2005 City Council Election Fun Facts

2005 Election Fun Facts (posted Mar 26, 2006)

I recently read some analysis of the Cambridge 2005 election on a few local “blogs.” Since my biennial look at the Cambridge municipal election is long overdue and there is some clear misinterpretation out there, here are a few morsels for the politically obsessed.

This was the first municipal election since 1941 in which the CCA (essentially defunct) made no endorsements. This apparently made no difference in the election, except possibly as one of many factors in a record low voter turnout. Depending on what definition you choose for “progressive,” one could argue that “progressives” gained ground with the election of Craig Kelley and the defeat of David Maher. A more likely explanation, however, is that the ongoing demographic shift in Cambridge toward higher income professionals does not benefit “townies” like Anthony Galluccio, Michael Sullivan, Tim Toomey, and David Maher. Ultimately, there just wasn't enough vote out there to elect four, and one of them was going to be defeated, namely David Maher.

Of course, if one “Independent” incumbent is defeated, this means that one challenger has to be elected, in this case Craig Kelley who had run before and who had a base to build on. The election results show that Sam Seidel could have been the one to pick up the seat with a little more directed campaigning, but you know what they say about 20-20 hindsight. In any case, Seidel came out of the race with his credibility intact, and he'll be a strong candidate if he chooses to run again.

I've heard it said that some “progressive” slates played a role in the defeat of David Maher. I don't see it. In fact, what seems abundantly clear is that the demographic shift that led to Maher's defeat probably would have produced the same result in the 2003 election, except that there was a rent control referendum question on the ballot that year. Significant money was spent to defeat that question (it lost in a landslide) and the campaign brought out a disproportionate share of voters who (a) hated rent control and (b) loved Anthony Galluccio. This gave Galluccio a large surplus of votes to transfer and which boosted Maher over challengers like John Pitkin, Craig Kelley, and Matt DeBergalis. It is very likely that in the absence of the rent control question, David Maher would have ceded his seat to John Pitkin in that election. [DeBergalis did well in #1 votes, but picked up few transfer votes.]

Here are some numbers (derived from the 2005 ballots) on two “progressive” slates I know about. The main one, from the Progressive Democrats of Cambridge (PDC)  had 7 endorsees, including 5 incumbents: Davis, Decker, Gordon, Murphy, Reeves, Seidel, and Simmons. With so many endorsed (and victorious) incumbents, one might argue that this slate elected a majority of the City Council. Someone actually made this nonsensical claim at a City Council meeting, by the way. In fact, the 2005 election was really all about individual campaigns, shifting demographics, and (in the case of the School Committee) the desire for some new blood. There was also a “Pitkin Slate,” so named only because John Pitkin reportedly urged people who had voted for him in the past to vote for 4 challengers (Adkins, Gordon, Kelley, and Seidel) in order to unseat an incumbent. Were either of these slates effective? Did many voters “vote the slate” before considering other candidates? Here are some numbers:

For the PDC slate of 7 candidates, 104 voters listed all of these candidates before ranking any others. There were 16,070 valid ballots cast in the election. The “Pitkin Slate” of 4 challengers yielded 46 voters who listed all of these candidates before ranking any others. A portion of the 46 likely voted this way based on other factors. This is not a statement about either PDC or John Pitkin. It's a statement about voters making choices without paying too much attention to any slate.

Order of exit matters

One of the things I do after every municipal election is to determine (directly from the ballot data) who would replace each of the elected councillors in the event of a vacancy. This is really just an academic exercise, because even if a councillor were elected to another office, e.g. Anthony Galluccio or Michael Sullivan, it is probable that he would finish out his Council term. If just one vacancy did occur in the current City Council, the replacement would be either David Maher (for Galluccio, Murphy, Reeves, Simmons, Sullivan, or Toomey) or Sam Seidel (Davis, Decker, or Kelley). You might expect that Brian Murphy's replacement (and perhaps that of Ken Reeves or Denise Simmons) would most naturally be Sam Seidel (the only one strong enough to have a realistic chance), but the “McSweeney Effect” makes it otherwise.

The McSweeney Effect is that the last candidate to be defeated in the PR Count is disadvantaged in the filling of a vacancy. Voting experts call this a “violation of monotonicity” - doing better should not lessen a candidate's chance of being elected. This effect led to Anthony Galluccio replacing City Councillor Bill Walsh (when he was sentenced to prison) in 1994 rather than Jim McSweeney - even though Galluccio did more poorly in the election and they were both drawing from a common voter base. Galluccio was counted out in the 16th Round (with 1262 votes - 366 to McSweeney and 270 to Walsh), followed by Ed Cyr (with 1645 votes) and then Jim McSweeney (with 1861 votes). By being defeated earlier, Galluccio was able to effectively “store” ballots with Walsh by having them transfer to him when Walsh still needed votes. When McSweeney was counted out, he gave some votes to Walsh (274), but once Walsh was elected, relatively few ballots were transferred to other candidates and most of the ballots (1384 of them) were “exhausted.” McSweeney filed a lawsuit that went to the Mass. Supreme Judicial Court who decided against McSweeney. Those of us who followed the case closely (I went to each of the court hearings and consulted with McSweeney's lawyers on several occasions) felt that the wrong issues were argued before the Court and that the proposed remedies were even more problematic.

