January 19, 2007
Falling Eagles
I'm sure you've heard by now about Boston College kicking Sean Williams and Akida McLain off the team. McLain hardly played this year, but Sean Williams was one of the top shot blockers in the country and had had several monster games recently. Clearly the loss will hurt BC, but as Ken Pomeroy points out in this great observation piece, the loss will very likely be overstated. BC is 5-0 in the ACC, but four of those games have been against the four worst teams in the league. Their two toughest ACC games, Virginia and Maryland, were both at home.
The Eagles are now entering a part of their season where their schedule gets noticeably tougher. According to Pomeroy's stats, BC would be favored in only three of their eleven remaining ACC games and one of those (home against Clemson) is basically a push. And that's using the Eagles' rating with Williams.
It's debatable how much losing Sean Williams will hurt BC, but it's quite likely that their season would have appeared to be heading down whether he played or not.
http://www.scacchoops.com/forms/ToughestSchedule.asp?UseOverall=1
While BC's schedule is tough, so is Duke's, but Pomeroy says Duke is underrated. I'm not sure the exact rationale behind that, but it is interesting that he seems to favor Duke despite their tough remaining schedule.
| scacchoops.com wrote: |
| Hey Dave, I went ahead and tried to calculate a way of determining which teams have the toughest schedules remaining. Not as complicated as Pomeroy's predictor, but still telling...
http://www.scacchoops.com/forms/ToughestSchedule.asp?UseOverall=1 While BC's schedule is tough, so is Duke's, but Pomeroy says Duke is underrated. I'm not sure the exact rationale behind that, but it is interesting that he seems to favor Duke despite their tough remaining schedule. |
Nice. I'm not sure I like the linear weighting though. How about using Pomeroy's power ratings instead? I guess that would be much harder to automate.
| Dave wrote: |
| Nice. I'm not sure I like the linear weighting though. How about using Pomeroy's power ratings instead? I guess that would be much harder to automate. |
That's pretty sweet, scacchoops, but I agree with Dave here. How about weighting by Pythagorean Winning Percentage (PWP)? You're already calculating all of the necessary tempo-free stats, so you could just go one step further and calculate the PWP. Maybe you then normalize it to a scale from zero to one to smooth things out a bit....
http://www.scacchoops.com/forms/ToughestSchedule.asp
| scacchoops.com wrote: |
| I liked the idea about Pythagorean Winning Percentage, so I went that route. Check it out, here is the link once again...
http://www.scacchoops.com/forms/ToughestSchedule.asp |
Very nice.
| Dave wrote: | ||
Very nice. |
One more question. Does it take into account home/away?
It doesn't take into account the home and away games, although I do have the counts listed. Hypothetically it could, but I would have to estimate how much a home or away game would help or hurt the given winning percentage.
BrightCar Car Maintenance Software - Software To Manage Your Car Maintenance Schedule
