Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[LSJ] What is "honoring a deal" means exactly?

5 views
Skip to first unread message

Joscha Duell

unread,
Nov 18, 2004, 12:03:43 PM11/18/04
to
In a post in the Dutch ECQ report LSJ wrote:

"A deal can be made at any time.
If the making the deal with the intention of honoring it is legal
within the play-to-win rules at the time it is made, then the
player is allowed to keep the deal when it comes time to do so,
even if breaking it would be better under the play-to-win rules.
If not, if making the deal is not "playing to win", then the
rules do not recognize the deal as an excuse to avoid "playing
to win" when the time comes to honor it -- that is, it must be
broken if honoring it at the time it comes to honor it is not
a "playing to win" move."

Now we are discussing for days now what that means exactly.

Maybe this example helps:

Player A has ousted already one Meth and is now preying on B. With
them on the table are C and D.
Player B feels too weak to keep A at bay so he offers A a deal. A
should help him to oust C and D (how is not important I think, maybe
with unactivity, maybe with votes or rushes against C and D,
whatever). For that he will roll over after both are ousted, thus
giving A the GW (for 2 VP for himself).
Now assume the judge judges the deal as "illegal", aka known as not
playing to win => ptw., because he sees B strong enough to cope with
A. So he will intervene "at the time it comes to honor" the deal and
tells A and B to ptw.

My qustion is: When is A and B honoring the deal?

- My mate says, honoring the above deal would be, if A damages C in
whatever way, because he does not ptw but honors the deal with B.

- I say honoring the deal is when B rolls over, after C and D was
ousted.

o I think, if my mate is correct, dealing will never be like it was
before.
o If I'm correct, deals are a problem for judges, because a judge has
to check deals, when they are made, to know if they could be honored
later or have to be "broken"/dealpartners have to ptw.

Thanks a lot.
Joscha

LSJ

unread,
Nov 18, 2004, 1:29:44 PM11/18/04
to
"Joscha Duell" <joscha...@gmx.de> wrote in message
news:8b87b6c9.04111...@posting.google.com...

> In a post in the Dutch ECQ report LSJ wrote:
>
> "A deal can be made at any time.
> If the making the deal with the intention of honoring it is legal
> within the play-to-win rules at the time it is made, then the
> player is allowed to keep the deal when it comes time to do so,
> even if breaking it would be better under the play-to-win rules.
> If not, if making the deal is not "playing to win", then the
> rules do not recognize the deal as an excuse to avoid "playing
> to win" when the time comes to honor it -- that is, it must be
> broken if honoring it at the time it comes to honor it is not
> a "playing to win" move."
>
> Now we are discussing for days now what that means exactly.

If the deal an example of playing to win, then it can be kept.
That's all.

> Maybe this example helps:
>
> Player A has ousted already one Meth and is now preying on B. With
> them on the table are C and D.
> Player B feels too weak to keep A at bay so he offers A a deal. A
> should help him to oust C and D (how is not important I think, maybe
> with unactivity, maybe with votes or rushes against C and D,
> whatever). For that he will roll over after both are ousted, thus
> giving A the GW (for 2 VP for himself).
> Now assume the judge judges the deal as "illegal", aka known as not
> playing to win => ptw., because he sees B strong enough to cope with
> A. So he will intervene "at the time it comes to honor" the deal and
> tells A and B to ptw.
>
> My qustion is: When is A and B honoring the deal?

The only time the question matters is when A or B attempts to do
something that is not "playing to win" (such as "rolling over"
when the time comes).

> - My mate says, honoring the above deal would be, if A damages C in
> whatever way, because he does not ptw but honors the deal with B.

If the deal is not an example of ptw, and if A damaging C is not ptw
in and of itself, then A damaging C is not ptw.

This is assuming that A realizes that the deal is not ptw.
It could be that only B knows this.

> - I say honoring the deal is when B rolls over, after C and D was
> ousted.

Any violation of ptw must be reconciled. Either it is allowed as
part of a ptw deal or it is disallowed.

