CPA Rankings - NCAA & NFL (TM)

CPA Retrodiction Rankings - NCAA & NFL (TM)

(c) 2003 Steven Wrathell, C.P.A., P.C.

Computer football Power rankings designed for Accuracy (TM) GOD BLESS AMERICA! Pray for American troops everywhere. Pray for the wisdom of America's leaders.

Click on one of the following links to move to one of my other pages:

Preseason NCAA I-A & I-AA rankings & conference predictions

Current NCAA I-A & I-AA rankings and ratings

Current NFL rankings and ratings

Final rankings for prior years, NCAA (since 1997) & NFL (since 1998)

NCAA Basketball ratings

My CF computer rankings study

Essay: "Why Computer Rankings are Better than Polls"

Links to the sites of others are below.


My original system, CPA Rankings, uses victory margins; CPA Retrodiction Rankings does not use victory margins and it meets all of the publicly-stated requirements for computer ranking systems for the Bowl Championship Series (BCS). CPA Rankings is primarily a predictive system that has some retrodictive characteristics. CPA Retrodiction is primarily a retrodictive system that has some predictive characteristics -- even though it does not use victory margins. How these systems work is described below. Both systems are used for NCAA Division I football and for the NFL.

AWARDS AND ACCOMPLISHMENTS: The CPA Rankings computer football rating system won several of Todd Beck's Prediction Tracker Awards (and honorable mentions) in several categories, predictive and retrodictive, for both the NCAA I-A and the


2003 NCAA IA - 2nd Place - Best Overall Retrodictive System (Large Sys.)
2003 NCAA IA - 2nd Place - Best Retrodictive Mean Error (Large Sys.)
2003 NCAA IA - 2nd Place - Smallest Retrodictive Bias
2003 NCAA IA - 3rd Place - Most Retrodictive Wins (Large Sys.)
2003 N F L  -  WINNER!!! - Smallest Predictive Bias, Entire Season
2003 N F L  -  WINNER!!! - Smallest Predictive Bias, Second Half
2003 N F L  -  WINNER!!! - Smallest Retrodictive Bias
2003 N F L  -  2nd Place - Best Against the Spread, Entire Season
2003 N F L  -  2nd Place - Best Retrodictive Mean Square Error (Large Sys.)
2003 N F L  -  2nd Place - Best Overall Retrodictive System (Large Sys.)
2003 N F L  -  3rd Place - Best Mean Absolute Error, Second Half
2002 NCAA IA - 2nd Place - Most Retrodictive Wins
2002 NCAA IA - 2nd Place - Best Retrodictive Mean Absolute Error
2002 NCAA IA - WINNER!!! - Best Retrodictive Mean Square Error
2002 NCAA IA - WINNER!!! - Smallest Retrodictive Bias
2002 N F L  -  Co-WINNER - Best Straight Up Winners, Second Half
2002 N F L  -  2nd Place - Smallest Predictive Bias, Second Half
2002 N F L  -  2nd Place - Best Overall Retrodictive System
2001 NCAA IA - Hon.Mentn.- Smallest Predictive Bias, Entire Season
2001 NCAA IA - WINNER!!! - Most Retrodictive Wins
2001 NCAA IA - WINNER!!! - Best Retrodictive Mean Square Error
2001 NCAA IA - WINNER!!! - Smallest Retrodictive Bias
2001 N F L  -  WINNER!!! - Best Against the Spread, Entire Season
2001 N F L  -  2nd Place - Best Retrodictive Mean Absolute Error
2001 N F L  -  2nd Place - Best Retrodictive Mean Square Error
2000 NCAA IA - Co-WINNER - BEST OVERALL SYSTEM (Pred. & Retrod.)
2000 NCAA IA - 2nd Place - Best Mean Absolute Error, Entire Season
2000 NCAA IA - 2nd Place - Best Mean Absolute Error, Second Half
2000 NCAA IA - WINNER!!! - Most Retrodictive Wins
2000 NCAA IA - WINNER!!! - Best Retrodictive Mean Absolute Error
2000 NCAA IA - WINNER!!! - Best Retrodictive Mean Square Error
2000 NCAA IA - WINNER!!! - Smallest Retrodictive Bias
2000 N F L  -  WINNER!!! - Best Straight Up Winners, Entire Season
2000 N F L  -  WINNER!!! - Most Retrodictive Wins

