2012 November Special Election : Analysis

Introduction

At the end of the 2012 November Special Election, I sought out to do a unique analysis using data generated from the Gameplay Alignment Test as well as publicly available voter behavior. Not all voters were willing to take the Gameplay Alignment Test, but the reduction of them from the analysis is not anticipated to establish a significant bias (4 additional voters out of 35). I am -not- a statistician, I am first and foremost a student political scientist simply interested in a rudimentary empirical guide to ideological voter behavior -- thus, take my results with a grain of salt.

Analysis

Voters in the 2012 November Special Election on The Gameplay Alignment Test:


On the onset, one may as well conclude that ideology did not absolutely determine voter behavior by any means -- there were other factors at play. The 2012 November Special Election would have looked vastly different if voters had chosen their "ideologically ideal" candidate. Every voter's ideologically ideal candidate was determined by noting the candidate with the closest difference to the ideological scores (both R/D and R/C) to them.

For example, if we take example myself, Unibot (-18, -13). The difference between my scores and Mcmasterdonia ( -2, 4.5) is -16, -17.5; if these two differences are "neutralized" with ?X² and then added together, the sum is 33.5. This same result for Tim (-8.5, -9.5) is 13, Romanoffia (-9, 3 ) is 25, Blue Wolf (14, 1) is 46 and Kingborough (-3,-4) is 24. The lowest sum (13) reflects my ideologically preferred candidate and nearest candidate to me on the Gameplay Alignment Compass: Tim. Obviously, my less ideal candidates range as the following from most to least: Kingborough, Romanoffia, Mcmasterdonia and Blue Wolf. Largely speaking, I myself am an ideological voter, since I would personally say that this reflects my ordinal ranking of the candidates in what order I would vote for them.

Despite my ideological convictions however, largely speaking voters did not chose their "ideal" ideological candidate. Thus, if one repeats my method for determining the ideological ideal candidate for every voter, they will find that the "predicted" election would be thoroughly unrealistic, as shown:


Obviously, there are some notable differences. For one thing, the winner of this ideological election would be the least preferred candidate in the actual election and the worst performer in this ideological election is the most preferred candidate in the actual election. It's essentially inverted results.

From this we can determine some candidate strengths and weaknesses:

Strengths: Mcmasterdonia made -huge- gains by capturing the ideological centre of The North Pacific away from Kingborough and the centre-right away from Blue Wolf, while also drawing in the prototypical "old guard" of The North Pacific which could have been predicted to go to their Regionalist counterpart, Romanoffia; Tim faired better than expected by mainly protecting the ideological far-left from opponents such as Romanoffia and Kingborough. Overall, Mcmasterdonia managed to unite the centre and the right against the far-left.

Weaknesses: Kingborough was obviously unable to hold his ideological advantage against Mcmasterdonia.

Although there are huge disparities between the "Ideal Vote" and the "Real Vote", do note: 61% of voters did not sacrifice more than 2 ideological points (on the Gameplay Alignment Test) from their ideal candidate's R/D value (this is to say the candidate with the closest R/D value), R/C values were demonstrably less valuable to the 61% of voters who sacrificed more than 2 ideological points difference from their ideal candidate's R/C value (this is to say the candidate with the closest R/C value). Where voters cared the most ideologically (military ideology), Mcmasterdonia and Kingborough were virtually identical at a mere point away, thus this was well within a window of compromise for Kingborough's theoretical base of support who favored Mcmasterdonia's credentials.

Conclusions

I interpret the following data as suggesting that how a candidate aligns, Raider or Defender can play a key role in how a vote turns out, but only on a secondary level. One could predict for example, that neither Tim nor Blue Wolf would have likely won the Election due to both facing a steep disadvantage in trying to ideologically "pinch" the middle where more competitive candidates such as Mcmasterdonia and Kingborough were strategically placed.

However, it is likely that voters are willing to sacrifice minor ideological disparities on the basis of leader perception and experience which I would posit as the downfall of Kingborough. Although Kingborough would theoretically resonate closer with most voters in The North Pacific by closely mirroring the average Gameplay Alignment in The North Pacific (-2.6, -2.6), the moderate defenderist-regionalist, Mcmasterdonia had to rely on his experience as Minister of Defense and others' leader perceptions of him to swoon the very moderately defenderist-cosmopolitan political centre with a political performance that focused on culture and civility policies that mirrored Kingborough's platform, while also maintaining the moderate defenderist-regionalist status quo from Eluvatar's regime on highly controversial issues such as military and WA affairs policy.

