[EM] Burlington 2009 -- analysis + responses to attacks by IRVpropagandists

[Terry Bouricius]

Warren (and all),

The amount of commitment you have to finding fault with IRV (and with me) 
is impressive. I think my Blog refutations of your analysis of the 
Burlington IRV election are correct. Readers can find them here
http://www.fairvote.org/blog/2009/03/response-to-faulty-analysis-of-burlington-irv-election/
and here
http://www.fairvote.org/blog/2009/03/more-on-warren-smiths-and-anthony-gierzynskis-flawed-analysis/

You have now offered a lengthy rebuttal to my rebuttals of your original 
analysis. Unfortunately I simply can't justify spending the amount of time 
it would take to repeatedly refute all of your repeated claims. I will 
just respond to a few of the opening claims as they are personal attacks 
that specifically impugn my integrity and competence. Although your errors 
in these sections are typical of errors you make throughout your entire 
essay, I have  more important work to do, than to correct all of your 
erroneous statements.

1. You have several factual errors regarding the rate of valid ballots in 
the Burlington IRV election (which was in fact better than the recent 
presidential election in Burlington). Setting aside your rude tone, let me 
simply correct you on the following...firstly, there were not eight 
invalid ballots. The four ballots in the "exhausted ballot" category were 
not four additional problem ballots. They were the exact same four invalid 
ballots the machine tally showed and were described as "invalid." That is 
an understandable error on your part, as the software and others did not 
use consistent terminology (e.g. "over-vote," "exhausted" or "invalid") to 
describe these four ballots that the scanner determined had two candidates 
marked as first choices.

But there were not in fact even four invalid ballots (first round 
exhausted ballots). You were apparently unaware that the recount performed 
by the Burlington Board of Civil Authority found that three of these 
invalid over-votes (all in Ward One) in the Burlington IRV election, were 
in fact valid votes, with stray marks in the write-in oval-- such that 
human counters unanimously agreed that voter intent was clear (two were 
for Wright and one for Kiss). That left one invalid ballot, which 
unfortunately was in a ward that was not recounted (Kurt Wright, who had 
called for the recount called it off half-way through saying he was 
satisfied the election was counted properly on election night), so the 
actual number of invalid ballots may well have been zero, but I did my 
calculation with one presumed invalid ballot, which works out to a valid 
ballot rate of 99.99% (compared to 99.96% in the non-IRV Presidential 
election you reference.)

You also point out that a few voters marked their ballot with only a first 
and fifth choice, and I would agree that these handful of voters may have 
intended to indicate W>anyone>K. FairVote, and I, recommended to the city 
in 2005 to use an algorithm that exhausts a ballot if more than a single 
ranking is skipped, but the city council adopted a regulation that ignores 
all skips. However, in nearly every case these odd ballots ended up 
counting for the voters' first choice in the final runoff tally, so these 
skips did not invalidate any votes.

The point I was making was that claims by the defenders of plurality 
elections that ranked ballots are "too complicated"  for voters and would 
result in vast numbers of spoiled ballots are not supported by the 
evidence. I stand by that.

2. You make a big point of questioning my integrity, (Even if you doubt my 
research skills, such that you wouldn't use the term "IRV expert" or 
"election method analyst" I suspect your choosing to call me a 
"propagandist" is calculated).  You state that I did not "disclose" my 
political history, or my founding of the company that wrote the software 
that counted the votes. You are incorrect on all counts. While it is true 
that I do not begin every essay I write by listing my prior political 
history, it is certainly no secret. There is a bias among many mainstream 
politicians against members of third parties, so I don't stress my third 
party history in my work with Democratic and Republican elected officials 
around the country (but neither do I deny it). However, people in 
Burlington are well aware of my partisan history (indeed, I used to be 
president of the city council), my bio on the FairVote web site lists 
these offices, and my March 17 Blog refuting your analysis, which you 
object to, for example states: "I am used to personal attacks having 
served a decade on the city council and another decade as a state 
legislator as a member of the Progressive Party. I have also worked as an 
election administrator for non-profit organizations when not working as an 
analyst for FairVote." Also, I never served with Bob Kiss in the 
legislature  (he was elected after I left the legislature).

Also I did not found and have no connections to the company that wrote the 
IRV tallying software as you repeatedly assert (as if that were a conflict 
of interest). Again, this is an understandable error on your part (I just 
wish you weren't so nasty about expressing it). Your mistake stems from 
the fact that the software was developed by a company called Voting 
Solutions, and I had a completely different consulting business with a 
similar name (but no overlap of personnel) called Election Solutions. I 
should also note that I have no connection to Premier Election Solutions 
(the new name of the former Diebold), so that you don't start spreading 
that false rumor across the Internet. Once your kind of erroneous 
statements get into cyberspace they seem to live on forever, and it gets 
tiring refuting old errors over and over again as new people read and 
repeat them.

Again, I have more important work to do, than refute all of your claims, 
but I assure readers that nearly all of them are simply false, or 
debatable at best. We have different notions about how to make meaningful 
progress improving American elections, of the usefulness of Bayesian 
Regret as a standard, as well as different interpretations of facts, but I 
will attempt to be civil. We could all use a little more of that, I 
suspect

Terry Bouricius


----- Original Message ----- 
From: "Warren Smith" <warren.wds at gmail.com>
To: "election-methods" <election-methods at electorama.com>
Cc: "kathy.dopp" <kathy.dopp at gmail.com>; "wdh3" <wdh3 at riseup.net>; "Jack 
Nagel" <nageljh at sas.upenn.edu>; "Anthony Gierzynski" <agierzyn at uvm.edu>
Sent: Friday, March 27, 2009 3:02 PM
Subject: [EM] Burlington 2009 -- analysis + responses to attacks by 
IRVpropagandists


The analysis of the IRV pathologies in the Burlington VT 2009 Mayor
race:

http://RangeVoting.org/Burlington.html

Subsequent attacks on that analysis by
IRV-propagandists, and our responses to those
attacks:

http://RangeVoting.org/BurlResponses.html


-- 
Warren D. Smith
http://RangeVoting.org  <-- add your endorsement (by clicking
"endorse" as 1st step)
and
math.temple.edu/~wds/homepage/works.html
----
Election-Methods mailing list - see http://electorama.com/em for list info