This forums is for questions, answers, and discussion about First Edition rules, formats, and expansions.
User avatar
First Edition Rules Master
By BCSWowbagger (James Heaney)
 - First Edition Rules Master
 -  
Community Contributor
#557948
Boffo97 wrote: Wed Jul 14, 2021 11:59 pm
BCSWowbagger wrote: Wed Jul 14, 2021 9:33 pm (INTEGRITY! MEANS! COMMITMENT! TO! PRINCIPLES! OTHERWISE! ITS! JUST! HONOR! AND WE ALREADY! HAVE! THAT! SKILL!)
The big problem though is that it might mean that NOW but Decipher has been less than consistent with it.

Otherwise, we have a whole ton of, say, PAQ Romulans whose scores indicate that they couldn't care less about the Empire with on-screen evidence either lacking or to the contrary.

But I suppose the counter argument to that is that PAQ was just weird and a lot about that era makes little sense now.
I contend that TNG Romulans are in fact not really all that committed to principle (outside the Underground). They seem committed to not getting executed by the Tal Shiar -- which is fair, but is less a principle. I don't see a lot of Rommies in TNG sacrificing a whole lot for their principles (Jarok being a clear exception).

But you're right, because INTEGRITY does measure one other thing: how honest you are. If you lie and connive and dissemble and subvert treaties and break laws (yes, you, TOMMIE T. TOMALAK), Decipher gave you low INTEGRITY. If you're honest and forthright and go through proper channels even when you feel strongly, high INTEGRITY. Not sure I agree with that choice, but it seems like almost the only way to explain relatively low Romulan INTEGRITY scores.

I had an answer for phaser about what Honor and Treachery represent, but then I did some extra research and now I'm not so sure of myself.
User avatar
First Edition Rules Master
By BCSWowbagger (James Heaney)
 - First Edition Rules Master
 -  
Community Contributor
#557949
PantsOfTheTalShiar wrote: Thu Jul 15, 2021 12:31 am Thanks for crunching these numbers. This is important stuff to think about. I'd be interested to see the affiliation averages weighted by number of copies of each personnel played in the past X months (ideally just OTF decks). (EDIT: Yes, that's number of copies total, not number of decks total. Not sure how easy it is process that info.) That'll show the extent to which affiliation attribute differences actually show up at game time.
Good idea! It's easy enough to pull card counts from deck lists, although the query will probably churn a bit as that's a lot of records. Filtering to just OTF-played decks is a good deal more complex, but we're already doing similar code in the "recently played in X decks" box, so maybe I can scavenge that...

Maybe tomorrow I'll try one. Any particular affiliation interest you?
User avatar
 
By PantsOfTheTalShiar (Jason Tang)
 - Delta Quadrant
 -  
#557951
BCSWowbagger wrote: Thu Jul 15, 2021 12:37 am
PantsOfTheTalShiar wrote: Thu Jul 15, 2021 12:31 am Thanks for crunching these numbers. This is important stuff to think about. I'd be interested to see the affiliation averages weighted by number of copies of each personnel played in the past X months (ideally just OTF decks). (EDIT: Yes, that's number of copies total, not number of decks total. Not sure how easy it is process that info.) That'll show the extent to which affiliation attribute differences actually show up at game time.
Good idea! It's easy enough to pull card counts from deck lists, although the query will probably churn a bit as that's a lot of records. Filtering to just OTF-played decks is a good deal more complex, but we're already doing similar code in the "recently played in X decks" box, so maybe I can scavenge that...