> o I think, if my mate is correct, dealing will never be like it was
> before.

I can't comment on this, since I don't know what your local situation
was before.

> o If I'm correct, deals are a problem for judges, because a judge has
> to check deals, when they are made, to know if they could be honored
> later or have to be "broken"/dealpartners have to ptw.

Only as necessary.

--
LSJ (vte...@white-wolf.com) V:TES Net.Rep for White Wolf, Inc.
V:TES homepage: http://www.white-wolf.com/vtes/
Though effective, appear to be ineffective -- Sun Tzu

Daneel

unread,
Nov 18, 2004, 4:19:25 PM11/18/04
to
On Thu, 18 Nov 2004 13:29:44 -0500, LSJ <vte...@white-wolf.com> wrote:

> If the deal is not an example of ptw, and if A damaging C is not ptw
> in and of itself, then A damaging C is not ptw.
>
> This is assuming that A realizes that the deal is not ptw.
> It could be that only B knows this.

This topic is like honey to a bee. ;)

I kind of think that what you are hinting at here is that there are two
sides to the deal. A deal may be legal from one POV and be illegal from
another. In fact, most table-split deals are probably legal from the
intended victor's POV.

Now it's interesting how B can perfectly well kill A if he wishes to,
but decides to the contrary, and fools A to take a deal that will
strengthen B further (and what's more important, when in the end B does
kill A, it will mean an ousted prey instead of an ousted predator).

--
Bye,

Daneel

lehrbuch

unread,
Nov 18, 2004, 8:23:16 PM11/18/04
to
[Daneel]

> A deal may be legal from one POV and be illegal from
> another. In fact, most table-split deals are probably legal
> from the intended victor's POV.

There's only one point of view that matters, the judge's.

Players can say anything they like (make any number of deals, that
relate to anything, in-game). It is what the players do that is
important. Making a deal is *not* doing something - it is just talk.

As I understand it, a player cannot do something that the judge deems
to be not be an attempt to "play to win" - the mitigating factors are
when they are fulfilling a deal (that *was* a "play to win"), or when
the player's position is "lost".

* lehrbuch

Darky

unread,
Nov 18, 2004, 9:30:53 PM11/18/04
to
<snip>

> > Maybe this example helps:
> >
> > Player A has ousted already one Meth and is now preying on B. With
> > them on the table are C and D.
> > Player B feels too weak to keep A at bay so he offers A a deal. A
> > should help him to oust C and D (how is not important I think, maybe
> > with unactivity, maybe with votes or rushes against C and D,
> > whatever). For that he will roll over after both are ousted, thus
> > giving A the GW (for 2 VP for himself).
> > Now assume the judge judges the deal as "illegal", aka known as not
> > playing to win => ptw., because he sees B strong enough to cope with
> > A. So he will intervene "at the time it comes to honor" the deal and
> > tells A and B to ptw.

My question would be:

Should the judge intervene if this was 'just' a honest error of
judgement from player B? Disallowing him to keep the deal, assuming he
intends to, after player A has fulfilled his part (who also thinks B
is ptw when going for 2 VP) does seem rather.. unfair.

-Bram Vink

Daneel

unread,
Nov 19, 2004, 3:14:44 AM11/19/04
to
On 18 Nov 2004 17:23:16 -0800, lehrbuch <lehr...@gmail.com> wrote:

> [Daneel]
>> A deal may be legal from one POV and be illegal from
>> another. In fact, most table-split deals are probably legal
>> from the intended victor's POV.
>
> There's only one point of view that matters, the judge's.

I wasn't referring to Point of View as "opinion".

I was referring to the two different sides of the equation.

A typical table split deal (3-2) will be always legal from the
intended victor's perspective. It will not always be legal from
the would-be loser's perspective. However, an unlikely 2/2 split
can well be legal for one player and illegal for another.

