Latest topics
» *knock knock*
by Welux Wed Jul 22, 2015 1:49 pm

» Spiral Knight's Wiki List of Useful Links
by Ankoku Flare Sun Mar 15, 2015 6:25 pm

» Fehzor's guide to farming crafting supplies
by Ankoku Flare Thu Mar 12, 2015 6:28 pm

» [Brushes off the Dust]
by Darkreaperwolf Sat Feb 21, 2015 8:39 pm

» Anyone else looking for the new recipes
by Urameshiya Wed Dec 24, 2014 11:49 am

» The long awaited...
by ReadReact Fri Dec 05, 2014 1:33 am

» I think it's time for an old knight to retire (Giveaway raffle thing - winners announced!)
by Kupoo Thu May 01, 2014 6:46 pm

» New content!
by Welux Mon Apr 28, 2014 8:28 pm


a bunch of numbers

View previous topic View next topic Go down

a bunch of numbers

Post  alphappy on Sun Feb 17, 2013 6:07 am

A bunch of numbers concerning the Coliseum bribe and the prizewheel, for any curious minds out there. The graphs will probably extend beyond the edge of the spoiler; right-click them and select "view image"(or something similar) to see them properly.
Spoiler:




Figure 1: Chance for certain bribe prizes. It would appear that:
• the chance of a winning-team-only bribe is roughly equal to the chance of a all-players bribe.
• the chance of a mask is roughly 2.69%(2.5%, or 1 in 40, might be the actual chance).
• the chance of a single mask purchasable from Sullivan is not much different than one that cannot be purchased from Sullivan.
• the ratio of all players receiving 1 coin to 2 coins to 3 coins is about 12:4:1.
• the ratio of the winning team receiving 2 coins to 3 coins to a special item is about 18:2:1(three coins is not much more common than a mask or tank).
• assuming that 1)you win 50% of your games, 2)50% of the bribes are all-players, and 3)the above ratios are exact, a bribe gets you ~1.17625 coins on average(worth around 198cr with current enamorock prices). Note: this calculation excludes FFA BN(and actually BN in general), since your winning chances are either far lower or far higher based on your skill.

The purple line in figures 2-5 indicates the amount of data(and thus the accuracy) for any given depth. The two spikes are RJP and t3 prestige, respectively, with the latter's spike trailing off into FSC's other three depths. Where this line is very low, data is not reliable(especially d1-Cool. The number of data points in RJP is 477/354/325, while FSC is 776/704/355/297/162. Every other depth has less than 100 data points. Slight spikes exist at d9-d10 because of Winterfest and d19 because of Basil runs.

Figure 2: Chance for certain prizewheel prizes, excluding Full Health and Mecha Knight Kit(whose chances are about 0.59% and 0.21%, respectively) due to technical difficulties. The gray shadings on the bottom are crowns, while the ones on top are the materials. A stacked area chart. It would appear that:
• the chance of receiving any number of crowns rises from ~55% in RJP to ~65% in FSC.
• the chance of receiving a heart drops from ~20% in RJP to ~15% in FSC.
• the chance of receiving a belt item drops from ~13% in RJP to ~11% in FSC.
• the chance of receiving a material remains steady at 10% from RJP to FSC.

Figure 3: Chances for different crown values from the prizewheel. This shows a clear connection between depth and potential crown value.

Figure 4: Chances for different materials, grouped star value, from the prizewheel. Note: this assumes that all mats of a certain star value have an equal chance to appear, which may or may not be the case. This shows that the chances do not seem to change much from RJP to FSC, save for the 5* chance which is nonexistent in the former.

Figure 5: Chance for different belt items, excluding the Mecha Knight Kit. Note: the prizewheel will never give you a belt item you cannot carry. As such, this data is technically influenced by what was already on my belt at the time. Note also that this chart starts at d3 rather than d1, because no belt items have dropped for me in the first two depths since I began collecting data. This shows that:
• the chance for both types of pills was probably intended to be the highest individually, though only barely so.
• although it appears as such, fire is not more likely than other vials(note that fire only spikes when the accuracy line is low).
• while most vials have roughly the same chance, the chances of curse and sleep are roughly 1/5 the others.
• conversely, there is a very small chance that poison's drop rate is higher than than the others. It's generally the most useless status effect in PvE, so why not?
avatar
alphappy
Alumni
Alumni

Posts : 487
Join date : 2012-06-22
Age : 22
Location : Alabama, USA

View user profile http://alphappy.tumblr.com

Back to top Go down

Re: a bunch of numbers

Post  FallingSnow on Sun Feb 17, 2013 3:38 pm

*Applauds*

I'm sure that took quite some time to all sort out.
avatar
FallingSnow
Veteran
Veteran

Posts : 54
Join date : 2012-10-08
Age : 20
Location : United States

View user profile

Back to top Go down

Re: a bunch of numbers

Post  MushyMushy on Mon Feb 18, 2013 7:44 pm

quite interesting :l
avatar
MushyMushy
Alumni
Alumni

Posts : 737
Join date : 2012-09-28
Age : 18
Location : USA

View user profile http://www.ign.com/

Back to top Go down

Re: a bunch of numbers

Post  alphappy on Sat May 18, 2013 5:35 pm

Turns out that most of this data is inaccurate for multiple reasons: if you have full health, the wheel will not give you heart or full health; if you have three of any given belt item, the wheel will not give you that item; if you have a full belt, the wheel will not give you any belt items at all.
avatar
alphappy
Alumni
Alumni

Posts : 487
Join date : 2012-06-22
Age : 22
Location : Alabama, USA

View user profile http://alphappy.tumblr.com

Back to top Go down

Re: a bunch of numbers

Post  Sponsored content


Sponsored content


Back to top Go down

View previous topic View next topic Back to top

- Similar topics

 
Permissions in this forum:
You cannot reply to topics in this forum