One alternative way to fill a vacancy would be to re-run the election with the vacating councillor excluded. Of course, you would also have to stipulate that no elected councillor could be “unelected” by this process. We don't have this option, by the way, and any modification to Cambridge's election system can only be made via home rule petition to the State Legislature. Even this could be problematic because the original M.G.L. Chapter 54A that prescribes our election method was repealed in 1972.

If we were to re-run the election to determine who would fill a vacancy, the results would be as follows, where initials give the order of finish of the candidates:



w/o Decker: AG,HD,MS,DS,TT,KR,BM,CK,SS

w/o Galluccio: MS,MD,TT,HD,DS,KR,DM,BM,KR

w/o Kelley: AG,MD,MS,HD,TT,DS,SS,BM,KR

w/o Murphy: AG,HD,MD,MS,DS,TT,KR,SS,CK

w/o Reeves: AG,DS,MD,MS,TT,HD,BM,CK,SS (neither Kelley nor Seidel reach quota)

w/o Simmons: AG,MD,HD,MS,KR.TT.BM,CK,SS (Seidel does not reach quota)

w/o Sullivan: AG,TT,MD,HD,DS,KR,DM,BM,CK

w/o Toomey: AG,MS,MD,HD,DS,KR,DM,BM,CK

w/o Galluccio and Sullivan: TT,MD,HD,DM,KR,DS,BM,CK,SS

Note that using this method, Seidel would replace 6 councillors (Davis, Decker, Kelley, Murphy, Reeves, Simmons) and Maher would replace 3 councillors (Galluccio, Sullivan, Toomey). In the event of a double vacancy of Galluccio and Sullivan, the replacements would be Maher and Seidel.

This really gets interesting if you consider what would happen (under the existing rules) if Sullivan exited first and then Galluccio, or if they exited in the reverse order. In the first case, Maher would replace Sullivan, and Seidel would then replace Galluccio edging out Jesse Gordon on a 176-154 vote. In the second case, Maher would replace Galluccio, and Gordon would then replace Sullivan, edging out Seidel on a 154-140 vote. In other words, the order of exit produces different replacements!

Again, this is all academic! -- Robert Winters

More 2005 City Council Election Fun Facts (posted Mar 28, 2006)

Number of candidates ranked:
87 voters ranked 0 candidates -- 0.547%.
1190 voters ranked 1 candidate -- 7.34%.
1350 voters ranked 2 candidates -- 8.33%.
2761 voters ranked 3 candidates -- 17.04%.
2379 voters ranked 4 candidates -- 14.68%.
1987 voters ranked 5 candidates -- 12.26%.
1527 voters ranked 6 candidates -- 9.42%.
1188 voters ranked 7 candidates -- 7.33%.
966 voters ranked 8 candidates -- 5.96%.
1186 voters ranked 9 candidates -- 7.32%.
367 voters ranked 10 candidates -- 2.27%.
155 voters ranked 11 candidates -- 0.96%.
104 voters ranked 12 candidates -- 0.64%.
72 voters ranked 13 candidates -- 0.44%.
45 voters ranked 14 candidates -- 0.28%.
46 voters ranked 15 candidates -- 0.28%.
77 voters ranked 16 candidates -- 0.48%.
124 voters ranked 17 candidates -- 0.77%.
590 voters ranked 18 candidates -- 3.64%.
1 voter ranked 27 candidates -- 0.006%.

Voting Success:

There were 16202 ballots cast (including invalid ballots).
12748 voters elected their #1 choice -- 78.68%.
2384 voters elected their #2 choice -- 14.71%.
432 voters elected their #3 choice -- 2.67%.
111 voters elected their #4 choice -- 0.69%.
46 voters elected their #5 choice -- 0.28%.
21 voters elected their #6 choice -- 0.13%.
12 voters elected their #7 choice -- 0.07%.
14 voters elected their #8 choice -- 0.09%.
9 voters elected their #9 choice -- 0.06%.
425 voters did not elect any candidates -- 2.62%.


Note that 78.68% of voters elected their 1st choice, 93.39% of voters elected either their 1st or 2nd choice, and 96.06% of voters elected either their 1st, 2nd, or 3rd choice.