OR "the death of a contest and the birth of 2 more".
 
With all you theoretical physicists, statisticians,
mathematicians and computer science professors out there,
I would have hoped that one of you would have stepped in
to save me from my own worst enemy, me. I would have
hoped when I published my contest rules last week, that one
of you would have responded to contest #3 with some reference
to the rules not making sense. But no, you just gave me
enough rope....
 
From the orginal rules:
 
 Additional points will be awarded as follows:
    a) predict the exact number of goals scored by either team
       and 3 points will be awarded (picking both teams'
       score gets you 6 points).
    b) if your predicted number of goals for either team is off by
       one,  2 points are awarded
    c) if your predicted number of goals for either team is off by
       two,  1 point is awarded
 
YOU NEED NOT PICK THE WINNING TEAM TO GET POINTS AWARDED FOR NUMBER OF GOALS
 
For example, suppose RPI faces North Dakota in the 1st round, and you choose
North Dakota to win 4-3. If RPI wins 5-4  :-) :-) :-), the scoring would be
calculated as :
0 points (didn't predict the winner)
+ 3 points(for predicting exactly that UND would score 4 goals)
+ 1 point since you predicted RPI would score 3 and they scored
     5 (hence, you were off by 2)
 
This makes perfect sense in the first round. It makes no sense
thereafter since it doesn't address the issue of how scoring takes
place if the team you picked isn't in the game whose score you predict.
For example, how does the score reflect your 2-1 prediction of the game
between North Dakota and Northern Michigan, when NMU gets eliminated
in the 1st round and North Dakota plays BC instead?
 
Since this isn't easily remedied, I scrapped contest 3
and bastardized the original contest into  contests 3A and 3B:
 
3A is simple (famous last words). You've picked the scores. Additional
points are awarded based on the goal differential between the actual
goals scored and your pick for both the winner and the loser. It doesn't
matter who you picked. It doesn't matter whose in the game. For example,
you pick Denver and CC to play in a semifinal game and Denver to win 2-1.
If UNH beats MSU 2-1, you get 3 points for picking the winners score
and 3 points for the losers score. You get zero points for failing to
pick the winning team. If MSU beats UNH 2-1, you still get the same
3+3+0 points
 
3B isn't so simple. I will use your choice for the 2 teams in each
game and award the additional points based on whether those teams are
actually playing. (Maybe an example will help).
 
Suppose you pick UNH and MSU to play one another in the semifinals
and you choose UNH to win 3-2. Suppose MSU wins 2-1.
The scoring is :
0 points for failing to pick the winning team.
+3 points since you picked MSU to score 2 goals which they did
+1 point for picking UNH to score 3 goals and they only scored 1
 
As a second example, suppose you picked UNH vs CC and you chose UNH 3-2
Now suppose that MSU wins 2-1.
The scoring is:
0 points for failing to pick the winning team
+0 points for CC since they aren't in the game
+1 point for picking UNH to score 3 and they only scored 1.
 
Sorry for the confusion (both for the original contest 3 as well
as any new confusion from that last explanation). It's the best I could do
while still trying to retain my original intent.
 
I've already managed to automate contests 1 2 and 3A using Lotus 1-2-3;
now I have to attack 3B.  I have a feeling that by next Monday,
I'll know how many levels of nested if's are allowed in 1-2-3 :-(
 
Mark Lewin
RPI '69
 
HOCKEY-L is for discussion of college ice hockey;  send information to
[log in to unmask], The College Hockey Information List.