The downfall of Blue Wolf is less clear, one possibility is that he simply alienated himself from the centre-right by seeming like a niche choice with no electoral prospect -- so much that his base of support turned to their second ideal candidate, Mcmasterdonia. An alternative possibility is that centre-right voters felt threatened by the idea of Tim, the far-left candidate, being elected as his campaign was legitimized with incumbency, so they voted strategically with the front-runner and their second ideal candidate, Mcmasterdonia. These said possibilities are not mutually-exclusive. However, far-left voters in the defenderist camp did not appear to estimate the threat to their candidate, Tim, or alternatively, saw the emerging electoral unsuccess of Kingborough and Romanoffia and bet their electoral success on Tim -- his support base, solidly in the far-left was built on those in the far-left and from the edge of Kingborough's less moderate voters who were dissatisfied or alienated by the electoral juggernaut, Mcmasterdonia, who had already by that time conquered the centre and the centre-left rather decisively. This was however a strictly good play of demographics with the centre under control, Mcmasterdonia dominated the centre-left and drawed enough from the right to win the Election handily without the support of the far-left -- thus during this Election, the far-left was largely left alienated.

Ultimately, a good candidate under these demographics to remain competitive against Mcmasterdonia in a future election would need to 1) have the appropriate leadership perception, 2) the experience in The North Pacific and 3) be strategically placed in the -2.6, -2.6 corner where the demographics are favorable and Mcmasterdonia is most hindered from capturing the centre-left. In a race where Mcmasterdonia was blocked from the centre-left by such a challenger, the demographics would simply not work in his favor although it would be a pressured tight-finish -- the centre and the right altogether cannot beat out the centre-left and the far-left without a third candidate to split the left. In other words: Hileville for Delegate 2013!

- Unibot

Unibot, interesting analysis, but one thing it does not consider is the personality and leadership of the candidates.

You may be an ideological voter but others will vote based on the effective leadership style of the candidates above and beyond their own ideological preferences.

Also - Kingborough probably did not know he held an ideological advantage amongst the voters and not knowing this he would be unable to proactically take advantage of it. Your analysis may make it easier for him to run in future elections though because people are reading that Kingborough's ideology most closely resembles a majority of TNP'ers.

but - of course - this is all spin doctoring and that's the beauty of politics. If you can frame the debate and the question effectively, no matter what your actual position is, you're likely to win the argument. But, I'd like to see King run again and run on the platform that his ideology matches most TNP voters.

I mentioned "personality and leadership of the candidates" quite a bit.

However, it is likely that voters are willing to sacrifice minor ideological disparities on the basis of leader perception and experience which I would posit as the downfall of Kingborough. Although Kingborough would theoretically resonate closer with most voters in The North Pacific by closely mirroring the average Gameplay Alignment in The North Pacific (-2.6, -2.6), the moderate defenderist-regionalist, Mcmasterdonia had to rely on his experience as Minister of Defense and others' leader perceptions of him to swoon the very moderately defenderist-cosmopolitan political centre with a political performance that focused on culture and civility policies that mirrored Kingborough's platform, while also maintaining the moderate defenderist-regionalist status quo from Eluvatar's regime on highly controversial issues such as military and WA affairs policy.

Click to expand...

For King to have another go at it, he'd need be able to pump up his leadership credentials but also poke holes in Mcmasterdonia's current base support using ideological differences -- namely issues that divide King from Mcmasterdonia on R/C issues. The most well discussed R/C issue in TNP is WA Policy -- where Mcmasterdonia sides with the idea that players should be given unnecessarily disclosed information to use as dirt. Opposing Mcm on this issue would woo the far-left, but you've got to have a more fresh R/C debate to divide the centre-left from Mcmasterdonia and isolate him into the Regionalist pocket as he is in the "Ideal Votes" -- but these debates seldom emerge and by the time elections occur, what was a pressing and divisive issue can often become a tired and "NOT THIS AGAIN" issue.

Interesting analysis. I can't say I agree with all of it, nor do I believe that it can all be measured on the gameplay alignment test. I think that fundamentally it comes down to issues of a) experience/leadership credentials b) time in TNP and c) trust. Not to mention that as I understand it, peoples results on the gameplay alignment test can vary significantly from day to day, and mood to mood.