CPA Retrodiction Rankings:
2003 NCAA IA - 2nd Place - Most Retrodictive Wins (Large Sys.)
2003 N F L  -  WINNER!!! - Best Against the Spread, Entire Season
2003 N F L  -  2nd Place - Smallest Predictive Bias, Second Half
2003 N F L  -  2nd Place - Most Retrodictive Wins
2003 N F L  -  2nd Place - Smallest Retrodictive Bias
2003 N F L  -  3rd Place - Smallest Predictive Bias, Entire Season
2003 N F L  -  3rd Place - Best Against the Spread, Second Half
2003 N F L  -  3rd Place - Best Overall Retrodictive System (Large Sys.)

Note that CPA Retrodiction had the highest win % in the NFL in 2001, but did not win a P.T. Award, since it began near mid-season. Likewise, for the chart above, in the predictive categories of mean error, bias, & mean square error, a small number of results of other systems were ignored where a system was not measured for the whole season or whole second half. Note that beginning with 2003, retrodictive award categories (other than bias) have been divided into two groupings, "large" systems (those with more complex HFA factors, including both of mine) and "small" systems. The large system winners have better retrodictive results than the small system winners, due to a statistical advantage.

The CPA Rankings/CPA Retrodiction Rankings preseason conference predictions have done VERY WELL for accuracy, on Chris Stassen's website, which primarily rates preseason magazines such as Athlon, Lindy's,Phil Steele, Sports Illustrated, The Sporting News, Street & Smith, etc. CPA Rankings/CPA Retrodiction Rankings has a free site, while the magazines sell for around $5 to $8 each. No magazine has I-AA 1-123 rankings. How well did CPA Rankings do?

CPA Rankings was #1 for 2003!

CPA Rankings was #1 for 2000!

CPA Rankings was #1 for the 5-years of 1999-2003!

CPA Rankings was #1 for the 3-years of 2001-2003!

CPA Rankings was #1 for the 3-years of 2000-2002!

CPA Retrodiction Rankings was the first system to be #1 in Ranking Violation Percentage, in the final, post-bowl/playoff listings, in both Division I-A and I-AA in the same year, 2002, as calculated by Kenneth Massey in his CF Comparisons.

Year  Dvsn  CPA Retro  CPA Rkgs  Total Systems   BCS Computers' Range
2003  I-A   #2  .130   #4  .146       97        .166-.187 (7 BCS systems)
2003  I-AA  #2T .118   #2T .118       34        .141-.156 (3 BCS systems)
2002  I-A   #1  .126   #3  .141       80        .156-.198 (7 BCS systems)
2002  I-AA  #1  .159   #8T .188       31        .184-.186 (4 BCS systems)
2001  I-A   #2  .140   #3  .152       72        .164-.176 (8 BCS systems)
2001  I-AA  #1  .105   #3T .128       31        .128-.142 (5 BCS systems)
Lower ranking violation numbers are better.

In 1998, I was 208-46 (.819) in the Clickoff contest (, for college picks (tied for 17th place), finished tied for 32nd place for NFL picks, and finished tied for 10th place (out of over 12,000 who participated during the year) for combined college and NFL picks (regular season only).

Links to Beck's Prediction Tracker, and its Awards; Stassen's Preseason Accuracy site; and Massey's Comparisons, I-A and I-AA are listed among the links below.


My straight-up picking records follow. Note that CPA Retrodiction began in mid-2001 and its records are listed after CPA Rankings' starting with 2002:

2002 NCAA     519-187  (.735) / 518-188 (.734) (IA v IA)
2002 NCAA     507-179  (.739) / 503-183 (.733) (IAA+ v IAA+/IA)
2001 NCAA     486-168  (.743) (IA v IA)
2001 NCAA     487-153  (.761) (IAA+ v IAA+/IA)
2000 NCAA     476-165  (.743) (IA v IA)
2000 NCAA     486-162  (.750) (IAA+ v IAA+/IA)
1999 NCAA     508-178  (.741) (IA v IA/IAA)
1998 NCAA     519-147  (.779) (IA v IA/IAA)
2002 NFL      167-98-1 (.630) / 164-101-1 (.618)
2001 NFL      166-93   (.641)
2000 NFL      171-88   (.660)
1999 NFL      174-85   (.672)
1998 NFL      169-68   (.713) * Clickoff record only

* The only record I have of my 1998 NFL results is my Clickoff contest record (with playoffs, 13-3). Clickoff excludes Monday night and Thursday games.

