• Dear forum reader,
    To actively participate in our forum discussions or to start your own threads, in addition to your game account you need a forum account. You can
    REGISTER HERE!
    Please ensure a translation in to English is provided if your post is not in English and to respect your fellow players when posting.

Feedback Winter Event 2022

MooingCat

Viceroy
Spoiler Poster
Last year was 5 stars per quest for the rush quests as well, as far as I can tell there is only a 65 star reduction in milestones:

Starting: 20
Daily Login: 10x33 = 330
Main Quests: 5x38 = 190
Daily Quests: 15x32 = 480 (will have to verify this Update: confirmed).
Milestones: 140 => 25 + 50 = 75

Total: 1160 => 1095 Stars

1.1k stars should just about be enough on its own, with incidents it should be enough :)
 
Last edited:
Last year was 5 stars per quest for the rush quests as well, as far as I can tell there is only a 65 star reduction in milestones:

Starting: 20
Daily Login: 10x33 = 330
Main Quests: 5x38 = 190
Daily Quests: 15(?)x32 = 480 (will have to verify this).
Milestones: 140 => 25 + 50 = 75

Total: 1160 => 1095 Stars

1.1k stars should just about be enough on its own, with incidents it should be enough :)
We will see. I will complete every quest, collect incidents every day and see how many upgrades I can obtain without spending diamonds. I am not hopeful :(
 

Owl II

Emperor
Last year was 5 stars per quest for the rush quests as well, as far as I can tell there is only a 65 star reduction in milestones:

Starting: 20
Daily Login: 10x33 = 330
Main Quests: 5x38 = 190
Daily Quests: 15x32 = 480 (will have to verify this Update: confirmed).
Milestones: 140 => 25 + 50 = 75

Total: 1160 => 1095 Stars

1.1k stars should just about be enough on its own, with incidents it should be enough :)
Math is not my strong suit. Tell me where I'm wrong, please:
1100 stars is 110 matches. 110 matches are 220 lights (on average). 220 lights are only 6 upgrades from GP. But we need 8. That is, we need to get 400 more stars somewhere. "Somewhere" - in incidents. Or you should beg shuffle to fall out more often.

The same story as with the fall event, I think. Everyone is on edge. If you are not lucky with shuffle or with garbage, you will not get a GP
 

xivarmy

Overlord
Perk Creator
Math is not my strong suit. Tell me where I'm wrong, please:
1100 stars is 110 matches. 110 matches are 220 lights (on average). 220 lights are only 6 upgrades from GP. But we need 8. That is, we need to get 400 more stars somewhere. "Somewhere" - in incidents. Or you should beg shuffle to fall out more often.

The same story as with the fall event, I think. Everyone is on edge. If you are not lucky with shuffle or with garbage, you will not get a GP
It's not just shuffle that refunds stars. Every board has a +14, a +3 (show-2), and shuffle. This is about 27/180 that gets refunded. Plus I believe stars from key fragments if you do your dump on a single day. And sometimes they might be doubled.

It is hard to do the math on how unlucky you have to be to miss, but if mooingcat's simulations are on the mark, it should be fine.

You are on the edge in the sense of "don't skip anything".
 

Owl II

Emperor
It's not just shuffle that refunds stars. Every board has a +14, a +3 (show-2), and shuffle. This is about 27/180 that gets refunded. Plus I believe stars from key fragments if you do your dump on a single day. And sometimes they might be doubled.

It is hard to do the math on how unlucky you have to be to miss, but if mooingcat's simulations are on the mark, it should be fine.

You are on the edge in the sense of "don't skip anything".

It is hard to do the math? Impossible, I think. I spent about 400 stars and got 3 shuffles and 2 reward of 14 stars. And it's disgusting. The result in the event does not depend on your skills or diligence, but solely on your luck. Or even not so. Exclusively from the RNG FOE
 

drakenridder

Overlord
Perk Creator
The result in the event does not depend on your skills or diligence, but solely on your luck.
In the distant past, when this event was still somewhat new, the minigame was memory. They showed the rewards getting packed into different presents and getting shuffled in front of you. If you were somewhat good with the minigame memory, you could house your skills and getting whatever you were chasing. Later they've changed it into this gamble minigame where the rewards getting shuffled out of side. I thought when they changed into what it is now, they've done it cause players were "too" successful in memory. The letter however is speculation.
If they would change back this minigame to a fair game of memory, everyone at least has a shot at getting the Callander and thus the main building, with this few stars. This however is really wishful thinking. As they've removed the event box since the summer event, it's clear that this new "balance" is based off "just" enough event currency on average.
 