--
Bye,

Daneel

James Coupe

unread,
Nov 19, 2004, 3:20:33 PM11/19/04
to
In message <1100827396....@c13g2000cwb.googlegroups.com>,

lehrbuch <lehr...@gmail.com> writes:
>Players can say anything they like (make any number of deals, that
>relate to anything, in-game). It is what the players do that is
>important. Making a deal is *not* doing something - it is just talk.

That's not true.

Play which would be typically not playing to win (rolling over, say) can
be legitimised by that talk and the state of the game at that time.
That's the point of many of these questions. It's not *just* about
play.

--
James Coupe
PGP Key: 0x5D623D5D Who's ever heard of that, though!
EBD690ECD7A1FB457CA2 Designing a deck that just calls votes.
13D7E668C3695D623D5D That's crazy talk, there.

Ira

unread,
Nov 19, 2004, 8:05:58 PM11/19/04
to
Hello LSJ,

> This is assuming that A realizes that the deal is not ptw.
> It could be that only B knows this.

If A wants to make the deal with B, can A call over the judge and
require an answer to the question, "Is this a legally made deal?" If
no, then A won't know if it was a legally made deal until it comes
time for B to oust himself, correct?

Can B call over the judge and require an answer? If yes, can B
require that the judge answer publicly if B wants?

Private answers from the judge and keeping A in the dark makes such
deals a bit riskier; on the other hand, B could simply break the deal
anyway, so it doesn't seem like a big change. And I'm sure if B wants
to honor the deal, he can find a way to lose to A that the judge
can't/won't stop.


Switching gears, I'm now thinking about this issue from the judge's
perspective. It seems like a hassle for the judge to make a good
ruling about a gamestate that is long gone. Is it correct that the
judge could potentially be first called over at the time B tries to
oust himself? The judge would need to decide if the deal was legal
based on the gamestate at dealtime.

LSJ, for you, what is the difference between ruling based on a
gamestate that is long gone, and ruling (or at least illuminating)
based on a case study? You seem to have no problem with the former,
but studiously avoid the latter. Perhaps because the former is
required while judging, and the latter will simply incite discussion
and make more work for yourself. :)

Have a great weekend,

Ira

Joscha Duell

unread,
Nov 20, 2004, 4:03:17 AM11/20/04
to
"LSJ" <vte...@white-wolf.com> wrote in message news:<3047ttF...@uni-berlin.de>...

> The only time the question matters is when A or B attempts to do
> something that is not "playing to win" (such as "rolling over"
> when the time comes).
>
> > - My mate says, honoring the above deal would be, if A damages C in
> > whatever way, because he does not ptw but honors the deal with B.
>
> If the deal is not an example of ptw, and if A damaging C is not ptw
> in and of itself, then A damaging C is not ptw.

I still don't get it (sorry for that): If it comes to you judging this
situation would you
- hinder A KRC C for three, PS C and giving the pool to B or rushing
all of C's minions because it is not ptw (which it is obviously only
because A and B have a deal, which was not ptw in judge's POV at the
time it was made in this example)

- or would you only intervene at the time B wants to roll over for A
getting the GW?

I know you'd like to see decks and the concrete situation to judge
that. I think the above description should suffice to get clear when
honoring a deal is to be supervised by a judge.

LSJ

unread,
Nov 20, 2004, 8:40:02 AM11/20/04
to
Ira wrote:
> Hello LSJ,
>>This is assuming that A realizes that the deal is not ptw.
>>It could be that only B knows this.
>
> If A wants to make the deal with B, can A call over the judge and
> require an answer to the question, "Is this a legally made deal?" If
> no, then A won't know if it was a legally made deal until it comes
> time for B to oust himself, correct?

A can call for a judge any time. A cannot require the judge to do
anything, certainly not divulge hidden information.

> Can B call over the judge and require an answer? If yes, can B
> require that the judge answer publicly if B wants?

B can call the judge any time, like A.

> Switching gears, I'm now thinking about this issue from the judge's
> perspective. It seems like a hassle for the judge to make a good
> ruling about a gamestate that is long gone. Is it correct that the

It is not a hassle, no. It's judging.