The records above include bowls and playoffs. I did not keep track of my picking record in 1997.


My NCAA systems include both Divisions I-A and I-AA teams. In 2002, there were 65 games between I-A and I-AA teams. I-AA teams played 109 games against teams in the NAIA and NCAA Divisions II and III. Some systems rate teams in all of these division (over 700 of them), which is beyond what I do. Other systems solve for inter-divisional games by having a "Wild-Card U" "team," which presumes all I-AA teams, or all non-Division I teams, are equal. Other systems ignore all games that do not involve just I-A teams. By rating I-AA teams, I consider that my I-A ratings are more accurate, especially early in the season. When an "upset" occurs in an inter-divisional game involving a rated team, to ignore it or to ignore that the unrated team is much better or much worse than the average unrated team will mean that the rated I-A or I-AA team will not receive the proper increase or decrease to its rating. To keep my I-AA ratings as accurately as possible, I also determine ratings of all non-Division-I opponents, primarily using others' ratings. I don't publish those ratings until the final listing, although I have treated some Division II teams as I-AA teams, such as Cal Davis (which joins I-AA in 2003).


In the CPA Retrodiction Rankings, it does not matter whether team A beat team B by one point or by 99 points, etc. What matters is who won and where each team was previously rated. Although HFA factors are listed, these have no effect on the calculation of the ratings in Retrodiction. When a favorite beats a substantial underdog, there is no effect. When a favorite beats a team with a rating that is close to it, the favorite's rating improves and the loser's rating decreases. What moves teams the most in both systems is an upset, which has greater weight. This helps to get the best possible ranking violation. In each week's ratings, prior games gets recalculated. In CPA Retrodiction, the weight of each game is not affected by the time of the season it was played. Strength of Schedule (SOS) is a primary component of both systems, as it is with any good computer ranking system. There are different ways to look at SOS in non-victory-margin systems, though. Consider this simplified example: Undefeated A has beaten C, D, E, & F, which are ranked, respectively, #3, #40, #110, & #111. Undefeated B has beaten G, H, I, & J, which are ranked, respectively, #25 #45, #50, & #60. CPA Retrodiction Rankings would consider A's SOS to have been tougher than B's, sinceits best opponent was much better than B's best opponent and playing against the 2 weak opponents would not hurt A (unless, of course, A lost). Other non-victory-margin systems calculate an average SOS (including the BCS's SOS, although the BCS's QW factor does adjust for this) and might rank B above A, since the average rankings of the opponents for B was #45 and was just #66 for A.

Utilizing preseason ratings as a starting point keeps the ratings more accurate, especially early in the season. Beating a team that is 2-0 that was in the I-A top 25 in my preseason rankings is much more impressive than beating a non-scholarship "Mid-Major" I-AA team with the same 2-0 record. The preseason ratings become meaningless for the top teams once they have played other top teams. This prevents errors seen in some non-victory-margin systems that overrate weaker undefeated teams, such as Tulane in 1998 and Marshall in 1999 (each of those teams were rated #2 by a few systems).


CPA Rankings utilizes victory/loss margins, schedule strength, and figures in team-specific home-field-advantage factors (except for neutral-site games). A major purpose of the system is to predict games and, thus, later games may have more weight than earlier ones, but I even out the weighting near the end of the season. I combine several calculations to get the ratings, and wins and losses (relative to the competition) are considered. Teams' rankings are adjusted to a greater degree for "upsets."

I have cap factors, for the scores, to essentially eliminate the effect of blowouts. When the BCS utilized computer systems that used victory margins capped at 21 points, CPA Rankings complied with that cap. Now that a 21-point cap has no official meaning, CPA's cap varies somewhat during the year, but it is always stated on the current NCAA ratings page. In the NFL, a fixed 30-point cap has been used.

Although CPA Rankings is primarily a predictive system, my system has been the best, or close to it, in the Prediction Tracker's "Retrodictive" calculations and I have had good scores in Kenneth Massey's "Ranking Violation" calculations.