Owl II

Emperor
If they would change back this minigame to a fair game of memory, everyone at least has a shot at getting the Callander and thus the main building, with this few stars. This however is really wishful thinking. As they've removed the event box since the summer event, it's clear that this new "balance" is based off "just" enough event currency on average.
I finished collecting a calendar for 4 doors. :D:D:D
That is, I will get some more master keys from DC, of course. But 340 stars for one key from the board is a lot. I bought the missing keys for diamonds in the living world in the first year, it was cheaper. But I'm not sure I want to do it now.
 
Last edited:

drakenridder

Overlord
Perk Creator
I finished collecting a calendar for 4 doors. :D:D:D
That is, I will get some more master keys from DC, of course. But 340 stars for one key from the board is a lot. I bought the missing keys for diamonds in the living world in the first year, it was cheaper. But I'm not sure I want to do it now.
Can you share where we can get the bottled luck item in the game?
 

MooingCat

Viceroy
Spoiler Poster
Let's dig deeper into the numbers :)

There are 4 ways to get back stars per turn: +14 stars, +3 stars (show two), +10 stars (key fragment), +10 stars (shuffle). Additionally, the double up will double these numbers.

So if we naively think each board is played fully, you will get back 29 if you haven't found the key yet, 39 if you have, and if we say you have an 1/17 chance of the double applying, that's 30.7 / 41.3 stars. That's a 17.1% / 22.9% increase in stars, making 1.1k stars worth 1281 / 1346 stars. Of course, looking at it like this is quite foolish and not realistic at all, you don't always get all of them, but getting the shuffle sooner will also improve the value of the stars you have picked up. This is guaranteed to make the shuffle worth more on the vast majority of boards, but on the flip side getting shuffle sooner might also mean you don't get all the stars (might still be a better ratio than playing full board, though). Also, the show two ads a bit of agency, so sometimes you have the ability to pick up more.

That's why I made my simulator last year, to make something that simply plays through the game a bunch of times, and that way we can gather the data we want. Here's a description of how it works, if you are interested:

You start with a given number of stars, and a random board is generated based on the known rules of the game (one of each of the mentioned ways of getting stars, with the rest being random rewards). The simulator then starts playing, paying 10 stars each time to pick a random present. If this gives stars, those are added back to the amount of stars you have left to play. If it's a shuffle, the board is reshuffled. If it's a show 2, 2 random rewards are "revealed", and now there is a bit more complexity.

If the daily special is among the two, that is picked first. If any of the two revealed are 14 stars or a key (+10 stars after 1st key, so worth picking up regardless if it's fragment or not), that is picked. If the shuffle is among the two, it will only be chosen after the daily special has been found from the regular presents. If neither of the show two are worth anything, they are ignored and a random present is chosen from the rest.

This continues until all stars are spent. Additionally, the simulator has two different "modes", one focusing on daily specials (described above), and one focusing on keys. The one focusing on keys differ slightly, in that it "stops" playing after the first key is found on a day. This is simulated by shuffling the board after the key is found, up to 32 times. The thinking here is that if you aim to complete the calendar, your best option is to stop playing each day after you have found the key.

This is then repeated a bunch of times, which allows us to get more accurate average values, but also allows us to save each individual "playthrough", which can then be displayed in graphs showing the spreads of where people are expected to end up.

So, using the simulator I've done two simulations with 1095 stars, for the two different modes I mentioned (focusing on daily specials vs focusing on keys). I've also done some simulations with additional stars from incidents, so as you can see each 100 stars from there drastically improves your chances.

