• Play
  • About
  • News
  • Forums
  • Yppedia
  • Help
Welcome Guest   | Login
  Index  | Recent Threads  | Register  | Search  | Help  | RSS feeds  | View Unanswered Threads  
  Search  


Quick Go »
Thread Status: Normal
Total posts in this thread: 181
Posts: 181   Pages: 7   [ First Page | Previous Page | 1 2 3 4 5 6 7 | Next Page | Last Page]
[Add To My Favorites] [Watch this Thread] [Post new Thread]
Author
Previous Thread This topic has been viewed 23908 times and has 180 replies Next Thread
Gillie017

Member's Avatar


Joined: Aug 3, 2006
Posts: 1043
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

You introduced a LOT of bias into your sample.

1. You used only one point in time. It is unlikely that this is representative of the activities of all pirates. You should have sampled multiple times over different times of the day and days of the week.

2. You sampled only pirates on ships. The stats of the pirates in your sample are therefore biased towards the type of voyages out at the time.

3. You sampled from one archipelago only. The activities in different archipelagos can be vastly different at any given time.

4. You got bored and stopped collecting data half way through a frig. If this had been a SMH your results would have been very different had you completed the entire ship.

Your sample was in no way representative of the population of pirates. If you were to come up with a sensible sampling scheme to address the biases above, your results would be worth looking at but as they stand, they are meaningless.

The only time you considered potential bias, you were wrong. No bias can be introduced by checking ships in alphabetical order since ships are named randomly.
----------------------------------------
If a bird looks like a duck, swims like a duck and quacks like a duck, then it's probably a Huntard .
[May 9, 2009 8:08:46 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
wrs1864b

Member's Avatar


Joined: Apr 10, 2006
Posts: 6314
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

Scervy wrote: 
how did this thread deviate from a discussion on whether or not we should hold Monarch meetings again to arguing how to determine the "average" pirate

It goes to the question: Do the monarchs have an accurate perception of what the average pirate is like, or are they biased toward the problems of the elite? If the typical monarch (and forum poster) thinks that they average pirate is about a Master, when it could very easily be that the average pirate is an Able, you can tell that they aren't good representatives.

Duckling wrote: 
1. You used only one point in time. It is unlikely that this is representative of the activities of all pirates. You should have sampled multiple times over different times of the day and days of the week.

I may do that, but I don't think that saturday evening is that out of whack. You don't see a lot of people commenting on how unusual that is.

 
2. You sampled only pirates on ships. The stats of the pirates in your sample are therefore biased towards the type of voyages out at the time.

Again, that was deliberate because people were explicitly saying that you can't sample the whole population because that doesn't represent jobbers. So, I sampled jobbers. Again, if we want to know the entire population, well, we have that. Able is 50% of the whole population.

 
3. You sampled from one archipelago only. The activities in different archipelagos can be vastly different at any given time.

On other oceans, yeah, but not so much for Viridian/Cobalt because the archs are very large (only three on the ocean) and laid out to be symmetrical.

 
4. You got bored and stopped collecting data half way through a frig. If this had been a SMH your results would have been very different had you completed the entire ship.

As I said earlier, I intended to sample about 100 pirates because you really don't need a much larger sample than that to get good results. It was well over 10% of the online population and big enough based on things like the law of large numbers and the central limit theorem. Yes, it surprises a lot of people that you can get accurate measurements from when you poll only a few thousand registered voters for a country the size of the USA, but it works.

 
The only time you considered potential bias, you were wrong. No bias can be introduced by checking ships in alphabetical order since ships are named randomly.

I was wrong? What? I didn't think there would be a bias based off the names, and you are agreeing with me, right? And as I have said a couple of times now, I considered arch bias when I chose Viridian. (That, and Gothmog wanted a green ocean.)
----------------------------------------
Algol can not assert the truth of all statements in this post and still be consistent.
----------------------------------------
[Edit 1 times, last edit by wrs1864b at May 9, 2009 8:41:57 PM]
[May 9, 2009 8:29:37 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Burnt_Water

Member's Avatar


Joined: May 16, 2006
Posts: 3092
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

Monarchs feel special for getting feedback from the almighty even if they're just repeating GD's voice.

People in GD don't feel as special without the constant feedback but have a desire to change how the game works, usually this can result from a short complaint/suggestion to a large detailed complaint/suggestion to the game mechanics. Also you're unlikely to get feedback to this compared to conferences.

Shockingly, the almighty have the final decision to game mechanics. Despite that, they would have felt pressured to implement few irrational suggestions from the monarchs of 'powerful' big flags compared to the 'vast and organised' information posted since the dawn of PP.

Why else do you think GD Forum, surveys and GD threads seem better for the game?