Team-Specific Home-Field-Advantage (HFA) factors are stated for in both systems, for both the NCAA and the NFL. Remember, of course, that for CPA Retrodiction, the HFA's do not affect the computation of its ratings. Add the HFA's of both teams to the predicted margin of the home team to determine the prediction from the home team's perspective. If you don't want to do the math, Todd Beck's Prediction Tracker lists these calculations with the calculations of other systems.

The fact that home-field advantage does affect the scores in football games is undeniable and most predictive systems list an HFA factor. My system, and a small number of other systems, lists team-specific HFA's to acknowledge that some teams are more affected by this than are other teams. A high or low number is not a reflection on the team's playing ability: that is reflected by its rating. An above-average HFA means that the team's home success exceeds its road success more than average. If a team has a negative HFA, it actually plays better on the road than it plays at home. I have noticed that teams that play on artificial turf often have higher HFA factors. Northern teams may have greater HFA's against Southern teams when it snows. The HFA factors will fluctuate year-to-year and differ between my two system (and, as previously stated, it is not used in determining ratings for CPA Retrodiction).

Note that in my listing of the HFA's, I am listing, effectively, each team's "half-HFA." Thus, if my listing for team X has the HFA at 1.50 and it is 2.00 for team Y, then my calculation is that the average total HFA in X's games is 3.00 and it is 4.00 in Y's games. The total HFA for their game is 3.50, the average of 3.00 and 4.00. But this is more simply calculated by the user by adding 1.50 plus 2.00. If I listed the full HFA's, as some sites do, then the user would need to also divide the sum in half. To complete the example, presume X is rated at 90.00 and Y is rated at 88.00. To calculate Y at X: 90.00-88.00+1.50+2.00 = 5.50: X is favored by 5.50 points. For X at Y: 88.00-90.00+1.50+2.00 = 1.50: Y is favored by 1.50 points. Thus, note that with closely-rated teams, the HFA can mean that a lower-rated team can be favored over a higher rated team, as in the example, where Y is favored if it is at home against X. If my systems ever picks a tie, add .01 to the home team.


No HFA factors are applicable for neutral-site games. Some games are obviously neutral games, but others are not as obvious. I try to list upcoming games and recent games that are either neutral or may appear to be neutral but are not, but only when involving teams I rate. .As I may occasionally not get one of these listed on my site in a timely manner, I want to make some clarifications. If a game is at a neutral site and I did not note it at all, consider it neutral. I will list it in this are as soon as possible.

Is it neutral? Note that my team descriptions include the state the schools are in. If two teams from two different states play in a third state (unless the site is VERY close to one of the teams), it will be considered neutral. If 2 teams from different states that are NOT adjacent to each other play in the state of one of the teams, the home state team will usually be considered to have an HFA. If the teams are in adjacent states (TX-OK, FL-GA, etc.), or if both teams are in the same state, the game will be considered neutral if the site is either reasonably midway between the 2 schools or if it is some distance from both schools (even if one school is closer). If the site is close to one and not the other, the close team will have an HFA. If both schools are in the same metropolitan area, the circumstances will have to be examined. If one team is treated as the home team for ticket sale purposes, it will have an HFA (e.g., UCLA-USC). If tickets are available evenly to both teams, it will be neutral.

Bowl games may have "natural" home teams, and there have been years where the home-field advantage has seemed to work in bowl games, but other years in which using HFA's for such games does not better reflect the results. For bowls, fans are willing to travel to "road" games in large numbers and often the teams arrive at the bowl's city well in advance of the game, eliminating the fatigue factor. Accordingly, I have decided, beginning with the 2001-2002 season, that ALL bowls, no matter what, will be considered to be neutral games. In the NFL, the only neutral game is the Super Bowl.


I am a CPA (certified public accountant) who rates college and pro football teams during that long period in between tax seasons. If you aren't far from Warren, MI, and you want your individual income tax return prepared (at a competitive fee) by someone who can talk football with you, send me an e-mail (but I take only honest clients). My B.A. is from the University of Michigan (Ann Arbor) (1982). My master's degree in accounting is from Walsh College (Troy, MI) (1985). I have been doing tax returns and accounting since 1983 and I received my CPA license in 1987. I have worked in a few small CPA firms before starting my own firm. I also taught cost accounting at Oakland Community College in 1991. For 2 years, I operated a review course for tax practitioners that were taking IRS Enrolled Agent Exam and I have taught segments of tax seminars. I opened an office for my own firm, Steven Wrathell, CPA, PC, (a Michigan professional corporation) in 1992.