> judge could potentially be first called over at the time B tries to
> oust himself? The judge would need to decide if the deal was legal
> based on the gamestate at dealtime.

The judge can be called any time.

> LSJ, for you, what is the difference between ruling based on a
> gamestate that is long gone, and ruling (or at least illuminating)
> based on a case study? You seem to have no problem with the former,
> but studiously avoid the latter. Perhaps because the former is
> required while judging, and the latter will simply incite discussion
> and make more work for yourself. :)

"Long gone" and "part of the current game" are mutually exclusive.

--
LSJ (vte...@white-wolf.com) V:TES Net.Rep for White Wolf, Inc.

Links to V:TES news, rules, cards, utilities, and tournament calendar:
http://www.white-wolf.com/vtes/

LSJ

unread,
Nov 20, 2004, 8:44:02 AM11/20/04
to
Joscha Duell wrote:
> "LSJ" <vte...@white-wolf.com> wrote in message news:<3047ttF...@uni-berlin.de>...
>>The only time the question matters is when A or B attempts to do
>>something that is not "playing to win" (such as "rolling over"
>>when the time comes).
>>
>>>- My mate says, honoring the above deal would be, if A damages C in
>>>whatever way, because he does not ptw but honors the deal with B.
>>
>>If the deal is not an example of ptw, and if A damaging C is not ptw
>>in and of itself, then A damaging C is not ptw.
>
> I still don't get it (sorry for that): If it comes to you judging this
> situation would you
> - hinder A KRC C for three, PS C and giving the pool to B or rushing
> all of C's minions because it is not ptw (which it is obviously only
> because A and B have a deal, which was not ptw in judge's POV at the
> time it was made in this example)

I don't know what the state is, so I couldn't say which would be the
case, so that's why I gave the answer I did.

If A damaging C is ptw in and of itself, then it is allowed under
the ptw rule.

If it is not, but it is part of a deal that was ptw, then it is
still allowed under the ptw rule.

If it is not ptw itself and it is either part of no deal or part
of a deal that is not ptw, then it is not allowed.

> - or would you only intervene at the time B wants to roll over for A
> getting the GW?

The judge should be called to intervene as soon as ptw is violated.

> I know you'd like to see decks and the concrete situation to judge
> that. I think the above description should suffice to get clear when
> honoring a deal is to be supervised by a judge.

Yes, it does suffice to the the clear answers above.

--
LSJ (vte...@white-wolf.com) V:TES Net.Rep for White Wolf, Inc.

Joscha Duell

unread,
Nov 21, 2004, 4:50:14 AM11/21/04
to
LSJ <vte...@white-wolf.com> wrote in message news:<CSHnd.941288$Gx4.7...@bgtnsc04-news.ops.worldnet.att.net>...

> > I still don't get it (sorry for that): If it comes to you judging this
> > situation would you
> > - hinder A KRC C for three, PS C and giving the pool to B or rushing
> > all of C's minions because it is not ptw (which it is obviously only
> > because A and B have a deal, which was not ptw in judge's POV at the
> > time it was made in this example)
>
> I don't know what the state is, so I couldn't say which would be the
> case, so that's why I gave the answer I did.
>
> If A damaging C is ptw in and of itself, then it is allowed under
> the ptw rule.
>
> If it is not, but it is part of a deal that was ptw, then it is
> still allowed under the ptw rule.
>
> If it is not ptw itself and it is either part of no deal or part
> of a deal that is not ptw, then it is not allowed.
>
> > - or would you only intervene at the time B wants to roll over for A
> > getting the GW?
>
> The judge should be called to intervene as soon as ptw is violated.

Is it correct that A is violating ptw, if he is honoring the deal with
B, which was ptw for him at the time made, but not ptw for B at the
time made, aka as illegal or no deal?

I'm just asking this because actually there are two sides of a deal,
the winning side (in this case A, who gets the GW and for that IS ptw
in keeping it) and B, who was not ptw (because he misjudged his own
position or because he wants to 'break the deal TM' in the future).
Until now my opinion was, that if A is damaging and ousting C for B he
would ptw, because he was "on the road to GW" because of the deal.
But now, I as a judge will step in and say "stop ousting C, that is
only helping B and that is not ptw for you, because B made an
illegal/non-ptw deal" (only if any player calls me to the table of
course ).

