12 replies to this topic

[Spectral]

The Idea: Track Coconut Shy Outcomes for Fun and Profit!

Based on Dreww's topic detailing Bagatelle outcomes and how outrageously unprofitable it is, I thought to do the same thing but for Coconut Shy. Reasoning wise, CS has only five outcomes, which makes our data much easier to track. Furthermore, it's known for giving out more prizes and generally being less of a massive waste of money, which means we may be able to get into the distribution for the possible prizes won by Coconut Shy.

Basically, Coconut Shy has costs 100 NP per play, at 20 throws per day. So, a daily cost of 2,000 NP per day. It has five outcomes:

1. You missed. (-100 NP)
2. The coconut hardly moved at all! (-50 NP)
3. Great shot, but the coconut did not fall down. (250 NP)
4. Knock down the coconut (9950 NP, Evil Coconut, avatar)
5. KABLAM! SUPER JACKPOT! (499950 NP, Evil Coconut)

Most Evil Coconuts are worth about 500,000 NP, which leads me to believe that unlike Bagatelle prizes with wildly varying levels of value, they drop at the same rate. Even if we can't test this out, by figuring out some sort of distribution for the main levels and the net profit over enough time, you can make an informed decision as to whether its worth it or not. If it is, it can be official for us: your daily 20 throws of Coconut Shy a day makes the money come your way! If not, steer clear and just buy those damn coconuts with your AB money.

My hypothesis is that it follows a normal distribution, however, the lower bound is entirely filled by the lowest option, missing (from 0 to sigma). Almost knocking off corresponds from 1 to 3 sigma, the jackpot from 3 to 4 sigma, and the explosion to 4 to 5 sigma. Therefore, we'd expect a distribution like so:

Spoiler

NB: The general normal distribution function will be shown once in mathematical form, then shortened to "f(x)". 
NB: Notation for math is done like so int[f(variable) d(variable), lower bound, upper bound]

Outcome             Probability      Mathematical Notation
Miss                ~15.8%           int[(1/sqrt(2*pi*sigma^2))*e^(-(x-mu)^2/(2*sigma)^2) dx, 0, mu - sigma]
Hardly Moved        ~68.2%           int[f(x) dx, mu - sigma, mu + sigma
Almost              ~15.7%           int[f(x) dx, mu + sigma, mu + 3*sigma
Win!                ~0.1%            int[f(x) dx, mu + 3*sigma, mu + 4*sigma
SUPER JACKPOT       ~0.00003%        int[f(x) dx, mu + 4*sigma, mu + 5*sigma

Which, if we multiply by our expected profits, we should get this after 1 million throws.

Spoiler

Outcome               Probability       Hits (out of 1M)       Profit                   Total  
Miss                    ~15.8%              158,000          -15,800,000 NP
Hardly Moved            ~68.2%              682,000          -34,000,000 NP
Almost                  ~15.7%              157,000           41,100,000 NP
Win!                     ~0.1%                1,000          510,000,000 NP
Explosion!           ~0.00003%                 0.3%              300,000 NP

                                                                                     501,600,000 NP!!!

If this is true, then due to just how obscenely profitable a win of Coconut Shy is at about 510,000 NP's worth, and a relatively high chance of happening, Coconut Shy would be a must-do daily the likes of which would seriously shake up Neopia. Doing this calculation makes me suddenly very skeptical of this, but I guess we'll see?

 

TL;DR: Getting The Data

Luckily, we have a program on the Program Manager that will not only do Coconut Shy for you, but it'll even give you a recording of your throws! However, it does include date-times, which we really want to avoid for security reasons (please let me know if this is not a concern). So, when you copy paste your data from your daily throws, just remove the date-time. But this is tedious. Therefore, we'll see to just replace the entire line with a number corresponding to the order above. That means 1 for a miss, 2 for a 'Hardly Moved', 3 for a 'Great shot', 4 for a win, and 5 for an exploding coconut. Here's an example below.

