ballot. Relative Turnout means the percentage of people who cast some vote who also voted, either yes or no, on Measure S.
As you can see, 3626 has an undervote that is almost 5 times higher than the average for the entire District. It’s certainly possible that voters in this precinct, which probably has one of the smallest school age child populations in the District, simply didn’t care about Measure S one way or the other. On the other hand, there are other precincts that also have relatively fewer school age children, and their undervotes were much lower.
I was wondering what might have caused this undervote when I heard about a problem a school supporter had voting for Measure S in another precinct. Briefly, this voter showed up at her precinct on Election Day looking to vote for Measure S, but couldn’t find it on her electronic voting machine display. After some discussion with the poll workers, it turned out they had given her an incorrect code to use with the voting machine. Once she was given the correct code Measure S appeared and she was able to cast her vote.
This got me thinking. Why would a different code cause a voting machine to display a different ballot? That’s when I remembered that polling places often serve multiple precincts. Rather than dedicate expensive voting machines to specific precincts in a single polling place, the Elections Office may load all the precincts served by a polling place into each voting machine, and then use the codes to display the appropriate precinct ballot.
If true, any polling place which serves precincts on the “edge” of the District is vulnerable to displaying the wrong ballot. This wouldn’t be a risk for polling places serving precincts “inside” the District because all the surrounding precincts are also in the District.
3626 is an “edge” precinct, as it adjoins a reasonably heavily-populated area of Belmont. So is the precinct where that voter encountered her problem. Perhaps that’s why 3626 shows such a large Measure S undercount.
It’ll be interesting to see whether any of this speculation is on target. I’ve pointed out the anomaly to the County Elections Office, but have yet to hear back from them. Stay tuned.