Focus on Daily SpecialFocus on Keys
trinket_plot.png
trinket_plot.png
+100 Stars from Incidents
trinket_plot.png
+100 Stars from Incidents
trinket_plot.png
+200 Stars from Incidents
trinket_plot.png
+200 Stars from Incidents
trinket_plot.png
+300 Stars from Incidents
trinket_plot.png
+300 Stars from Incidents
trinket_plot.png

So based on those results, I have a few takeaways:
  • Don't go for Calendar. Not only is it impossible to complete, but you get less stars back, because:
  • Spending stars on few days will give you more stars back than spending on many days
  • Getting the main building fully leveled should be quite achievable if you pick up some incidents
Below you can find the actual simulator, it's a slightly different version than the one I initially posted. Nothing in mechanics has changed, but changed it a bit to only output the grand prize graphs:

Code:
import random
import math
import numpy as np
from matplotlib import pyplot as plt

# Number of stars you start off with
startingStars = 1095+300

# If True, simulate stopping after finding each daily key. If False, spend all the same day.
goForKeys = True

# Number of keys required for calendar (you can get ~11 for free from daily challenges)
numKeys = 21

# Number of Simulations
simNum = 100000

# Show plots
showPlots = True

# Don't change anything below here (unless you know what you're doing)
######################################################################

fragments = 5

def getRandomList():
    # 4 - shuffle, 5 - key, 15 - daily, 20 - double
    l = [3,14,4,5,15,20,0,0,0,0,0,0,0,0,0,0,0,0]
    random.shuffle(l)
    return l
def pickPresent(days,l,l2,i2,dsFound,s,d,ds,k):
    i = 0
    t = False
    if(len(l2) > 0):
        if sum(l2) > 0:
            if (dsFound and (not goForKeys or (days > 32))) and 4 in l2:
                i = l2.index(4)
                t = True
            elif 4 in l2:
                if sum(l2) == 4:
                    try:
                      i = random.choice(range(len(l)))
                    except:
                      print(days,l,l2,i2,dsFound,s,d,ds,k)
                      exit()
                else:
                    i = l2.index(sum(l2)-4)
                    t = True
            else:
                i = l2.index(max(l2))
                t = True
        else:
            i = random.choice(range(len(l)))
    else:
        i = random.choice(range(len(l)))
    if t:
        res = l2[i]
        del l2[i]
        del i2[i]
    else:    
        res = l[i]
        del l[i]
    if res == 0:
        # nothing
        #print(0)
        d = 1
    elif res == 3:
        i2 = random.sample(range(len(l)),min(2,len(l)))
        l2 = [l[i2[x]] for x in range(len(i2))]
        del l[max(i2)]
        if len(i2) > 1:
            del l[min(i2)]
        s += 3*d
        #print(3*d,"stars + show 2:",l2)
    elif res == 4:
        l2 = []
        i2 = []
        l = getRandomList()
        dsFound = False
        s += 10*d
        #print(10*d,"stars + shuffle:",l)
        return days, l,l2,i2,dsFound,s,d,ds,k
    elif res == 5:
        k += 1 if goForKeys else 1/fragments
        if d == 2:
            k += 1/fragments
            s += 10
        if goForKeys and (days <= 32):
            days += 1
            l2 = []
            i2 = []
            l = getRandomList()
            dsFound = False
        else:
            s += 10
        #print(d,"key(s) +",d*10,"stars")
    elif res == 14:
        s += 14*d
        #print(d*14,"stars")
    elif res == 15:
        ds += 1*d
        dsFound = True
        #print(d,"Daily Special(s)")
    elif res == 20:
        d = 2
        #print("Double")
    if res == 20:
        d = 2
    else:
        d = 1
    
    return days, l,l2,i2,dsFound,s,d,ds,k

keys = []
dailies = []
candles = []
gpProgress = []
for x in range(simNum):
    if x % round(simNum/10) == 0:
        print(round((x/simNum)*100),"%")
    starting = startingStars
    s = starting
    ds = 0
    k = 0 if goForKeys else 1
    l = getRandomList()
    l2 = []
    i2 = []
    dsFound = False
    d = 1
    c = 0
    days = 0
    while s >= 10:
        s -= 10
        c += 1
        days,l,l2,i2,dsFound,s,d,ds,k = pickPresent(days,l,l2,i2,dsFound,s,d,ds,k)
    