I don't mind another conference, but they might.
----------------------------------------
Hard and big guide to whack off Brigands
How to make a pirate feel good
[May 9, 2009 8:56:36 PM] Show Printable Version of Post        Send Private Message    Bored en route? Bring a book or puzzle next time. [Link]  Go to top 
vnork



Joined: Jul 22, 2006
Posts: 1004
Status: Offline

Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
As I said earlier, I intended to sample about 100 pirates because you really don't need a much larger sample than that to get good results. It was well over 10% of the online population and big enough based on things like the law of large numbers and the central limit theorem. Yes, it surprises a lot of people that you can get accurate measurements from when you poll only a few thousand registered voters for a country the size of the USA, but it works.

This is where you're wrong. Your method of choosing 100 pirates is very inefficient, and is open to a lot of sampling error. Therefore, you need a sample of much larger than 100 to get accurate results. The polls you're referring to use much more efficient sampling methods than you used, so that comparison is invalid.

Illustrative example: Let's say Gallup decided to do a presidential poll by randomly selecting an American, then polling 5,000 people who live closest to that random American. This is a form of random sampling, but it's prone to error. You would have to repeat this many times to achieve the accuracy of individually selecting 5,000 people at random from across the nation. Not all forms of random sampling are made equal.

Now imagine a /w list of ships in an archipelago, expanded to a list of jobbers. Just as an example, let's say there's 200 jobbers on ships on Viridian. There's a SMH with 75 pirates that's screened to where 95% of jobbers are above able. The rest of the ocean is about randomly distributed with only 25% above able. You sample the first 100 jobbers on the list of ships. Keep in mind that the 75 jobbers on the SMH are clumped together on the list. Even though your sample is half the population, it is pretty much only representative of the entire population when the frame samples roughly 35 SMH jobbers and 65 other jobbers. Any other time, you will have large amounts of sampling error. Twenty percent of the time, a data point will report that 75% of pirates are able, when in fact only about 50% are. Another twenty percent of the time, it will report that 22% of pirates are able. To get accurate statistics, you would have to repeat the sampling process many times. This is not an efficient sampling method.

Instead, imagine having a list of jobbers taken from the /w list of ships. By randomly selecting every 4th jobber along the entire list, you can get a random sample with a much, much, much lower margin of error with only one sample of 50 jobbers. This is a much more efficient sampling method.

-Vnork
----------------------------------------
[Edit 8 times, last edit by vnork at May 10, 2009 12:16:40 AM]
[May 9, 2009 8:57:46 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Gillie017

Member's Avatar


Joined: Aug 3, 2006
Posts: 1043
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
 
1. You used only one point in time. It is unlikely that this is representative of the activities of all pirates. You should have sampled multiple times over different times of the day and days of the week.

I may do that, but I don't think that saturday evening is that out of whack. You don't see a lot of people commenting on how unusual that is.

You "think" it is no different but you don't know for sure. I "think" it could be quite different but I don't know for sure. The multiple sampling I suggested would eliminate any potential bias from choosing just one point in time.

 
 
2. You sampled only pirates on ships. The stats of the pirates in your sample are therefore biased towards the type of voyages out at the time.

Again, that was deliberate because people were explicitly saying that you can't sample the whole population because that doesn't represent jobbers. So, I sampled jobbers. Again, if we want to know the entire population, well, we have that. Able is 50% of the whole population.

People were saying that you cannot simply sample all pirates who have been active over the past 10 days because this give equal weighting to greenies who log on and never come back, labour alts and addicts who are on all the time. This is the flaw with the numbers generated by Lizthegrey. Sampling from those online at a particular point in time overcomes those flaws. You sampled jobbers only and biased your results towards those on ships which are in no way representative of the "average" pirate.

 
 
3. You sampled from one archipelago only. The activities in different archipelagos can be vastly different at any given time.

On other oceans, yeah, but not so much for Viridian/Cobalt because the archs are very large (only three on the ocean) and laid out to be symmetrical.

The uniformity of the arches is irrelevant. You sampled people who were in the arch at the time, not those who call it their home. If a popular SMH chart had one or two frigs on the board in a different arch to where you did your sample then there were a lot of pirates who could not have been counted by you. Similarly if the arch you were in had a lot of greenie WB pillages out, you over-represented them in your sample.

 
 
4. You got bored and stopped collecting data half way through a frig. If this had been a SMH your results would have been very different had you completed the entire ship.

As I said earlier, I intended to sample about 100 pirates because you really don't need a much larger sample than that to get good results. It was well over 10% of the online population and big enough based on things like the law of large numbers and the central limit theorem. Yes, it surprises a lot of people that you can get accurate measurements from when you poll only a few thousand registered voters for a country the size of the USA, but it works.

It only works if the sampling frame is not biased.
 
 
The only time you considered potential bias, you were wrong. No bias can be introduced by checking ships in alphabetical order since ships are named randomly.

I was wrong? What? I didn't think there would be a bias based off the names, and you are agreeing with me, right? And as I have said a couple of times now, I considered arch bias when I chose Viridian. (That, and Gothmog wanted a green ocean.)


