Sunken Probability - Best Rods for Sunken (It's Luring Bronze and I'll prove it)

Game Discussion
1

I don’t like writing so lets just jump into the math aspect.

Some initial premises to begin: I’m assuming that the odds of catching any Sunken is 1/2000, and and once you catch a sunken it’s a 1/4 chance of it being a sword.
Pity is assumed to be at 3000 fish and no sunken. updated to account for pity only resetting when catching sunken with it
Increased item chances like Magnetic and Collector Rods are assumed to reduce the 1/2000 chance to 1/(2000/item chance)

Since I actually suck dick at math and statistics, I just ran some simulations using the above numbers in C++. The numbers then obviously aren’t perfect, but they’re pretty close because I averaged about 10,000 instances.

Using the above method, I found the average number of fish needed to be caught before getting a sword, including pity.
With a item catch rate of 1.5, it took about 3800~ fish. (Magnetic Collector)
With a rate of 1.2, it took 4550~ (Luring Collector)
With a rate of 1, it took 4750~ (Idk you pick)
With a rate of .8, it took 5700~ (Luring Bronze)

But to make a definitive choice on which rod was best, you’d have to how long it takes each rod on average to catch a fish. And unfortunately, I can’t simulate that so I had to actually run some tests in game, so my sample size is way smaller and prone to human error. That being said, here are the numbers.

With .8 luring, or Magnetic Collector and across 34 fish, each fish averaged about 38.6 seconds each. Considering it takes about Magnetic Collector Rods about 3800 fish to get a sword, that’ll be 146,680 seconds or ~41 hours. Fun.

With 1.1 luring, or luring Collector and across 20 fish, each fish averaged about 31.1 seconds. With respect to the 4550 fish per sword Luring Collector Rods averaged, it’ll take 141,505 seconds or ~39 hours. Slightly better.

With 1.5 luring, or Luring Bronze and across 30 fish, each fish averaged about 24.2 seconds each. Further tests found 23.8, but I thought it was close enough so I kept it. You know the drill - 5700 fish required for a Luring Bronze on average to catch a sword, so 137,940 seconds or 38 hours. We have a winner!

Some other cool data I found when running these sims was that on average, within every 200 sunken swords caught, the fastest someone will have caught it in as low as 40 fish, and someone will have taken as long as 35,000. However unlucky you feel like you are, you probably aren’t the 35k guy.

Remember in WOM when disenchanting wasn’t a thing and you’d have to do this 3 times (on average) if you wanted a strong one? :slight_smile:

haha i can name this anything i want ANYTHING amo gus
Magnetic Collector (1.5, 0.8) Luring Collector (1.2, 1.1) Luring Bronze (0.8, 1.5)
# of fish 3800 4550 5700
time per fish 38.7 seconds 31.1 seconds 24.2 seconds
total time per sword in seconds 146,680 141,505 137,940
total time per sword in hours 41 39 38

*adjusted for how pity actually works now

15 Likes
2

Finally someone did the math, thank you! I’d like to note though, pity is at 3000 regardless of whether you catch a sunken or not, as I’ve gotten pity twice, both times I’ve caught a sunken item within the last 3000 fish.

1 Like
3

fr? now i gotta do this shit again?
thanks for letting me know lol

4

props to you for all these calculations but this brought back ptsd from an old fishing odds thread :sob:

5

yeah it was a sight to see
face1

6

cuz like what does this even MEAN

image
image717×364 10.9 KB

7

i think his method was interesting, but not that great. doing it mathematically is better than simulating it sure, but i didn’t cause it’s hard.

a couple issues with his method are that A, sunken pity doesn’t reset when you catch a sunken, only when you catch one with PITY. meaning you get a guaranteed one every 3k fish, regardless of whether or not you fished one up earlier.

and B, he seems to be assuming that luring is a flat time decrease, but it’s not. Due to the time it takes inbetween casts and whatnot, luring isn’t actually as effective over magnetic as it might seem on paper.

but not like it matters bc we get the same numbers in the end.

i see level typing lmao
also FUCK FUCK FUCK I JUST TESTED MORE NUMBERS AND GOT 27.4 SECONDS FOR LURING BRONZE WHY IS IT DIFFERENT

8

Ah, that’s me and that’s really old.