    
    dailies.append(ds)
    keys.append(math.floor(k))
    candles.append(c)
    
    gp = 0
    for i in range(c):
      gp += random.choice([1,2,3])
    gpProgress.append(gp)

if showPlots:
    labels = ["Keys","Dailies","Grand Prize Progress"]
    results = [keys,dailies,gpProgress]
    fig, ax = plt.subplots(1,1,figsize=(10,6))
    #fig, ax = plt.subplots(3,1,figsize=(10,15))
    for i in range(2,3):
      label = labels[i]
      rmin = min(results[i])
      rmax = max(results[i])
      ax.hist(results[i],edgecolor='white', label=label, bins=range(rmin,rmax,1))
      #ax.set_title(label)
      if i == 1:
        ax[i].axvline(x = numKeys, color = "r", label = str(numKeys) + " Required for Calendar")
      if i == 2:
        ax.axvline(x = 300, color = "r", label = "Required for Full Building")
      ax.legend()
      

      #bin_centers = np.diff(bins)*0.5 + bins[:-1]
      #n = 0
      #for fr, x, patch in zip(freq, bin_centers, patches):
      #    height = (freq[n]/len(results))
      #    plt.annotate("{}".format(height),
      #                xy = (x, height*len(results)),             # top left corner of the histogram bar
      #                xytext = (0,0.2),             # offsetting label position above its bar
      #                textcoords = "offset points", # Offset (in points) from the *xy* value
      #                ha = 'center', va = 'bottom'
      #                )
      #    n = n+1
    #plt.legend()
    c = np.average(candles)
    gppU = [k for k in gpProgress if k < 300]
    gppO = [k for k in gpProgress if k >= 300]
    gpU = round(100*(len(gppU)/len(gpProgress)),2)
    gpO = round(100*(len(gppO)/len(gpProgress)),2)
    print(gppU)
    if len(gppU) > 0:
      avgM = round(300 - min(np.average(gppU),300),2)
      meanM = int(300-min(np.median(gppU),300))
    else:
      avgM = 0
      meanM = 0
    maxM = max(300 - min(gpProgress),0)
    
    
    mode = "(Daily Special Focus)" if not goForKeys else "(Keys Focus)"
    fig.suptitle("Expected Grand Prize Progress w/ " + str(startingStars) + " Starting Stars " + mode + "\nActual stars spent (average): " + str(round(c*10,2)) +
                  " | Chance of Full Building: " + str(gpO) + "%\nMissing Progress (if not full building): Max: " + str(maxM) + " | Average: " + str(avgM) + " | Median: " + str(meanM))
    plt.show()
# For last game
printStats = True
if printStats:
    c = np.average(candles)
    ds = np.average(dailies)
    k = np.average(keys)
    gp = np.average(gpProgress)
    kU = [k for k in keys if k < numKeys]
    kO = [k for k in keys if k >= numKeys]
    cU = round(100*(len(kU)/len(keys)),2)
    cO = round(100*(len(kO)/len(keys)),2)
    
    gppU = [k for k in gpProgress if k < 300]
    gppO = [k for k in gpProgress if k >= 300]
    gpU = round(100*(len(gppU)/len(gpProgress)),2)
    gpO = round(100*(len(gppO)/len(gpProgress)),2)
    if len(gppU) > 0:
      maxM = 300 - min(gpProgress)
      avgM = round(300 - np.average(gppU),2)
    else:
      avgM = 0
      maxnM = 0
    