I said before the sampling scheme is extremely important. Your sampling failed in a number of important ways, this is why your results are meaningless.
----------------------------------------
If a bird looks like a duck, swims like a duck and quacks like a duck, then it's probably a Huntard .
----------------------------------------
[Edit 3 times, last edit by Gillie017 at May 9, 2009 9:05:30 PM]
[May 9, 2009 9:01:42 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
basso

Member's Avatar


Joined: Mar 19, 2006
Posts: 3156
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

Sorry slight derail/question here. Able is half of all total pirates, or all pirates currently active in a puzzle? I mean, I have a raw score and standing for sails, even if I am currently dormant in sails. So if I am ranked respected yet I am dormant, am I taking up a spot or not?
----------------------------------------
Montage of Sage
Mads wrote: 
OK, now I'm convinced. The problem here is that you cannot understand plain English.

[May 9, 2009 10:27:51 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Gillie017

Member's Avatar


Joined: Aug 3, 2006
Posts: 1043
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
Sorry slight derail/question here. Able is half of all total pirates, or all pirates currently active in a puzzle? I mean, I have a raw score and standing for sails, even if I am currently dormant in sails. So if I am ranked respected yet I am dormant, am I taking up a spot or not?


It depends what you are measuring.

The percentages that Lizthegrey did (and are quoted somewhere above with 50% ables) were based on all pirates who were non-dormant for that puzzle. So your dormant respected pirate would not take up a spot. Lists such as this include everyone who has played that puzzle in the past 10 days - this includes recent greenies who log on, do the basic tutorials and never come back.

The "average" pirate is better measured by taking a random cross-section of pirates online. The correct sampling for which has been discussed ad nauseam above.
----------------------------------------
If a bird looks like a duck, swims like a duck and quacks like a duck, then it's probably a Huntard .
----------------------------------------
[Edit 1 times, last edit by Gillie017 at May 9, 2009 10:54:42 PM]
[May 9, 2009 10:50:47 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Gothmog1065

Member's Avatar


Joined: Apr 12, 2004
Posts: 11593
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

If viridian is anything like Cobalt, their arches are NOT uniform (Garnet is much smaller than Onyx, which is smaller yet than Jade).

If you are gathering data for statistics, then you cannot use one sample set and call it the end. Hell, data 15 minutes from when you took that sample would be different.

Also, something your data will never show (Because you don't have access to their creation date), is that two trends will occur:

As the pirate age increases,
- The Population will decrease
- Average stats will rise

How do you know that out of 100 of those pirates, 20 of them aren't pure green, was made just before you took your sample?

We can argue statistics all day long, but until someone actually runs multiple samples from multiple days from multiple areas, then your statics will always be lacking in information.

Also Liz's numbers are hard fact. But those numbers again don't include those who do the puzzle and leave, and other facts.
----------------------------------------
Nemesis wrote: 
You ruined my joke, Goth.
Arakael wrote: 
Goth is right, as usual.
Blobbles wrote: 
Goth is right. *shudders*

[May 10, 2009 8:01:06 AM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Grinfish

Member's Avatar


Joined: May 24, 2004
Posts: 7984
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 

As the pirate age increases,
- The Population will decrease
- Average stats will rise

I didn't think it was possible for Average stats to rise, as the system always makes sure that the lowest 50% of active players in a puzzle are Able, and so on. If you're including mothballed pirates then yeah, on paper there would be more. However, until they're active again, they don't "count" to the live population, and the population's stats will adjust to suit.
----------------------------------------
Whitewyvern. Distinctly Limey.
Retired, No regrets.
Unretired, still No regrets.
Available in Cerulean and Obsidian flavours.

Briggs wrote: 
StuManchu puts the "sensual" back in "Nonconsensual"

[May 10, 2009 8:44:24 AM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Gothmog1065

Member's Avatar


Joined: Apr 12, 2004
Posts: 11593
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
 

As the pirate age increases,
- The Population will decrease
- Average stats will rise

I didn't think it was possible for Average stats to rise, as the system always makes sure that the lowest 50% of active players in a puzzle are Able, and so on. If you're including mothballed pirates then yeah, on paper there would be more. However, until they're active again, they don't "count" to the live population, and the population's stats will adjust to suit.

No, as active pirate age increases (Last 10 days), then the number of pirates in that set will decrease, and the average stats of that set will increase.

Fictional numbers. You poll 1000 pirates.

Overall stats:
~ 50% are < distinguished (Able).
500 pirates have been playing for < 1 year
300 have been playing 1yr < x < 2 years
200 have been playing > 2 yrs.

I bet you that out of the 1 yr or less, more than 50% of them are able.
I bet you that out of the 1<x<2 pirates, the average stat moves up to respected or even higher.
I bet you that out of the >2yr pirates, the average stat moves up to GM or so.