In TNP I think it's likely for people to vote for individuals they trust to be their leader, this won't necessarily mean agreeing with them or even identifying with them ideologically. Trust is an important factor - i.e. respecting the democratic processes of the region, and always acting with the best interests of TNPers in mind and not selfish or personal pursuits. Agreeing with these decisions, or supporting them ideologically isn't always necessary.

Likewise, people who voted for Kingborough or Blue Wolf or Romanoffia, may have done so knowing that there candidate would not change the course of the election. Voted this way in order to recognize the work they've put into their campaign, or whatever, And may have been willing to support a Tim or Mcmasterdonia Delegacy as a second preference. Preferential voting, would make elections even more interesting.

Not to mention that Tim and I, as councillors agreed on nearly every single matter that came before the chamber, and had a good working relationship. The one difference of course, was the WA Voting matter, and even then, the council came together in the spirit of compromise to reach the policy we have now. I'm not convinced that politics, and voting, is as divisive as this analysis seems to suggest.

Nonetheless, thank you for putting the effort into doing this. It was an interesting read.

By "it" I meant the actual figures of your analysis do not take into consideration leadership, it only looks at ideology. And so, as an analytical tool used to forecast elections I think it will not prove useful. Even if you are simply trying to make the argument that people vote contrary to ideology, an effective leader with the same ideology as the masses could win in the future and your analysis would show that the electorate voted for the most ideologically aligned candidate. But, there wouldn't be much to infer from the analysis.

That's what I was trying to say.

I don't feel qualified to give an in depth analysis of your analysis, but I found it really frigging interesting.

First, thank you very much for this coherent and intriguing analysis. It is an avenue of inquiry worth deliberating.

unibot:

For example, if we take example myself, Unibot (-18, -13). The difference between my scores and Mcmasterdonia ( -2, 4.5) is -16, -17.5; if these two differences are "neutralized" with ?X² and then added together, the sum is 33.5. This same result for Tim (-8.5, -9.5) is 13, Romanoffia (-9, 3 ) is 25, Blue Wolf (14, 1) is 46 and Kingborough (-3,-4) is 24. The lowest sum (13) reflects my ideologically preferred candidate and nearest candidate to me on the Gameplay Alignment Compass: Tim. Obviously, my less ideal candidates range as the following from most to least: Kingborough, Romanoffia, Mcmasterdonia and Blue Wolf. Largely speaking, I myself am an ideological voter, since I would personally say that this reflects my ordinal ranking of the candidates in what order I would vote for them.

Click to expand...

You've made, I think, an error in mathematics here. If you're modeling our ideology with a two dimensional coordinate system, the distance between two people's ideology is not the sum of the absolute value of the differences in the two dimensions, but a two dimensional measure of distance between the two coordinates. So rather than taking the distance (-16,-17.5) and finding a scalar distance of |-16| + |-17.5| = 16 + 17.5 = 33.5, you ought perhaps use the standard distance formula of ?(?x²+?y²) which would be ?(^-16² + ^-17.5²) ? 23.7. Similarly, distance((-18,-13),(-8.5,-9.5)) ? 10.1, distance((-18,-13),(-9,3)) ? 18.4, distance((-18,-13),(14,1)) ? 34.9, and distance((-18,-13),(-3,-4)) ? 17.5. The difference between these distances and the ones in your analysis is significant: by adding the distance in two dimensions together arithmetically, you exaggerate combined distance, and this particularly increases the perceived distance for positions which differ in both dimensions. I do not know, however, what effect this would have on your overall conclusions as I do not know by what method you used to find the voter choices you argue the model would predict.

This is an interesting thread. I am a big fan of data-driven analysis and inference, which in part reflects to my RL training and occupation (I have a master's in statistics, and do research in machine learning).

Unibot, first a comment on methodology. The "difference" between two people's scores is essentially a distance metric between the two-dimensional vectors representing the scores, say (u,v) and (x,y). The distance you choose to use is what is called the l¹-norm: d = |u-x| + |v-y|, where |.| is the absolute value. I would recommend that you use instead the l²-norm, or Euclidean distance: d = sqrt( (u-x)^2 + (v-y)^2 ). I will not go into details about why this is better, but suffice it to say that it is by far the standard distance metric (whenever people use "variance" or "standard deviation", they are implicitly selecting to use the Euclidean distance). This choice would slightly change the order of your preferred candidates to (in descending preference): Tim, Romanoffia, Kingborough, Mcmasterdonia and Blue Wolf. I doubt it would make any significant difference for the rest of your analysis.