It was before AO release so I wouldn’t know how pity worked and for B, I mentioned that luring isn’t a flat time increase however the difference would be negligible if you use the method in that thread.

The method was to use luring collector’s, ignore fishes, and exclusively fish up item bags.

However, now that fishes matter, it might be worth it to just use luring bronze since that takes less brainpower anyways(now you only need to listen to audio rather than see) but you could still do that old method and only fish up exotic fish+ if you’re a masochist and memorized what fish are exotic+.

9

i actually read that method of fishing up exclusively items and im like “what the fuck”
like that’s really clever but the effort you’d have to put in for however long on end is agonizing

doesn’t seem like any of the rods really matter in the end though, it’s like 10% difference at best

10

Ya with the knowledge pity always works on the 3000th fish probs should just run lucky bronze for exotic+ fish while getting sunken

11

Where can I get the bronze fishing rod? I fought the only two available in the game would be the collectors and the wooden fishing rod.

12

Bronze rods are available at Palo Town’s fish monger!

1 Like
13

Thanks :+1:

14

Is this fr, and if so how did you find this out? Is there some difference between getting a sunken from pity and from regular fishing, did Vetex confirm it in Vetcord or in some other place I’m not in, or is there confirmation elsewhere?
It’s seems very weird that it would work like that, since it’s the opposite of what pity usually does in games.

Yup, there were some assumptions and rough “rounding” in those calculations, but I’m still confident the result is right (if pity works like it does in every other game, since I’m not sure how to factor it in as you here are saying it works).
I had some plans to account for these, as well as the other technique (see bellow), but all the math would get actually complicated (I still think that what I did already is simple), and the best I would get would be some 5-dimensional “plots” with multiple unknowns I would have to experimentally get a value for, and I was too lazy to both do that and then fish in WoM to confirm it.

All this time I thought the method is simply inferior as you wouldn’t be stacking up pity points which is what carried Luring Bronze in the first place. Now I am a bit sceptical but still would bet on it being less effective since pity would require catching fish, and relying on the base odds of catching sunkens might not cut it. Still, I have nothing to back any of that up, but I never saw anything in favor of it that wasn’t just talk, so I’m still biased against it.

Be careful when running these Monte Carlo simulations as they can depend a lot on the quality of the underlying pseudo-random generator. I believe std::rand() is the thrashiest implementation of a generator you can get that suffices only for stuff where the randomness is not actually what matters.
When generating large amount of random values bad generators start having patterns and even repeated sequence which no longer are really random.
There’s definitely be a library that has a better implementation you can just switch in and not worry about it if you want to do more of this stuff, there probably is one in one of the many standard headers for C++.

1 Like
15

fshish

16

I think that the base sunken chance is noticeably higher than 1/2000 or the item chance modifier scales exponentially. I’ve caught around 4,400 fish with magnetic collector rod and the result is below. (pity was a chestplate)
1 Sword
2 Helmets
3 Chestplate
1 Boot/Leggings

1 Like
17

There’s been a couple posts on here about people getting a sunken and then getting another one shortly after on their 3000th fish

I’ve kinda thought this is how it worked for a while, since in the patch notes vetex wrote “guaranteed sunken every 3000 catches” and not “guaranteed sunken every 3000 catches without a sunken” so I’m not too surprised that’s how it acc works

18

I’ve gotten a sunken piece twice, from two different saves, both at 3001 fish caught. Both times I’ve caught a few sunken pieces before the 3001st fish so it’s prolly safe to assume that pity happens regardless of if you catch a sunken beforehand or not

19

Too add onto this, I’ve heard this from a couple other people as well. Seems too improbable to be pure chance.

20

Real goofy pity system. I’m pretty close to a multiple of 3k fish caught so I won’t really complain.