13

Dear kind sirs and madams,

I have found participating in your tomfoolery once again, as in the past, to be quite invigorating! I cannot properly convey my appreciation using mere words, but may it suffice to say that this has made my birthday quite merry indeed!

Sadly, I find, however this year's buffoonery to be, shall we say, onerous.

Nevertheless, I shall persist, and with luck will win the internet!

Please provide herein those trade secrets and discoveries that may make this task more successful. One strategy per answer, please. With which we will convey the will of this site concerning its egg, and thus win the internet more than those numerous, but otherwise unworthy, stackoverflowians.

closed as off-topic by nicael, Werner, user259867, James, Martijn Pieters Apr 7 '15 at 13:29

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "The problem described here can no longer be reproduced. Changes to the system or to the circumstances affecting the asker have rendered it obsolete. If you encounter a similar problem, please post a new question." – nicael, Werner, Community, James, Martijn Pieters
If this question can be reworded to fit the rules in the help center, please edit the question.

  • All the stackeggs are on the same timer. Interesting. – Adam Davis Mar 31 '15 at 12:57
  • You can participate in as many stackagotchi's as your computer can handle. – Adam Davis Mar 31 '15 at 12:58
  • my strategy is the best one! Play on SharePoint overflow, when I am alone and get to make all the decisions :P – SPArchaeologist Mar 31 '15 at 13:06
  • 7
    Sadly, it's really easy to restart a site, requiring only half the players to get rid of all progress and ruin the game for the other players. And it appears there are a lot of people who are playing that way... – Adam Davis Mar 31 '15 at 13:08
  • 2
    This question on Arqade may help. – Powerlord Mar 31 '15 at 18:34
  • With the release of Stackagotchi's code, I have updated my answer to be exact. Unfortunately, it's really not all that useful anymore... – The Guy with The Hat Apr 2 '15 at 20:56
9

Private Beta

New Stats

  • Questions: Decreases over time; 0.3 every day. Ranges between -1.5 and 4.5. Can be increased by Ask.
  • Answers: Decreases over time; 0.3 every day. Ranges between -1.5 and 4.5. Decreases faster if Questions is low. Can be increased by Answer.

New Actions

  • Ask: Increases Questions by 1.
  • Answer: Increases Answers by 1.
  • Nothing: Does nothing. Ask or Answer are always better options than Nothing, because they at the very least will negate any decrease in their stat that might have happened.
  • Restart: Enters a new day during which the only options are Restart and Continue. If Restart is selected a second time, the site will restart out of Area 51; otherwise, it continues on as normal.

Optimal Strategy

The best strategy is to Ask on the fist day, Answer on the second day, and repeat. You should never do Nothing.


Public Beta

When entering public beta, Questions and Answers are decreased to 1.

New Stats

  • Users: Decreases over time; 0.3 every day. Ranges between -1.5 and 4.5. If Traffic is at 2, only decreases by 0.2. If Traffic is at 3, does not decrease or increase. If Traffic is at 4, increases by 0.2. Can be increased by Upvote, and decreased by Downvote.
  • Quality: Decreases over time; 0.3 every day. Ranges between -1.5 and 4.5. If traffic is at 3, decreases by 0.7. If traffic is at 4, decreases by 0.9. Can be increased by Downvote.

New Actions

  • Upvote: Increases Users by 1.
  • Downvote: Increases Quality by 1, decreases Users by 0.5.
  • Flag: Sets all stats to 1. Uses up 1 Flag. 1 Flag is obtained at the start of Public Beta and 1 at Site Graduation.

Optimal Strategy

If a stat is critical (red, with a warning), always do the usual action to increase it. After that, keep Questions and Answers full, but put priority on keeping Questions at or above 2. If they are both full, then Downvote. If Quality is full, Upvote. If all stats are full, either Ask, Answer, or Upvote; do not do Nothing.


Graduation

When graduating, all current stats are decreased to 1.

New Stats

  • Traffic: Increases or decreases by a certain amount (less than 1) depending on what Quality and Answers are at:

                          Answers
            |  -1  |   0  |   1  |   2  |   3  |   4  
      ——————+——————+——————+——————+——————+——————+——————
        -1  | -0.8 | -0.8 | -0.8 | -0.5 | -0.4 | -0.2 
    Q ——————+——————+——————+——————+——————+——————+——————
    u    0  | -0.8 | -0.8 | -0.8 | -0.5 | -0.4 | -0.2 
    a ——————+——————+——————+——————+——————+——————+——————
    l    1  | -0.8 | -0.8 | -0.8 | -0.5 | -0.4 | -0.2 
    i ——————+——————+——————+——————+——————+——————+——————
    t    2  | -0.5 | -0.5 | -0.5 | -0.2 | -0.1 |  0.1 
    y ——————+——————+——————+——————+——————+——————+——————
         3  | -0.4 | -0.4 | -0.4 | -0.1 |  0.0 |  0.2 
      ——————+——————+——————+——————+——————+——————+——————
         4  | -0.2 | -0.2 | -0.2 |  0.1 |  0.2 |  0.4 
    

    -1 means critical, displayed as red

    Ranges between -0.49 and 4.5.