If my answers above are correct I finally got it and know, what
honoring a deal exactly means for you.

LSJ

unread,
Nov 21, 2004, 9:57:52 AM11/21/04
to
Joscha Duell wrote:
> LSJ <vte...@white-wolf.com> wrote in message news:<CSHnd.941288$Gx4.7...@bgtnsc04-news.ops.worldnet.att.net>...
>>If A damaging C is ptw in and of itself, then it is allowed under
>>the ptw rule.
>>
>>If it is not, but it is part of a deal that was ptw, then it is
>>still allowed under the ptw rule.
>>
>>If it is not ptw itself and it is either part of no deal or part
>>of a deal that is not ptw, then it is not allowed.
>
> Is it correct that A is violating ptw, if he is honoring the deal with
> B, which was ptw for him at the time made, but not ptw for B at the
> time made, aka as illegal or no deal?

No. Not if A didn't realize the deal was not ptw for B (that is, if A
could reasonably assume the deal was ptw for B).
If A can expect B to honor the deal, then honoring the deal himself
is an example of ptw, according to the set-up given.

Ector

unread,
Nov 22, 2004, 5:37:20 AM11/22/04
to
LSJ <vte...@white-wolf.com> wrote in message news:<Q12od.44154$7i4....@bgtnsc05-news.ops.worldnet.att.net>...

> Joscha Duell wrote:
> > LSJ <vte...@white-wolf.com> wrote in message news:<CSHnd.941288$Gx4.7...@bgtnsc04-news.ops.worldnet.att.net>...
> >>If A damaging C is ptw in and of itself, then it is allowed under
> >>the ptw rule.
> >>
> >>If it is not, but it is part of a deal that was ptw, then it is
> >>still allowed under the ptw rule.
> >>
> >>If it is not ptw itself and it is either part of no deal or part
> >>of a deal that is not ptw, then it is not allowed.
> >
> > Is it correct that A is violating ptw, if he is honoring the deal with
> > B, which was ptw for him at the time made, but not ptw for B at the
> > time made, aka as illegal or no deal?
>
> No. Not if A didn't realize the deal was not ptw for B (that is, if A
> could reasonably assume the deal was ptw for B).
> If A can expect B to honor the deal, then honoring the deal himself
> is an example of ptw, according to the set-up given.

To my mind, players must PTW always. This means that player B that
just ousted players C and D MUST break his deal and fight with player
A for GW. This, in turn, means that accepting such deal would be
stupid for player A, and the deal becomes very unlikely. Wonderful!
Stop the damned table-splitting! Table-splitting is really bad thing
as it often involves "out-of-play" relations between players.
I really cannot understand why do you think that judge should examine
such things as deals between players. Players can make or break any
deals they want, and judge shouldn't care about that. Your ruling
makes judge's work extremely hard: imagine that players A and B
pretend to have a deal, but players C and D don't remember it, or they
remember another wording of the deal... Who is right?
Let's just prohibit "giving up" at any time, and the game would become
much better (IMHO)

Ector

LSJ

unread,
Nov 22, 2004, 6:41:53 AM11/22/04
to
Ector wrote:
> To my mind, players must PTW always. This means that player B that
> just ousted players C and D MUST break his deal and fight with player
> A for GW.

This has been covered before, so I refer you to a previous thread (and
the historical threads it, in turn, cites).
http://groups.google.com/groups?selm=2t2pv9F1qqe8jU1%40uni-berlin.de

Ira

unread,
Nov 22, 2004, 8:13:35 PM11/22/04
to
Hello LSJ,

Thank you, as always, for your patience with me (and all of us.) Feel
free to not answer this post if it's a waste of time, and I just don't
realize that yet. :)

> It is not a hassle, no. It's judging.