Second, some thoughts. Your analysis assumes that voters weight differences in the two ideological axes equally, by virtue of the distance metric you chose to use. It is reasonable to doubt this assumption. To offer two plausible counter-hypotheses: A citizen who is also, and primarily, a member of a major raider/defender organization is unlikely to vote for a candidate who is in the opposite side in terms or R/D, regardless of where they stand on the regionalism/cosmopolitanism axis. A long-term citizen who is detached from R/D is likely to vote for a similarly long-term "regionalist" candidate. This raises two interesting, I think, questions:

• Does your data provide evidence for any strong such trend among voters; i.e., do citizens strongly favor proximity on one axis over the other? One way to go about investigating this would be to examine the standard deviations between voters and their selected candidates separately in the two coordinates; a large difference between the two values would indicate that there is such a trend.
• Is there a weighted distance metric that can be used to explain voter behavior? Or, to pose the question in another way, what is the best weighted distance metric in terms of predicting voter behavior, and how good is it? It may turn out, for instance, that even this best metric is fairly bad at making predictions; this could in turn be interpreted either as further supporting evidence for the conclusion you drew in your analysis, or as an indication that the Gameplay Alignment Test is a poor predictor of ideology. Investigating this problem is harder, as it requires using a class of computational algorithms dealing with the problem known as metric learning. But, you can glimpse some information regarding weights describing the relative importance of the two axes using simpler methods, such as linear regression.

Another interesting, in my opinion again, issue has to do with how the amount to which ideology affects voting changes with the location of the voter on The Gameplay Alignment Test graph. The following is, I believe, a reasonable hypothesis: citizens that are near the boundaries of the graph, that is they are towards the extreme in at least one axis, factor ideological alignment more heavily into how they vote, compared to citizens near the center. This hypothesis is also not unrelated to your own conclusions. It would be interesting to see if your data provides any quantitative evidence for such a hypothesis. This can be investigated by separating voters in two categories, near the boundary and near the center, and comparing the variance in their votes.

EDIT: There have been a few new posts while I was typing this one. Elu's raised the same point as I do in my second paragraph.

Props to all our resident wizards.

I gather from the results that ideology is a less important factor than experience and trust. It will be interesting to see how the analysis plays out in future elections.

Great Bights Mum:

Props to all our resident wizards.

I gather from the results that ideology is a less important factor than experience and trust. It will be interesting to see how the analysis plays out in future elections.

Click to expand...

If unibot continues to do these analyses in del elections if I was someone who was helped by it, I'd use it as a propaganda tool.

I would find it useful if Unibot could highlight the candidates themselves on the graph. Whilst I'm interested in statistics I get lost in advanced maths, and a visual reference point would be convenient.

Unibot, first a comment on methodology. The "difference" between two people's scores is essentially a distance metric between the two-dimensional vectors representing the scores, say (u,v) and (x,y). The distance you choose to use is what is called the l1-norm: d = |u-x| + |v-y|, where |.| is the absolute value. I would recommend that you use instead the l2-norm, or Euclidean distance: d = sqrt( (u-x)^2 + (v-y)^2 ). I will not go into details about why this is better, but suffice it to say that it is by far the standard distance metric (whenever people use "variance" or "standard deviation", they are implicitly selecting to use the Euclidean distance). This choice would slightly change the order of your preferred candidates to (in descending preference): Tim, Romanoffia, Kingborough, Mcmasterdonia and Blue Wolf. I doubt it would make any significant difference for the rest of your analysis.

Click to expand...

My error must be more basic than that, I think. I was under the impression when I designed my excel tables that ?X² +Y² = ?X² + ?Y², obviously it does not work like this.

My formula was

D = sqrt((u-v)²) + sqrt((x-y)²)

I'll just recalculate the data, then. No sense in having junk facts.

A weighted version of the Test would be interesting because it seems like about 61% of voters did move more than 2 ideological points from their ideal candidate's R/D value -- so they weren't willing to sacrifice very much.

