We have detected that cookies are not enabled on your browser. Please enable cookies to ensure the proper experience.
Results 1 to 25 of 43

Hybrid View

  1. #1
    Join Date
    Jun 2007
    Posts
    305

    Probability of Completing "Fishy, Very Fishy"

    After failing this quest several times, it occurred to me that I should figure out whether it was actually possible before I tried to complete it. So I did some math.

    Short version: This quest, though theoretically possible, is not possible to complete in real life.

    Assumptions:
    * Although we are not given actual fail rate percentages for any but the last type of fish, it seems reasonable to assume that the fish show a progression like 10% fail rate, 20%, 30%, 40%, 50%. However, I tend to fail on the second type of fish, so I think these may be a bit high. For my calculations, I use: 5%, 10%, 20%, 25%, 50%. I believe that these are conservative, which means that the numbers in reality are probably much worse than the numbers you'll see below.
    * I have not gotten past the third tier of fish, so I don't actually know, but I'm assuming it's 10 fish for each tier up to tier 5.
    * There will be rounding errors here because there's no reason to be overly precise when we're making up numbers in the first place.

    Math:

    Round 1 (estimated: 95% success chance per easy fish)
    Chance of catching 10 easy fish in a row = .95^10 = approximately 60%
    Note: This seems low based on my own experience of failing on this round once out of about 10 tries. However, the sample size is too low to be sure.

    Round 2 (estimated: 90% success chance per simple fish)
    Chance of catching 10 simple fish in a row = .90^10 = approximately 35%
    Note: This seems to match my experience where I have only gotten past this round once out of about 8 or 9 tries.
    Cumulative chance of success: .60 * .35 = approximately 21%

    Round 3 (estimated: 80% success chance per possible fish)
    Chance of catching 10 possible fish in a row = .80^10 = approximately 11%
    Note: I have only gotten this far one time, so I don't have a sense of how accurate this is.
    Cumulative chance of success: .60 * .35 * .11 = approximately 2%

    Round 4 (estimated: 75% success chance per tricky fish)
    Chance of catching 10 possible fish in a row = .75^10 = approximately 6%
    Note: It seems like a big jump from 25% failure rate to 50% failure rate, so this is probably a much worse chance in reality.
    Cumulative chance of success: .60 * .35 * .11 * .02 = approximately .04%

    Round 5 (stated in-game: 50% success chance per unlikely fish)
    Chance of catching 10 unlikely fish in a row = .50 ^ 10 = approximately .1%
    Note: The game says there's an equal chance to catch a bad fish or an unlikely fish, so it is 50%. However nobody has gotten far enough to confirm that it requires 10 (at least, nobody that I'm aware of)
    Cumulative chance of success: .60 * .35 * .11 * .02 * .001 = approximately .00005%

    To put that in perspective:
    .00005% in math terms is .0000005 = 5/10,000,000 = 1 winner in every 2 million attempts.

    This year, you will get about 20 opportunities to try to win this. If you try every time it's available, you can expect to win once in the next 100,000 years.

    From another perspective, you are about 4 times as likely to win $10,000 from buying one powerball lottery ticket as you are to complete this quest.

    From yet another perspective, you are 10 times as likely to get killed by an asteroid falling from outer space and landing on your head as you are to complete this quest. (source: http://www.livescience.com/3780-odds-dying.html )


    So, will you be trying this quest again? WHAT IF YOU GET LUCKY?!?!?!?!?!?!?

  2. #2
    Join Date
    Sep 2010
    Location
    Washington
    Posts
    15

    Unhappy

    I have a lucky horseshoe so maybe???? I will do it until the event ends only because it breaks up the monotony of grinding levels as quick as you can before Rohan's release....................... .............but it still doesn't look good. I have reached tier 3 fishing once also.
    [charsig=http://lotrosigs.level3.turbine.com/0e2150000000156eb/01006/signature.png]undefined[/charsig]

  3. #3
    "Never tell me the odds." - Hans Solo
    [charsig=http://lotrosigs.level3.turbine.com/0920d00000004b1a9/signature.png]Killydor[/charsig]

  4. #4
    Join Date
    Jan 2007
    Location
    Boston
    Posts
    2,951
    Quote Originally Posted by killykil View Post
    "Never tell me the odds." - Hans Solo
    "Ja, never tell me ze odds. Zis is absolute verboten!"