"Judging" and "hassle" are not mutually exclusive. ;)

> "Long gone" and "part of the current game" are mutually exclusive.

I will rephrase. What is the difference between ruling based on a
past gamestate that is described by players in person, and ruling
based on a past gamestate (i.e. case-study) that is described by
players over e-mail? You seem to have no problem with the former, but
studiously avoid the latter.

Thank you!

Ira, who thinks that case studies by LSJ would improve judging quality

Daneel

unread,
Nov 23, 2004, 3:05:37 AM11/23/04
to
On 22 Nov 2004 17:13:35 -0800, Ira <ira...@gmail.com> wrote:

> I will rephrase. What is the difference between ruling based on a
> past gamestate that is described by players in person, and ruling
> based on a past gamestate (i.e. case-study) that is described by
> players over e-mail? You seem to have no problem with the former, but
> studiously avoid the latter.

IANLSJ, but I think the objectivity criterion is easier to meet
when every participant can be interviewed and the game state/discards
can be observed.

--
Bye,

Daneel

Ector

unread,
Nov 23, 2004, 4:05:14 AM11/23/04
to
LSJ <vte...@white-wolf.com> wrote in message news:<5gkod.47171$7i4....@bgtnsc05-news.ops.worldnet.att.net>...

> Ector wrote:
> > To my mind, players must PTW always. This means that player B that
> > just ousted players C and D MUST break his deal and fight with player
> > A for GW.
>
> This has been covered before, so I refer you to a previous thread (and
> the historical threads it, in turn, cites).
> http://groups.google.com/groups?selm=2t2pv9F1qqe8jU1%40uni-berlin.de

Sorry for wasting your time, but the question is really serious for
me.
These are some exerpts from the linked thread I cite just for
reference:

-------- Start of citing
--------------------------------------------
> A different question: may I (as player) propose a 3-2 split (where I get 2),
> intending to break that deal and getting at least 3VP? (In a situation where
> I might get more than 2VP on my own, but I stand to gain from luring the
> other player into believing the deal is genuine)
>
> I think I know the answer, but I'd like to be sure.

[LSJ]
Yes. The difficulty comes in making a deal if the other player knows
it isn't "playing to win" and therefore you will be obliged to break
the deal.
----------------------------------------------------------------------------
> For example, if your
> grand-predator at a 5 player table and you make a deal that he helps
> oust your prey and grand-prey if you help him oust his prey (your
> predator) and then roll over for him, giving him 3 VPs and 2 VPs, but
> secretly you plan to renege on the deal and nail him when he's most
> vulnerable.

[LSJ]
Then the only way you could legally make the deal is if you plan to
break it. Your partner in the deal will know that. Now, who would
be your partner in such situations?
----------------------------------------------------------------------

[LSJ]
The judge's judgment as to whether it is a legally-honorable deal for
the player accepting the 2VP (loss) may factor in information that
is hidden from the other player (the contents of hands, uncontrolled
regions, etc.).
The judge need only inform the players when it comes time for the
deal to be honored when honoring it runs contrary the play to win
rule if the deal was not "legally-made (to be honorable).
--------------------------------------------------------------------
[LSJ]
Explain the rules to them: that an illegal deal cannot be used to
circumvent "play to win" later.

The player making the deal will either be told that game state sans
his privileged information voids the deal (in the presence of the
other players) or will take the judge aside to discuss his privileged
information (cards in hand or whatever) and receive the ruling then.
--------------------------------------------------------------------
[LSJ]
The players' opinions as to what is or is not maximizing their VPs
are not part of the test for legality. The judge's job (or part of
it) is to "see" whether it is legal or not, based on his or her own
evaluation of the game state (he or she can solicit "logical
arguments" from the players if he or she likes, but ultimately
the decision rests with him or her, not with the players). It is
correct that the judge's job doesn't require him or her to show
(i.e., "prove") that a deal is legal -- he or she need only
declare his or her findings (judgment).
-------- end of citing ----------------------------------------

Please answer am I right in the following statements:

