lowpost.net/blog - Schedule Difficulty

<< Introducing lowpost.net NBA Weblog Tracking | Home | November's Best and Worst Performances >>

Permalink
[Statistical Analysis] Schedule Difficulty

My schedule strength post from a week and half ago prompted a question about whether home court advantage should be factored into the picture. This leads to a slightly different evaluation: determining the overall difficulty level of a schedule. Besides opponent strength and the home court advantage, there are a number of other factors that could conceivably affect the difficulty of a game: opponent's opponents' winning percentages, timing of prior games, whether opponent is coming off a loss or win, opponent's record over recent games, etc.

I decided to run a series of regression analyses to determine which factors are the most significant contributors to overall schedule toughness. I incorporated regular season results from the past four full NBA seasons (2001-2005) into each analysis. Each test actually included two runs: one to predict the home team's score, and another to predict the road team's score.

I examined as many factors as possible: Opponents' winning percentage (overall + home/road), opponents' opponents' winning percentages, opponents' average points scored/allowed (overall + home/road), home court advantage, timing of prior games, opponents' record over the last 1, 5, 10, 20, etc. games.

In the end, only a handful were useful in predicting the final result. Not surprisingly, the most important factor in determining the outcomes for a given team was its own strength, measured by its average points scored and allowed across all other games in the season. I kept these variables in the regression analysis to ensure meaningful results. Here are the remaining factors, which I combined to evaluate schedule difficulty:

Opponent average points scored in all other games in the season
Opponent average points allowed in all other games in the season
Home court advantage (worth 3.41 net points)
Road team playing 4th game in 5 days (worth 1.85 net points)
Road team playing 2nd of back-to-back games (worth 1.49 net points)

Interestingly, teams were not significantly worse when playing the 4th game in 5 days or the 2nd of back-to-back games at home. Apparently, travel and unfamiliar surroundings contribute more to fatigue than game action.

Given these factors, it is possible to calculate which teams have faced the most difficult overall schedules so far this year. To do this, I compared each team's actual schedule to an average league schedule--multiplying the regression coefficients by actuals and averages, then subtracting the overall difference. The result indicates how many extra net points a team is losing or gaining each game due to the difficulty of their schedule.

The Rockets need an easy stretch ASAP.

Most Difficult Schedules (through games of 11/27/05)

Rank	Team	Net Points/Game
1.	Denver	-1.90
2.	Houston	-1.37
3.	Dallas	-1.31
4.	San Antonio	-1.16
5.	Milwaukee	-1.12

Easiest Schedules (through games of 11/27/05)

Rank	Team	Net Points/Game
30.	New York	1.49
29.	Golden State	1.35
28.	New Jersey	1.33
27.	Portland	1.33
26.	LA Clippers	1.30

Title
Body	HTML : b, i, blockquote, br, p, pre, a href="", ul, ol, li
Math Quiz 6 + 5 =	(Helps stop blog spammers)
Name
E-mail address
Website
Remember me	Yes No
E-mail addresses are not publicly displayed, so please only leave your e-mail address if you would like to be notified when new comments are added to this blog entry (you can opt-out later).