0

Which Strength of Schedule formula is “correct”?

Since I created my draft order calculator an issue that has come up a couple of times is what formula you should use when calculating strength of schedule during the season for a given team. Should you only include the win-loss records of the teams that the team has played to that point or should you use a formula that includes the win-loss records of all 16 teams (10 teams played once, 3 teams played twice) on the team’s schedule.

Both approaches have their merits, but the method you use should depend on what you are trying to illustrate. I will use the example of the Jaguars and the Texans records at the moment (“the moment” being post Week 12, 2013) to show the differences in the approach. The Jaguars and Texans both have 2-9 records, tied for the worst record in the NFL. Depending on which method you use you will get a different answer to whom should select at #1 overall.

If you want to examine a scenario based on “if the season ended today” (ITSET) you should use the formula that only includes the win-loss records of teams played to that point. On the premise that the season is ending today, the future game will not be played meaning that those future games do not count. At the moment, using this method, the Jaguars’ opponents have a win-loss record of 70-51 (0.579) and the Texans’ opponents have a win-loss record of 66-55 (0.545). As the team with the weaker strength of schedule “wins” the tie breaker, if the season ended today the Houston Texans would pick #1.

However the other formula (ALL16) will give you a different answer. If you include all 16 opponents on the schedule the Jaguars’ opponents have a win-loss record of 92-84 (0.523) while the Texans’ opponents have a win-loss record of 97-79 (0.551). Using this formula the Jacksonville Jaguars are currently set to pick #1. The reason for this difference is because the Jaguars have a much weaker second half of the season, playing only one team with a winning record in their final 5 games, thus dropping their strength of schedule number quite a bit.

The draft order calculator I created was designed to give the user an idea of who is currently in pole position to select #1. It was not intended to give the user ITSET information. The ALL16 formula allows you to see who is currently “leading” the race for the #1 pick. A problem with the ITSET formula is that it can be skewed by particularly tough or easy stretches in the schedule for a particular team. For example, the combined win-loss record for the Jaguars’ opponents in the first 8 weeks of the season is 56-32 (0.636) while the combined win-loss record or the Jaguars’ opponents through the final 8 weeks of the season is 36-52 (0.409). This is obviously a huge difference.

Also, if a team meets an undefeated (or winless) team one week then their strength of schedule number can change drastically from one week to the next using ITSET. I prefer the ALL16 method as it gives a truer indication of how the season is progressing from week to week.

At the end of the day though, the formula you use doesn’t really matter as it’s all conjecture anyway. The only thing that matters is the true strength of schedule calculation that’s made following the final game of Week 17. That calculation decides who gets the #1 pick and who gets first shot at building their team of the future.