1) When players are going to make a deal that can possibly lead to
violating PTW rule, somebody should immediately call a judge to check
legality of the deal. Without it judge will be unable to determine the
legality in the future, as game state may change significantly. The
best way to avoid future problems is a call from the player that's
going to use the deal as a permission to violate the PTW rule (e.g. a
player who is going to "give up" in the future).
2) Judge should make a decision on the legality, but he shouldn't
declare it. If a player wants to know whether a deal he is going to
suggest would be legal, he can ask judge, and the judge can tell him
his decision made ONLY using the information available to that player
(again, this is the best way to avoid future problems). Judge should
consider a deal legal only if he believes that the player is unable to
get more VPs without it.
3) If a judge considered the deal illegal, or if he wasn't called and
now believes that the deal was illegal, he shall force the players to
PTW and break the deal.

Ector

LSJ

unread,
Nov 23, 2004, 6:29:03 AM11/23/04
to

Correct.

LSJ

unread,
Nov 23, 2004, 6:37:11 AM11/23/04
to
Ector wrote:
> 1) When players are going to make a deal that can possibly lead to
> violating PTW rule,

That's virtually every deal.

> somebody should immediately call a judge to check
> legality of the deal.

No. Only if there is a question about the legality of the deal.
The vast majority of the deals that players make are on the
clear-cut side of things (one way or the other).

> Without it judge will be unable to determine the
> legality in the future, as game state may change significantly.

Incorrect. For the clear-cut cases, the game state can be sufficiently
remembered to still be clear-cut. For the close-to-the-line cases,
you can call a judge and, even if you don't, still should be able to
regenerate the elements of the game state in question to satisfy the
judge. For elements that cannot be adequately reconstructed/agreed
upon, the judge will have to use his judgment.

> 2) Judge should make a decision on the legality, but he shouldn't
> declare it.

Depends on whether the decision was made based on any hidden
information.

> If a player wants to know whether a deal he is going to
> suggest would be legal, he can ask judge, and the judge can tell him
> his decision made ONLY using the information available to that player
> (again, this is the best way to avoid future problems).

Yes.

> Judge should
> consider a deal legal only if he believes that the player is unable to
> get more VPs without it.

... or if it leads to a GW.

With the appropriate "reasonably"s thrown in.

> 3) If a judge considered the deal illegal, or if he wasn't called and
> now believes that the deal was illegal, he shall force the players to
> PTW and break the deal.

Both the "if" and the "or if" are the same thing:
If the deal is illegal, it cannot be used as protection from PTW.

This needn't mean that the players must break the deal. On occasion,
the activities involved in keeping the deal are in and of themselves
PTW at the time it comes to do them.

Ira

unread,
Nov 23, 2004, 4:41:23 PM11/23/04
to
Hey Daneel,

> IANLSJ, but I think the objectivity criterion is easier to meet
> when every participant can be interviewed and the game state/discards
> can be observed.

Thanks for the answer; if it's not too much trouble, could you define
"the objectivity criterion" ? I don't know what that means.

IMO, in case studies the gamestate (including discards) and player
interviews could be described equivalently to the gamestate and player
interviews available to a judge 5 turns after a deal has been made and
three players have been ousted.

Furthermore, even if it's "harder" to judge based on a case study
(because they are intentionally difficult situations designed to
enlighten judges), the benefits justify the effort.

Perhaps that's where LSJ and I disagree?! I think case studies by
LSJ would provide a great benefit to judges, and perhaps LSJ thinks
the benefit would be minimal or even negative.

Ira

Daneel

unread,
Nov 23, 2004, 5:23:18 PM11/23/04
to
On 23 Nov 2004 13:41:23 -0800, Ira <ira...@gmail.com> wrote:

> Hey Daneel,
>
>> IANLSJ, but I think the objectivity criterion is easier to meet
>> when every participant can be interviewed and the game state/discards
>> can be observed.
>
> Thanks for the answer; if it's not too much trouble, could you define
> "the objectivity criterion" ? I don't know what that means.