I have been using computer spreadsheets long before I did football ratings. Unlike many of the CPA's I have worked for that prefer to spend their time balancing countless numbers on 13-column paper (and billing for those hours), I have always preferred the ease of letting the computer do the math. I have created spreadsheets for many purposes in accounting, including consolidations, tax worksheets, analytical ratios, audit calculations, and menu-driven budget comparisons and statements of cash flows, etc. I have modified a spreadsheet-based Medicaid Cost Report program, to add additional pages that previously had to be done manually. The use of computer spreadsheet programs for football rankings was the obvious way for me to rate the teams.

In reading the descriptions of some of the other systems, I sometimes read where the preparer has never played football and doesn't watch any games. I played football in only one season (ninth grade) until an injury tore my ligaments in my left knee (which still occasionally bothers me). I also played hockey and ran track and cross country (although years of accounting have taken away my previously lean appearance). My father played briefly on the University of Detroit football team until he joined the Air Force for the Korean War. My paternal grandfather played football at the University of Detroit (which was a good football school then) and then played professional football, for one season. When I was at Michigan, I had season tickets and I watched the Wolverines' games. Yet, my rankings do not favor Michigan or the Big Ten. I had Nebraska as #1 in 1997 even though U-M was undefeated.


The need for computer measurements of teams is more significant in the NCAA than in the NFL, since NFL schedules are much more (yet still imperfectly) balanced. Opinion polls of coaches and sports writers, despite the diligence of the participants, is tainted by bias and a lack of objectivity. If a I-A team plays just 11 of 117 teams in I-A, its strength of schedule can vary widely from that of another team playing 11 different opponents. The conference strengths of the BCS conferences differ greatly from the non-BCS conferences and non-conference games are self scheduled: some teams choose tough opponents and others choose weaker ones. Computer ranking systems are utilized by the Bowl Championship Series (BCS) as part of its method in selecting teams for the bowls in the BCS.

The primary purpose of a football ranking system, whether it is "predictive" or "retrodictive," is to provide an objective assessment of which teams are better, in accordance with a particular standard of measurement. Predictions can be made of future games from my systems and the systems of many others. For "predictive" systems, the accuracy of these predictions is considered to simply reflect the quality of the ratings. Who is better: team A or team B? My system chooses A, yours chooses B and if A beats B, then my system is considered to have more accurately utilized the available data. This is NOT in any way implying any form of fortune telling or any other occult practice. "Retrodictive" systems are not concerned with what WILL happen, but are designed to retroactively predict what has ALREADY happened. Most retrodictive systems ignore victory margins and are often good performers at minimizing ranking violation, although some predictive systems may also do well at ranking violation.

Just because a system is considered to be "predictive" does not mean that it is designed to be used in gambling. To use my systems, or other systems, as an aid in betting on games is NOT a wise thing to do. Please note that in Todd Beck's Prediction Tracker results for NCAA I-A, in 2000, EVERY system would have lost money betting against the spread, even though 7 of 25 systems were (barely) above 50% ATS (mine barely beat the 50%). It must be pointed out that there is a profit factor in the pay-outs in gambling, so a 50.0%-ATS-picking rate will LOSE money for you. Thus, I wish to emphasize that these rankings are not here to help you in gambling. I feel it's important to say that I do not bet on games and I discourage others from doing so. This site is also subject to this disclaimer, written by Todd Beck.

You may also be interested in clicking on the following links:

------------------- I-A sites: --------------------

Official BCS Rankings from ABC Sports

Kenneth Massey's College Football Ranking Comparison - I-A

PerformanZ Ratings, Prediction Tracker,P.T. Awards & more

YourLinx Ratings' Super List, interviews, & more

------------------- I-AA sites: --------------------

Kenneth Massey's College Football Ranking Comparison - I-AA

Gridiron Power Index (GPI), I-AA's BCS-style rankings For great info on I-AA football

------------------- Preseason sites: --------------------

Chris Stassen's CF Preseason/"> site (home)

See how CPA Rankings has done in the "AWARDS AND ACCOMPLISHMENTS" section above.

------------------- Other: --------------------



What is the meaning of life?