Do I have to start making graphs?
----------------------------------------
Nemesis wrote: 
You ruined my joke, Goth.
Arakael wrote: 
Goth is right, as usual.
Blobbles wrote: 
Goth is right. *shudders*

[May 10, 2009 9:46:31 AM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Grinfish

Member's Avatar


Joined: May 24, 2004
Posts: 7984
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

Nah, no graphs required, but I disagree with your betting strategy ;) There may well be a higher level of skill in older pirates, but not as high as you're claiming IMHO.
----------------------------------------
Whitewyvern. Distinctly Limey.
Retired, No regrets.
Unretired, still No regrets.
Available in Cerulean and Obsidian flavours.

Briggs wrote: 
StuManchu puts the "sensual" back in "Nonconsensual"

[May 10, 2009 10:55:13 AM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
wrs1864b

Member's Avatar


Joined: Apr 10, 2006
Posts: 6314
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
If viridian is anything like Cobalt, their arches are NOT uniform (Garnet is much smaller than Onyx, which is smaller yet than Jade).

The size of the population isn't important, it is the ratios of Ables to Ultimates. When an arch first opens, there is usually a rush by elites of the colonizing flags to set up shop there, thus shifting the curves. Where the arch goes from there depends on the difficulty of the routes, commodities, location in the ocean, etc. I'm claiming that, due to the three archs of Cobalt/Viridian being so similiar in structure and not being recently colonized, that their ratios are very similar.


OK, I did another survey, this time all three archs of Viridian and all ships in all archs (about 400 pirates). I'll spare you the details, but the results are pretty much the same as the first set I reported. Ables account for around 40-45%, distinguished another 15-25%. So, ok, the average jobber is not Able, they are closer to being a low Distinguished. They certainly are not Master, or even close to it.

I may well do another survey or two, but I probably won't post the results unless they are significantly different than the first two surveys.

In case anyone is interested, the perl script that snarfs the stats can be found here. You also need to have wget installed, or some other way of grabbing the raw HTML.
----------------------------------------
Algol can not assert the truth of all statements in this post and still be consistent.
[May 10, 2009 12:00:27 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Gillie017

Member's Avatar


Joined: Aug 3, 2006
Posts: 1043
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

The average jobber is not the same as the average pirate. I don't understand why you insist on only sampling pirates on ships. they aren't representative of the population of pirates.

Rather than spare us the detail, some more information would be useful. For example: What size ships were included in those you counted? What type of voyages were out? etc.
----------------------------------------
If a bird looks like a duck, swims like a duck and quacks like a duck, then it's probably a Huntard .
[May 10, 2009 12:36:44 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
vnork



Joined: Jul 22, 2006
Posts: 1004
Status: Offline

Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

That's much better data, but I have another concern: are you counting Novice/able stats? A lot of jobbers are have something like broad/able in one stat, and novice/able in everything else. If someone is novice/able in a stat, I think it should be excluded as they probably haven't even tried that puzzle.

-Vnork
[May 10, 2009 12:56:55 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
tanonev

Member's Avatar


Joined: Nov 18, 2006
Posts: 2361
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
The average jobber is not the same as the average pirate. I don't understand why you insist on only sampling pirates on ships. they aren't representative of the population of pirates.


I'm very confused by this statement.

wrs1864b's claim: At least 50% of (visible) jobbers are Able or Distinguished in bilge. One data point backing this up.
your claim: The claim is false for (visible) pirates.
a couple of steps of logic on your claim: Less than 50% of (visible) pirates are Able or Distinguished in bilge.
Pirates not on ships must be better on average at bilge than pirates on ships.

Can you explain to me why you believe the last statement is true? All three of those statements are equivalent statements of your objection, and that last version seems very odd. Not impossible, of course, but a very counterintuitive claim.

Yes, jobbers aren't representative of the population of pirates. However, wrs1864b's claim is about a mathematical inequality, and I suspect your objection is actually in the direction that helps that inequality.

@wrs1864b: Just checking to make sure you're throttling your pirate stat lookups. I forget how long we're supposed to wait between searches...
Also, out of curiosity, how are you grabbing the pirate names from /w and /vw? Are you just writing them down by hand, or do you have a way of dumping them from the client into a file?
----------------------------------------
Tanonev on all oceans; currently exploring Meridian.
Puppetar by Tilinka
[May 10, 2009 1:21:48 PM] Show Printable Version of Post        Send Private Message    http://www.alpha-slash.com [Link]  Go to top 
Squashbuckle

Member's Avatar


Joined: Dec 21, 2004
Posts: 6276
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

The average pirate is also not all-able. I think when people say "the average pirate is master," they further imply "in the duty puzzle they are likely to be playing."

