c4countdown

Go ORDINATES!

Ahh, very nice, nerdy and awesome. Interesting conundrums occur more often than numbers games.

Kai Laddiman wrote:Go ORDINATES!

How come it's top of the list of most disallowed words?

Phil Reynolds wrote:

Kai Laddiman wrote:Go ORDINATES!

How come it's top of the list of most disallowed words?

It's at the top of the Most Frequent Nines page actually. The reasons for which are obvious.

Charlie, this is a very useful addition to the site =) Fankoo!

Daniel O'Dowd wrote:

Phil Reynolds wrote:

Kai Laddiman wrote:Go ORDINATES!

How come it's top of the list of most disallowed words?

It's at the top of the Most Frequent Nines page actually. The reasons for which are obvious.

Yeah, I know that. But, at the time I posted, the same list was coming up whichever one I selected. I see Charlie's fixed that now, obviously with the sole aim of making me look stupid.

Ooh yes, these are the kind of stats I like.

Interesting stuff. I'm guessing that with most frequent 9s, some more obscure ones like UITLANDER will appear higher up than usual due to their frequent appearance in Goatdown. Interesting that AUDITOR is the most missed word as I used to always miss it. Actually, these two facts are probably linked. So not very interesting.

Junaid Mubeen wrote:Interesting stuff. I'm guessing that with most frequent 9s, some more obscure ones like UITLANDER will appear higher up than usual due to their frequent appearance in Goatdown. Interesting that AUDITOR is the most missed word as I used to always miss it. Actually, these two facts are probably linked. So not very interesting.

The stats don't include anything other than standard mode for exactly that reason. I might do a Goatdown table sometime although it might not be particularly interesting.

Great stuff Charlie, I love these kinds of stats. I'm sure there are some other good ones that can be added in future.

If you're at all interested, off the top of my head I can think of (I know some of these may be unfeasible depending on what data mining you're doing / expensive database queries / pure impossibility):

- most missed/solved conundrums
- numbers game player %maxes for the different formats (I always wondered whether 4 large or 6 small is easier overall)
- proportion of impossible numbers game by format
- top missed/spotted double (or more) darrens
- most missed words of players over/under an arbitrary rating - maybe simpler to have a personal "Bob's missed words" list for each player
- most offered words overall

More of a high scores/top players thing:
- top 10 highest proportion of letters / numbers / conundrums maxes per game (per format? where it makes sense)
- best goatdown pickers

- most missed/solved conundrums: The Raccoon and I have Big Plans for that so there will be something eventually.
- numbers game player %maxes for the different formats and proportion of impossible numbers game by format: http://www.apterous.org/statland.php?section=numbers (I think that's what you wanted.)
- most offered words overall: http://www.apterous.org/statland.php?se ... st_offered and http://www.apterous.org/statland.php?se ... _offered_9

The others are good ideas too, I'll see what I can do.

This is great!

I've spotted however that on the "Most offered" pages, the top of the numbers column says "Disallowed", which reduces my enjoyment no end.

Charlie Reams wrote:- most missed/solved conundrums: The Raccoon and I have Big Plans for that so there will be something eventually.
- numbers game player %maxes for the different formats and proportion of impossible numbers game by format: http://www.apterous.org/statland.php?section=numbers (I think that's what you wanted.)
- most offered words overall: http://www.apterous.org/statland.php?se ... st_offered and http://www.apterous.org/statland.php?se ... _offered_9

The others are good ideas too, I'll see what I can do.

Awesome.

Regarding the numbers, that's actually beyond what I had thought of as far as maxes are concerned, so nice job. In terms of difficulty I was actually thinking of whether the (human) player was actually able to solve the numbers game, or how far away was attained (so in games between two human opponents, there are in effect two numbers games to record).

If it's not too hard to find, what were the two impossible 1-large numbers games?

And great work on the new additions.

