Welcome to the CSL Ratings.  This site produces ratings for all 4 year college football teams.
The algorithm used to create these ratings rely on one simple principle, that the object of every game is to score more points than your opponent.  In other words, the primary driver of these ratings is wins and losses.  Of course, it would be easy to simply list the ratings as "standings", as they do in the NFL, but we all know that the population of teams is large as compared to the number of games played.  Therefore, simple standings are just that, overly simplistic.  The ratings must attempt to differentiate between the various schedules and strength of opponents.
The CSL ratings are composed of 3 factors.  The first is simply a team's wins minus that team's losses.
The second is an indication of the strength of one's opponents, It is measured as follows, The sum of the wins of the teams I beat minus the sum of the losses of the teams that beat me.
The third factor is much like the second, but goes one level deeper.  It is the sum of the wins of the teams beaten by the teams I beat, minus the sum of the losses of the teams that were beaten by the teams than beat me.
Each win will provide a positive impact on all three factors, and each loss will provide a negative impact on each factor.  Clearly, a win against a team with a "good" record will supply more of a reward in the second and third factors than win over a team with a "poor" record.  Similarly, a team is not penalized very much for losing to a good team, but is penalized more punatively for losing to a poor team.
Each factor is normalized within the range for the division and the three factors are weighted evenly and then summed.  The best rating would be 99.99.
Games between teams in different divisions are weighted in accordance with the overall record between the two divisions.  For instance if the FBS is 46-4 against the FCS, a FBS win over a FCS team is only worth .08.  However, a loss carries the same weight as any other loss.  Using this formula, wins against a lesser division team will still help, but at a much reduced rate than a win over a same division opponent.

NEW for 2020:

In the main stream of NCAA rating systems and polls, the is an extreme bias in favor of schools from the Power 5 conferences over the schools from the Group of 5 conferences.  I have become convinced that there is absolutely no way that a Group of 5 school, regardless of record and consistency over several years will ever make it into the current 4 team playoff bracket.  The view of the major polls, and some algorithmic nuances in most rating systems give favor to teams from Power 5 conferences and refuse to treat a win as win as a win, and a loss as a loss as a loss.  So, in 2020, I will treat P5 and Go5 as separate divisions within the universe of College Football.  The CSL Ratings will now rank in 6 Divisions: D1A FBS P5, D1A FBS Go5, D1A FCS, DII, DIII, and NAIA.  D1A FBS Independents will be assigned to P5 of Go5 status based on how many games they play against the respective sub-subdivisions.  Based on 2019 Schedules, Notre Dame will be the only Independent included in the P5 sub-subdivision,

I will publish separate ratings for P5 and Go5, and will also publish a merged FBS rating which includes both, but which will use actual results of P5 and Go5 cross sub-subdivision games.  In 2019, P5 teams went 87-28 against Go5 teams, so the highest rated Go5 team only shows up at #37 in the merged ratings.  Unless Go5 teams improve their head to head results with P5 teams, I suspect there will be no Go5 teams in my top 25.
It is not recommended that the ratings be used to handicap games.  Not taking into account the spread, these ratings correctly predict the winner in about 75% of the games at all levels.
If you have any questions about these ratings, please email me at c_loest@yahoo.com.