Another interesting, in my opinion again, issue has to do with how the amount to which ideology affects voting changes with the location of the voter on The Gameplay Alignment Test graph. The following is, I believe, a reasonable hypothesis: citizens that are near the boundaries of the graph, that is they are towards the extreme in at least one axis, factor ideological alignment more heavily into how they vote, compared to citizens near the center. This hypothesis is also not unrelated to your own conclusions. It would be interesting to see if your data provides any quantitative evidence for such a hypothesis. This can be investigated by separating voters in two categories, near the boundary and near the center, and comparing the variance in their votes.

Click to expand...

I would like to posit a different hypothesis, that far-ideological voters are just as willing to compromise and in fact did so, far-right voters were often likely to choose Mcmasterdonia over Blue Wolf.

You could argue this is because the far-left is less likely to compromise R/D values.

On the contrary, I think this suggests that this is simply a reflection of the voter options instead of some inclination-to-compromise among centrist voters; there was only one solidly "raiderist" candidate, two centrist candidates (one slightly leaning more defender) and two far-left candidates. Centrist or Left-Centrist voters had two options that were quite similar ideologically, enough so that they could vote according to their beliefs within a window of acceptability while also judging leadership perception, I think the far-left on the other hand felt that backing their ideal candidate would have been more successful than the left-centrist candidate (King) who was diminishing quickly out of the gun, bearing in mind that Mcmasterdonia was even picking up the right-of-centre and Tim had the advantage of being an incumbent. Alternatively, you can say that Mcmasterdonia and Tim "pinched" King in the left-centre, with Mcmasterdonia mostly being the victor in that battleground demographic, but to assume that centrists are anymore compromisers than the far-left would be illusory because they had more substantial choices within close ideological proximity to begin with! The fact that the far-right looked at the ideal candidate and went nuh-uh and voted for their second-ideal candidate (in a lot of cases), Mcmasterdonia, suggests to me that far-ideological voters will compromise quite a bit ideologically if there is a perceived electoral threat, a lack of credibility regarding their ideal candidate or a simple lack of choice.

I suspect overall, right-wing (including the centre-right and large parts of the far-right) voters in The North Pacific are often forced due to candidate options, to choose candidates further away from them ideologically than other voters are. Under this hypothesis, if Romanoffia and Tim hadn't been in the vote to separate the far-left (I suspect even removing Tim would have done this due to credibility and strategic voting which hurt Blue Wolf just as much as it probably would have hurt Romanoffia), voters would have migrated to Kingborough; the vote probably would have still have been a blowout Mcmasterdonia victory, but it would have been a bit more of a fight between Kingborough and Mcmasterdonia.

I think that fundamentally it comes down to issues of a) experience/leadership credentials b) time in TNP and c) trust. Not to mention that as I understand it, peoples results on the gameplay alignment test can vary significantly from day to day, and mood to mood.

In TNP I think it's likely for people to vote for individuals they trust to be their leader, this won't necessarily mean agreeing with them or even identifying with them ideologically. Trust is an important factor - i.e. respecting the democratic processes of the region, and always acting with the best interests of TNPers in mind and not selfish or personal pursuits. Agreeing with these decisions, or supporting them ideologically isn't always necessary.

Likewise, people who voted for Kingborough or Blue Wolf or Romanoffia, may have done so knowing that there candidate would not change the course of the election. Voted this way in order to recognize the work they've put into their campaign, or whatever, And may have been willing to support a Tim or Mcmasterdonia Delegacy as a second preference. Preferential voting, would make elections even more interesting.

Not to mention that Tim and I, as councillors agreed on nearly every single matter that came before the chamber, and had a good working relationship. The one difference of course, was the WA Voting matter, and even then, the council came together in the spirit of compromise to reach the policy we have now. I'm not convinced that politics, and voting, is as divisive as this analysis seems to suggest.

Click to expand...

I think as a candidate you are perhaps a little too overenthusiastic about how much your qualities had to determining your success; your leader perception certainly helped you greatly in challenging Kingborough but it's not so clear that gave you an edge against Tim (who had a lot of experience leading other regions).

