Here are the results and analysis of the mock election held by Columbia IRV during August 2001:

Introduction
Results
     First Count
     Second Count
     Third Count
     Fourth Count
Conclusions
Acknowledgement
 
 

Well the ballots from the Twilight Fest IRV booth's mock election have been tallied and the results proved quite interesting. Before announcing the results, however, I thought I would point out some factors regarding the interpretation of the election results such as sampling bias and related methodological issues. 

1. As with Earth Day, the Twilight Fest crowd who filled out ballots are not a random sample and thus the results are not a representative cross section of Columbia, much less the Nation. As with Earth Day, the people who filled out ballots at Twilight Fest tended to lean to the left...I think for two reasons...the first being that events like Twilight Fest held in Downtown Columbia attract a disproportionate number of progressives and second, the progressives are more likely to take the time to fill out an IRV ballot. 
I believe that Greens and Libertarians are probably the most aware and educated on Electoral reform issues like IRV and thus are on average more likely to vote at a mock election IRV booth (or in the case of our ballot, Greens in particular, since Browne was not on our sample ballot...see note 2 below) than say a typical Republican or Democrat (of course people who are "active" in the major parties are also likely to be educated on IRV, but currently what might be called the "average voter" is not yet familiar with IRV in my opinion). 

2. I owe an apology to anyone who would have chosen Harry Browne, John Hagelin or Howard Phillips as their Presidential choice. 
Clearly to best illustrate the need for Instant Runoff Voting ALL candidates on the ballot should be included, otherwise we are somewhat defeating the purpose of the very idea that we are advocating...fuller participation and a greater range of choices in elections. 
Unfortunately, because I am involved in a number of other activities and because I took a 17 day vacation in the Rocky Mountains during June I failed to print out new ballots for Twilight Fest and instead used the leftover four candidate ballots from the Earth day mock election. 
Initially we had gone with a four candidate ballot to keep the concept of voting being as simple as 1,2,3 consistent with a ballot that required voters to only rank order 3 choices. In light of Mitch Moore's valid criticism of our Earth Day ballot I have since come to the conclusion that a 7 candidate ballot would better highlight the necessity of IRV even if giving voter's 6 choices might seem confusing or cumbersome to some. 
Ultimately however, I failed to make time to print out new ballots which would have included the three additional candidates and for this I apologize. 

3. Since this a mock election we did not require voter registration. Consequently, the possibility exists that some of the voter's were underage, citizen's of another country, or even perhaps convicted felons. Some of the voter's were probably not registered voters and given that the polls were open on 3 different Thursday nights some people may have voted twice. We tried to prevent multiple votes but without intrusive ID checks there is no guarantee that nobody voted twice. 
4. Since our sign included the phrase "let's do it right this time", and since the ballots were only semi-secret (folded and we made no attempt to observe people's choices directly but clearly our booth did not conform to the level of secrecy that would be mandated in a real election) I believe Bush voter's are underrepresented (even beyond their underrepresentation in the Twilight Fest crowd to begin with). 
While the "let's do it right this time" refers to the procedure rather than the outcome of the presidential election of 2000, I think some people interpreted it as anti-Bush and may have chosen not to vote on that basis. 

5. Finally, while the IRV results are for the aforementioned reasons not a scientifically representative sample and therefore do not reflect the actual percentages the candidates would have received in Columbia...the results are very instructive on how IRV works. 

RESULTS....TWILIGHT FEST INSTANT RUNOFF VOTING (MOCK ELECTION) 

First Count: 
1. Gore 33 votes 
2. Nader 28 votes 
3. Bush 9 votes 
4. Buchanan 5 votes 
5. Alan Keyes (write-in) 1 vote 
76 votes cast...39 required for a majority. Gore falls 6 votes short of a majority...thus Keyes vote will be dropped and we will look at the Keyes' voter's second choice. The Keyes voter lists Buchanan as his second choice obtaining the following results once Keyes is dropped from the race: 

Second Count: 
1. Gore 33 votes 
2. Nader 28 votes 
3. Bush 9 votes 
4. Buchanan 6 votes 
76 votes cast... 39 still required for a majority. Now again we will drop the last place candidate (Buchanan) and look at the Buchanan voter's second choices (and incidentally since both Keyes and Buchanan have been eliminated...the original Keyes voter's third choice {Bush} will be added in our new tally since it is his first choice of the three remaining candidates and at this point the election is treated strictly as a three-way contest.). 
The 6 Buchanan voters next choices were as follows: 4 selected Bush as their next choice, 1 chose Gore and 1 chose Nader yielding the following results on the third count: 

Third Count: 
1. Gore 34 votes 
2. Nader 29 votes 
3. Bush 13 votes 
76 votes cast...39 required for a majority. Gore is still 5 votes shy of a majority. The 13 Bush voter's now will decide the election as the race narrows to two. 
With Bush (and Buchanan) eliminated, 6 of the 13 voters did not select between Gore and Nader (confronted with the dilemma of selecting between two liberals they chose to leave the column blank.). The remaining 7 chose Nader! 

Fourth Count: 
1. Nader 36 votes 
2. Gore 34 votes 
70 votes cast...36 required for a majority, and the Bush and Buchanan voters swing the Election to Nader !!? Nader wins! 
A couple of comments regarding the results: 
1. Third party strength: As with most IRV elections, alternative candidates run stronger than under the current plurality-winner takes all system. Not only did Nader fare well, but Buchanan was also relatively competitive trailing Bush by less than a 2 to 1 margin (In November's actual election Bush outpolled Buchanan nearly 50 to 1 for comparison). 
2. Gore's unpopularity among Republicans: Note that while Gore had a plurality of votes to begin with (33), he only managed to pick up 1 additional vote from all the Republican voters (Bush, Buchanan and Keyes) put together while Nader was able to ultimately gain 8 votes from the Republican voters. It is important to bear in mind that with IRV a candidate needs broad support to win...and while Gore was able to secure immediate support from 33 voters...it appears many other voters were voting against Gore above all else (i,e- ballots that read Buchanan-Bush-Nader....i.e. - anybody but Gore). 
3. In light of the fact that Republican voter's chose Nader over Gore as a second or third choice on their ballots, I think I should point out that with IRV voting as opposed to the current system it is important to vote FOR someone rather than AGAINST someone. 

Let me explain that more fully: 
Would Bush supporters really prefer Nader to Gore, given Nader's anti-corporate agenda and proposal for a $10 minimum wage? It seems to me that if I were a Republican I would prefer Gore's centrist politics to Nader's progressive activism. My guess is that many Republicans just couldn't stomach choosing Gore over anybody (many Democrats also chose Buchanan ahead of Bush on their ballots for what I imagine to be similar reasons) since he developed a negative image over last falls campaign and the endless recounts etc. 

However with IRV it is very important to select the candidates in the order you would actually prefer them...once you have chosen your first two choices and you are left with the dilemna of choosing between two objectionable third choices...you should select the candidate you would actually prefer to hold office because at this stage a symbolic protest against Gore, for example, translates into electing Nader....in a different context the same could be said that selecting Buchanan over Bush could ultimately end up with President Buchanan. 

Thanks again to everyone who participated in the mock election and thank you to the people who staffed and helped set up the IRV booth (Henry Lane, Amy Damashek, Mark Haim, Claire Garden and myself). Any comments, suggestions, criticisms or questions regarding IRV can be directed to me through e-mail at brekhus@verizon.net 
Keith L Brekhus 

back to campaigns