Gary Male wrote:If it's not too hard to find, what were the two impossible 1-large numbers games?

http://www.apterous.org/viewgame.php?ga ... und=231420 and http://www.apterous.org/viewgame.php?ga ... und=210560

No surprises, they both involve the 25, small workers and a large target.

Phil Reynolds wrote:

Daniel O'Dowd wrote:

Phil Reynolds wrote: How come it's top of the list of most disallowed words?

It's at the top of the Most Frequent Nines page actually. The reasons for which are obvious.

Yeah, I know that. But, at the time I posted, the same list was coming up whichever one I selected. I see Charlie's fixed that now, obviously with the sole aim of making me look stupid.

It still say 'Disallowed' rather than 'Hits' on the column headings of both the 'most offered' pages, which confused me momentarily.

But still a great resource! Thanks Charlie.

Simon Myers wrote:- best goatdown pickers

Ta-dah and ta-dum!

I think this mainly reveals that the bots are way too good at picking.

Charlie Reams wrote:

Simon Myers wrote:- best goatdown pickers

Ta-dah and ta-dum!

I think this mainly reveals that the bots are way too good at picking.

Very cool Charlie.
I presume the darrenicity measure relates to the number of maxes available given your pick? So that with the two measures you can differentiate players like Rex (great max picking with few co-max words) and myself (good max picking but with many max words available).

In that case, goatdown aficionados are free to use me as a stooge in the quest for high scores

I wonder how well someone who just picked S every time would perform in the max picking.

Jon O'Neill wrote:I wonder how well someone who just picked S every time would perform in the max picking.

Looking at the stats here, perhaps not that well. In fact, S wasn't even clearly much better than L or R given the small sample of games.

That has got me thinking - I had a game the other day where the first three letters in the selection were S, S, S; I mused at the time that if I didn't pick anything at the end, I would automatically have at least four Ss in the selection.
I wonder what the best pick in general is if there is already an S (or more) in the selection, and if it is largely different to the default ["pick S"] action, should it be changed in this (these) cases?

Charlie Reams wrote:

Simon Myers wrote:- best goatdown pickers

Ta-dah and ta-dum!

I think this mainly reveals that the bots are way too good at picking.

Please keep it that way! Well, you can mess with the others, but don't fuck with Prime

Cool stats btw.

How about the most common people to submit any given word?

Ian Volante wrote:How about the most common people to submit any given word?

That'll probably be me in most cases, I'm VERY common. Just ask Richard Brittain.

Jon Corby wrote:

Ian Volante wrote:How about the most common people to submit any given word?

That'll probably be me in most cases, I'm VERY common. Just ask Richard Brittain.

I wish your humour was less so...

At the bottom of the frequent 12s, it says, "Table includes only standard games in the normal variant." Yet this is clearly not true.

Kai Laddiman wrote:At the bottom of the frequent 12s, it says, "Table includes only standard games in the normal variant." Yet this is clearly not true.

I worded that deliberately to be ambiguous, because I'm lazy. For 12s, the normal variant is Hyper, so it makes sense. Kinda.

This doesn't look too much like the normal variant.

Kai Laddiman wrote:This doesn't look too much like the normal variant.

You're right. Now I have to fix it properly. Boooo.

I think the update code for the "best pickers" (maxes and darrens) isn't doing its job properly. I've noticed I've been stuck on 80 rounds picked since it was introduced (I think). It probably should've gone up a bit after all those Aptobash games!

Simon Myers wrote:I think the update code for the "best pickers" (maxes and darrens) isn't doing its job properly. I've noticed I've been stuck on 80 rounds picked since it was introduced (I think). It probably should've gone up a bit after all those Aptobash games!

Good spot.

Simon Myers wrote:I think the update code for the "best pickers" (maxes and darrens) isn't doing its job properly. I've noticed I've been stuck on 80 rounds picked since it was introduced (I think). It probably should've gone up a bit after all those Aptobash games!

This is now more fixed than a Stephen Maguire game. My original implementation had no few than 3 different bugs, causing its results to be utter bollocks; fixed version is now up.