When setting about to make rankings you have to ask yourself a few questions:
1) What data needs to be gathered?
2) What are the most important factors in measuring a team's prowess?
3) How can all this diverse data come together in a way that is responsible and actually telling?
After considering all these questions and sifting through a lot of possibilities, we chose to bring seven winning percentages together into one single rating. The choice to focus only on winning percentages was threefold: 1) Winning is the goal of basketball, 2) there are many different ways to win and fixating on certain numbers within a box-score can be deceiving, and 3) everything brought together needs to be of a similar quality so we are comparing "apples to apples."
The seven winning percentages we included:
1) Home winning percentage
- Measures a team's ability to protect its home-court advantage.
2) Road winning percentage
- This percentage was given slightly more weight than home winning percentage in order to capture the need to be able to win on the road in the playoffs. Strong road teams are more likely to "steal" games in the playoffs.
3) Close-game (3 pts or less) winning percentage
- This was chosen because winning close games is more than luck and requires BBIQ, veteran savvy, and clutch play from clutch players. There are many close games in the playoffs and the ability to find ways to win more often than not is important.
4) Winning percentage against teams with a +0.500 record
- This measurement is helpful for measuring a team's ability to "show up" for the big games against tough competition. The playoffs are inevitably against teams with good records and so this measurement can give some insight into a team's ability to play with "the big boys."
5) Pythagorean winning percentage
- Pythagorean winning percentage is the expected winning percentage based on points scored and points allowed. This is a derivative of "margin of victory" and measures that popular quantity in a winning percentage form. This is helpful because good teams tend to "put teams away" and get larger margins of victory. This is also one way of measuring "killer instinct." This winning percentage receives the largest say in our final rating because it has been shown that there is a strong correlation between Pythagorean winning percentage and playoff success.
6) Last 10 games winning percentage
- This winning percentage helps determine how "hot" or "cold" a team currently is. This is important for determining which teams are "putting it altogether" and which ones are "falling apart."
7) The winning percentage of all opponents
- This final winning percentage measures the level of competition that each team has faced during the year. This tends to even out towards the end of the year, but often there will be one conference, which is much stronger than the other, and this can help account for those differences. In our formula this winning percentage is used as a final adjusting factor for the previous six winning percentages.
So when we put these seven winning percentages together into our formula we get what we call "Adjusted Winning Percentage" or AWP. This winning percentage is meant to bring all of the above factors into one rating that can help show the strength of each team in the areas that usually correlate with playoff success (based on the last decade of results in the NBA).