But noting that you, Tim and Kingborough ran on virtually the same platform which focused on civility, culture and diplomacy, with only Tim mentioning military policy (favoring the status quo) which was identical to your proposed military policy -- I think knowledge regarding Tim's ideological defender-bent and your reputation as a near-centrist, played a big part to the far-left being alienated from you and the far-right compromising with you as their candidate.

punk d:

By "it" I meant the actual figures of your analysis do not take into consideration leadership, it only looks at ideology. And so, as an analytical tool used to forecast elections I think it will not prove useful. Even if you are simply trying to make the argument that people vote contrary to ideology, an effective leader with the same ideology as the masses could win in the future and your analysis would show that the electorate voted for the most ideologically aligned candidate. But, there wouldn't be much to infer from the analysis.

That's what I was trying to say.

Click to expand...

If I were to predict an Election, it would be by making an electoral ideological map, then giving seats to the more experienced TNPer within a certain spread of probability. In cases where candidates are ideologically very similar and both of an equal level of experience and leadership qualities, I'd just say "too close to call", but if you put a gun to my head, I'd say the one with the closer score to the sweet-spot (-2.6,-2.6).

In an election just between Eluvatar and Mcmasterdonia (without the recent history of him going inactive looming over Eluvutar), I think Eluvutar would win it about three-quarters of the time. Eluvatar would benefit from not being as vulnerable to Mcmasterdonia in the experience-category as King was, but lean more defender and not scare away voters as much with his regionalist credentials, enough so at least to make ground in the centre-left where Mcmasterdonia slaughtered his competition. One should note, the two (who used to be running mates) share similar ideological scores. Elu at -5,2 and Mcmasterdonia at -2,4.5.

The two divisive issues, WA Policy and Military Policy would be deciding factors I think. If just WA policy was brought up in this hypothetical election, Mcmasterdonia would narrowly lose, if Military Policy was brought up, Mcmasterdonia would narrowly win more often (due to his military position being more centrist) and if both issues were brought up, Eluvatar would scrub a win (in a tight show-down nonetheless), three-quarters of the time.

The rarity, one could conclude about The North Pacific voter demographics is the regionalist-defender minority of The North Pacific, which I would posit as traces of an earlier age where this minority was more popular. Regionalist-defenders are increasingly rare in NationStates, probably due to a lack of distinct defender regions.

Another thing is: we have a lot of candidates running in an election. Some of the more predictable and rudimentary votes are done with two or three candidates.

The classic example came in The Rejected Realms 2011 with Sedgistan v. Guy v. Durkadurkiranstan. Sedge was the left-of-centre candidate pledging full status quo in regards to military policy (we acknowledge the existence of the RRA, that's it) and was having to deal with the Devonitians scandal, Durkadurkiranstan was the right-wing candidate pledging an invader army in The Rejected Realms and finally, Guy made a serious electoral miscalculation and pledged to maintain both the RRA and a new invader (tagger) army -- this put Guy in the tight centre. The conclusion was predictable, centre and centre-right went to Guy, right went to Durkadurkiranstan and Sedgistan walked away with the left and the centre-left in a handbasket. Essentially Guy was pinched in between Sedgistan and Durkadurkiranstan with not enough moderate-right or centre voters to save him. Everyone who feared the idea of The Rejected Realms having an invader army ran out to vote for Sedgistan, so he united almost everyone on the left, which was a significant majority.

This makes me want to enable automatic double post merging

unibot:

The downfall of Blue Wolf is less clear, one possibility is that he simply alienated himself from the centre-right by seeming like a niche choice with no electoral prospect -- so much that his base of support turned to their second ideal candidate, Mcmasterdonia.

Click to expand...

I think that's code for me waiting to see if McMaster was winning and, seeing that he was, put no serious effort forth to win.

I think that fundamentally it comes down to issues of a) experience/leadership credentials b) time in TNP and c) trust.

Click to expand...

McMasterdonia is right. I also strongly feel that McMasterdonia met all three of these criteria and deserved to win. A long time before the election I decided to vote for Romanoffia because I perceived him (it's only an impression), having even more time in TNP, more experience and more of a cynical idealist, better placed to deal with plots against the state. Both candidates set out their policies well, Blue Wolf II didn't. These reasons are not measured on the gameplay alignment test.

Unibot your theories are still really interesting. Don't be put off by the complexity, get a clear definition of a metric and you'll find it really useful.

Chasmanthe:

I think that fundamentally it comes down to issues of a) experience/leadership credentials b) time in TNP and c) trust.

Click to expand...

