LAX: Cornell 8 Dartmouth 7 Final

[peterg]

Tie breaker is head to head, so a two way tie is simple.  Three way ties where the head to heads do not settle the issue are decided by....a coin toss.  That is how Dartmouth got the AQ two years ago.

Dartmouth still has to play Yale, Brown (both this week) and Princeton.
Brown has Dartmouth, Princeton, and Cornell.
Princeton has Brown, Cornell and Dartmouth.

All things considered, Cornell is in a very good position following today's game.  Great win on the road.

[nshapiro]

So i guess that means that a Cornell win over Princeton, and a Dartmouth win over Brown will clinch the Ivy title for Cornell, because worst possible result would be a two-way tie with Dartmouth, and we beat them

[Al DeFlorio]

[Q]nshapiro Wrote:

 So i guess that means that a Cornell win over Princeton, and a Dartmouth win over Brown will clinch the Ivy title for Cornell, because worst possible result would be a two-way tie with Dartmouth, and we beat them[/q]
Right.  

Dartmouth and Brown still have to play Princeton, too.

[Josh '99]

[Q]nshapiro Wrote:
So i guess that means that a Cornell win over Princeton, and a Dartmouth win over Brown will clinch the Ivy title for Cornell, because worst possible result would be a two-way tie with Dartmouth, and we beat them[/q]If I'm not mistaken, in that scenario (assuming, for the sake of argument, that Brown then beats us and we and Dartmouth both finish 5-1), it's a shared Ivy title but we get the automatic bid.

[Al DeFlorio]

[Q]nshapiro Wrote:

 So i guess that means that a Cornell win over Princeton, and a Dartmouth win over Brown will clinch the Ivy title for Cornell, because worst possible result would be a two-way tie with Dartmouth, and we beat them[/q]
Wouldn't want to see it, but I think it's possible to have a five-way tie at 4-2.  Would be some coin flip.  ::nut::

[French Rage]

[Q]Al DeFlorio Wrote:

 [Q2]nshapiro Wrote:

 So i guess that means that a Cornell win over Princeton, and a Dartmouth win over Brown will clinch the Ivy title for Cornell, because worst possible result would be a two-way tie with Dartmouth, and we beat them[/Q]
Wouldn't want to see it, but I think it's possible to have a five-way tie at 4-2.  Would be some coin flip.[/q]

For those of us not in the know for Lax, is there a historical reason for the coin-flip?

[peterg]

I've always assumed it was because with a three or more-way tie, a playoff is impractical given the length of the season and proximity of the tournament to the end of the regular season.

Or it could just be an Ivy thing.

[DeltaOne81]

[Q]peterg Wrote:

 I've always assumed it was because with a three or more-way tie, a playoff is impractical given the length of the season and proximity of the tournament to the end of the regular season.

Or it could just be an Ivy thing.[/q]
Yeah, but there's no reason they couldn't do a few more tiebreakers, like H2H goal differential. At least that would *probably* break any tie.

[jtwcornell91]

[Q]DeltaOne81 Wrote:

 [Q2]peterg Wrote:

 I've always assumed it was because with a three or more-way tie, a playoff is impractical given the length of the season and proximity of the tournament to the end of the regular season.

Or it could just be an Ivy thing.[/Q]
Yeah, but there's no reason they couldn't do a few more tiebreakers, like H2H goal differential. At least that would *probably* break any tie.[/q]

At least one conference uses overall RPI to award the auto-bid in case of a tie like that.

[peterg]

[Q]DeltaOne81 Wrote:

 [Q2]peterg Wrote:

 I've always assumed it was because with a three or more-way tie, a playoff is impractical given the length of the season and proximity of the tournament to the end of the regular season.

Or it could just be an Ivy thing.[/Q]
Yeah, but there's no reason they couldn't do a few more tiebreakers, like H2H goal differential. At least that would *probably* break any tie.[/q]

Like I said, it could just be an Ivy thing (discouraging the unsportsmanlike spectacle of running up a score against an opponent, etc., etc., etc.?).

[Jeff Hopkins '82]

Or not really caring about athletics enough to come up with a realistic tiebreaker - that's an Ivy thing.

[Hillel Hoffmann]

Hey, it's better than the way the Big 10 used to decide who to send to the Rose Bowl when teams tied for the conference championship in football. If there was a two or three-way tie, they'd send whoever had waited the longest since their last Rose Bowl appearance.

[KeithK]

While I don't necessarily disagree with you Jeff, it seems like good tiebreakers are  hard to come up with when you have a seven game schedule.  Let me rephrase that.  You can come up with plenty of valid criteria (goals margin, margin in H2H, record against top 4, etc.) but there's not a very large sample size. Using RPI means using out of conference games, which strikes me as the wrong approach when deciding the conference title (even if it's not strictly the title, just the AQ).  Goals margin H2H isn't much less random as a coin flip in a three way tie like we've discussed.   Record against Top 4, like the ECAC uses, devolves to record against #4 in this case.  My point is not to say that a coin flip is wonderfully fair.  It's just not as horrible as it might seem at first glance.

Presumably they'd do something like drawing straws with an odd number of teams.  But maybe not.  You could do something goofy like "who can pick more coins right out of 10 tries" with sudden death if needed after that. :-D

[Liz '05]

That actually makes a lot of sense for the conference as a whole (publicity for more teams)...and is at least easily predictable.  Not fair to the players, though.

[DeltaOne81]

While goal diff may very possibly be nearly as random as a coin flip, at least it relates to something that actually happened *on the field*, as opposed to the wind and a thumb motion of an official.

As far as the 3-way "coin flip" goes, if I remember correct from the 3 way Cornell/Dartmouth/Princeton tie a couple years back, I'm pretty sure they just used a random number generator or something of the sort. So that would change my previous statement to "as opposed to the motion of certain electrons."

A 3 way actual coin flip wouldn't be too hard to do though. Just flip two coins (either simultaneously or sequentially) and make H,H Cornell, H, T Dartmouth, T, H Princeton, and T, T redo or such. 1/3rd chance each way around. Of course, they'd probably just do the random number generator cause then you can't blame human bias/poor flips. And concerns about imperfections in random # generators are probably above the level of athletes/coaches, even of the Ivy variety.