One of my favorite rating systems for NBA teams is the Simple Rating System (SRS). A team’s SRS rating is made up of two things: average margin of victory and strength of schedule. The rating is denominated in points above or below average, where zero is average.
These ratings can be used to estimate the probability that home team A will defeat visiting team B. Let me explain how I went about the process of converting SRS ratings into win probabilities.
Continue reading Home Win Probability Using SRS
More than a few people were ready to declare Philadelphia Eagles head coach Chip Kelly a genius after one half of football, and now many of those same people have decided he is a failure.
The truth, as always, is more complex than that, and while the Eagles have struggled the last two weeks, the offense has certainly been impressive the first half of the season.
Continue reading Chip Kelly: Genius or Idiot?
On Monday I wrote I a piece for ESPN Insider that used multiple years of MVP voting to determine who was generally viewed as the best player in the NBA on a season-by-season basis.
This post is an extension of that idea, although I’m going to make the following tweaks:
- Win shares will be used rather than MVP award shares.
- Three seasons of data will be used rather than four seasons of data.
- Win shares in season n will receive a weight of 1⁄2, win shares in season n − 1 will receive a weight of 1⁄3, and win shares in season n − 2 will receive a weight of 1⁄6.
Without further ado, here are the players who — based on win shares, at least — had established themselves as the best players in the game:
Continue reading The Best Player in the NBA
The Keltner List is a series of subjective questions formulated by famed sabermetrician Bill James used to help assess whether or not a player deserves to be elected to the National Baseball Hall of Fame.
Although the system was designed to evaluate baseball players, with a few minor tweaks it can also be used to assess the Hall-worthiness of basketball players. Today I will assess the Hall of Fame chances of Sidney Moncrief, an All-Star shooting guard for the Milwaukee Bucks in the 1980s.
Continue reading Is Sidney Moncrief a Hall of Famer?
While doing some research for a future post I got sidetracked looking at players who were named to the All-NBA and All-Defensive teams in the same season.
I thought the results were interesting, so I’m going to post them here with some commentary sprinkled in. Keep in mind that the league began naming All-Defensive teams following the 1968-69 season, so many all-time greats — most notably Bill Russell and Wilt Chamberlain — will be missing from these lists.
Continue reading Dual All-NBA and All-Defensive Selections
My favorite contemporary writer, the incomparable Bill James, has used the “In a Box” concept in several of his books. Basically what James does is choose a topic (e.g., a baseball manager) and then makes an idiosyncratic list of the topic’s defining features. I think the format works well for a blog post, so today I would like to put the San Antonio Spurs franchise “In a Box”.
Continue reading The San Antonio Spurs Franchise In a Box
Earlier this week over at ESPN Insider I took a look at the average career value produced by each of the top 14 NBA draft picks (i.e., the lottery picks).
You have to be an Insider in order to read that piece, so let me briefly summarize what I did in order to get a career value for each player:
- A player’s season value is equal to his regular season win shares plus his postseason win shares.
- The player’s season values are ordered from best (highest) to worst (lowest).
- The player’s career value is equal to 100 percent of his best season, plus 95 percent of his second-best season, plus 90 percent of his third-best season, etc.
In the Insider column I went on to give the average value for each of the lottery picks as well as the top three players and bottom three players for each slot.
In today’s post I’m going to expand on that idea a bit by building a model to figure out the expected value for each lottery pick*, then use that model to find some of the best and worst draft picks since the ABA-NBA merger in 1976.
* From here on out I will use the term “lottery pick” to refer to a top 14 pick even though the draft lottery did not start until the 1985 NBA Draft, not to mention that not all top 14 picks were technically lottery picks.
Continue reading The Value of an NBA Lottery Pick
The other day I was glancing through a list of players who had won the NBA’s Most Improved Player award and it struck me that many of the winners were not able to sustain their success after their award-winning season.
In other words, these were just cases of players having fluke seasons, statistical anomalies in otherwise ordinary careers. That led me to the question that motivated today’s post: What were the biggest fluke seasons in NBA history?
I used a couple of formulas to help me along the way, but in the end this more art than science. When making the selections I eliminated cases where an established star was having his peak season. For example, Tracy McGrady’s 2002-03 season stands out in his career line, but it’s hard to call it a fluke given that McGrady was a seven-time All-NBA selection.
Here are my choices for the six biggest fluke seasons in NBA history, presented in chronological order. Please feel free to agree or disagree in the comments, and let me know if you feel that I’ve missed someone.
Continue reading The Biggest Fluke Seasons in NBA History
My favorite contemporary writer, the incomparable Bill James, has used the “In a Box” concept in several of his books. Basically what James does is choose a topic (e.g., a baseball manager) and then makes an idiosyncratic list of the topic’s defining features. I think the format works well for a blog post, so today I would like to put the Miami Heat franchise “In a Box”.
Continue reading The Miami Heat Franchise In a Box
As with many things in sports analytics, it started with Bill James.
James was looking for a way to translate a baseball team’s runs scored and runs allowed into wins and losses. He noticed that the relationship could be expressed as follows:
(W / L) = (RS2 / RA2)
The above could then be used to express a team’s won-lost percentage as a function of runs scored and runs allowed:
WL% = RS2 / (RS2 + RA2))
Because of the squared terms in the equation above, James chose to dub this the Pythagorean formula. James found that when he used this formula to try to predict a team’s win total given its runs scored and runs allowed, he would usually get within plus or minus four wins of the team’s actual win total.
Continue reading Pythagoras of the Hardwood