Close this discussion please

Discussion in 'General Discussion' started by doronos, Dec 15, 2020.

Thread Status:
Not open for further replies.
  1. doronos
    Offline

    doronos Well-Known Member

    Joined:
    Sep 29, 2016
    Messages:
    184
    Likes Received:
    227
    Gender:
    Male
    I don't know if I agree with that. How many new/poor players that got no money actually manage to reach that state? where they gather up 4 CS and FS, make all of them work to reach 14/1, all taking them up AND they got nothing. Very little in my opinion (just a hunch I didn't really check it, it might be true). I think it's a conscience decision that comes from the hate of losing rather than the greed. But, then again, it's just an opinion. I didn't come from the field of behavioral psychology (even though it's super interesting) I'm just speculating.
     
    FlawlessNa likes this.
  2. Kenny
    Offline

    Kenny Donator

    Joined:
    Dec 31, 2018
    Messages:
    4,729
    Likes Received:
    5,000
    Country Flag:
    IGN:
    Shadower
    Level:
    200
    You are still missing the point. I am not saying 1 slot necessarily = 1 attack but I am saying 1 slot adds value to the original item. From what I see in your model, you are still assuming 14/1 slots have the same value as 14/0 fs, which is incorrect.
    Another issue I found in your calculation: the cs outcome of "-1" gives a monetary outcome of "-2.5" instead of "-3.5". The price of 14att (15b) - the price of 13att (12.5b) is 2.5b, not 3.5b. This is a simple error of subtraction and yet hurts your credibility by a lot

    The price guide is constantly updating so you might comparing the price of 2020 to the price of 2019. And again, data reference is important. Anyway using the most updated data won't hurt. I got the reference price from this post and put them in a table for you
    upload_2020-12-15_7-22-52.png
    Assume your model outline is correct, your E(Y) based on the most updated price is actually -0.767, assuming 14/1 = 15/0. (Again, pls gives 14/1 some sort of value above than 14/0)

    The point I am trying to make is that you fix your model first before drawing the conclusion.
     
    Last edited: Dec 15, 2020
  3. doronos
    Offline

    doronos Well-Known Member

    Joined:
    Sep 29, 2016
    Messages:
    184
    Likes Received:
    227
    Gender:
    Male
    I'm sorry, I don't know how to value 1 slot other than the expected value of the difference in item price after the last slot was taken. I'm trying to avoid comparing 14/1 to 15/0 directly, but just by expanding what outcomes could 14/1 give. If you have and idea for a better model (or more correct one) please suggest it, I would love to see another point of view I will most certainly learn from that. I corrected the type you mentioned. It only looks as if I gave 14/1 same value as 14/0 because it is the last slot. But if I were to analyze 2/5 the price of "+0" would be 2/4 FS (which I will have to expand on since each slot isn't really valued, you will have to expand on each slot until you get 0 slots and price for that item). I'm really sorry it is pretty complicated but the goal is to eliminate the "slots" and get only the expected value of the scrolling. Our goal in all of this is to see weather or not we should settle for 1 att higher or put the last scroll ourselves.

    It is true that prices change, and maybe that guide isn't accurate today, but just remember that it is the Christmas event, many people are making socks and the prices are changing due to supply and demand and not probabilities and rarity of them. (Remember that the rarest item that no one wants is worthless)

    I feel like I'm being yelled at and it's not nice, I shared thoughts and findings I researched and I'm opened for criticism and questions if something isn't clear. If you have a counter proposition just write it and we can discuss it.

    Luckily for me I made a disclaimer that I'm not a math professor, there might be errors in the way I approached that matter, I'm opened for discussion as I do want to learn, I am passionate about math in general. unfortunately typos happen, and I have corrected it. Most of my typos are getting red marks so I'm used to the editor pointing me to them but it doesn't mark numbers. Maybe one day AI will be smart enough to point those out too. Anyway, if you don't believe me, it's alright I don't believe myself most of the time. If I told myself that I'm going to defend myself in a Maplestory forum for a typo I made I would never believe that :)

    As for your request:
    FS
    19 att FS = 57b (+42b at 1/11 chance)
    18 att FS=43b (+28b at 1/11 chance)
    17 att FS=32b (+17b at 1/11 chance)
    16 att FS=25b (+10b at 1/11 chance)
    15 att FS =20b (+5b at 1/11 chance)
    14 att FS = 15b (0 at 1/11 chance)
    13 att FS=12.5b (-2.5b at 1/11 chance)
    12 att FS=9b (-6b at 1/11 chance)
    11 att FS=8b (-7b at 1/11 chance)
    10 att FS =6b (-9b at 1/11 chance)
    9 att FS = 4.4b (-10.6b at 1/11 chance)

    The expected value of change in the price of the item (assuming 0 slots all the process) is:
    (-10.6-9-7-6-2.5+0+5+10+17+28+42)/11 = 6b. The price it cost to activate that transform is 1WS+CS which has expected value of 1.666b. It makes the shoe worth, according to my model, 15+6-1.6 = 19.4b And currently you will get that it is worth to make that specific trade, HOWEVER, the stronger the item the less it worth to make that trade.

    Keep in mind that we are in an event, and the prices of shoes are currently dropping rapidly (it's just for now) sooner or later we will get back to the same prices we've already seen in that guide. Unless a new item is brought into the pool, or rates change etc.... we will revolve around the same item in the next few months.
     
