Golfers are familiar with the idea of handicaps. A golfer who regularly shoots 78 on the local 6100-yard course with big greens and not much trouble is a fine recreational player, but not ready for prime time. A golfer who regularly shoots 78 on Pebble Beach set up to U.S. Open conditions has some serious game.

The U.S.G.A. recognizes this reality by giving golf courses a "rating" -- which is an assessment of how difficult the course is for "scratch" (zero handicap) players -- and a slope, which is essentially a first-order linear correction to the rating. Essentially the slope recognizes that difficult courses get progressively more difficult for progressively higher handicap players. Essentially the slope and the rating recognize the reality that our first golfer would probably not break 100 on Pebble set to U.S. Open conditions and our second golfer would regularly break par at the local course.

So, what does this have to do with football coaching? In certain respects, a football coach is like a golfer playing a course. The "course" is the University's football program. Because of the fan base, facilities, exposure, scheduling, resources and the like, some places are more difficult places to win, just as some courses are more difficult places to shoot good scores. Just as the golfer can't do anything about the difficulty of the course, a coach can do little, if anything, about the natural constraints on a program. If you were looking for good golfers, you'd look for players shooting good scores on difficult courses. With football coaches, there's some of the same. Ara was .507 at Northwestern, Spurrier was .603 at Duke and Mack Brown .597 at North Carolina. When they went to Notre Dame, Florida and Texas respectively, Ara and Spurrier were both about .830 and Brown is .750, and likely to improve on that mark in the next few years.

To try to assess the relative ease and difficulty of Notre Dame and Stanford in this regard, I looked at their last 50 years of record, i.e., back to 1952. Rather than going year by year, however, I looked at 3-year "rolling" averages of their records. Each 3-year "window" is a pretty good indicator of where the program was at this point. Looking at in 3-year intervals has the advantage of smoothing out random events -- injuries, lucky wins, unlucky losses, schedule variations -- that can make a season look better or worse than it should be. For instance, in retrospect, N.D.'s 2000 team was extremely lucky to be 9-3 and the 1986 team was extremely unlucky to be 5-6. If those two teams could be put in a time machine to play each other, I'd put a lot of money on the 1986 squad.

Going back 50 years gives 48 of these 3-year "windows" -- because, obviously, we don't have the results yet from 2002 and 2003 that would be necessary to complete the windows beginning in 2000 and 2001. I then collected the top and bottom 12 (the upper and lower 25%) and list them below.

Notre Dame -- Best 12

1. 1988-90: 33-4-0 (.892) -- Holtz
2. 1953-55: 26-3-1 (.883) -- Leahy-Brennan
3. 1952-54: 25-3-2 (.867 -- Leahy-Brennan
3. 1964-66: 25-3-2 (.867) -- Ara
5. 1987-89: 32-5-0 (.865) -- Holtz
6. 1972-74: 29-5-0 (.853) -- Ara
6. 1973-75: 29-5-0 (.853) -- Ara-Devine
8. 1991-93: 31-5-1 (.851) -- Holtz
9. 1971-73: 27-5-0 (.844) -- Ara
10. 1968-70: 25-5-2 (.839) -- Ara
11. 1965-67: 24-4-2 (.833) -- Ara
11. 1966-68: 24-4-2 (.833) -- Ara

Stanford -- Best 12

1. 1969-71: 25-8-1 (.750) -- Ralston
2. 1968-70: 22-8-2 (.719) -- Ralston
3. 1970-72: 24-11-0 (.686) -- Ralston-Christiansen
4. 1976-78: 23-12-0 (.657) -- Christiansen-Walsh
5. 1971-73: 22-12-0 (.647) -- Ralston-Christiansen
6. 1977-79: 22-12-1 (.643) -- Walsh-Dowhower
7. 1990-92: 23-13-0 (.639) -- Green-Walsh
8. 1967-69: 18-10-2 (.633) -- Ralston
9. 1975-77: 21-12-2 (.632) -- Christiansen-Walsh
10. 1999-01: 22-13-0 (.629) -- Willingham
11. 1991-93: 22-14-0 (.611) -- Green-Walsh
12. 1973-75: 18-12-3 (.591) -- Christiansen