    print("---------")
    if goForKeys:
      print("Average Totals (going for Keys):")
    else:
      print("Average Totals (going for Dailies):")
    print("Starting stars:",startingStars)
    print("Stars spent:",round(c*10,2),"(x" + str(round(c*10/startingStars,3)) + ")")
    #print("Stars Multiplication Factor: x" + str(round(c*10/startingStars,3)),"( 10 ->",str(round(c*100/startingStars,2)) + ")")
    print()
    print("Matches:",round(c,2))
    print("Stars* per Match:",round(startingStars/c,2))
    print()
    print("Daily Specials:",round(ds,2))
    print("Stars* per Daily Special:",round(startingStars/ds,2))
    print()
    print("Grand Prize Progress:",round(gp,2))
    print("Stars* per Grand Prize Progress:",round(startingStars/gp,2))
    print("Change of getting Full Building:",str(gpO) + "%")
    print("Maximum Missing Progress:",str(maxM))
    print("Average Missing Progress:",str(avgM))
    print()
    print("Keys:",round(k,2))
    print("Stars* per Key:",round(startingStars/k,2))
    print("Chance of getting",numKeys,"keys:",str(cO) + "%")
    print()
    print("")
    print()
    print("*Starting Stars")
 

Owl II

Emperor
Let's dig deeper into the numbers :)

There are 4 ways to get back stars per turn: +14 stars, +3 stars (show two), +10 stars (key fragment), +10 stars (shuffle). Additionally, the double up will double these numbers.
Ok. Since I'm bad at math, I'll try to simplify this for mysel
So based on those results, I have a few takeaways:
  • Don't go for Calendar. Not only is it impossible to complete, but you get less stars back, because:
  • Spending stars on few days will give you more stars back than spending on many days
  • Getting the main building fully leveled should be quite achievable if you pick up some incidents
If
So if we naively think each board is played fully, you will get back 29 if you haven't found the key yet, 39 if you have, and if we say you have an 1/17 chance of the double applying, that's 30.7 / 41.3 stars. That's a 17.1% / 22.9% increase in stars, making 1.1k stars worth 1281 / 1346 stars. Of course, looking at it like this is quite foolish and not realistic at all, you don't always get all of them, but getting the shuffle sooner will also improve the value of the stars you have picked up.
And add some math else. We need 1500 stars (300 lights) to get 8 GP upgrades. If we are so naive that we are counting on 1346 stars, given the fact that we will get ~250 stars on the board, then we must additionally get 150 stars from the trash incidents.
 

MooingCat

Viceroy
Spoiler Poster
Ok. Since I'm bad at math, I'll try to simplify this for mysel

If

And add some math else. We need 1500 stars (300 lights) to get 8 GP upgrades. If we are so naive that we are counting on 1346 stars, given the fact that we will get ~250 stars on the board, then we must additionally get 150 stars from the trash incidents.
No, just ignore those numbers, they are not reliable. The simulation numbers are more accurate, so use those :) Based on those, you need a bit of luck if you don't pick up incidents, but if you pick up enough incidents it's quite easy to get it :)
 

Ironrooster

Baronet
  • Don't go for Calendar. Not only is it impossible to complete, but you get less stars back, because:
  • Spending stars on few days will give you more stars back than spending on many days
  • Getting the main building fully leveled should be quite achievable if you pick up some incidents
So it looks like my plan to ignore the calendar and only play when the Daily Special is one of the few I want, is the way to go.

Thanks for the analysis
 

MooingCat

Viceroy
Spoiler Poster
I just tested spending 1095 stars, this was my result:
  • 1530 Total Stars Spent, for 153 Matches
  • Exactly 300 grand prize progress from those 153 Matches
  • 12 Daily Specials
  • 2.6 Keys (1 + 8 fragments)
That fits very well with my simulation :) Daily specials and keys are a bit higher than expected, but definitely within expected ranges. Stars spent and GP progress is pretty much bang on :)
  • 1530 vs 1503 Stars Spent = +27 (+0.18%)
  • 300 vs 300.58 GP Progress = -0.58 (-0.02%)
  • 12 vs 9.67 Daily Specials = +2.33 (+24.1%)
  • 2.6 vs 2.37 Keys = +0.23 (+9.7%)
 

Owl II

Emperor
No, just ignore those numbers, they are not reliable. The simulation numbers are more accurate, so use those :) Based on those, you need a bit of luck if you don't pick up incidents, but if you pick up enough incidents it's quite easy to get it :)
Umm. I counted my stars, matches and lights at the moment. After spending 320 stars, I got 92 lights, which corresponds to your calculations, +-. Ant yet I feel a little uneasy when "a little luck" is mentioned in math calculations. The casino always wins
 
Top