McMasterdonia is right. I also strongly feel that McMasterdonia met all three of these criteria and deserved to win. A long time before the election I decided to vote for Romanoffia because I perceived him (it's only an impression), having even more time in TNP, more experience and more of a cynical idealist, better placed to deal with plots against the state. Both candidates set out their policies well, Blue Wolf II didn't. These reasons are not measured on the gameplay alignment test.

Unibot your theories are still really interesting. Don't be put off by the complexity, get a clear definition of a metric and you'll find it really useful.

Click to expand...

Your vote was a major anomaly though, the two voters for Romanoffia made little to know sense (he picked up two Cosmopolitan raiders). My explanation is: Romanoffia's actual ideological score has nothing to do with what he preaches depending upon the time of day -- at least there's enough inconsistency on his part to confuse me, my test, you, every other voter in The North Pacific and the whole of Gatesville. Amen.

I think that's code for me waiting to see if McMaster was winning and, seeing that he was, put no serious effort forth to win

Click to expand...

I was being charitable.

I voted for Romanoffia because my personal ideals from a game perspective clash with what I believe is best for TNP at this particular moment.

I believed that a more regionalist approach was the best way and I certainly trusted Romanoffia to do just that even though, if I had my druthers, I'd prefer a much more cosmopolitan and raider friendly candidate. But I'm not exactly convinced that that is what TNP needs.

I think BW is much more like my own ideology (not sure where he falls on the cosmo scale) but honestly, I believed a more regionalist approach was needed. That is the cause for my vote and divergence with my ideology score.

unibot:

(he picked up two Cosmopolitan raiders)

Click to expand...

As I posted on another forum, my score has moved from (7, -2) to (6, 4) which takes me out of cosmopolitan territory for the time being.

punk d:

I voted for Romanoffia because my personal ideals from a game perspective clash with what I believe is best for TNP at this particular moment.

I believed that a more regionalist approach was the best way and I certainly trusted Romanoffia to do just that even though, if I had my druthers, I'd prefer a much more cosmopolitan and raider friendly candidate. But I'm not exactly convinced that that is what TNP needs.

I think BW is much more like my own ideology (not sure where he falls on the cosmo scale) but honestly, I believed a more regionalist approach was needed. That is the cause for my vote and divergence with my ideology score.

Click to expand...

BW scored "14,1".

Interestingly, McMasterdonia is recorded as 4.5 regionalist, compared to Romanoffia's 3 regionalist. Thus McMasterdonia is the one with the more coherently regionalist principle system according to the Gameplay Alignment Test and I think you can see this in practice. Romanoffia flipped flopped quite a bit on the WA Policy, whereas McMasterdonia fought hard to try to get the status quo to be tougher on citizens and then begrudgingly accepted compromise after compromise which undermined his initial "victory".

R/D-wise you wouldn't have gained or lost anything between McMasterdonia or Blue Wolf, they're both "8" ideological points between you on the R/D scale. Whereas you chose the candidate least representative of your R/D views to compliment your vision of TNP as lead by a regionalist. Just as a note.

Unibot:

Interestingly, McMasterdonia is recorded as 4.5 regionalist, compared to Romanoffia's 3 regionalist. Thus McMasterdonia is the one with the more coherently regionalist principle system according to the Gameplay Alignment Test and I think you can see this in practice. Romanoffia flipped flopped quite a bit on the WA Policy, whereas McMasterdonia fought hard to try to get the status quo to be tougher on citizens and then begrudgingly accepted compromise after compromise which undermined his initial "victory".

Click to expand...

Wow. I love how you jump on any attempt to criticize me. There was no initial victory, if you recall, I showed up half way through the meeting, and it was Eluvatar that proposed the policy change. I did not initiate it, but I supported the decision to change the policy. This was despite you claiming it was all a mcmasterdonian vendetta, shortly before I had even woken up for the meeting. Not only that, I intitiated the compromise discussions with Cormac, and together we put the idea to Eluvatar. I did not do it begrudgingly at all. Generally, it was out of an interest, in stopping people from being upset and hurt by our decision, and to move on from it. The compromise was not good enough for you, but unfortunately it's impossible to keep everyone happy. You were not even there for the discussions, so for you to say, that the compromise was reached begrudgingly, or even to attempt to say that you understand how these discussions took place is misleading.

