Analysis and Numbers

04/07/99 Defensing a Team's Top Option
UConn won their national title by double-teaming Duke's Elton Brand, preventing him from scoring a lot. Is it a generally good strategy to stop a team's top scoring threat? This is a basic look at the question.
11/10/98 What Do Individual Win-Loss Records Mean?
There are several different ways to evaluate an "individual win-loss" record. Each way has unique meaning and value in characterizing players' contributions.
10/14/98 The Need for Individual Win-Loss Records
The fundamental basis of evaluating teams is a win-loss record. This column outlines the need for translating wins and losses to individual performances.
11/14/97 Are They Meeting Expectations
Early records on the year are a weak indicator of whether that team will live up to predictions. Here is a statistical test that decides how well the predictions are doing -- early in the season or at the end.
05/20/97 Toying With 'Em
The Bulls will either blow you out and play their scrubs in the fourth quarter or they will play a good team tight until the fourth quarter, then turn it up. By doing this, their numbers do not reflect how good they really are... until now. This article explains how to look past the Bulls' tendency to play down to opponents.
01/20/97 Of Point Spreads and Predictions, Using a Kalman Filter
Predicting games and point spreads is not one of the goals of this journal. However, this simple tool that does just that can also be used to analyze players' overall games and may prove to be one of the best tools available for evaluating individual defense. If that doesn't convince you to read this, I should also mention that this "Kalman Filter" is also used in many engineering disciplines, in finance, in law, and, though they don't talk about it on ER, even in medicine.
01/03/97 Response to Critiques
Almost two years ago, I posted an article to a newsgroup where I chose the Rookie of the Year using methods from JoBS. A follow-up post protested the use of any sort of statistics for evaluating players. This is my response to that criticism.
11/16/96 Can the Chicago Bulls Be Perfect?
Michael Jordan stated that he wanted to have "the perfect season". These were there odds... This features the interesting mathematical result showing that if you blow out a team by 30 points, that can actually mean you are a worse team than if you blew them out by 20 points.
11/12/96 The Bridge to the 21st Century
The NBA is skimming the best players out of college basketball anticipating that they will make them strong into the next century. Take a look at what is really happening.
11/02/96 The Effect of Bad Referees
Finally I've figured out how to do it. Now I know how to strike back at the refs, to let them know how they can change the odds of winning or losing if they are doing a bad job. I finally figured out what it means for a ref to do a bad job.
10/13/96 Making Your Teammates Better: Measuring It For the First Time
We can now measure how an individual's performance is correlated with his teammates. It implies numerous things, including whether that player makes his teammates better, as they say about all of the greats.
04/30/96 Playoff Home Court Advantage
The playoffs give teams a day of rest between games and extra preparation time to stand up to that brutal home court advantage. That's the traditional logic, at least. This article examines whether there really is an effect of the playoff format upon the home court advantage.
04/23/96 The Playoff Odds
Using some statistical methods developed by Bill James and Dallas Adams, some basic odds of winning the first round playoff series can be found. The method is shown here.
04/15/96 Shooters vs. Dunkers: Their Effects on Winning
The streaky three point shooter can get you back into a game or take you out of a game. The inside player who always gets to the line or hits layups may not be as dramatic, but is the workhorse to carry you through games. Are there cases when one player is more preferable than the other?
04/07/96 The Charlotte Hornets' Decision of Whether to Trade Rex Chapman
In 1990, I was asked to analyze a proposed trade sending Cleveland's Larry Nance and Steve Kerr to Charlotte for Rex Chapman. The trade never happened, but six years later, the projected futures of the players involved look pretty good... An overview of some basic talent evaluation techniques.
04/01/96 April's Fool: The Devil in Dennis Rodman
It is said that sportswriting is made up of two things: Knowledge and BS. Most of the articles written about Dennis Rodman have fallen into the latter category. This one doesn't.
03/25/96 What Strategies Are Risky?
What exactly are risky strategies in basketball and how do they help an underdog or hurt a favorite? The implications of the method.
03/17/96 The Fundamentals for Analyzing Basketball
is an overview of the basic methods used in JoBS. Read this if you are a new reader or if you want to understand some of the theory behind the methods I use.
03/10/96 Measuring the Effects of Consistency on Winning for Individuals
Only on the web can you actually interact with state-of-the-art research. This article includes a JavaScript calculator specially designed to evaluate how often a player with a certain and wins a game. The mathematics are not very difficult, but the results are occasionally counterintuitive, which is why experimenting with the calculator is a good idea. For example, you can show that a player with a worse floor percentage and offensive rating can actually have a winning percentage higher than another player with better numbers. This occurs when the worse-looking player uses more possessions in a game... But it is hard to generalize, which is one reason why evaluating talent is difficult even with good numbers.
02/16/96 The Effect of Controlling Tempo
Theoretical statistics to show how slowing the game helps an underdog's chances. Also, a brief discussion of how the pace of the NBA has changed in the past 20 years.
02/05/96 Jordan vs Olajuwon: Who's Better? Using a Scientific Method
A scientific comparison of two of the best players in the game. Includes thorough explanation of how to derive individual win-loss percentages.
05/--/95 A Comparison of Rating Methods
This was a comparison of the team and individual rating methods seen used in various basketball news groups.
--/--/91 New Measurement Techniques and A Binomial Model of the Game of Basketball.
This article was written for a statistical journal and is thus probably the most technical one here. However, this article also introduces the Possession Scoring System, which is the backbone of all methods here.
--/--/91 Individual Win/Loss Records
A basic article that talks about how individuals create wins and losses for their team and how we can measure them.