From: milivella on
Sid:

> I am thinking of a good rule base to run an office competition for the
> world cup.

I've a project for an unsophisticated game that I will probably submit
to the ng, but maybe you can use the idea if you like it. Basically,
for each NT the gamemaster predict the last stage reached (winner,
final, semifinal, etc.) [1]. You can pick any number of teams. For
each team picked by you, you get +1 point for each stage reached more
than the predicted one, -1 point for each stage not reached (e.g.
Italy is predicted round of 16 but go out in the group = -1 point).

I guess that it respects your conditions:

> 1. Be 'simplish' (the sophisticated format will not work)
> 2. Include both group and knockout stages
> 3. be fire-and-forget (something that does not need contestants to
> come back after the group stages)
> 4. Needs a little bit of effort (not just pick the overall winner)
> 5. Not too much effort (predict the score for every game)
> 6. Obviously a point system which covers all stages. (there may be a
> minimum amount of money involved - say $10 an entry).
"7. The competition should allow for unlimited entries. (i.e. a system
where someone
bids for and gets a team or a pool of teams would not work). We have
200-300 employees in this office."

More, result computations takes an instant: you just have to fill a
spreadsheet (or let each contestant compute his score! ;) ). And, most
important, is should "make folks a bit more interested in the whole
tournament and some of the non-top teams."

But the ideas by Jesus and Luiz rock!

[1] I'll use objective predictions like
http://www.castrolfootball.com/predictor/

--
Cheers
milivella
From: milivella on
milivella:

> Basically,
> for each NT the gamemaster predict the last stage reached (winner,
> final, semifinal, etc.) [1]. You can pick any number of teams. For
> each team picked by you, you get +1 point for each stage reached more
> than the predicted one, -1 point for each stage not reached (e.g.
> Italy is predicted round of 16 but go out in the group = -1 point).

Sorry, I forgot to add: you also "pay" 1 point for each picked team.
Otherwise, everybody would pick all the teams that are predicted to
exit at the group stage, because it could do no harm to your score.

--
Cheers
milivella
From: Clément on
On May 13, 5:53 pm, milivella wrote:
> milivella:
>
> > Basically,
> > for each NT the gamemaster predict the last stage reached (winner,
> > final, semifinal, etc.) [1]. You can pick any number of teams. For
> > each team picked by you, you get +1 point for each stage reached more
> > than the predicted one, -1 point for each stage not reached (e.g.
> > Italy is predicted round of 16 but go out in the group = -1 point).
>
> Sorry, I forgot to add: you also "pay" 1 point for each picked team.
> Otherwise, everybody would pick all the teams that are predicted to
> exit at the group stage, because it could do no harm to your score.

If I understood it correctly, there's still one more issue. I could
pick all the strongest teams and predict them to exit at the group
stage. Then potentially I would get a very high score, assuming the
strongest teams avance to the later stages.

In fact, thinking about it, the prediction becomes pointless, because
I should predict every team I pick to leave at the group stage,
because then my score can only grow.

From the top of my mind, there are 2 alternatives to it:

1) The penalty for getting the prediction wrong could be -1 per round,
regardless of whether the team left earlier or later than you
predicted; or

2) You simply pick a maximum X teams, and earn a point for each round
they reach after the group stage. In order to avoid everyone picking
the same X favorite teams, each pick could have a different cost,
based on the team's strength (the criteria to assign this might not be
that simple). In this alternative, predictions would not be necessary;
each player's goal would be to pick the best X finishers, while paying
a minimum cost for them. Spain will likely get you many points, but is
also very expensive. New Zealand would be a cheap pick, but probably
would bring you zero points while wasting one of your limited picks.


Abraço,

Luiz Mello
From: milivella on
Clément:

> On May 13, 5:53 pm, milivella wrote:
>
> > milivella:
>
> > > Basically,
> > > for each NT the gamemaster predict the last stage reached (winner,
> > > final, semifinal, etc.) [1]. You can pick any number of teams. For
> > > each team picked by you, you get +1 point for each stage reached more
> > > than the predicted one, -1 point for each stage not reached (e.g.
> > > Italy is predicted round of 16 but go out in the group = -1 point).
>
> > Sorry, I forgot to add: you also "pay" 1 point for each picked team.
> > Otherwise, everybody would pick all the teams that are predicted to
> > exit at the group stage, because it could do no harm to your score.
>
> If I understood it correctly, there's still one more issue. I could
> pick all the strongest teams and predict them to exit at the group
> stage. Then potentially I would get a very high score, assuming the
> strongest teams avance to the later stages.
>
> In fact, thinking about it, the prediction becomes pointless, because
> I should predict every team I pick to leave at the group stage,
> because then my score can only grow.

Sorry, I haven't explained the concept well. Let me retry.

It's *the gamemaster* that predicts the most likely stage reached by a
team. Such a prediction is the base for your score.

I.e. the gamemaster publish a list like this (fictional):
- Final: Brazil, Spain
- Semifinal: Argentina, England, Germany, Netherlands
- ecc.

Then you play, i.e. you choose what teams to pick; you don't predict
what stage they will reach.

E.g. you pick Brazil and Netherlands. Then, if Brazil perform as
predicted (by the gamemaster!) and reach final, you get -1 point
(price). Similarly, if Netherlands win the cup, you get +2 points.

So the game is basically picking the NTs that are in your opinion
underrated (by the gamemaster, that hopefully reflects common sense).

As you see, each team can hurt your score (even a team predicted to
exit at the group stage, since you pay 1 point to pick it), but can
benefit your score as well (assuming that, in the improbable case a
team is predicted to win the cup - i.e. has 50% or more chances to do
it! -, it can't be picked, i.e. is out of the game). So it's not true
that in some conditions your score can only grow.

I hope that I've explained the idea a bit better this time, but feel
free to ask me about it.

(Your #1 proposal would be an interesting simple game. Your #2
proposal is, I guess, practically equivalent to my game!)

--
Cheers
milivella
From: Clément on
On May 13, 7:34 pm, milivella wrote:
> Clément:
>
> > If I understood it correctly, there's still one more issue. I could
> > pick all the strongest teams and predict them to exit at the group
> > stage. Then potentially I would get a very high score, assuming the
> > strongest teams avance to the later stages.
>
(...)
>
> Sorry, I haven't explained the concept well. Let me retry.
>
> It's *the gamemaster* that predicts the most likely stage reached by a
> team. Such a prediction is the base for your score.


No, I re-read your original post and you were perfectly clear. It was
me who missed the gamemaster bit. Thanks for correcting me!

Abraço,

Luiz Mello