Truth, morality & "tolerance" - Absolutism vs. Relativism

The Ten Commandments

Scriptural Sabbath vs. Sunday tradition

Do the Jewish Scriptures mention Hell?


Could God became a man?

Isaiah 53 (& 52): Did Isaiah describe Jesus (Yeshua) on the cross 700 years before the event?

Psalm 22: Did David describe Jesus (Yeshua) on the cross 1000 years before the event?

Prophesies about the Messiah - Various

Prophesies about the Messiah - Various (more)

Prophesies about the Messiah - Crucifixion and Resurrection

Jesus and the Jewish (Biblical) Holidays

Jesus (Yeshua) in Bible codes (Isaish 53, Psalm 22, & more)

What is Messianic Judaism?

Congregation Shema Yisrael (a Detroit, MI area Messianic congregation with lots of great aritcles and more about the Messiah). (Note: in addition to articles in English, articles are also on the site in German, Russian, and Spanish.)

The Association of Messianic Congregations

Jews for Jesus - Believers' page (Witnessing helps, etc.) Jews for Jesus - Seekers' page (Evidence for Jesus, Objections, etc.)

Zola Levitt Ministries (Messianic ministry)

The Messianic Times (quarterly newspaper)


The American Red Cross

The Salvation Army


Seventh-Day Baptists (NOT Adventists) (keeping Sabbath since 1650)

Christian Research Institute - The Bible Answer Man (creation vs evolution, false teachings, etc.)

Billy Graham Evangelistic Association

Thru the Bible Radio, with Dr. J. Vernon McGee (teaching the whole Bible to the whole world)

Translating the Bible into languages that have never had Bibles before


Focus on the Family, with Dr. James Dobson (guidance for families and information on traditional values)

Dr. D. James Kennedy, Truths That Transform (Ministry & Church-State issues)

Bob Duco's links (WMUZ-FM, Detroit Christian radio) (Many Christian topics: abortion, creation science, religious rights, apologetics [answers to Bible questions], etc.)


Morality (Right & Wrong): Christianity vs. Darwinism

Darwin's "Natural Selection" and the Columbine massacre

Darwinism's influence on Hitler, other eugenics movements & more

You may be an atheist if... (intelligent Christian humor)


The text of The WRITTEN U.S. Constitution (not the Supreme Court version)

Quick points:

* The "legislative" (law-making) authority of the federal government is given to the Congress, not the courts. (Art. 1, Sec. 1)

* All states must have a republican form of government (which could not be compatible with a dictitorial system led by appointed judges. (Art 4, Sec. 4)

* The Constitution can be amended only thru procedures requiring the consent of 3/4 of the states and 2/3 of both houses of Congress. This precludes "amendments" by lone judges, 6-3 Supreme Court votes, etc. (Art 5)

* The Constitution is the supreme law of the U.S., not judicial precedents, even those of the Supreme Court. (Art 6, par. 3)

How Congress felt about religion in the schools in 1787:

"Religion, morality, and knowledge, being necessary to good government and the happiness of mankind, schools and the means of education shall forever be encouraged." (Article 3, Northwest Territories Act)

The Mayflower Compact (the first American constitution)

The Federalist Papers (Constitutional writers explain the provisions' meanings)

The Anti-Federalist Papers (which led to the Bill of Rights) (See #'s 78-82 on the power of the judiciary.)

Alabama's ex-Chief Justice Roy Moore's controversy

Biblical displays by U.S. Government (including displays of Moses in the U.S. Supreme Court building) (Also quotes from the founding fathers)

How did Thomas Jefferson feel about judicial control of The Constitution?

Can the U.S. Constitution be amended by European judicial precedents?

Which liberal movement (& "Constitutional" right) of today was based on black genocide?

The ACLJ (American Center for Law and Justice) (Jay Sekulow) - defenders of the WRITTEN Constitution (not to be confused with their adversaries, the Anti-Christian Lawyers Union a.k.a. the Anti-Constitutional Liars Union)

Stop judicial tyranny


Contemporary Christian music top 30 songs, per WMUZ-FM, 103.5 FM, Detroit, MI

ZERANOVA (Detroit-area piano & acoustic guitar duo -- great singers!) with... Laura Tackett and Amy Boberg (Listen to sample songs on the web!)

There is also a lot of great info & countless links at David Wilson's site. Click on the "Parent Directory," next:

<- Parent Directory

Steve Wrathell /