I'm a distinguished THer, not because I'm no good in TH, but because I don't enjoy SMH/flotilla enough to ever play that puzzle. Similarly, I have a lousy stat in TD, able spades, and able hearts, but I don't play those puzzles. In the puzzles I choose to play, my skills are much higher. I suspect that most pirates put time into gaining skill in puzzles they enjoy and are thus likely to be at least master in whichever puzzle they play most often (assuming they are also at least solid - it's difficult to get to master before solid in duty puzzles), even if many other stats are able.
----------------------------------------
Polly
Hunter
Captain of Flocktarts

Avatar by Hedvvig
[May 10, 2009 1:35:37 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Gillie017

Member's Avatar


Joined: Aug 3, 2006
Posts: 1043
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

My concern is that we don't know what types of voyages were counted.

If, for example, there were no SMH out at the time then the pirates with the higher stats would be unlikely to be out jobbing on regular pillages, they'd be standing around on land waiting for a SMH to load and therefore not be included in the sample.

Taking a random cross-section of ALL pirates puts all these concerns to rest and will give information about the "average"pirate rather than the "average jobber" - and it was the notion of the average pirate that sparked this discussion.
----------------------------------------
If a bird looks like a duck, swims like a duck and quacks like a duck, then it's probably a Huntard .
[May 10, 2009 1:36:43 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Fiddler

Member's Avatar


Joined: Mar 4, 2004
Posts: 4619
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

I don't understand; shouldn't a random sampling of an ocean's entire population show a distribution really close to the one detailed by LiztheGrey? That is, 50% of the sample ought to be able.
----------------------------------------
Orsino, Viridian ocean
Forum-mute: your best friend
Whitewyvern wrote: 
The only high end goals are those you set for yourself. What happens to anyone else is irrelevant.

[May 10, 2009 2:06:21 PM] Show Printable Version of Post        Send Private Message    Avatar by Stimmhorn [Link]  Go to top 
vnork



Joined: Jul 22, 2006
Posts: 1004
Status: Offline

Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
I don't understand; shouldn't a random sampling of an ocean's entire population show a distribution really close to the one detailed by LiztheGrey? That is, 50% of the sample ought to be able.

That is solely used to determine active pirates that count towards the scoring curve. In determining the average pirate, there's no reason why you should consider pirates that have puzzled once in the last 10 days instead of some other arbitrary time frame. With a longer sampling frame, you're weighting greenies that logged on once to try the game and neophyte/able alts equally with more dedicated pirates.

-Vnork
----------------------------------------
[Edit 2 times, last edit by vnork at May 10, 2009 2:15:59 PM]
[May 10, 2009 2:14:42 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
abacadafa

Member's Avatar


Joined: Jun 13, 2006
Posts: 493
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
I don't understand; shouldn't a random sampling of an ocean's entire population show a distribution really close to the one detailed by LiztheGrey? That is, 50% of the sample ought to be able.


Not necessarily.

Liz's search grabs all recent pirates, online or not. A survey based on who's online at a moment is far more likely to grab people who play a few hours a day.

If people with higher rankings play more often than people with all-ables, you could reach a situation where 50% of all pirates are able but far less than 50% of the pirates online at any moment are able.
----------------------------------------
.
[May 10, 2009 2:18:49 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Gothmog1065

Member's Avatar


Joined: Apr 12, 2004
Posts: 11593
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

I need to find someone to write me a script.

Data points include:
- Total online for each instance of data collected.
- Total on ships.
- Going island to island and doing a /who there.
- Pulling trophy information (Namely jolly rogers to get a timeframe)
- Publishing a table and a graph that shows the percentage of ranks as a function of time (in years, there's no other way to tell how long that pirate has been around).

Stats can be averaged by placing a numerical value to each stat (1 Able, 2 Distinguished, 3 Respected, 4 Master, 5 Renowned, 6 GM, 7 Leg, 8 Ult). Infact it would be best to only include piracy stats (As those are pretty much the standard of "how good" a pirate is, and some don't even touch shoppe puzzles, and some don't mess with most parlor games, etc). Chop chop programmers.
----------------------------------------
Nemesis wrote: 
You ruined my joke, Goth.
Arakael wrote: 
Goth is right, as usual.
Blobbles wrote: 
Goth is right. *shudders*

[May 10, 2009 2:44:47 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Kotetsu534



Joined: Sep 7, 2007
Posts: 1406
Status: Offline

Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

A better way to grade would be any skill with a certain level of experience. That way those who don't play, for example, DNav don't get their rating artificially lowered.
----------------------------------------
Nomura, SO of Innocent, Member of Crimson Tide, Midnight.
[May 10, 2009 2:53:16 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
wrs1864b

Member's Avatar


Joined: Apr 10, 2006
Posts: 6314
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

OK, remember folks, the whole point of this "average pirate" and "Master" thing was the assertion that monarchs and forum posters have a very skewed perception of what an "average pirate" is. That they "think master is an average stat." This proposition wasn't well specified, trying to dig too deeply into proving/disproving it via technicalities misses the point.

