Mechanics of Item phases in The Haunting

Discussion in 'Strategy' started by tmh, Dec 14, 2011.

  1. Trying to make up for the fail back when Krullian and I tried so hard to precisely derive the then-new, now-current ally plunder calculation. We had it worked out really well, R value was pretty close to 1 but not close enough... But we never cracked it. I'm gonna try and figure this one out as best I can. And I do love me some analysis. :)
     
  2. Well this proves your the real tmh, back and testing game mechanics. 
     
  3. A little update... The two EB's I reference in the first post, where one paid 60% higher than the other... The one that paid more had 3917 total actions in it, the one that paid less had 4457. So that seems like it might be a promising indicator. BUT. A third EB, our most recently completed one, had only 3807 actions, the fewest of the lot, but its payout was in between. So that kinda pokes a hole in the idea of pay per item being directly related to your number of actions vs total actions.

    I'm going to do a first look at damage to earning ratios now.
     
  4. Hit me up if you need help.
     
  5. one of the rare good threads in forum. keep it up op!
     
  6. Great job! I don't have the patience to do it. This should be stickied, in my opinion.
     
  7. Don't know if this will help or not and if you want I'll give it a try with this account as well but my thus far I've noticed with my other two accounts, that your build does play an indicator in your item payout, in the clan my other two accounts are we try to put a limit on how many items a player uses during the haunting, while using this method I've used preciously 20 items (with no other actions) on both accounts and the account with smaller stats gets a bigger payout every time.
     
  8. It seems items get a % of plunder. More assassinations in eb means less item plunder and less total plunder[/cool]
     
  9. Damn typo in color. Too early lol
     
  10. Damage ratio seems will give you some headaches use some spy pots for atak build. Good luck tmh 
     
  11. I've got a running trend of how much items pay vs 1) total plunder in EB, 2) total actions in EB, and 3) average $ per action [total $ of EB / total actions in EB]

    There appears to be no correlation between $/item and total plunder, or total actions, BUT, so far the $/item DOES track similarly to #3, the average $ earned per item.

    The deeper I get into this the more I get the feeling that I'm going to need to organize a few Grand Test sort of EB's, to produce some 0 assassination EB's and , one some 0 attack EB's. And to refine it even more, could do 0 attacks with all large builds, 0 attacks with all small builds, and 0 attacks with a range of build sizes. This could help isolate the variable of the cap. The small builds, which are below the cap, might make it easier to do some regressions. The cap messes that all up.

    The EB we have running now is only the second where I have a record of the exact damage done by each item used. Unfortunately it's for two different builds. Next EB I'll try and get a second set of full damage data for both of them. That might open some doors.
     
  12. Oops. Meant to write "average $ per *action*", not "average $ per item", in the second paragraph of that last post.
     
  13. And that correlation only applies to capped players, from what I've found so far.

    Keep in mind, the first post of this thread is still preliminary, and everything I'm posting outside of that first post isn't just preliminary, it's speculative. I'm sharing my thought process and my methods in these posts. Once I have something worked out well enough that I'm pretty certain I've got that aspect of the mechanics locked down, I'll add it to the first post. For now, everything outside that post is super super rough and not at all certain. Just wanna say that explicitly.
     
  14. Damage does play a roll on payout. On my osf I only unload spies for a easy 500m
     
  15. I don't know if this helps but here goes. One haunting just only atk/ass I accumulate around 20 mill-ish a successful action. Then the next one I got in only 30-35 items just to see how the payout goes. I got 700 mill for it. Any data on that?
     
  16. Guys keep in mind we are talking here about only the item phases, which means 0 attacks and 0 assassinations throughout the entire EB.
     
  17. Your numbers may not be as off as you think there TMH. Keep in mind ATA set variables a while back on everything so our ally plunder didn't end in 0000.

    You can see this evidenced in how one pot will do 50 damage then the next 76 (or whatever your rage might be)

    I think you're on the right track with using capped and uncapped players.
     
  18. If ya need any help with testing in an EB I'm sure that anyone here would be likely to help :)

    ******* genius
     
  19. Nice post I do seems to notice a much smaller build getting comparable end plunder compared to a build wi much higher stats even thou the smaller build did much less damage. Both did 1 full unload on the entire eb. Hope this rings some bells on yr tests
     
  20. @tmh:

    i never have the luxury to conduct really focused research and trial like yours, but i'd like to share some of my observation on the payout mechaninc of EB.

    1.There are 2 types of income, direct plunder (yes, those ka-ching when u hit skelly) and non-direct plunder AKA bonus (those appears when u finish EB)

    2. My theory is non-direct plunder AKA BONUS is affected by TOTAL DAMAGE you deliver to skelly, including assn, direct hit and pots

    3. My theory is the TOTAL DIRECT PLUNDER is the same amount of TOTAL NON-DIRECT PLUNDER that are distributed as BONUS, an example is if the whole clan's TOTAL DIRECT PLUNDER is 50bill, then the TOTAL BONUS distributed is als0 50 bill.

    Some of you might ask where the heck i get that theory.

    - If u all remember, during few days of EB launching, each time u finish EB the recorder plunder history is only 40-50 bill ish, then after awhile the record saying 90-100 bill ish along with the bonus distributed in the news, this is my first source of theory
    - If u guys ever notice, a clan with more hansel than hitters usually the total EB record is not as high as EB with more hitters, same thing also applied if u got LB players joining your EB, the EB might ended with only 70-80 bill ish total plunder recorded in history, this is supporting my theory even tho not so direct.

    4. My theory is TOTAL NON DIRECT PLUNDER AKA BONUS is distributed proportionally based on your TOTAL DAMAGE multiplied by TOTAL DIRECT PLUNDER COLLECTED. to make it easier the example is like this:
    if TOTAL DIRECT PLUNDER is 50 bill and you contribute 10% of the TOTAL DAMAGE DEALT then you will have bonus 5 bill (Assume that you do haunting with 10 peeps only, i've done that and the result more or less support my theory)

    5. Referring to theory #3 and 4, it is almost impossible to know the exact mechanism or formula for each of your action BECAUSE damages to skelly is always RANDOM, same with POTS, all RANDOM. Any trials or research conducted by some peeps such as 'i only throw 1 pot and no action', 'i only throw 20 pots and no action' i think CANNOT justify that 1 pot = .... gold or to decide that more pots thrown means less gold, WHY is it so?

    - You cannot be sure if your first, second and so on research is based on the same TOTAL DIRECT PLUNDER
    - You cannot be sure if your testcase EB got the exact same TOTAL DAMAGE PROPORTION of yourself compared to others, on first trial u might deliver 2% of TOTAL DAMAGE, second might be 3%, and so on.

    CONCLUSION:
    1. No offense for you tmh but judging that more pots thrown meanings less eficiency for the whole clan is too early to say...
    2. The best way to figure out the mechanic is by having EB with less people full crystal (i've done with 10 peeps and finished in 3 or 4 hours or so, yes we are no LB..). have all the members NOTE DOWN each of DAMAGE DEALT and DIRECT PLUNDER, put up on excel to know EACH TOTAL DIRECT PLUNDER and DAMAGE DEALT, you will have the proportion and can work out from that..

    I myself quite satisfied with my theory and not interested to put up some heavy research furthermore, the effort is too much for me.. daily work startin to heat up and im gettin busy..

    Best of Luck for you TMH, i can see that u're being such a nice guy sharing your precious research, thus i triggered to share also for the community.