Unibot:

I think as a candidate you are perhaps a little too overenthusiastic about how much your qualities had to determining your success; your leader perception certainly helped you greatly in challenging Kingborough but it's not so clear that gave you an edge against Tim (who had a lot of experience leading other regions).

Click to expand...

Perhaps as you opposed my candidacy and generally everything I do, you are blinded by that, and ignore what others may actually see in me as a candidate for Delegate. Tim had experience in other regions yes, I think where to my benefit, was that I had been in the region longer, and had more experience here. Until you are able to at least recognize what other voters might see, your negative campaigns will continue to fail, and I think sadly for Tim, this effected his final results. I think he would have faired better, had it not been for blatant campaigning and aggressive electioneering conducted, by yourself.

I would also say that a clear part in my election, was how I handled myself in the campaign. Particularly against aggressive posts, and repetitive questioners when I gave an answer they did not like. As well as the fact that Eluvatar and I, had worked closely for all of his term of office, as his Vice Delegate and a Minister. Not only my role in ensuring the security of the NP and my work as a minister, It could be argued, that while people reluctantly recalled Eluvatar, they did not necessarily wish for the entire government to be flipped on its head, only improved. So I was the logical choice to flow on from that.

Lastly, while I find the gameplay alignment test interesting, as I have said, peoples results vary a lot depending on the time that they take the test and their mood at the time. I retook this test on your request, to see how my results had changed from the initial time I took the test. Not only that, some of the questions, are far too hard to answer, as it depends on which context they are applied.

Again, thank you for the analysis. It's quite biased, I think, but that's to be expected.

unibot:

R/D-wise you wouldn't have gained or lost anything between McMasterdonia or Blue Wolf, they're both "8" ideological points between you on the R/D scale. Whereas you chose the candidate least representative of your R/D views to compliment your vision of TNP as lead by a regionalist. Just as a note.

Click to expand...

Are you trying to handcuff me to McMasterdonia, Unibot? It seems you've been making quite a lot of comparisons between us in this thread.

Interesting analysis, Uni.

Bear with me on the back story of what I will later conclude:

Back in the 1930's a group of German behavioral scientists, psychologists, social engineers and mathematicians came up with a field of statistical analysis to not only analyze after the fact, but to predict mass behavioral patterns of humans. This pseudo-science was dubbed 'Cybernetics'. On the surface, they system appeared to have validity, but it failed to take into account (or rather quantify and qualify) the random nature of human behavior in terms of mass-behavior of humans.

At any rate, Josef Goebbels in his various tomes and articles on propaganda and the manipulation of mass-behavior (Die Zeit ohne Beispiel in particular) concluded that people think in either rational/logical/utilitarian terms or irrational/emotional terms, but rarely both at the same time. Goebbels in a not so round about way concluded that if a political leader could appeal to both modes of thinking simultaneously, then that leader will always win.

On a second element, pier pressure or pier behavior is the main motivation for mass behavior - most people vote the way that they thing their piers will vote - and a concentration on making people think that a majority of people will vote affects who people will vote for. we see this in political polls - polls in real life have more of an effect of driving mass behavior rather than reporting on likely mass behavior. Goebbels was a reprehensible little bastard, but his analyses on mass political behavior was spot on.

Reducing it to NationState, TNP in particular, you can look back at past elections, see who voted for who, and who won those elections to determine who will win a given election once certain particular people have cast their votes (if those votes were cast openly).

This is largely reflected in legislation passed by the RA. Most people appear to vote the way certain particular other people vote and never actually read the legislation or take the time to understand any given item of legislation and its consequences, positive or negative. Oddly enough, the particular people whom everyone else seems to follow in voting behavior aren't exactly who you think they might be.

But, by looking at who is voting for whom, long before even a majority of votes have been cast, you can generally tell who is going to win a TNP election once particular people cast their votes openly.

I have concluded that if we held elections via an "Australian Ballot" (totally anonymous voting) we'd get substantially different results each and every time rather than the predictable results that occur when people know who is voting for who.

In a nutshell, who wins an election (in NS or RL) is not really based upon the qualifications of the candidate or the candidate's proposed program, or their track record in implementing a proposed program, but rather upon who seems more 'sympathetic' to the individual voter's emotional needs. Logic, reason and rationality never determined an election but emotional appeal always does. ;D

We currently have "australian ballot" in our Executive Council elections.