Favorability feature sneak-peek

The Favorability feature for President Infinity is coming along. Here you can see how Romney’s Favorability scores in 2012 for Utah vs. Washington, D.C. are distributed vs. % of eligible voter.

These numbers were generated from the existing 2012 campaign, using only the percentages data available. Although the algorithm for doing this is still in early stages, they still look generally reasonable at this point. You can also see the overall Favorability rating in the top right.

The electorate is now modeled with ‘persons’ who in turn represent 1/10,000th of the eligible voters. This means the electorate is modeled to 0.01 of a % (1/100th of 1%). This is a huge change from the previous game engine, which only had basically 4 numbers for each candidate (% committed, % leaning, % undecided, and % alienated), and involved simply adding or subtracting numbers from those categories based on momentum.

Each of these ‘persons’ in the new game engine has Favorability ratings for each candidate, and the %s are then generated by adding up each ‘person’s’ first place candidate. Favorability in the game is a bit different from how the word is usually defined in real life. In the game, it means just how much a person prefers a given candidate. So, whichever candidate has a higher Favorability score gets that person’s vote, unless it falls below a certain threshold, in which case they become alienated.

The strength of this approach is in more realistically modeling electoral dynamics, such as sudden collapses or meteoric rises, independents choosing between parties based on the eventual nominees, and so on. It also brings the game engine a big step closer to allowing electoral blocs.

Feedback welcome!

29 thoughts on “Favorability feature sneak-peek

  1. Can’t wait, this will help make my 2018 Ohio Governor and Senate races complete and I can finally update it.

  2. Sounds awesome! Just a little confused on the graphs. So for the DC graph the Romney voters only barely support him? Veruses the Utah graph where a lot of his voters fall into the right hand side of the graph are more likely to vote for him?

  3. So I assume this means that campaigning/ads etc will increase/decrease favorability ratings which will in turn impact vote shares? How does momentum work with this new system?

  4. @Jacob,

    Ya, that graph should be clearer in the release form. Left to right are the categories of support (so, voters where Favorability is 0-10, 10-20, 20-30, and so on). The numbers on top of the bars indicate the % of the eligible voters in that category.

    So yes, the Utah graph shows many more eligible voters on the right, indicating higher levels of Favorability. The D.C. graph shows many more eligible voters on the left, indicating lower levels of Favorability. Obama’s charts would be opposite.

  5. @Matthew,

    The Favorability system plugs into the existing momentum system, but instead of momentum changing raw %s for each candidate, momentum instead changes Favorability scores for each ‘person’ (each ‘person’ represents 1/10,000th of the eligible voters), which in turn change %s.

    It used to be

    Activities, ads, and so on -> Momentum -> %s.

    Now, it is

    Activities, ads, and so on -> Momentum -> Favorability scores -> %s.

  6. Will Two round system be released with this? Also looks great! Yeah I think it could be shown much clearer,I think it’d be cool to have a universal fav too. Taking each states views into account. Having a tick option for “universial favorbility” for example,as shown here,because of the states views to Romneys views,he’d have much higher fav in say Kansas,than Vermont. (But this can just be a tick option,u could also use this version to)

  7. @TMP,

    No, but two-round (like preferences) will be much closer to implementation because of Favorability.

    What do you mean by universal favorability exactly? In the game as it is (I didn’t show this), if you select the country, you see the overall average Favorability number.

  8. Oh okay that’s what I was a bit confused about,it will be easy to see other characters favorbility I hope?

  9. @TMP,

    Yes, a player will be able to select Favorability from the drop-down box on the Main Screen, and view all players’ Favorability scores.

  10. Any ETA next couple weeks or so? Also is primary pop vote on your list of features to add?
    Thanks for the quick responses,A+ dev rating 🙂

  11. Anthony – This looks great. Looks like the new engine will fix basically all the issues people have been having recently.

    Just wondering, how does favorability work in practice for candidates who drop out in the primaries? It seems every simulated voter (1/10000) has an array of favorability ratings for each candidate. Do you think we’ll see more bitter voters making protest votes/voting for the other party/not voting?

  12. @TMP,

    The best guess release is Dec. 5th, but it really depends, because this sort of work is finicky, involving the core of the game engine and then all sorts of things that plug into that.