The point is, there are lots of players that find YPP, puzzle a bit, maybe even start their own crew and never make it much past solid and often not past solid/respected. They then leave the game before they are much noticed by anyone. These are people OOO needs to care about and cater too also, in particular, how can we keep them from leaving the game? And, there are a lot more of these people than I think most old-timers imagine. Talking about how "if you puzzle enough, you will get up to master" is missing the point too, OOO needs to think about all those people who won't puzzle enough because they don't find YYP interesting enough.

duckling wrote: 
The average jobber is not the same as the average pirate. I don't understand why you insist on only sampling pirates on ships. they aren't representative of the population of pirates.

I'm quite certain that if I had listed all pirates on the ocean, people would tart about how shopkeepers/parlor tarts are irrelevant to the bilging standing. The subject also involved duty reports, after all.

If there was an easy way of limiting the bilging (or sails or carp) stats to only those that were actually doing that puzzle on a ship, that would be even better, but I can't think of a way to do that. However, looking that the sails/bilge/carp stats for jobbers on a ship seems fairly reasonable to me.

Another useful stat would be what the typical pirate who applies for a pillaging job looks like. I can't get that information either, but these stats are what people apparently accept for jobbers. If some pirates can't get on ships because of their stats, well, that probably means that the average jobber is even worse.

If you want to see different stats, well, go collect them.

 
Rather than spare us the detail, some more information would be useful. For example: What size ships were included in those you counted? What type of voyages were out? etc.

I don't know of a way to tell the size of a ship from stuff like /vwho. Stuff on notice board pops on and off far too quickly for me to be able to gather that kind of data. There was at least one WF, and several other "larger" ships though.

Vnork wrote: 
That's much better data, but I have another concern: are you counting Novice/able stats?

Yes, I'm including novice/able mostly because it is easier, plus I'm not sure what else to exclude too. Is neophyte that much better? throw out everything below broad? Mind you, OOO thought getting to broad was such an achievement that they give out a trophy for it. All those players that pass through YPP without disturbing the elite realms often don't make it to broad.

I posted the script I used, I can PM the list of pirates I ended up checking, you can reproduce the data I have and play with it any way you want.

tanonev wrote: 
Just checking to make sure you're throttling your pirate stat lookups. I forget how long we're supposed to wait between searches...
Also, out of curiosity, how are you grabbing the pirate names from /w and /vw? Are you just writing them down by hand, or do you have a way of dumping them from the client into a file?

I've been typing the /vw <shipname> by hand, and then typing the names of the pirates on the ship by hand into my script (actually a wrapper script around the perl script.) As such, I'm not worried about putting too much load on OOO's servers.
----------------------------------------
Algol can not assert the truth of all statements in this post and still be consistent.
[May 10, 2009 3:10:12 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
basso

Member's Avatar


Joined: Mar 19, 2006
Posts: 3156
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 


The point is, there are lots of players that find YPP, puzzle a bit, maybe even start their own crew and never make it much past solid and often not past solid/respected. They then leave the game before they are much noticed by anyone. These are people OOO needs to care about and cater too also, in particular, how can we keep them from leaving the game?


Excellent point. Furthermore, many players are familiar with the concept of "leveling up". When I first started, I was pumped when my bilge went up to "neophyte". I had no clue what it meant of course, but it still felt like I did something. My point is that having 50% of players at the bottom makes no sense in terms of retaining new players. Why not have the first few standings be rather easy to obtain? Yes it would be trivial progress, but newer players would feel a sense that progressing is possible. It would also be helpful to see who is trying, and who is truly hopeless at a puzzle. I mean currently a high able can actually perform pretty well, but they are lumped in with the lazers and boochers.

Basically I am saying the bottom few standings could mirror the top few in terms of percentages. 1% could be able, which everyone starts at. Then a 4% group, then the next 5%, etc. Distinguished would still mean breaking into the top 50%, so nothing would change except that the range of able would be broken up and given different words.
----------------------------------------
Montage of Sage
Mads wrote: 
OK, now I'm convinced. The problem here is that you cannot understand plain English.

[May 10, 2009 4:15:37 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
marundel

Member's Avatar


Joined: Jan 7, 2007
Posts: 2955
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

It's been a long time since I've had a statistics class... but if I remember correctly, having a curve skewed to the lower end does not drive the average to the lower end. It also depends on whether you are defining "average" as "mean", "median", or "mode".

If it's the mean, and 50% are Able, with a scaling down to 1%-2% Ultimate, the mean very well may be in the high Respected to low Master area.

If it's mode, then Able is definitely the average.

If it's median, than the average would likely be somewhere in the high Distinguished arena.