Spoiler

2
2
2
2
2
2
2
2
1
2
1
2
2
2
1
3
1
2
2
3

I'm not sure how to make a webtool to do this for you, but if someone is willing to help, I'd appreciate it. It's easy enough to do in R or Stata, maybe I can make a widget using ShinyApps?

Every two weeks, assuming we have any new data, I'll try to push it through some programs to see what I get. When I get to 1,000 throws, or 50 days worth, I'll do a more detailed analysis. If we get to 10,000, or 500 days worth, I'll be more able to construct a good picture, and ideally even get into prizes. I'll also bump data myself.

[Spectral]

Today's throws:

Spoiler

2

2

1

1

2

2

2

2

3

2

2

2

1

2

1

2

3

2

2

2

[Spectral]

Spoiler

3

2

3

3

1

2

1

2

1

3

2
2
2

1

2

3

1

2

1

2

Spoiler

3

2

2

2

2

2

2

2

2

2

2

1
1

3

2

3
2

2
2
2

Spoiler

1

2

1

2

1

3

3

2

2

2

1

1

1

2

3

1

1

3

2

3

The more you post the better data we will get!

Edited by Spectral, 21 April 2021 - 10:36 AM.

[Spectral]

Spoiler

2

1

2

1

2

1

2

2

1

2

2

2

2

1

2
2

1
1
1

2

[Spectral]

Spoiler

2

1

2

2

2

2

2

2

3

2

2

2

2

2

2

2

1

1

3

2

[Spectral]

Sorry, my computer literally exploded. Did you know Windows 10 doesn't have the registry key to make backups of your registry, unlike any other Windows system? I just migrated from 7 so I had a great day yesterday.

Spoiler

2
1
2
2
3
2
1
2
2
3
2
2
3
1
1
3
2
1
2
2

Spoiler

2

2
2

1
1
1

2
2
2
2

1

3

2
2
2

3

1

2

2

1

So I did a tiny bit of preliminary number crunching and found that the numbers trend in the direction of:

Miss: 25%
Hardly Moved: 60%
Almost: 15%, which I didn't expect but is interesting. Due to how the profit works, that means that at that rate you'll lose about 25% of what you put in, not counting falling coconuts, which I haven't seen yet. (I did get one already though )

[Spectral]

Spoiler

3

2
2
2

3

2

3

2
2

3

2
2

1

2
2
2
2

3

2
2

Spoiler

1
1

3

2
2
2

2
2
2

2
2

1

3
3
3

2
2
2

1
1

Edited by Spectral, 29 April 2021 - 02:53 AM.

[Spectral]

Spoiler

2

1

3
3

2
2
2
2

1
1

2

1

2

3

1

2

1

2
2
2

[Spectral]

Spoiler

2
2
2
2
2

1
1

2
2
2

2
2
2

2
2
2
2

1
1

Spoiler

2

1

2
2

3

1
1
1

2
2
2
2
2
2

3

2

1

3

2

1

Alright folks, after finals I'll do the first data-analysis post.

[Spectral]

Spoiler

2
2

3

2
2
2

3

2

3

1

2
2
2

3

2

1

2

1
1

3

[Spectral]

Spoiler

2

1

2

1

2
2

3

2
2

1
1

2

1

2
2

3

2
2
2

3

[Spectral]

oops finals

Spoiler

1
1

3

1
1

2
2
2

3

1
1

2
2
2

1
1

2

1

2
2

[Spectral]

Been a while but I might do this again, it'll just require me to find a way to take the Neopets website non-flash version's text and turn it into something useful, or I guess manually do it. I do wanna see what's the general probabilities. Coconut Shy I am pretty sure is genuinely profitable because of the Stamps. Of course, to the extent this game is still playable, IDK if it matters. There's other probabilities to run stuff on, but idk.

Well, it'll give me something to do. Is there a way to not have this notify people?

 

Spoiler

3
2

2

1

3

1

1

2

3

3

2

1

1

2

2

1

2

2

2

2