Fishing

[cdgreg]

http://i47.tinypic.com/jkf1co.png

30 Caught/71 Broke/ 204 Pugils used** / 4 Trips
1/29 shawl

Food for thought:
7catch/17break 43 bait (first trip 106 skill + stall + clothes) 29% 7/24
11catch/34break 94 bait (second trip 107 skill + stall + clothes) 24% 11/45
3catch/9break 30 bait (third trip 108 skill + stall + clothes) 25% 3/12
9catch/11break 37 bait (fourth trip 109 skill +stall + clothes + The Big One + Net & Lure) 45% 9/20

On paper it would seem that The Big One and Net & Lure drastically improved break rates. I know someone did some testing previously and showed similar results.

Though obviously 101 bites is a small sample size.

God damnit Urat, please stop with the small sample size-I stumbled onto a major discovery bullshit. I've seen it a number of times on the last couple pages, this is how terrible rumors start, and then blossom into accepted truths. Next up, overwhelming water energy really did help!

edit: the other person I was thinking of that posted how terrible dark day is for line breaks was you as well...

just stop

[Karah]

Just posting stats homey, chillax. I really don't contribute anything worthwhile so I just post what I come up with.

I keep literal track of everything, every single fish ive caught since 0 skill. Forgive me for assuming some relevant info was worth posting.

[cdgreg]

Just posting stats homey, chillax. I really don't contribute anything worthwhile so I just post what I come up with.

That is the point, what you are posting is not worthwhile. There are a number of folks who scan these threads rather than read them carefully. They then take those worthless claims and start turning them as truth to lsmates and other forums. Before you know it, we are attempting to refute their absurd claims while their reply is "NO, I'M RIGHT UNTIL YOU PROVE ME WRONG BRO."

if you truly believe you've stumbled onto something and want help designing a test let us know, or come back and post once you have a reasonable sample size. Simply posting X day is terrible for line breaks or X furniture drastically helps line breaks with no evidence to support it is counter productive to everything we attempt to do here.

[Karah]

Quite clearly stated the sample size was indeed small, and that there was prior testing done with a much larger sample.

07-29-2012 04:42 PM #7266

Join Date
Apr 2011
Posts
3,360
FFXI Server
Bahamut

Default

Hakryu parses

All right to preface this all of this was fishing using sinking minnow (my minnow sample size is too small to share as I only get them when sinkers are high which almost never here lol), and as you'll see from the sample size not all my characters were all fishing at the same time. But my mules were almost always there together and if my main was there both my mules were. Hope that doesn't really throw things off too much. Also sorry I don't know how to do tables and shit on here.

Character 1:
Ebisu, puffin ring, noddy ring, fishermans smock, crafted fishing gear, megamoglification and 100 skill

Casts: 28207
Catch: 2263
Breaks: 1255
hakryu bites/casts: 12.47%
hakryu catch/casts: 8.02%
break/bite: 35.67%

Character 2:
Everything character 1 had but with 110 skill

Casts: 26202
Catch: 2460
Breaks: 1240
hakryu bites/casts: 14.12%
hakryu catch/casts: 9.39%
break/bite: 33.51%

Character 3:
Same as character 2 except: Waders instead of crafted boots, and in addition to the stall a signboard and the big one.


Casts: 17665
Catch: 1803
Breaks: 731
hakryu bites/casts: 14.34%
hakryu catch/casts: 10.21%
break/bite: 28.85%

I don't know enough about statistics to know if those are big enough differences or if I need a bigger sample size but it does appear that 100-110 skill does provide a decent boost in bite rate and decrease in break rate. And that either the small skillboost from waders or the extra furniture helps with break rate. If someone with more statistics knowledge then me could help better interpet the data that would be swell thanks.

[cdgreg]

^ I would not refute that an increase of 10 skill probably has a positive effect on bite rate and decrease in break rate. I would however refute that adding 1 skill and 2 pieces of furniture which have not been proven to add any effect when coupled with stall decreases line break rate 20%. Agree or disagree on that?

edit: you could test fairly easily if adding the big one/net and lure set adds skill when coupled with stall.

[Karah]

I did not blindly say that it would show a 20% increase, as i stated, merely posted what happend.

There is a sizable difference between claiming ZOMG I PROMISE YOU'LL NEVER BREAK YOUR LINE IF YOU BUY THIS!!!!!!!!!!

And posting what literally happened.

Just wanted to get the point across that when -I- added (things) -I- saw an improvement, if anyone were to ask if they were (worth) getting, I'd have to say, They are.

[cdgreg]

Are you sure you saw an improvement, or did you simply see a small sample size? I just flipped a coin 3 times while standing on my left foot, all 3 times it landed on heads. What can you conclude from this?

(other than I'm stupid for flipping a coin while standing on one foot!)

[Karah]

Regardless of the size of the sample, it's still valid. Just not as valid as a 1000, or a 10,000 size would be. Which again, I did state. I can't emphasize this enough.

100 is enough to establish a percentage, and in a semi (idealistic) way 100 (bites) should be enough to obtain anyone's shawl.

Had this been a fish that we catch 200 daily, like haks, liks, gugs, I wouldn't bother posting it.

As for comparing your coin flipping, apples meet oranges. We already know that matsya break rate is obscenely poor, any chance of improvement should be welcomed.