Well, IASNLJS, but IMHO judges need to be impartial. They may not rely on
the accounts of a single perspective. They also usually need to actually
see the table (as most judges are theoretically experienced players as
well, they can use the same instinct that guides them through the game).
Debating a case in theory will always remain theory, because if a
theoretical case is clear-cut, then there is no problem judging, if it is
hazy or cornercase, then judging would probably rely heavily on the
"player instinct" most judges are probably keenly aware of.

> IMO, in case studies the gamestate (including discards) and player
> interviews could be described equivalently to the gamestate and player
> interviews available to a judge 5 turns after a deal has been made and
> three players have been ousted.

Exactly my point. But for an impartial judgement more acuracy is needed.

> Furthermore, even if it's "harder" to judge based on a case study
> (because they are intentionally difficult situations designed to
> enlighten judges), the benefits justify the effort.

What exactly are you looking for? I mean, really?

If a case is obvious then it does not warrant a case study at all, because
it is obvious.

If a case is not obvious, then being theoretical won't enhance the issue.

IMHO the only thing that could do this matter any justice is to discuss
_actual_ examples of fair play judgements without even the _slightest_
intent of establishing any sort of unofficial precedential code. Because,
while discussing actual cases may help future judges face the perils of a
demanding judgement, it falls short of defining any sort of clear-cut
reference, because hazy cases can seldom be made clear-cut.

> Perhaps that's where LSJ and I disagree?! I think case studies by
> LSJ would provide a great benefit to judges, and perhaps LSJ thinks
> the benefit would be minimal or even negative.

I think I don't know what LJS thinks, but I can imagine that demanding any
sort of case-study producing initiative from him is stressing his work
time. I really encourage you (and other forumites) to provide any
judgements they make (or encounter) for debate on this board. Because
debating actual cases (preferably observing all points of view) can shed
more insight on judging than theoretical examples ever would.

Of course, IASNLSJAINWBLSJE. ;)

--
Bye,

Daneel

Ector

unread,
Nov 24, 2004, 3:50:36 AM11/24/04
to
I assume that a judge who noticed a player violating the PTW rule
should "make the player play to win". Can the judge issue a penalty
instead, and what penalty?

LSJ

unread,
Nov 24, 2004, 6:22:07 AM11/24/04
to

Any penalty he likes. The main purpose is to ensure that no player
gains any benefit from the illegal activity.


--
LSJ (vte...@white-wolf.com) V:TES Net.Rep for White Wolf, Inc.

LSJ

unread,
Nov 24, 2004, 7:55:27 AM11/24/04
to
"Ira" <ira...@gmail.com> wrote in message
news:fa0c3aac.04112...@posting.google.com...

> Perhaps that's where LSJ and I disagree?! I think case studies by
> LSJ would provide a great benefit to judges, and perhaps LSJ thinks
> the benefit would be minimal or even negative.


Perhaps.

In either case, any and all judges who feel that case studies would be
beneficial are free to make them.

--
LSJ (vte...@white-wolf.com) V:TES Net.Rep for White Wolf, Inc.

Ira

unread,
Nov 24, 2004, 3:22:51 PM11/24/04
to
Hey LSJ,

> In either case, any and all judges who feel that case studies
> would be beneficial are free to make them.

I feel that case studies that include your insights would be very
beneficial; my insights are trivially beneficial. The insights of
vastly experienced judges would be more beneficial than mine, but
still less valuable than your words and thoughts.

Therefore, I'm reluctant to embark on case studies without your
participation. :)

"Insights" could be anything, but I'd love to learn: what you would
think about and consider in the given situation, what you'd do to
avoid the situation in the first place, what you'd do if such a bad
situation somehow managed to occur, other thoughts that judges should
think about in a similar situation, the key parameters that (if
changed) would cause you to rule in a different way, etc.

All of my judging questions can be answered via precise yes or no
questions to you, but giving you the freedom to respond somewhat
free-form to a case study is a better teaching method IMO.

Regardless, thank you again for all the work you do for this great
game, by whatever methods you choose! :)

Ira

0 new messages