Things you really want to see in "Cantr Statististics

[Thetaris]

You idiot, do you know how much work and effort would have to go into that?

It's not like it's blatantly obvious for the people who do it, it's not flashing up on their screen.

It's a massive database with thousands of little numbers.

There are over 5,000+ figures they'd have to put together, not to mention find in the first place.

Most of these suggestions are pure hope and will never happen. This thread is quite ignorant of the fact it's putting more labour on the people who already put enough effort into the current statistics.

[Chris Johnson]

Thetaris , there is absolutely no need to insult the posters in this thread

As one of the key people who extracts the data for HF's use in his Cantr Statistics column , I'm very aware of what is involved (and indeed, working as a professional statistician for the past 20 or so years I'm reasonably clued up on analysing data).

Yes it is a massive database . Being a database the data is organised in such a way that it is extremely easy to extract the answers to most of the queries people have requested in this thread.

For instance the previous request before your post could be answered by a 40 or 50 character SQL statement.

This doesn't mean that all these will be followed up, some pose potential
CRB issue, others are difficult to post analyse and some (as I'm sure HF will testify) turn out to be very boring when run, showing little or no discrimination .

[Hellzon]

EDIT: Mostly pwned by Chris. But I don't bother with that.

You idiot

Chill...

Most of these suggestions are pure hope and will never happen. This thread is quite ignorant of the fact it's putting more labour on the people who already put enough effort into the current statistics.

Well, yeah. But what's the harm in suggesting things for upcoming articles (beyond the closest one)? As for "putting more work" - it's not as if we're standing outside farmer's house with pitchforks and DEMANDING that he does all of these suggestions.
Besides, those of us who aren't on ProgD or RD (the ones who have access to these statistics, I assume) can't know what's easy or not to generate. We can make an educated guess, yes, but that's all. (And indeed, the educated guess about Okuds and my suggetion is that it would be practically impossible to do, but we can hope, can't we?)

/Hellzon - spreading the love...

[WojD]

Burying (high in the Polish region? )

Almost all bodies buried... With showel this takes a moment

[formerly known as hf]

I should probably make a point by noting that I haven't even started the next stats issue, or thought about it.

I'm in the middle of actually trying to get my dissertation done, as well as getting a proposal for another research project done, and getting that reserach underway. Not to mention it being the begining of a new semester and all the bumpf that goes with that.

Oh, and my motherboard shorted, so I'm without a computer unless I'm in a Uni cluster all week...

All in all, I doubt there I will be able to write a stats article for the next issue, unless the next issue is much much more delayed than it's looking to be at the moment (and it's already looking like it'll be fairly heavily delayed)

[Just A Bill]

When you get some free time, while in IRC, we were talking about getting new players to stick with the game, and thought the following might be interesting statistics..

Lifespan of first character vs time the player is active (days from account activation till deactivation)
Cause of death of first character vs time player is active

The thought is that perhaps a significant number of players has a bad experience, exploring, picking up things, accused of stealing and killed and this might turn the player off cantr. Personally I think you'll see a lot of heart attack/starvations for short term players representing players that come in for a day, decide the game is not for them and never come back...

[SekoETC]

I was thinking, deaths sorted by age. It would look something like a loooooooong one for first year, then I'd guess the bars going shorter rather linearly. But if there were peaks in like 23 years or 28 years when your character has done exploring the region or settling in a community and life gets boring, this would be interesting to find out. Then we could see what's the average time of getting bored with a character.

[Talapus]

Or how about the number of keys in peoples inventories?

[SekoETC]

lol, That would be cool. How about the most rare tools/objects (wooden bows, apotechary kits, brooms...)?

[formerly known as hf]

I'd like to see how many of the total number of keys are useless because of broken locks...

[Amlin]

I wish something that isnt simple query, but i wish to be stored... I hope to see time diagram of current statistics page. I.e.: how many players joined language group, how much changed characters numbers, how fast was translation made, etc.. I think about something like time/number diagram instead of simple numbers "now is"... I hope i explained not very confusing...

[Sho]

Is that data stored, though? So far all statistics (as far as I can recall) have used only data in the current state of Cantr.

[Chris Johnson]

Some of this is (and some time series stats are stored by Jos as a matter of course) - All of Amin's suggestions are derivable from the database as it stands as we have language group and time of joining / spawning for players/ characters - Also whilst not exactly the same , every translated text is date stamped so something approaching % translation over time could be extracted.

[formerly known as hf]

The next issue (if I get 'round to it) will be based on some time-series data.

[TatteredShoeLace]

I didn't read the first page, but perhaps most common note in inventory? Money on Treefeather or Hobbes I'd think.