    On-topic: so it looks like we're talking about the same as the chances of getting all of the stat times via drops.
    solien
    armor-plated since SoA alpha 3
    arkenstone: hall of fire & into the west

  5. #5
    Join Date
    Oct 2010
    Posts
    107
    Nice analysis, Zelxyb, thanks for that.

    My recommendation to Turbine's chief of development: get someone on your QC team who really understands probability and statistics, and ensure their directions explicitly include looking for, and helping developers with, quests like this one involving probability series.

    Can't really fault the dev who came up with this quest, because prob & stats is often very counter-intuitive. Flip a coin 10 times in a row and have it come up heads? Sure, no problem, might take me a few tries, but I can do that, doesn't sound hard at all. The poor dev probably thought he/she was setting a quest up that would challenge players, but would be beatable by the end of the faire, for those who stuck with it diligently. Little did s/he know, the cumulative chance of success through 50 tosses of the fishing line would be so improbably low that there's a good chance not a single player, on all the worlds/servers, will win over the course of the festival.

    Calls for a dedicated "prob and stats" guru who will help guide devs through issues like this, help them adjust their numbers so that the odds match their expectations better. During the QC/QA process seems the best time to tackle it. Doesn't really need to be any earlier, as the chances should be simple to adjust last-minute.

    Cheers!
    Last edited by Lothran; Aug 07 2012 at 06:20 PM.

  6. #6
    Youre ofc basing this on each tier asking for 10 fishes. We dont know that but you still have a nice analysis going on there.

  7. #7
    Quote Originally Posted by Bicyclerepairman View Post
    Youre ofc basing this on each tier asking for 10 fishes. We dont know that but you still have a nice analysis going on there.
    I can confirm that you need 10 fish for at least the first 4 rounds. Farthest I have gotten was 9/10 tricky fish (and that was with basic pole, low skill, and no horseshoe so I doubt there is any ingame method to improve your odds - have not tried sacrificing a virgin to unholy gods yet). That was also the only time I reached round 4; I usually fail on rounds 1 or 2.

    I agree with Bicyclerepairman that the dev who designed this quest probably did not intend for it to be this impossible but failed to properly calculate the actual probability of getting that many consecutive "wins" from the RNG. I also doubt that doing the quest at high or low ebb for a server would help either.

    As it stands now I think the only way anyone will finish the quest is to have more luck than the gods (if I do finish it, I will immediately hop in my car and go buy a lottery ticket). More realistically it will probably require the devs to realize what they did and change the quest before the faire ends (by severely cutting the number of fish required for each round, or something even more drastic).
    Fortunas Songspinner
    Minstrel and Tailor
    [url=http://www.hallofcrafters.com]Hall of Crafters[/url]
    Arkenstone Server

  8. #8

    Some sample data

    Since the fishing bug lets you keep fishing up quest fish even after the quest is gone, I decided to fish for while when I finally reached tricky fish again (round 4).

    Out of 50 casts (still a pretty small sample size but enough to get a feel):
    38 tricky fish (runs of 2, 1, 3, 2, 4, 9, 4, 12, 1)
    12 bad fish (runs of 1, 2, 1, 1, 1, 2, 1, 1, 2)

    Based on above, Zelxyb's guess of 75% success rate for tricky fish is pretty darn close.

    Given that I was able to get 10 fish once, they could probably fix this quest by breaking it into 5 separate quests where completing an earlier quest unlocked the next in chain (ala the 3 versions of the beer bash quest during the anniversary event). They could then have their 24 hr retry on failure for each of them, but you could pick up again from highest quest on your next attempt. Would take most folks several days to complete but at least doable.
    Fortunas Songspinner
    Minstrel and Tailor
    [url=http://www.hallofcrafters.com]Hall of Crafters[/url]
    Arkenstone Server

  9. #9
    I feel bad for anyone who completes this quest.

    Less than once in a lifetime odds, and I get lucky on a pointless quest in LOTRO, rather than on one of the rare times I buy a Lotto ticket? I'd be furious!