New Actions

  • Close: Increases Quality by 2, and decreases Questions by 2.

Optimal Strategy

If a stat is critical, always do the usual action to increase it. After that, keep Questions and Answers full, but put priority on keeping Questions at or above 3. If they are both full, then Downvote. If Quality is full, but either Traffic is not full or Users is not at or above 3, Upvote. If all stats are full, either Ask, Answer, or Upvote; do not do Nothing.

  • 1
    I was playing alone and it worked for me. Went to graduation much faster using your strategy. – Zanon Mar 31 '15 at 23:01
12

You could think of this as a juggling game. Each of your stats decays at a rate much slower than 1 heart per turn (but you have multiple stats). The rate of decay of a stat can be influenced by the game state. The resource you are given is 1 action per turn from a limited pool of actions. Each action awards a fixed 1 or 2 hearts, which cannot be increased. Since your resource income is fixed, it stands to reason that a good strategy is to promote game states that minimize the total decay rate of your stats.

A low question stat appears to increase the decay rate of the answer stat, which represents an added value of the question stat.

A low quality stat does not appear to significantly affect the decay rates of stats other than traffic. In addition, the only actions that can increase the quality stat (downvote, close) have negative effects on other stats.

One approach to decreasing the total decay rate of your stats initially is to leave one or more stats at 0. The idea is that a stat at 0 can't decay any further—though this depends on some assumptions about the engine. Leaving any stat at 0 for too long can cause the engine to complain about impending doom, after a few turns of which you are forced to use a flag, so you'll want to pay attention to messages and deal with warnings by increasing the appropriate stats. From what I've seen so far, you can stay in this warning state for at least two turns without suffering a consequence.

Since a low quality stat seems relatively harmless early on and increasing it is costly, my approach has been to leave quality at 0 and work on building up questions as my first priority. This seems to have the best "bang for your buck" as it ostensibly reduces the decay rate of the answer stat (which must be "fed" every now and then) and doesn't seem to be negatively impacted by any other stat, or any action other than closing.

Other observations:

  • The question and answer stats are a good place to build up some inertia. Users and quality are less stable because of their interaction with downvotes and upvotes (apparently a tradeoff with net benefit in one or the other direction).
  • I'm not convinced of the value of the close action (lose 2 question, gain 2 quality). Perhaps it has some value if your question score is very low; since my strategy has been to prioritize question score, I haven't tested this.
  • It may or may not be useful to add to a stat during a round when all your stats are full. You definitely still pay the cost of voting or closing, though, so given that you may not receive the benefit of an action when a stat is full, I tend to favor questions or answers when all stats are full.
  • I have never observed a stat not to increase when taking an action that increases that stat. In other words, if question score would decay in the next turn, choosing the question action seems to have the added bonus of negating that decay. This points to a potentially effective timing strategy. Depending on how the stats are represented behind the scenes and whether decay has a probabilistic component, it could be easy or difficult to take advantage; generally speaking, I tend to avoid performing the same action twice in a row wherever convenient.
  • When votes are tied, the winner is not chosen based on who acted first. Most likely a random tiebreaker.
  • Doing nothing doesn't seem to have a benefit.
  • Near the end of the game, users will start to take care of themselves, provided you keep questions, answers and quality up (and thus, traffic).

Note that none of this is based on direct analysis of game data, just on my observations during a play-through on Engineering, where I had the benefit of being solo for most of the turns.

  • Ties seem to be chosen by who acted last, even clicking after having voted on your choice seems to make you the last to act (I've noticed this often but could be random and I've been lucky). – Jonathan Drapeau Mar 31 '15 at 18:50
  • @JonathanDrapeau: ties are definitely broken at random. – Marti Mar 31 '15 at 19:58
  • I had three rounds during public beta where all stats were full; twice I picked Q and lost a single A, and the third time I picked do nothing and lost both a Q and an A. So my anecdotal evidence suggests you do want to pick something. – mmyers Mar 31 '15 at 20:23
  • I found closing to be helpful when the questions stat is full, but the quality stat is one heart away from running out. – Kevin Apr 1 '15 at 0:06
6

Behold my most successful strategy yet: I downvote every single time, as quickly as possible upon start of a new day.