Notre Dame -- Worst 12

1. 1959-61: 12-18-0 (.400) -- Kuharich
1. 1960-62: 12-18-0 (.400) -- Kuharich
3. 1961-63: 12-17-0 (.414) -- Kuharich-Devore
4. 1958-60: 13-17-0 (.433) -- Brennan-Kuharich
5. 1956-58: 15-15-0 (.500) -- Brennan
5. 1984-86: 17-17-0 (.500) -- Faust-Holtz
7. 1985-87: 18-16-0 (.529) -- Faust-Holtz
8. 1999-01: 19-16-0 (.543) -- Davie
9. 1981-83: 18-15-1 (.544) -- Faust
10. 1962-64: 16-13-0 (.552) -- Kuharich-Devore-Ara
11. 1983-85: 19-15-0 (.559) -- Faust
12. 1955-57: 17-13-0 (.567) -- Brennan

Stanford -- Worst 12

1. 1958-60: 5-25-0 (.167) -- Curtice
2. 1959-61: 7-23-0 (.233) -- Curtice
3. 1960-62: 9-21-0 (.300) -- Curtice
4. 1983-85: 10-23-0 (.303) -- Wiggin-Elway
5. 1981-83: 10-22 (.313) -- Wiggin
6. 1982-84: 11-22-0 (.333) -- Wiggin-Elway
7. 1987-89: 11-20-2 (.363) -- Elway-Green
7. 1988-90: 11-20-2 (.363) -- Elway-Green
9. 1957-59: 11-19-0 (.367) -- Taylor-Curtice
10. 1961-63: 12-18-0 (.400) -- Curtice-Ralston
10. 1956-58: 12-18-0 (.400) -- Taylor-Curtice
12. 1962-62 13-17-0 (.433) -- Curtice-Ralston

It's interesting to see the cream rise to the top and the sour milk fall to the bottom. Holtz's mind-bending 33-4 run from 1988-90 (fueled by a 23-game winning streak) stands as one of the greatest stretches in N.D. football. To find anything better you have to go back to some legendary teams: Leahy had 26-0-2 and 29-0-1 stretches from 1946-48 and 1947-49 respectively; Rockne went 28-1 from 1919-21 and there was a 20-0-2 stretch from 1911-13. Ara's appearance on N.D.'s top 12 seven times is testament to the consistent excellence of his teams.

Notice that Devine appears only once, and that one time he was riding Ara's coattails as Ara's 21-2 record for his last two years coupled with Devine's 8-3 first year record was good enough for 6th on the list. Devine's best stretch on his own was 28-7 (.800) from 1975-77, good for 15th place. Devine's four "windows" that were solely on his watch were all good, however; he was never worse than .729.

Neither Davie nor Faust come close to the top 12. Davie's best stretch on his own was 23-13 (.629) from 1998-2000, which ranks 29th best -- i.e., well into the bottom half. Faust's best stretch all on his own 20-14-1 (.586) ranks 35th best, and just misses the bottom quarter.

If we turn our attention to Stanford's top 12, we see some well-known coaches who were among Stanford's best. Undoubtedly, Stanford's best coach in this era was John Ralston. Ralston won back-to-back Rose Bowls after the 1970 and 1971 seasons (the Jim Plunkett era) that powered Stanford to its greatest success since Pop Warner's days. (Stanford did not return to the Rose Bowl until Willingham took them after the 1999 season.) Ralston left Stanford to become head coach and G.M. of the Denver Broncos, where he enjoyed occasional but not consistent success. Walsh was 17-7 for two years in the late 1970s, which helped fuel some of Stanford's better stretches. Walsh, of course, left for the 49ers to become one of the great pro coaches of all time. Denny Green -- who also had considerable success as a pro coach up until this year -- had only a 16-18 record for 3 years, but his 8-3 final season coupled with Walsh's 10-win season in his first year in a return stint generated another of Stanford's better stretches.

