Up until this point, the most troubling part of my project was gathering contract details, especially ones from the 1970’s and 1980’s.  I thought the data would be readily available and all contained in one place.  However, I was mistaken.  Baseball Reference, the site that I was expecting to find the details on, was inconsistent.  Sometimes it would show a player’s annual salary, but other times it would not.  There were also instances where Baseball Reference would give a salary but mark it with an asterisk to signify different reported amounts.  The other big problem with Baseball Reference is that it only has the annual salary, not the details of the full contract.  There were times when players would earn incentives that were built into the contract but not included as guaranteed money.  I am focusing on the guaranteed money only, so sometimes Baseball Reference could be a little misleading.  (This is one of my only knocks on Baseball Reference; I love that website.)  Cot’s Baseball Contracts, a Baseball Prospectus webpage, contained exactly what I was looking for.  There was a nice spreadsheet with most free agent contracts from 1991-present, and it listed length and total guaranteed money.  I relied heavily on that spreadsheet for the free agent deals that have occurred in the past 30 years.  Unfortunately, Cot’s had no data from the 1970’s and 1980’s.  I found that generally speaking, the best sources for data from those decades were newspaper articles.  The New York Times had a good amount of information, as did United Press International.  But scouring these databases and others like them was a tedious task.  There were times when I had to remove a free agent from my data set and replace him with someone else because I couldn’t find any contract details.  

Unless I make a drastic change, I am done collecting contract information.  I was able to find details for 10 free agents from each free agent class since the inaugural one after the 1976 season.  I chose these free agents by sorting by WAR3 (wins above replacement over the past three years) and choosing the top ten.  This wouldn’t always result in the largest contracts, but it should theoretically result in the most productive players over the past three seasons.  I’m not sure if I could have done anything differently.  Every project has some data that is really tough to find, and there’s not really much a researcher can do except be methodical and look for the information as thoroughly as possible.