  10. #10
    Join Date
    Jun 2007
    Posts
    305
    Quote Originally Posted by Fortunas View Post
    Given that I was able to get 10 fish once, they could probably fix this quest by breaking it into 5 separate quests where completing an earlier quest unlocked the next in chain (ala the 3 versions of the beer bash quest during the anniversary event). They could then have their 24 hr retry on failure for each of them, but you could pick up again from highest quest on your next attempt. Would take most folks several days to complete but at least doable.
    Assuming they aren't lying about it being a 50% chance to get an unlikely fish and it does require 10 in a row, you would still only complete the last tier once every 1000 attempts, which is still one win every 50 years.

    However, I like the idea in general if they also reduce the number needed per tier.

  11. #11


    "Don't even bother." -Turbine
    [CENTER][B]Yalras - Burglar[/B]
    [SIZE=1]Eldar to Evernight[/CENTER][/SIZE]

  12. #12
    Join Date
    Oct 2007
    Location
    Australia
    Posts
    953
    I assumed that the last round only need one fish, hence pure 50/50? After 4 rounds, it is down to sudden death of either completing or not completing.

  13. #13
    Join Date
    Jun 2010
    Posts
    7,583
    Quote Originally Posted by duvelmoortgat View Post
    I assumed that the last round only need one fish, hence pure 50/50? After 4 rounds, it is down to sudden death of either completing or not completing.
    I dont even know if anyone's finished. Certainly the info on lotro-wiki suggests the best we've done is get to some point in the 4th round (And yes. I've gotten to the fourth round..maybe twice.)
    Crell-L85-Champion - Riddermark ; Swego-L85-Burglar ; Kotvi-L95-Runekeeper
    Delego-L85 Hunter ; Stodden-L51-Captain ; Edhul-L61-Loremaster
    Deglorion - SoA XP Disabler

  14. #14
    Quote Originally Posted by Zelxyb View Post
    From yet another perspective, you are 10 times as likely to get killed by an asteroid falling from outer space and landing on your head as you are to complete this quest. (source: http://www.livescience.com/3780-odds-dying.html )
    Nobody wants to be that lucky LOL

  15. #15
    Join Date
    Dec 2011
    Location
    Grand 'Ole NY
    Posts
    444
    I think the probability also relies on placement. Last year I noticed that few spots have drastically higher probability to get a good catch. Hint: I am usually the only one standing in those spots. Everyone else is no where's close to where I stand.

    Why am I not giving away these locations? Just like in irl fishing, if everyone fishes there, the probability of catching what you want decreases.

    Don't follow the crowd, find a spot where no one is and go for it.

    P.S. My two tries this year (baby to blame for low number) were 46/50 and 38/50 towards completion. Will be giving it another try soon.

  16. #16
    Join Date
    Jun 2011
    Posts
    76
    Newest poll results in 36 tries.
    - Tier 1: 255 good, 16 bad fish
    - Tier 2: 138 good, 9 bad fish
    - Tier 3: 45 good, 10 bad fish
    - Tier 4: 1 good, 1 bad fish
    - Tier 5: 0 good, 0 bad fish

    case 1)
    439 good fish, 36 bad fish, equals 7.5789% chance breaking quest each cast
    Probability finishing all 50 casts in a row: 1.9433%

    case 2)
    - Tier 1: 255 good, 16 bad fish, 5.9041% chance to fail each cast, 54.4138% chance to survive 10 casts
    - Tier 2: 138 good, 9 bad fish, 6.1224% chance to fail each cast, 28.9285% chance to survive 20 casts
    - Tier 3: 45 good, 10 bad fish, 18.1818% chance to fail each cast, 3.8889% chance to survive 30 casts
    - Tier 4: 1 good, 1 bad fish, 50.0000% chance to fail each cast, 0.0038% chance to survive 40 casts
    - Tier 5: 0 good, 0 bad fish, 99.9900% chance to fail each cast, 0.0000% chance to survive 50 casts

  17. #17
    Join Date
    Jun 2011
    Posts
    76
    After two days fishing some updated result.