In any case, I don't think any monarch out there can speak for the "average" player game-wide. However, I think any monarch of an active, productive flag can speak for the average player in their flag... especially if they poll their flag before the conference. Even with an awkward ranking system, I believe it would be safe to say that a majority of the flags in the "top 20" would be of sufficient experience and game-wisdom to recognize the difference between representing their flag appropriately and representing only the elite members.
----------------------------------------
Pizzahutpete on the Cerulean Ocean
Prince, Super Awesomeness
SO, Boochin' Drunks

Pizzahutpete everywhere, thanks to the merge
[May 10, 2009 7:09:13 PM] Show Printable Version of Post        Send Private Message    http://www.thehomebrewstore.com    meadbrewer [Link]  Go to top 
Gillie017

Member's Avatar


Joined: Aug 3, 2006
Posts: 1043
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

Remember that the "50% ables" is based on OOO data from people who have puzzled in the past 10 days - an arbitrary time frame that gives equal weighting for everyone from people who log on once and never return through to full-time players. Taking cross-sectional data removes this problem and will reflect the standings of people online at any given time.

Scores for any given puzzle are not normally distributed and Marundel has remembered statistics classes well. The only data we have access to is a categorical representation of a numeric score. OOO could answer the question of the "average" pirate easily using real data.

I still object to using only jobbers on ships to determine the "average" pirate. Pirates who never get on ships are still part of the pirate population and therefore must be included in sampling to be able to claim information about the "average" pirate rather than the "average" jobber.
----------------------------------------
If a bird looks like a duck, swims like a duck and quacks like a duck, then it's probably a Huntard .
[May 10, 2009 7:52:32 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
wrs1864b

Member's Avatar


Joined: Apr 10, 2006
Posts: 6314
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

Statistics are fun, but it is always important when looking at an answer to try to remember what the question was, and the purpose of the question.

OOO has collected data from everyone who has puzzled in the last 10 days, and divided that up into the standings we all see from Able to Ultimate. This is a useful statistic for the purposes it is being used for.

People who have played in the last 10 days will be a different population with a different scoring curve than people who are online right now (or typically online at any given time.) Personally, I can't think of what the value of this statistic would be, unless you like to go up to random pirates and compare stats.

A captain of a ship may well be quite interested in "what will the average jobber who applies look like?" That will likely be different than both the online-population and the 10-day-puzzled statistics. This information can't easily be collected in an accurate way. If labor-alting wasn't such a huge problem, shopkeepers would be interested in similar information about who might take jobs.

Another question is "what is the average score from doing a puzzle?" This is what you will see on duty reports, tournament results, crafting puzzles, etc. Again, this information is very hard to collect but is quite likely to be different than any of the other statistics mentioned above.

Pizzahutpete is right that there are even three common definitions of average. For the purposes of how OOO defines the standings, they are:
Mode: The most common value. This almost has to be Able, by definition.
Median: The point where half the population is higher and the other half is lower. This has to be the Able/Distinguished line, by definition.
Mean: There are actually three common ways of defining the Mean, but the most common is the Arithmetic Mean. That is the sum of all scores, divided by the number of scores. Depending on how the the underlying score curve looks, this can indeed be different than the other two definitions of average, but for most real-world cases, it rarely is that far away. (See the central limit theorem for more details.)

Personally, I strongly suspect that all these statistics mentioned above, from the 10-day-puzzled, to the duty reports, are much more similar than many people seem to think. Yes, it is possible that 99% of the players log on for 5 minutes to puzzle once to get a Booched, and the only people who play constantly are the Ults, thus making the average duty report an Incredible, but I see no evidence of that. And, no, handwaving assertions do not count as evidence.

Now, the data I collected shows that the average pirate on a ship has a low-Distiguished/high-Able standing in the primary ship-board puzzles (bilge, carp, and sails). This is probably very closely related to what a captain/XO will see when hiring jobbers, the duty reports you will see on the ships, and the standings of people who regularly do those puzzles. No, it won't be an exact match, but again, I suspect it is much much closer than what many people who read the forums seem to expect.

Now, the data I collected has three primary sources of bias that I can think of, all of which skew the average up above Able.

First, I'm picking only those pirates that actually made it onto a ship, and we all know that many bnavvers/XOs filter out jobbers. I suspect it doesn't skew it a huge amount since these jobbers will likely get jobs on ships that are less restrictive, but it almost certainly skews the average up some.

Second, I can not get information about pirates on navy ships and training missions. This is where players typically go to learn a puzzle and get their stats up so they can get hired. I suspect that this doesn't skew the average up that much since you find a fair number of leg/utls on navy ships just puzzling to improve their scores/techniques/etc.

Third, between the time I do a /who to get a list of ships on the ocean and when I do a /vwho to get a list of pirates on the ships, some ships will have finished their pillages. This can happen quite quickly, the last time I did a survey, the very first ship I queried had already ported, but in general, the ones toward the end will be ships that stay out longer and those will likely be higher-end pillages.