    Primary popular vote is not planned for the Autumn 2018 release cycle.

  13. @Thr33,

    If a candidate drops out of the primaries, that candidate is removed from a person’s list of possibilities, and their vote then switches to the next highest candidate in their list. It very well could be someone from a party that isn’t the same as their previous candidate’s.

  14. @admin,On your reply to Thr,that is awesome. Wasn’t expecting stuff that in depth,but you’re doing it! Great job. Am excited

  15. So, as I understand this, favorability slides from one candidate to another as the preferred candidate drops out or is otherwise eliminated. Will viability play into that as well? I guess, my question would be what determines how favorable a candidate is? Many times, you hear someone say they agree with or like a certain candidate better than others, but they do not feel that candidate has a chance to win.

    Also, with favorability, if there are candidates A, B, and C. Then candidate A drops out, but his people are split 50/50 between B and C. If he endorses candidate C, will greater than 50% of his people go to candidate C or will they split evenly as was the way it would have occurred before? I guess another way to put that would be do endorsements affect favorability?

  16. If sales permit, is it possible that there will be any discounts on any of the election games/ subscriptions on cyberMonday? Just wondering? 🙂

  17. Discounts are usually available if purchasing one game for other games purchased at the same time. No CyberMonday discounts planned.

  18. @anthony

    Could use the Iowa Caucuses as a means to describe how favorability works? I think that might help aid in understanding it.

  19. @Josh,

    “what determines how favorable a candidate is? Many times, you hear someone say they agree with or like a certain candidate better than others, but they do not feel that candidate has a chance to win.”

    Favorability in the game doesn’t perfectly match favorability as it is typically used for polling and so on. Rather, Favorability is simply the rating a person has towards a candidate, where the highest rated candidate gets that person’s vote (if they vote – they can be alienated where the Favorability rating is below a threshold). This roughly matches favorability as typically used, but as you noted people in the real world might find a candidate favorable but still not vote for them for whatever reasons.

    Initially determining Favorability requires taking the specified startup %s and translating those to Favorability scores. This isn’t simple, and there is no straightforward way to do this (startup %s are ambiguous). However, the startup Favorability scores when then read back at the beginning of a game preserve those %s while trying to model relative platform distances and candidate strengths, and so giving a depth to the electorate it did not have previously.

    “do endorsements affect favorability?”

    Yes. If candidate A withdraws and endorses candidate B, B gets a boost in Favorability for people who supported A.

  20. @Jonathan,

    In the Iowa caucuses, each candidate receives the votes of all the eligible voters where that candidate has the highest Favorability score. So if 8% of eligible voters (who aren’t alienated) rank candidate A as most Favorable, candidate A gets 8% of the vote. This is also modified by GOTV and last-minute momentum, as it is currently, so typically it won’t be exactly 8%.

  21. Will VP candidates have favorability– particularly by state? Wondering if that could factor into the Veepstakes.

  22. @Mackenzie,

    VP candidates currently have possible % bonuses for regions, and so Favorability ratings will be adjusted based on the VP selected.

  23. @Will,

    No, Favorability is automatically generated based on the %s specified. In the future, there might be tools added for campaign designers to customize Favorability ratings further.

  24. This is excellent, Anthony! Thanks for all the effort you’ve obviously put in. Quick question, and apologies in advance if my utter lack of knowledge of programming makes this question sound naive. Is there a chance that an unfavorable rating could be incorporated into this new system?

    I only ask because for me, a favorability rating can lack context without an unfavorable rating. For example, say someone has a 30% favorable rating. That sounds pretty bad if they are well known and there unfavorable rating is 60%. But it’s pretty good if they’re less well known and their unfavorable rating is only 15%.

    Either way, I’m excited to play this.

  25. @Brandon,

    Favorability doesn’t work exactly like it’s usually used in real life. It’s a score given to each candidate by an eligible voter. So, you can consider Favorability the net of favorability – unfavorability.

  26. Looks cool, although I paid for the game with the understanding that this feature would be added. Now my subscription has expired before the feature was ever implemented.

  27. @Jeff,

    I can only give best estimates for when features are to be added. Subscriptions are required to allow me to invest in making substantive updates – otherwise, there would be no major updates to the Infinity games.

Leave a Comment