This may surprise those of you having personal knowledge of my voting habits, but bear with me whilst I explain the beautiful logic contained in this exercise:

By selecting one stat and voting early and consistently, other players quickly learn that someone will always vote on that task, leaving them to choose more freely what might need more specific tending. It needn't be downvoting, indeed, I'd suggest each player choose according to his desires after viewing the voting rounds for some days.

This strategy of a necessity will require people who change their vote each round, but I've come to the conclusion that many people come and go, and generally participate in such short time periods as to merely chase those lacking stats. Thus this position is seemingly always filled.

This strategy, therefore, must needs be fulfilled by those with a degree of time and freedom to spend that time however they may.

Notably, near the end of the game after graduation, the traffic did not increase until quality was at minimum three hearts. Thus we perceive that downvoting is consistently necessary towards the end.

  • Notably, I've only attempted this on small games with fewer than ten players. Would be interesting to see if this was useful on larger games. Maybe I'll play on SO later today. – Adam Davis Mar 31 '15 at 13:07
5

Users are the least important item for winning. Get everything else up in this order: Questions, Answers, Quality. Traffic will populate rapidly after that. Then keep everything at four hearts, and only boost Users when everything else is maxed. Constantly trying to keep everything at an equal number of hearts will keep you going in circles endlessly.

4

If both Users and Quality are at zero, downvote first, then upvote. Otherwise, you're shooting yourself in the foot. (Downvoting adds one to Quality and subtracts one from Users. However, if Users are at zero, there's nothing to subtract. Thus, downvote-then-upvote results in +1 Users, +1 Quality. If you do it the other way around, the downvote will subtract the Users stat you just added, resulting in +0 Users, +1 Quality.)

  • 2
    Downvoting doesn't always subtract one from users. I suspect it's some percentage chance, or perhaps it depends on something else entirely. – mmyers Mar 31 '15 at 20:41
  • @mmyers: IME, if you have three or four hearts in Users, then downvoting doesn't always subtract one. But that doesn't invalidate this stratagem: what this is counting on is that stats are truncated at zero, so if you're going to subtract whatever it is, do it when the stat is already at zero, because then the subtraction will have no effect. – Marti Mar 31 '15 at 22:35
  • I'm not trying to contradict you; I've been following the same strategy. But I'm almost certain I've seen a single heart in Users that wasn't removed by downvoting. – mmyers Mar 31 '15 at 22:39
  • The source code reveals that downvoting subtracts a flat 0.5 users (should have guessed the stats were floating-point, not integer). This is indeed the optimal strategy for this situation. – mmyers Apr 2 '15 at 16:17
1

I have attempted, but found little success, in waiting until the end of the voting day to choose what seems lacking. In this attempt I expected to be able to make sure that unvoted for, but currently succeeding stats, had consistent pressure behind them despite other participants generally voting only for the least successful stat.

Perhaps participants would find this strategy more successful should they see other players employing companion strategies that would make this valuable to all.

1

Griefing is apparently being practised by some people. A few people are even forming groups and organizing in chat in order to grief other sites - they can both improve their site's stats, and make it difficult for other sites to improve theirs.

The main tactic I'm seeing right now is restarting other site's eggs, sometimes in an organized manner.

  • 2
    At Arqade, we just won within 251 days... we gamers are gaming! – Jonathan Drapeau Mar 31 '15 at 18:31
  • 2
    The number of votes required depends on the number of people who voted in the last round. Keeping the popup open has no influence on the quorum. – balpha Mar 31 '15 at 18:43
  • Related: meta.stackexchange.com/q/252393/266359 – Shokhet Mar 31 '15 at 20:48
1

I got the lowest time to winning as of this posting, at 248 days with both flags remaining.

Used a combination of the strategies mentioned here. Once it got to about 4 stars on everything, it was just a matter of keeping up. Seemed towards the end that I was mostly alternating between downvoting and asking, with the occasional upvote or answer to keep things happy.

I did not use Close at all, and did not use Flag at all, during the entire process. If Users have been upvoted to 4 in the past 2 turns (possibly more, not totally sure) then downvoting doesn't appear to reduce the Users score. This actually goes against the previous advice to downvote first and upvote second - if you do them sequentially and Users is high, it's better to upvote first it appears.

Not the answer you're looking for? Browse other questions tagged .