    Last edited: Dec 15, 2020
    atlas4 and Henray17 like this.
  4. Kung
    Offline

    Kung Well-Known Member

    Joined:
    Jan 20, 2018
    Messages:
    557
    Likes Received:
    626
    IGN:
    Schnegge
    As far as we can observe our surroundings at the moment, nothing seems to be made for a purpose. Your view of evolution does not reflect what it looks like at the moment.
    There seems to be no purpose in things.
     
  5. haiki
    Offline

    haiki Donator

    Joined:
    Jan 11, 2020
    Messages:
    27
    Likes Received:
    130
    Gender:
    Male
    Country Flag:
    IGN:
    haiki
    Level:
    66
    Guild:
    Cozy
    I feel smarter now after reading this
     
    doronos likes this.
  6. MoriForest
    Offline

    MoriForest Well-Known Member

    Joined:
    Sep 30, 2016
    Messages:
    1,573
    Likes Received:
    12,149
    Country Flag:
     
    Eli and Shnang like this.
  7. Pleh
    Offline

    Pleh Member

    Joined:
    Nov 14, 2020
    Messages:
    10
    Likes Received:
    39
    Gender:
    Male
    Country Flag:
    IGN:
    Woad
    This is exactly what it is. Generally speaking, humans LOVE the feeling of accomplishment and that little fuzzy feeling in the back of your head from the dopamine rush of (for example) landing a positive chaos scroll, which is exacerbated by just how much of a positive it is. Humans however have this weird emotion called regret and the desire to not feel that emotion is typically greater than the little dopamine rush we get. I'm not sure where you could fit this into your statistical data, but it most likely has a lot to do with the anomalies.
     
    doronos likes this.
  8. Kenny
    Offline

    Kenny Donator

    Joined:
    Dec 31, 2018
    Messages:
    4,729
    Likes Received:
    5,000
    Country Flag:
    IGN:
    Shadower
    Level:
    200
    You are open for criticism and I am giving my criticism. I am not yelling at you. I am pointing out the flaws in your model and conclusion. Let me explain the flaw again word by word (the color text is my comments):
    As for your request:
    FS scrolling of 14/1:
    19 att FS = 57b (+42b at 1/11 chance) 14/1 -> 19/0, you subtract 14/0 att FS from 19/0 att FS, which is 57b - 15b =+42b
    18 att FS=43b (+28b at 1/11 chance) 14/1 -> 18/0, you subtract 14/0 att FS from 18/0 att FS, which is 43b - 15b= +28b
    17 att FS=32b (+17b at 1/11 chance) 14/1 -> 17/0, you subtract 14/0 att FS from 17/0 att FS, which is 32b - 15b = +17b
    16 att FS=25b (+10b at 1/11 chance) 14/1 -> 16/0, you subtract 14/0 att FS from 16/0 att FS, which is 25b - 15b = +10b
    15 att FS =20b (+5b at 1/11 chance) 14/1 -> 15/0, you subtract 14/0 att FS from 15/0 att FS, which is 20b - 15b = +5b
    14 att FS = 15b (0 at 1/11 chance) 14/1 -> 14/0, you subtract 14/0 att FS from 14/0 att FS, which is 15b - 15b = +0b !!!
    13 att FS=12.5b (-2.5b at 1/11 chance) 14/1 -> 13/0, you subtract 14/0 att FS from 13/0 att FS, which is 12.5b - 15b = -2.5b
    12 att FS=9b (-6b at 1/11 chance) 14/1 -> 12/0, you subtract 14/0 att FS from 12/0 att FS, which is 9b - 15b = -6b
    11 att FS=8b (-7b at 1/11 chance) 14/1 -> 11/0, you subtract 14/0 att FS from 11/0 att FS, which is 8b - 15b = -7b
    10 att FS =6b (-9b at 1/11 chance) 14/1 -> 10/0, you subtract 14/0 att FS from 10/0 att FS, which is 6b - 15b = -9b
    9 att FS = 4.4b (-10.6b at 1/11 chance) 14/1 -> 9/0, you subtract 14/0 att FS from 9/0 att FS, which is 4.4b - 15b = -10.6b

    Do you see the fundamental flaw here? you don't need to be a math professor to see the flaw here as it's solely game knowledge. In a simple way to say: if your 14/1 went 14/0, you lose money. If your 14/1 went 15/0, you cut even. (assuming we may still exchange 14/1 for 15/0 which should be true in the market now)
    Your center point for your expected value calculation should be the value of 14/1, in this case and realistically speaking, is equivalent to the value of 15/0. Even if you don't agree with 14/1 ==15/0, give 14/1 a value that is greater than 14/0. You do need a reason for the value tho.

    The price thread is created on Oct 31th, 2020 so it's probably unaffected by the Christmas event

    Please don't bring evolution into this...
     
    Last edited: Dec 15, 2020
    Shnang likes this.
  9. Mori0
    Offline

    Mori0 Well-Known Member

    Joined:
    Mar 31, 2020
    Messages:
    50
    Likes Received:
    36
    Country Flag:
    big brain
     
  10. Dave Deviluke
    Offline

    Dave Deviluke Forum Moderator

    Joined:
    Oct 5, 2017
    Messages:
    10,997
    Likes Received:
    10,569
    Gender:
    Male
    Location:
    MapleRoyals Discord
    Country Flag:
    IGN:
    CygnusQueen
    Level:
    110
    Guild:
    WorldTour
Thread Status:
Not open for further replies.

Share This Page