[cdgreg]

Regardless of the size of the sample, it's still valid. Just not as valid as a 1000, or a 10,000 size would be. Which again, I did state. I can't emphasize this enough.

100 is enough to establish a percentage, and in a semi (idealistic) way 100 (bites) should be enough to obtain anyone's shawl.

Had this been a fish that we catch 200 daily, like haks, liks, gugs, I wouldn't bother posting it.

As for comparing your coin flipping, apples meet oranges. We already know that matsya break rate is obscenely poor, any chance of improvement should be welcomed.

1.) It is not valid to draw conclusions from

2.) How is 100 bites enough to obtain anyone's shawl? You can obtain a percentage from 1 data point, it does not make it valid to draw conclusions from.

3.) So decreasing line break rate is only worth posting on a certain fish?

4.) My point back to #1

[Karah]

1.) It is not valid to draw conclusions from

2.) How is 100 bites enough to obtain anyone's shawl? You can obtain a percentage from 1 data point, it does not make it valid to draw conclusions from.

3.) So decreasing line break rate is only worth posting on a certain fish?

4.) My point back to #1

Clearly you just want to argue. But I'll indulge you anyway.

1) It was not a test, for the 5th(?) time? just results from 4 days attempts, valid enough. closeenough.jpg

2) I said the word "ideally" you should look this word up in the dictionary. Most people obtaining shawls do so in a "reasonable" (look this word up too) amount of fish. Obviously, there are outliers.

3) It was stated in again reference to size, had it been the aforementioned fish, a sample that small I wouldn't post. But being the most sought fish (atm) and the most difficult to obtain fish (again atm) some info > none, yes?

4) Point 3 re-iterated, anyone who cared about your coin flipping test, would be interested, I am not. Interested in helping people who are fishing (this thread is about fishing) catch the fish.

[cdgreg]

Meh, enough of that bitchfest though, I'll gladly agree to disagree.

[Corkadile]

People making claims off statistically insignificant data pools make scientists cry.

Please, think of the scientists.

[Karah]

No claim was made, they're all good~

[Uselessaccount2]

2) I said the word "ideally" you should look this word up in the dictionary. Most people obtaining shawls do so in a "reasonable" (look this word up too) amount of fish. Obviously, there are outliers.

Just wanted to know how you know "most" people obtain that annoying cape in a "reasonable" amount of fish. Also, how much is "reasonable"?

Anyway when my mule was level 102, with AH fishing gear and the stall, I pulled up 50 matsya with about 400 bait.
Fishing at the same time was my main, level 110 fishing with apron, waders, other AH gear, and the stall. With 400 bait I got 21 matsya.

Conclusion: Matsya sucks ass and random is random.

[Karah]

A'ight, I know better than to bother. Y'all win the internet. Do not buy any furniture for fishing and forget I posted at all. catch/break/quit idc. I got my shawl. I'm good. *washes hands*

[Tyrath]

.

Conclusion: RNG sucks ass and random is random.

FIFY

[nibu]

That is the point, what you are posting is not worthwhile. There are a number of folks who scan these threads rather than read them carefully. They then take those worthless claims and start turning them as truth to lsmates and other forums. Before you know it, we are attempting to refute their absurd claims while their reply is "NO, I'M RIGHT UNTIL YOU PROVE ME WRONG BRO."

if you truly believe you've stumbled onto something and want help designing a test let us know, or come back and post once you have a reasonable sample size. Simply posting X day is terrible for line breaks or X furniture drastically helps line breaks with no evidence to support it is counter productive to everything we attempt to do here.

Well, thank God we nipped that one in the bud then and avoided a lengthy, angry, unnecessary debate!

[barber2006]

I don't think there is anything wrong with posting stats. If 50 people post stats with an insignificant sample size then all of a sudden you have some real numbers to work with. That's how we got zaldon gut rates. One person didnt fish up 100,000 lik and get 1,000 silks. The information was compiled from a lot of users.

Way back during rusty caps I remember I posted something about it might not matter if you had one rusty bucket or 100 in your moghouse so long as you had the moghancement. I posted some stats from 8 days of fishing (4 faint, 4 overwhelming) showing very little difference and got totally raked over the coals for a 'small sample size'. And that was on Alla where the most brazen witchcraft theories prevail. Two months later I had a large enough sample size to convince myself there was no difference when it came to pulling up rusty caps but I wasn't about to post anything then. Those guys could kill their mog space with buckets while I just kept one for all I cared. That was what? 2005? I haven't posted stats since then because they are always shot down for insignificance. They are insignificant, but with input from other users it becomes significant. Do I think net and lure does anything? No, I bought it for completion sake. But in a 400 page thread it wouldn't hurt to have more people posting numbers and less people bitching about zaldon. Maybe, just maybe, with an environment of encouragement somebody might stumble onto something we otherwise would not have known.

[sluggo]

Wait... greg... weren't you the one who said ebisu had 9000some skill????!!

[dasva]

Yeah there was a reason that I didn't post those parses until I had like 20k casts each lol. Because I knew that not only would it be meaningless but that this would happen. I am currently actually adding to them even more. And while my data does suggest that either 1 skill, extra furniture or both help break rate we are talking rather small amounts that you could probably argue isn't worth the effort of getting them. At least not the gp ones