Maybe tomorrow I'll try one. Any particular affiliation interest you?
How about [1E-Rom] or [Kli] ? Something classic that's supposed to have definite strengths and weaknesses, but not such a big dataset as [Fed] .
User avatar
First Edition Rules Master
By BCSWowbagger (James Heaney)
 - First Edition Rules Master
 -  
Community Contributor
#557953
PantsOfTheTalShiar wrote: Thu Jul 15, 2021 12:46 am
BCSWowbagger wrote: Thu Jul 15, 2021 12:37 am
PantsOfTheTalShiar wrote: Thu Jul 15, 2021 12:31 am Thanks for crunching these numbers. This is important stuff to think about. I'd be interested to see the affiliation averages weighted by number of copies of each personnel played in the past X months (ideally just OTF decks). (EDIT: Yes, that's number of copies total, not number of decks total. Not sure how easy it is process that info.) That'll show the extent to which affiliation attribute differences actually show up at game time.
Good idea! It's easy enough to pull card counts from deck lists, although the query will probably churn a bit as that's a lot of records. Filtering to just OTF-played decks is a good deal more complex, but we're already doing similar code in the "recently played in X decks" box, so maybe I can scavenge that...

Maybe tomorrow I'll try one. Any particular affiliation interest you?
How about [1E-Rom] or [Kli] ? Something classic that's supposed to have definite strengths and weaknesses, but not such a big dataset as [Fed] .
Makes sense to me! I like [1E-Rom] for this.
 
By phaserihardlyknowher (Ben Daeuber)
 - Beta Quadrant
 -  
#557958
BCSWowbagger wrote: Thu Jul 15, 2021 12:34 am I had an answer for phaser about what Honor and Treachery represent, but then I did some extra research and now I'm not so sure of myself.
Well, honor and integrity are synonyms according to my buddy Roget, and least for some definitions of honor. That's why a skill distribution would be interesting. I'd guess that personnel with Honor have a higher integrity and personnel with Treachery and Greed have lower.

Are there other skills or classifications that track? SECURITY and Strength? Cunning and Physics? Exobiology and Integrity?
User avatar
 
By Takket
 - Delta Quadrant
 -  
#558011
Armus wrote: Thu Jul 15, 2021 12:06 am Sirol and Dorian Collins are two sides of the same jacked up coin. At least Sirol has the excuse of being a PAQ era card....

And speaking of PAQ era attribute jacked-upedness, do we need to go any farther than Benjamin Maxwell and Norah Satie and their respective Integrity of 9 each?!
I guess if Joseph McCarthy ever gets a STCCG card, he'll have integrity of 9 too
User avatar
First Edition Rules Master
By BCSWowbagger (James Heaney)
 - First Edition Rules Master
 -  
Community Contributor
#558071
BCSWowbagger wrote: Thu Jul 15, 2021 12:52 am
PantsOfTheTalShiar wrote: Thu Jul 15, 2021 12:46 am How about [1E-Rom] or [Kli] ? Something classic that's supposed to have definite strengths and weaknesses, but not such a big dataset as [Fed] .
Makes sense to me! I like [1E-Rom] for this.
UPDATE: This is trickier than I thought. I have a weighted average that works correctly if there is only one copy of each card in each deck. But it doesn't properly account for quantity. If there's a deck with 8 D'vanos in it, my queries are either counting 1 copy or like 5.5 of them.

I'll keep at it, but that's why there's no results yet.

EDIT: ...on the other hand, perhaps "only count 1 copy per deck" is the right thing to do, since counting each copy in a deck would overweight the impact of large decks. In which case I could post results tonight. Any thoughts on this, anyone?
 
By phaserihardlyknowher (Ben Daeuber)
 - Beta Quadrant
 -  
#558077
BCSWowbagger wrote: Sat Jul 17, 2021 4:48 pm EDIT: ...on the other hand, perhaps "only count 1 copy per deck" is the right thing to do, since counting each copy in a deck would overweight the impact of large decks. In which case I could post results tonight. Any thoughts on this, anyone?
One per. If what you want to know is how the current meta weights a specific attribute or affiliation, you need to make an apples to apples comparison and each card was counted only once in your initial results.

However, if you're curious how the meta weights attributes *in general* all, regardless of deck size or universal status. That still indicates what people are hoping to accomplish, large deck size or not.
User avatar
 
By nobthehobbit (Daniel Pareja)
 - The Center of the Galaxy
 -  
Moderator
#558080
Maybe you could do some sort of weighted average whereby you count each personnel as a proportion of its share of each deck it appears in, or something, then renormalize?
User avatar
First Edition Rules Master
By BCSWowbagger (James Heaney)
 - First Edition Rules Master
 -  
Community Contributor
#558083
nobthehobbit wrote: Sat Jul 17, 2021 8:24 pm Maybe you could do some sort of weighted average whereby you count each personnel as a proportion of its share of each deck it appears in, or something, then renormalize?
One could do this. I could not. I am not smart enough.

