![]() ![]() Xy = (x, height*len(results)), # top left corner of the histogram bar ![]() Label = "Keys" if keys else "Daily Specials"įreq, bins, patches = plt.hist(results,edgecolor='white', label=label, bins=range(0,50,1))īin_centers = np.diff(bins)*0.5 + binsįor fr, x, patch in zip(freq, bin_centers, patches): # a histogram returns 3 objects : n (i.e. L = ĭef pickPresent(l,l2,i2,dsFound,s,d,ds,k): # 4 - shuffle, 5 - key, 15 - daily, 20 - double The main reason I made the script was for daily specials, so for those interested I also simulated games going for daily specials (same "show 2" strat as above, except it picks "shuffle" if daily special has been found), here's the distribution for that: You did a great job improving it from last winter to archaeology/halloween, in those events it's actual a valid strategy to go for it, but for this event it's practically impossible to get it for free. Only a bit better than 1/3 chance if you have that many stars? I think it clearly shows that the calendar is very flawed in it's current state. 41% of the game got 23+ keys, but only 0.5% got 32+. ![]() If we bump the starting stars up to 2000 stars, which I think is a very generous amount from incidents, the story is a bit better. in that case a whopping 0.6% of the simulated games managed to get enough keys. Of course, you also have the daily challenges, which should give you 9 keys on average, so you "only" need 23 from the minigame. I simulated 100k games with 1355 starting stars, which are the free stars you can get. In other words, optimal play if you want to go for 1 key each day. The script "stops" each day after a key is found, so that the player gets a free shuffle each midnight. So the script takes that into consideration, picking the double, daily special (because you would go for this even if your goal is keys), key or stars if those are visible, if not ignoring the two shown rewards. The only part of the minigame that's not purely random that the player has any impact on is the choices made when the show 2 is picked. Because it's purely down to chance and because each board is the same when it comes to keys (same number of keys and other rewards that impact keys/stars you get per board), it should be accurate. To take it a step further in regards to keys, I wrote a simple script to simulate the minigame. ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |