When I read Hound Dog's article on his ranking system, I mostly thought it was very good. However, he designed the system to rank a limited number of players for each year. The idea was to redesign the system to rank all of the players. Starting with the players currently playing, then adding all of the players listed in the mini biographies on the LPGA web site. All official tournament winners were then added. Additional players were then added using the players listed in the career money list. Even more players were added by looking at the Qualifying School Results to find those who qualified to play on the tour. A lot of the data was taken from the LPGA web site. The second source was the Golf Observer web site. Finally, names of players are being  taken from the players shown in individual tournaments on the Golf Observer web site. Currently, the database has data on over 900 players. The biggest problem in developing a comprehensive ranking system is the limited data from 1950 thru 1962. Another smaller problem is the conflicting data between the the two sources.  Also, the Golf Observer web site has some tounaments with two sets of results on the same date. However, the following describes the current ranking system, which must be about Version 5.0. A rating is calculated for each year there is data, an average yearly rating is calculated and a value for the All Time Best List is calculated.

In developing a ranking system, at some point a value judgement must be made about how many criteria are to be included in the analysis and what is most important to least important. Starting with Hound Dog's rating system and looking at the requirements for entry into the LPGA Hall Of Fame a modified system was developed.

1)  If you are an elite athlete, nothing is more important than winning. 30 percent of the rating is based on wins. However, not all wins are the same. The LPGA awards two points for a Major Win for entry into the Hall Of Fame and only one point for all other wins. This rating system doubles the value of Major Wins. In the early 50's there were only about 25 LPGA professonal touring players. Since 1984 there have been at least 150 players per year who have played at least 10 tournaments (the LPGA for the Hall Of Fame requires 10 tournaments to be a full year  for Hall Of Fame minimum criteria of 10 years). The number of players increased slowly from 1950 through 1972. I have no idea how they approved new players to join the LPGA during that period. In 1973 the LPGA started running two Qualifying Schools per year to add new players. One at the start of the year and one mid year. In 1983 there were three Qualifying Schools, which increased  the number of players up to the point that in 1984 the tour became essentially what it is today. In 1984 there was only one Qualifying school toward the end of the year, which continues through today. In the 1950's the average number of wins by the players who won was about 12. In the 1960's the average number was about 9 wins. In the 1970's the number was down to about 5 wins on average. In the 1980's the number was 4 wins on average and has been just under 4 since. As competition increased the average number of wins decreased. Therefore, the rating system must value wins in the 50's less than those in the 60's, and the 60's wins less than the 70's wins and less than the 80's wins.

2) The ladies are professional players. They play to earn money. So money becomes the second criteria and accounts for 20 percent of the rating. Over time the amount of money has gone up well over the rate of inflation. So, the money rating is based on the average prize money per tournament during the year being calculated. As the number of players increased the distribution of the prize money changed. More players getting a cut means the money rating must be a function of average purse as well as the number of players winning money.

3) Points are awarded by the LPGA toward the Hall Of Fame for winning the Player Of The Year (POY). The POY is a points based award with points given for finishing in the top 10 in a tournament. This system awards 20 percent of the rating for Top 10 finishes. However, the calculation has two parts. The rating is a function finishes in the Top 3 places (percentage of tournaments in contention) and a function of finishes in the Top 10 (percentage of tournament playing very well). Because it was much easier to get a Top 3 or Top10 with less players this calculation is also function of the number of players playing in the minimum number of tournaments.

4) Points are awarded by the LPGA toward the Hall Of Fame for winning the Vare Trophy (low stroke average for the year). The stroke average criteria is 15 percent of the rating. The stroke average on the LPGA has fallen from the 1950's thoughout the years.  Golf course agronomy has improved so the courses are better, resulting in lower scores. The design and production of golf balls and golf cubs has improved resulting in better equipment and lower scores. The target stroke average for maximum rating is reset when the winner of the Vare Trophy breaks through to a lower stroke average. The zero percent rating is for target stroke average plus 6 strokes. Anyone greater than 6 strokes above target gets a negative value, but no worse than -15.

5) The remaining 15 percent is based on the percent of cuts made (tournaments finished) or in the early years percent of tournament in which money was won. In the early years players evidently played the whole tournament then only the top players won money. That was continued at least through 1983, except some tournaments started introducing cuts in the 70's. The zero percent taget for years after 1984 is making the cut 50 percent of the time, because half the field misses the cut on full field tournaments.  Before 1984 the zero percent target varies as the percent of players earning money to players entered into the tournaments. The rating can be no worse than -15. This statistic show how consistantly a player plays from week to week. The top players will miss very few cuts.

 Mickey Wright, Kathy Whitworth, and Annika Sorenstam each have at least 6 years where they won at least 6 tournaments, noboby else has more than three. Kathy Whitworth has seven seasons where she won at least 6 times, but only one season when she won at least 10 times. Sorenstam has two seasons with at least 10 wins, while Wright has 4 seasons with at least 10 wins. Almost all of the best players in LPGA history have played at least 15 years. Almost everyone who has an informed opinion on who is the best LPGA of all time, agrees that one of those three players is the best.  Using the data of those three players and the data on the number of players competing, a series of coefficients for the equations to determine the ranking for wins and POY rankings were determined. The idea was that the best statistics for those players should determine the values for maximum ratings of the various criteria. The result of this approach is that all other players are compared to the best of these three players.

To get the best comparison, the decision was made to calculate a rating based on 15 years of a players career (the prime of their career).  No players that I checked has a better yearly rating  if a calculation is made for greater than 15 years. Only years when a player has played at least 10 tournaments is considered. When a player goes past 15 years in their career, a year is eliminated based on the following criteria. If a player had a season in which they played the minimum number of tournaments, but the rating was bad because they were pregnant. The player rating should not suffer because they chose to have a child. Second, if a player played the minimum number of tournaments, but the year was affected by significant injury or illness.  Third, the players first or last year will be eliminated, depending on which gives the best result. The longer a player plays, the more years eliminated. The idea is for the most part to keep the years continuous except for pregnancy or injury or illness. Some players took several years to learn how to play and have several years at the start of their career eliminated. Some players played well past their prime and have years eliminated at the end of their career, some at the start and the end. This rewards the players with long careers. The flip side is to punish the rating of those with short careers. Hee-Won Han, Lorena Ochoa, and Paula Creamer all started in the 2000-2009 decade. They all have higher average yearly ratings than the final rating given in the published Top 50 list. Hee-Won Han's rating is lowered because she does not have 15 years (7 years). Ochoa's rating is lowered because she does not have 15 years (6 years) and slightly because she does not have 150 tournaments played (146). Creamer's rating is lowered the most because she only has 4 years and 102 tournaments. To have a calculated rating for the All Time Best list, a player must have played 100 tournaments in qualified years. The sums of the ratings for wins and money for the All Time Best list are divided by 15 instead of actual years. The ratings for player of the year statistics, stroke average and cuts or money made percent are multiplied by the tournaments played divided by 150 until the 150 tournaments are played. This explains why Babe Zaharias who played at a very high level for 6 years rates 33 on the list. This system rates on actual long term performance, not potential. Two of the criteria are allowed to calculate negative numbers rather than stop at zero. This was done to determine the lowest rated players, the young Not Ready For Prime Time Players and the older Past Their Prime Players. Young players who probably will lose playing privilages and older players who should maybe consider retirement.

This rating system was done as a fun mental exercise for a retired engineer. I tried to keep my opinions out of the system as I would require more than 10 tournaments per year and other changes. I have no axe to grind in this system. I have my favorite players but was more curious about where they would rank than trying to tailor a system to rank them higher. I have no problem with Sorenstam being ranked the best and Mickey Wright and Kathy Whitworth ranked second and third. Making the numbers of players an important  consideration in the rating is logical based on observed trends in the data and the statistical fact that with 30 players one third would make the top10 and one in ten makes the Top 3 versus one sixth in the Top 10 and one in 20 makes the Top 3 for 60 players and progressively lower percentages for additional players. I am still collecting data to improve the coefficients in the equations, Version 5.1. Based on comments I have seen on various golf sites, many people feel that the depth of players is better today than in the past. If that is true, then a statistical trend should be found in the data since 1984 (years with greater than 150 players playing minimum tournaments). As more data is analyzed, such trends would prompt a Version 6.0 System with an added factor in the equations. I am open to suggestions on how to improve the rating system if anyone has any. Hopefully people will understand what I wrote, sometimes what I am thinking does not come out in the words I write (sometimes a hazzard of having an engineer's mind).