=>I ranked the % chance of any given matchup based on how many times out of 20 a team would beat every other team in the MAC. Ex. Miami would beat BUGS 15 out of 20 times thus Miami gets .75 vs. BUGS and BUGS gets .25 vs. Miami. I then calculated on a spreadsheet the odds of the teams meeting based on the tourney bracket and came up with % chances of any team advancing to each round.
% CHANCE OF REACHING THE QUARTERFINALS
TOLEDO 100% (bye)
EMU 50%
BSU 50%
MIAMI 100% (bye)
OHIO U. 65%
BUGS 35%
AKRON 100% (bye)
CMU 55%
BUFFALO 45%
KENT 100% (bye)
WMU 65%
NIU 35%
% CHANCE OF REACHING THE SEMIFINALS
TOLEDO 90%
EMU 5%
BSU 5%
MIAMI 62%
OHIO U. 29.25%
BUGS 8.75%
AKRON 90%
CMU 5.5%
BUFFALO 4.5%
KENT 77%
WMU 19.5%
NIU 3.5%
% CHANCE OF REACHING THE FINALS
TOLEDO 59.55%
EMU .82%
BSU 1.14%
MIAMI 24.96%
OHIO U. 12.72%
BUGS .81%
AKRON 58.61%
CMU 4.29%
BUFFALO 3.52%
KENT 28.495
WMU 4.78%
NIU .31%
% CHANCE OF WINNING THE TOURNEY
TOLEDO 30.87%
EMU .12%
BSU .18%
MIAMI 9.17%
OHIO U. 3.79%
BUGS .12%
AKRON 39.37%
CMU .62%
BUFFALO .61%
KENT 16.53%
WMU 1.41%
NIU .03%
Where are you guys getting your figures from? I know there are is an obvious distance between the haves and have-nots, but to see 6 teams have less than 1/3 of 1% of a chance at winning a tournament and two others with only a 4% and 3& chance just doesn't seem right when in March Madness any team can get hot and roll.
Polish Hammer Wrote:Where are you guys getting your figures from? I know there are is an obvious distance between the haves and have-nots, but to see 6 teams have less than 1/3 of 1% of a chance at winning a tournament and two others with only a 4% and 3& chance just doesn't seem right when in March Madness any team can get hot and roll.
=>I explained mine above. Take Miami, I have them winning over BG 75% of the time and over Ohio U. 55% of the time. I have Ohio U. beating BG 65% of the time. Thus Miami's chances of making the semifinals are a variable of the % odds of playing each team and the odds Miami would beat each team. Clear as mud?
I understand, but how are you assigning these probabilities? Are you just arbitrarily saying Miami should beat BGSU 3 out of 4 tries or is there some exact scientific method with assigning these figures?
thegeneral Wrote:OMFG already done n00b! :)
http://www.ncaabbs.com/forums/mac/phpbb/...hp?t=44415
moll's been around these parts at least as long as I have (since 1997), kid. You might also check how close his judgmental probabilities are to the Ken Pomeroy stats you stole. mollautt is a seasoned MAC basketball savant, and he's probably forgotten more about the league than you know.
LOL -- take it easy buddy -- all tongue in cheek.
And I gave those numbers to Kenpom, not the other way around.
Polish Hammer Wrote:I understand, but how are you assigning these probabilities? Are you just arbitrarily saying Miami should beat BGSU 3 out of 4 tries or is there some exact scientific method with assigning these figures?
=>Arbitrary. Yes, it is my own subjective determinations on the initial probabilities subject to my own bias. I then just plug it into the spreadsheet and that gives me the final figures for each round. You may have a different analysis. If you would tell me what each team would do head-2-head against every other team in the MAC, I can insert that data into the spreadsheet and give you your own breakdown.
What you will notice though, is if you add up the %s for each round, they will equal 1 for the champion, 2 for the finals, 4 for the semis, and 8 for the quarterfinals (they may be a little off due to rounding error). You can change a teams chances around, but if you give a better chance to Kent, someone else has to go down. That is, the probabilities have to be in that range of 1, 2, 4, and 8 above for each round.
I also have a spreadsheet that does the same thing mollaut. The only difference from what you and I did is the probabilities that one team will beat another. It's pretty cool stuff, IMO.
I enjoy seeing how the percentages will change after the first round and also after every game.
I did similar calculations for the CCHA tournament, but that was much more complex because they reseed after the first round. There were a 2^4 different combinations of first round results.
I think I'm going to do the same sort of thing for the NCAA and NIT tournaments.
And I hope you realize that my initial post in this thread was tongue-in-cheek. I'm the noob here and don't want to ruffle any feathers. I used your "if the season ended today" thread to determine the matchups.
thegeneral Wrote:I also have a spreadsheet that does the same thing mollaut. The only difference from what you and I did is the probabilities that one team will beat another. It's pretty cool stuff, IMO.
I enjoy seeing how the percentages will change after the first round and also after every game.
I did similar calculations for the CCHA tournament, but that was much more complex because they reseed after the first round. There were a 2^4 different combinations of first round results.
I think I'm going to do the same sort of thing for the NCAA and NIT tournaments.
And I hope you realize that my initial post in this thread was tongue-in-cheek. I'm the noob here and don't want to ruffle any feathers. I used your "if the season ended today" thread to determine the matchups.
=>I did not take any offense at all. I gathered that you were just pointing out that you also did one as well.
=>After the 1st Round, the %s have now changed to the following:
% CHANCE OF REACHING THE SEMIFINALS
TOLEDO 90% stays at 90%
EMU 5% => 10%
MIAMI 62% => 55%
OHIO U. 29.25% =>45%
AKRON 90% stays at 90%
CMU 5.5% =>10%
KENT 77% =>70%
WMU 19.5% =>30%
% CHANCE OF REACHING THE FINALS
TOLEDO 59.55% =>56.78%
EMU .82% => 1.45%
MIAMI 24.96% =>22.28%
OHIO U. 12.72% => 19.8%
AKRON 58.61% => 59.30%
CMU 4.29% => 7.35%
KENT 28.495 => 25.9%
WMU 4.78% => 7.35%
% CHANCE OF WINNING THE TOURNEY
TOLEDO 30.87% => 27.79%
EMU .12% => .16%
MIAMI 9.17% =>7.7%
OHIO U. 3.79% => 5.53%
AKRON 39.37% =>40.28%
CMU .62% =>1.08%
KENT 16.53% =>15.25%
WMU 1.41% =>2.21%
=>After the quarterfinals, the %s have now changed to the following:
% CHANCE OF REACHING THE FINALS
TOLEDO 59.55% =>56.78%=>65%
MIAMI 24.96% =>22.28%=>35%
AKRON 58.61% => 59.30%=>60%
KENT 28.495 => 25.9%=>40%
% CHANCE OF WINNING THE TOURNEY
TOLEDO 30.87% => 27.79%=>28.6%
MIAMI 9.17% =>7.7%=>10.15%
AKRON 39.37% =>40.28%=>39.15%
KENT 16.53% =>15.25%=>22.10%