Willingham's 22-13 stretch over his final 3 years is interesting, and hopefully encouraging for his prospects at N.D. This stretch was the first time since Ralston that a Stanford coach on his own, playing entirely with his recruits, had managed to generate a top-quarter stretch at Stanford. Moreover, given that Willingham took over the team after it had gone 4-7 and 3-7-1 in Walsh's last two years in his return stint, it's clear that Willingham's run was generated through his own efforts. Also, the 1999-2001 stretch was the only 3-year stretch of Willingham's Stanford career with his current offensive and defensive coordinators. After a disappointing 5-6 1997 campaign and a poor 3-8 record in 1998, Willingham did a considerable amount of house cleaning in his staff with positive results. Willingham's first three years (19-15-1, .557) also amounted to a good stretch by Stanford's standards, ranking 17th best.

If we look now to N.D.'s bottom quarter, we see the usual suspects. Davie's more frequent appearance on the list is avoided mostly by Holtz's decent teams his last two years (9-3 and 8-3), but Davie's 19-16 for his last 3 years and its place in N.D.'s history speak volumes. Occasionally, good coaches get dragged into the list by either bad predecessors or bad successors. Ara thus appears once for the 1962-64 window, even though he went 9-1 in 1964. Holtz also gets dragged in cleaning up Faust's mess. Only Devine's 9-2-1 final year avoided having the 1980-82 window fall into the bottom.

Essentially the same thing occurs at Stanford. Interestingly, Walsh's 7-14-1 performance his final 2 years would've put him in danger of making the list, but he was bailed out by Willingham's 7-4-1 record in his first year.

Now, if we attempt to compare Stanford and N.D., the results are revealing. Stanford's best stretch (.750) would be the 24th best stretch at N.D., i.e., the dead middle. The bottom of Stanford's top quarter (.591) would just miss making N.D.'s bottom quarter. Stanford's median (.508) would be right in the middle of N.D.'s bottom quarter.

Of course, it could be that N.D. has just had better coaches over this period. N.D.'s roster includes three legends -- Leahy, Ara and Holtz -- and one near-legend, Devine. But N.D. has had its share of adequate or worse coaches: Faust, Davie, Kuharich, Devore and perhaps Brennan. Stanford, moreover, has had its share of coaches who either were successful at Stanford, such as Ralston and Walsh, or demonstrated in other contexts that they were capable football coaches. Green has already been mentioned, Wiggin was the coach of the K.C. Chiefs (though with a sub .500) record before coming to Stanford and Elway won at about a .630 clip at San Jose St. before coming to Stanford. Perhaps, on the whole, N.D. has had better coaching than Stanford, but it seems very unlikely that it alone would account for the huge difference in success.

At just about every step -- top, middle and bottom -- N.D.'s records are about .200 or more above Stanford's. Thus, being conservative and giving credence to the possibility that N.D. has had somewhat better coaching over the last 50 years, coming from Stanford to N.D. should be worth about 2 games per season and going from N.D. to Stanford would cost a coach about 2 games per season. Thus, hypothetically, had Davie been Stanford's coach for the last five years, he would probably have had a record of 25-35 (.416). Willingham's 7-year Stanford record would been about the equivalent of going 58-22-1 (.731) at N.D. Willingham's 22-13 record for his last three years would be about the equivalent of 28-7 (.800) at N.D.

Now, of course, I want to emphasize that none of the "proves" anything about how Willingham will do at N.D., just as we can't prove that our hypothetical second golfer who shoots 78 at Pebble will tear up the local course. We'd just have to get him out there to see. Moreover, sometimes your game fits one course better than another, no matter what the rating and slope. British Open courses have always favored low ball hitters like Watson because of the wind, and the Masters has always favored high ball hitters like Nicklaus and Woods. But, of course, all three have them were good enough to win both the Masters and the British Open, whether or not it exactly suited their games.