borodkin@uiucdcsb.CS.UIUC.EDU (09/26/85)
/* Written 10:27 pm Sep 22, 1985 by gpb@msdc.UUCP in uiucdcsb:net.sport.football */ /* ---------- "Son of George's Picks" ---------- */ Week five in College Football 1985 finds a new number one in George's poll, the Maryland Terps blanked a supposedly tough West Virginia team, 28-0, thus vaulting them up the ladder. George's Top Twenty (or so) - Week five 1. Maryland Terrapins 2. Air Force Falcons 3. Nebraska Cornhuskers 4. Auburn Tigers 5. Ohio State Buckeyes George P. Burdell George's Football Picks Atlanta, Georgia {gatech,ihnp4,mcnc,ncsu,akgua}!msdc!gpb /* End of text from uiucdcsb:net.sport.football */ Something funny is going on around here. I was mildly amused to see George rank #1 the once beaten Terps (at home against Penn State). But then I saw the AP poll this week, and the SEVENTEENTH-ranked Terps got one FIRST place vote. So what gives? Is George on the AP panel of sportswriters? Is George THAT influential? I'd like George to respond. Stephen L. Borodkin USENET: ...!{pur-ee,ihnp4}!uiucdcs!borodkin CSNET: borodkin%uiuc@csnet-relay.arpa ARPA: borodkin@uiuc.arpa
beth@gymble.UUCP (Beth Katz) (09/29/85)
While listening to the Maryland-Michigan game (in which Michigan held the Terps to ZERO points) on the campus radio station, one of the sportscasters wondered who had given the Terps a first place vote. After the way they played today, they don't belong in the top 20. I wonder whether they belong in the top 40. It will be interesting to see where they land in George's poll. Michigan sounded very good. Maybe they should be number one. I believe they received another wayward first place vote in the AP. Maybe it wasn't so wayward. As for me, I'm going back to cheering on my alma mater, William and Mary. They may be only I-AA, but at least they try to win. Maryland, Clemson, and the Washington Redskins hardly seem to be trying. Oh well, maybe next week. Beth Katz {seismo,allegra}!umcp-cs!beth
gpb@msdc.UUCP (George P. Burdell) (10/04/85)
>> But then I saw the AP poll this >> week, and the SEVENTEENTH-ranked Terps got one FIRST place vote. >> >> So what gives? Is George on the AP panel of sportswriters? >> Is George THAT influential? I'd like George to respond. >> >> Stephen L. Borodkin Stephen (and the rest of the net): No, George is not THAT influenetial. We only wish that we get paid the way thoses sportswriters do for doing this stuff. Although, we'd like to use this as a way of getting rid of most of our local sportscasters and writers. But that's another story. I imagine that the sportswriter you mentioned in your article uses the same algorithms for determining his college rankings that George uses. The basic algorithm is: given two power ratings of some teams, predict the outcome. Now take the real outcome of the game and adjust the power rating of the two teams. Ex: Team A has a power rating of 113; Team B has one of 108. Teams A plays at Team B: 113 - (108 + 3 for home field) Team A is favored by two points. Now take some hypothetical results: 1. A beats B by 10. Real margin of victory is 8. Add one point to A's power rating, subtract 1 from B's 2. A beats B by 40. Real margin is 38. Adjust each team by 4 points. 3. A beats B by 1. A actually loses due to the spread but because the gap is close, no adjustment is made 4. B beats A by 10. Actually margin is 12 in B's favor Add two points to B; subtract two from A This is why we post George's spread results in each week's article. 0-4 point margin causes no change in the power ratings; 5-9 causes one point difference; 10-14 = 2 points; 15-19 = 3 points; and 20-up causes a 4 point difference. Also, late in the season you do not adjust the power rankings as much. i.e. use 3 point adjustments instead of 4. It seems that the Vegas oddsmakers do something similar because George is remarkably close with their lines. Last year we did predictions using George on college basketball games (but not on the net) and he did very well. This year we will post George for basketball games (more on that in net.sport.hoops in the next couple of weeks). I'm glad to see some people out there are actually reading George's picks and enjoying them. We have fun working them out and we hope you have some fun gleaming over his prognastications. BTW, we apologize for not getting George's picks to the net in time for this week's games. George's owners have had several projects due this week and haven't had time to put the picks together. However, the next article will have his picks. For the record, they are being posted Friday morning (not that any of them will be perfect anyway). Now to quote another expert football picker, "Get me out of here, Percy" George P. Burdell George's Football Picks Atlanta, Georgia {gatech,ihnp4,mcnc,ncsu,akgua}!msdc!gpb
lor@ucla-cs.UUCP (10/05/85)
In article <355@gymble.UUCP> beth@gymble.UUCP (Beth Katz) writes: >While listening to the Maryland-Michigan game (in which Michigan >held the Terps to ZERO points) on the campus radio station, one >of the sportscasters wondered who had given the Terps a first >place vote. After the way they played today, they don't belong >in the top 20. I wonder whether they belong in the top 40. It >will be interesting to see where they land in George's poll. > > Beth Katz > {seismo,allegra}!umcp-cs!beth As a Maryland alum, it's really sorry to see this Maryland team. With an exceptional offensive line, good running backs, receivers, and a solid defense, this may be our best chance to win the national championship. However, Stan Gelbaugh, the starting quarterback, is one of the worst I have ever seen. Just think what team we would have had if Boomer Esiason or Frank Reich were still there. Well, we better wait 'till next year. After Gelbaugh has graduated, we will have a sure shot at beating Penn St., or even a shot at the national championship. -- Eddy Lor ...!(ihnp4,ucbvax)!ucla-cs!lor lor@LOCUS.UCLA.EDU Computer Science Department, UCLA