So, argue how you want, but until I see some data that shows that the average pirate is higher than a low-Distinguished, I doubt that you will change my mind.
----------------------------------------
Algol can not assert the truth of all statements in this post and still be consistent.
[May 11, 2009 8:06:36 AM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
marundel

Member's Avatar


Joined: Jan 7, 2007
Posts: 2955
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

All of which still does not address the OP, which is the initiation of a discussion on the value of another Conference of the Monarchs.

Average pirates and duplicate posters aside, so far it seems to me to be about an even split between those who feel it is not only appropriate but beneficial and those who feel it is either inappropriate or pointless.
----------------------------------------
Pizzahutpete on the Cerulean Ocean
Prince, Super Awesomeness
SO, Boochin' Drunks

Pizzahutpete everywhere, thanks to the merge
[May 11, 2009 8:33:45 AM] Show Printable Version of Post        Send Private Message    http://www.thehomebrewstore.com    meadbrewer [Link]  Go to top 
vnork



Joined: Jul 22, 2006
Posts: 1004
Status: Offline

Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
Now, the data I collected has three primary sources of bias that I can think of, all of which skew the average up above Able.

Heh, those three may be sources of bias, but I find it very highly unlikely that they're the only primary sources of bias.

Here's more sources of bias, and ones that would overreport the number of ables. I think all of them are just as strong as the three you gave in your last post.

1. Standings in a puzzle default to able. Ideally, standing would remain empty until the pirate had puzzled enough for a meaningful puzzling standing. When I tried /who-ing people on ships, a large number were novice/able in many piracy stats. This has a huge effect on your data, and there's no easy way to get rid of the effect.

2. You took your data on the weekend. With two blockades scheduled on Viridian, skilled jobbers are more aware of the schedule and likely to budget their puzzling time accordingly.

3. You took your data on the weekend. It's possible that many casual players only play on the weekends, when they have plenty of free time.

4. You took your data on Sunday, I believe. Poker is free on Sunday. Greenies may not have the PoE to use in poker, and are more likely to run out of PoE if they do try. Many established players like to play poker on free days because they perceive that more unskilled players play on free days.

5. Carousing and stall-keeping usually require doubloon sinks, and thus are more accessible to established players, which keeps them on land.

6. Experienced players have more extensive social connections, and crews and flags more suited to their personal tastes. Therefore, they find it more fulfilling to stay on land to chat than the newer pirate.

7. SMHs take longer to load, while green pillages will often sail with 3 people aboard. While a SMH is loading, some jobbers will stay on land until it's close to being filled, to manage stalls or play carousing games. This is much more rare with pillages.

8. SMHs and Flotillas suddenly force all jobbers to land if the voyage is unsuccessful. Also, SMHs and Flotillas spend longer time in between rejobbing periods than pillages, and experienced jobbers will wait on land until they exit the board.

 
Now, the data I collected shows that the average pirate on a ship has a low-Distiguished/high-Able standing in the primary ship-board puzzles (bilge, carp, and sails).

If you collected data where the average pirate was Able, you didn't post it here. Just based on the numbers you posted earlier, the median pirate would be low-to-mid distinguished.

-Vnork
[May 11, 2009 11:22:13 AM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
wrs1864b

Member's Avatar


Joined: Apr 10, 2006
Posts: 6314
Status: Offline
Re: Conference of the Monarch (Thoughts) Reply to this Post
Reply with Quote

 
 
Now, the data I collected shows that the average pirate on a ship has a low-Distiguished/high-Able standing in the primary ship-board puzzles (bilge, carp, and sails).

If you collected data where the average pirate was Able, you didn't post it here. Just based on the numbers you posted earlier, the median pirate would be low-to-mid distinguished.

IIRC, one stat (sailing?) on onyx yesterday that had like 55% able and another had like 50%, but it was a very small sample size so I probably shouldn't have mentioned it.

I did another survey today, all ships on all archs of viridian. Not much different. The average pirate on the ships was in the low distinguished range.away the

Garnet had two SMHs going on, with not much else, that pushed the average up to high-distinguished. Onyx had even few pirates out on the ocean, only 12, and only one able sailor.

I'm fully prepared to hear people claim that well, Saturday's survey didn't count, and Sunday's didn't count, and today didn't count, and how master *is* an average stat. When this thread started, I thought that monarchs and forum posters weren't that out of touch, but all these posts defending the idea that master is an average stat has changed my mind. I don't think OOO should hold another conference of monarchs. Too many of the people that OOO needs to hear from probably aren't even in a flag, or at least not one anyone in the top 10 flags has heard of.
----------------------------------------
Algol can not assert the truth of all statements in this post and still be consistent.
[May 11, 2009 4:56:25 PM] Show Printable Version of Post        Send Private Message [Link]  Go to top 
Posts: 181   Pages: 7   [ First Page | Previous Page | 1 2 3 4 5 6 7 | Next Page | Last Page]
[Show Printable Version of Thread] [Post new Thread]

Puzzle Pirates™ © 2001-2016 Grey Havens, LLC All Rights Reserved.   Terms · Privacy · Affiliates