I guess this increases the strength of the case for a public API so it's not just a handful of people who can do this stuff.
User avatar
First Edition Rules Master
By BCSWowbagger (James Heaney)
 - First Edition Rules Master
 -  
Community Contributor
#558088
Alright, this was several hours for a very boring conclusion about a question Pants didn't quite ask.

This is average attributes, by affiliation, weighted by use in tournament-played OTF decks since 1 January 2016. But it only counts one copy per deck of each personnel. So if 8 decks at the same tournament each used 1 copy of Taibak, Taibak is counted 8 times. If 1 deck used 1 copy of Taibak, but that deck was then played at 8 different tournaments, Taibak is counted 8 times. But if 1 deck used 8 copies of Taibak... only the first copy is counted here. I regret the omission. I just couldn't figure out the math in the time available today.

(FWIW, in a small-scale test with a specific (but quite generic) TNG Romulan deck, I found that the per-copy weighting didn't matter very much -- even if I had gotten it to work correctly, it just didn't nudge the results significantly, even in a deck that ran multiple copies of multiple cards. But I still wish I'd gotten it to work. But I didn't. So here are my actual results.)

The result is fairly boring. Weighting by use in decks made only a small difference:

Image

Image

Image

Image

(MQ cards are excluded from all these figures except -- obviously -- for the KCA and TE measurements.)

The affiliation differences and averages are largely the same, although there were a few tenths-of-a-point changes that (for example) broke up the four-way tie for second place on CUNNING. ( [Rom] turn out to be in 5th place on CUNNING by this measure, which seems terrible.) There is some jostling in the exact rankings, but the top-tier and bottom-tier on each attribute remained pretty constant.

On average, by filtering against which cards are actually being used (and, to the extent possible, how often they are being used), each affiliation gained 0.05 points each of INTEGRITY, CUNNING, and STRENGTH compared to their overall averages, with an overall gain of 0.15 points. So, yes, players are definitely systematically favoring and stocking personnel with higher attributes over personnel with lower attributes -- this is a significant and consistent effect. But it's small and pretty consistent across affiliations, so it doesn't really change much from my earlier posts.

The exception to that is [Hir], which, surprisingly, jumped 0.9 average total attribute points when I weighted by which [Hir] cards are actually being stocked. They gained 0.1 INTEGRITY, 0.5 CUNNING, and 0.3 STRENGTH with the weighting! (This was enough to put them ahead of Dominion as the champions of CUNNING.) I can only speculate why this is, but, glancing at the Hirogen personnel list, I notice that their overall average CUNNING is 7... but several of their personnel with CUNNING below that are also (coincidentally?) the least-used Hirogen by far: Konuric, Mellis, Zivan, and Janeway. (Two of those cards are also quite new, which hurt them a lot in this measurement.)

Overall, we as a player base are clearly leaving some low-attribute personnel in the binders, but the overall differences and rankings hold (mostly) true to what you'd expect just looking at all available cards.

I've added this data to my spreadsheet:

https://docs.google.com/spreadsheets/d/ ... sp=sharing

P.S. Data integrity note: I'm not quite certain I did a correct apples-to-apples for [TE] [Fed] here. In this query, I looked for [TE] [Fed] only; last time, I think I looked for "all [TE] except [SF]," which includes people like McKay and Commander Leeta. This doesn't seem to have made a lot of difference, so I am not going to take the time to correct it.
1EFQ: Game of two halves

Honestly, I don’t think I’ve re[…]

HAPPY BIRTHDAY!!!!

Happy birthday to @Takket ! :D :thumbsup: […]

Opponents turn

Remodulation

It started in mid-2013. At that time it became sta[…]