    Newest poll results in 63 tries.
    - Tier 1: 469 good, 25 bad fish
    - Tier 2: 243 good, 20 bad fish
    - Tier 3: 89 good, 15 bad fish
    - Tier 4: 8 good, 3 bad fish
    - Tier 5: 0 good, 0 bad fish

    case 1)
    809 good fish, 63 bad fish, equals 7.2248% chance breaking quest each cast
    Probability finishing all 50 casts in a row: 2.3529%

    case 2)
    - Tier 1: 469 good, 25 bad fish, 5.0607% chance to fail each cast, 59.4921% chance to survive 10 casts
    - Tier 2: 243 good, 20 bad fish, 7.6046% chance to fail each cast, 26.9752% chance to survive 20 casts
    - Tier 3: 89 good, 15 bad fish, 14.4231% chance to fail each cast, 5.6824% chance to survive 30 casts
    - Tier 4: 8 good, 3 bad fish, 27.2727% chance to fail each cast, 0.2352% chance to survive 40 casts
    - Tier 5: 0 good, 0 bad fish, 99.9900% chance to fail each cast, 0.0000% chance to survive 50 casts

    Assuming, my next try would be a success, the probalilities would change as follows:
    Code:
    Hypothetical results in 64 tries.
    - Tier 1: 479 good, 25 bad fish
    - Tier 2: 253 good, 20 bad fish
    - Tier 3: 99 good, 15 bad fish
    - Tier 4: 18 good, 3 bad fish
    - Tier 5: 10 good, 0 bad fish
    
    case 1)
    859 good fish, 63 bad fish, equals  6.8330% chance breaking quest each cast
    Probability finishing all 50 casts in a row:  2.9048%
    
    case 2)
    - Tier 1: 479 good, 25 bad fish,  4.9603% chance to fail each cast, 60.1243% chance to survive 10 casts
    - Tier 2: 253 good, 20 bad fish,  7.3260% chance to fail each cast, 28.0950% chance to survive 20 casts
    - Tier 3: 99 good, 15 bad fish, 13.1579% chance to fail each cast,  6.8538% chance to survive 30 casts
    - Tier 4: 18 good, 3 bad fish, 14.2857% chance to fail each cast,  1.4671% chance to survive 40 casts
    - Tier 5: 10 good, 0 bad fish,  0.0000% chance to fail each cast,  1.4671% chance to survive 50 casts
    

  18. #18
    Join Date
    Jun 2011
    Posts
    76
    Newest poll results in 98 tries.
    - Tier 1: 742 good, 38 bad fish
    - Tier 2: 364 good, 37 bad fish
    - Tier 3: 106 good, 20 bad fish
    - Tier 4: 8 good, 3 bad fish
    - Tier 5: 0 good, 0 bad fish

    case 1)
    1220 good fish, 98 bad fish, equals 7.4355% chance breaking quest each cast
    Probability finishing all 50 casts in a row: 2.1000%

    case 2)
    - Tier 1: 742 good, 38 bad fish, 4.8718% chance to fail each cast, 60.6866% chance to survive 10 casts
    - Tier 2: 364 good, 37 bad fish, 9.2269% chance to fail each cast, 23.0496% chance to survive 20 casts
    - Tier 3: 106 good, 20 bad fish, 15.8730% chance to fail each cast, 4.0928% chance to survive 30 casts
    - Tier 4: 8 good, 3 bad fish, 27.2727% chance to fail each cast, 0.1694% chance to survive 40 casts
    - Tier 5: 0 good, 0 bad fish, 99.9900% chance to fail each cast, 0.0000% chance to survive 50 casts

    I will update my statistic after the end of the farmer's faire.
    Honestly, I think Turbine should change that quest, either lowering the fail-chance, or introduce steps, so you don't have to restart from scratch, lower the chances, if you fished enough fishes, allow to fish more than one bad fish, etc. Something should be done so that quest is finishable, which is in my opinion virtually impossible.

  19. #19
    Quote Originally Posted by Peresbert View Post
    I think the probability also relies on placement.
    Your hypothesis has merit. If you wish to test it, I suggest sharing your spot with a few trustworthy people on other servers and getting a lot more data. In my own experience, location has little or no affect on results.

 

 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

This form's session has expired. You need to reload the page.

Reload