Roboscout and the Four Factors of Basketball Success

Dean Oliver, Author Basketball on Paper
Dateline: 06/01/04

LET'S TALK ABOUT WHAT WINS IN BASKETBALL. It gets taken for granted a lot because, at the surface, what wins is pretty much common sense. What defines a win in the game of basketball - whether you're in the U.S., in Australia, in Belgium, in a pro league or in a high school league - is scoring more points than those guys in the other color uniforms. Each team goes back and forth trading possession of the ball hopefully for points, sometimes for nothing. Sometimes they run up and down the court, sometimes they walk. Sometimes they press, sometimes they sit in a zone, sometimes - as in the case of Grinnell College in Iowa - they try to press but are only partially effective and they give up a lot of layups very quickly. But all these defenses are just trying to get the ball back. In the case of Grinnell, they're willing to give up easy layups to get the ball back so that their offense can go attempt a three-pointer, something they do 60 times a game.

So, returning to the theme here, what wins a basketball game is also scoring more points per possession than the guys in the other color uniforms. This, while adding some italics, doesn't add a tremendous amount of insight into how to win a basketball game. Basically, it says that a team needs to be more efficient than its opponent to win. Or, using the language in Basketball on Paper, my book on using stats to improve player evaluation and strategy, a team that has a higher offensive rating (points scored per 100 possessions) than defensive rating (points allowed per 100 possessions) wins the game.

What Basketball on Paper also adds is that these ratings can be broken down into four elements of the game: shooting, turnovers, rebounding, and getting to the foul line. It is those four elements, or "Four Factors," that really start allowing a strategic understanding of the game. Basketball on Paper describes those Four Factors, but only touches on some of the strategy that can be built by looking at them. This article is meant to expand on that to illustrate strategic advantages that can be gained with a thorough understanding of these factors.

Let's describe the Four Factors before talking about strategy. First of all, you need four of them. You can't really describe winning in full without at least four distinct factors. They are four different skills and they are pretty much independent of each other. The first factor, shooting the ball, is the most important. The game of basketball was set up that way more than one hundred years ago, where the objective of that first game in Massachusetts with two peach baskets was nothing more than getting the ball into those baskets. In that essence, the game hasn't changed. Whether it's 3-foot shots or 3-point shots, shooting the ball from the field remains the dominant means of scoring points before giving it back to opponents. (See Box for details of how the Four Factors are calculated, not that it's hard.)

The second factor is taking care of the ball, or avoiding turnovers. This factor can be very important at lower levels of basketball, such as with young kids, where dribbling and passing skills aren't very well developed. They may not be able to get the ball over half court if these skills aren't refined, which means that they're not even able to take shots. But at professional levels, each team often has several players who can bring the ball across half court without significant concerns about losing it. Full court pressure may change this, but it isn't commonly used much at the highest levels because it isn't as effective. Nonetheless, turnovers are an underappreciated aspect of pro basketball, with traveling and shot-clock violations rarely getting the outrage (from coaches or commentators) that a bad shot gets.

The third factor is offensive rebounding. If a team can get back its missed shots, it can partially make up for a problem with that first factor. It still eventually has to put the ball in the basket, but giving itself multiple opportunities allows a team a chance when its gunners from the outside are misfiring.

The fourth factor is getting to the foul line. I phrase this intentionally as "getting to the foul line," not "making foul shots" or "free throw percentage" or "free throws." This is because the biggest aspect of "free throws" is actually attempting them, not making them. Teams that get to the line more are more effective than teams that make a higher percentage of their free throws. Game-by-game exceptions can definitely exist - there are plenty of games that are lost by a team missing its foul shots - but over the long haul, just getting to the line frequently wins a lot more games than missing a few freebies will lose.

So those are the Four Factors, but don't forget that it is four each for both the offense and defense. A team's offense must shoot well, but its defense must also shut down an opponent's shooting. A team's offense must follow its own misses with offensive boards, but it also needs to keep its opponents off the glass by getting defensive boards. Our favorite team has its offensive rebounding percentage and the bad guys have their offensive rebounding percentage, which is effectively the same as our team's defensive rebounding percentage. Knowing one gives you the other but both are important.


Measuring the Four Factors

The preferred ways to measure the four factors are highlighted in this box.

1. Shooting. The simplest way to evaluate shooting is through an effective field goal percentage, one that combines 2-point shots with 3-point shots. This is given as

Eff FG% = (FGM + 0.5*FG3M)/FGA

2. Turnovers. The impact of turnovers is through a turnover percentage, or the number of turnovers a team commits per possession. In fact, all the measures you will find here are effectively percentages of some total. In this case, possessions can be counted in a game or they can be estimated as

Poss. = FGA - OR + TO + 0.4 * FTA

Evaluated this number for both an offense and a defense, then averaging, gives a pretty good estimate. Dividing turnovers by this number gives a turnover percentage.

3. Offensive rebounding. Measuring a team's ability to get offensive rebounds is simple. It is just their offensive rebounding percentage, a measure shown in NBA official stats, which is found through this formula.

OR% = OR / (OR + Opponents Def Reb)

4. Getting to the foul line. There are multiple ways to assess this. Since actually getting fouled is the more variable skill than making foul shots, you can measure this skill as

FTA/FGA

But I often prefer to make some account for ability to make the foul shots, too, and use

FTM/FGA

Those are the measurements of the Four Factors referred to herein.

The Value of the Four Factors

Saying above that shooting is "the most important" of the four factors hints at what this section is about: The Four Factors aren't all equivalent in value. You can do better at your opponent in three of these factors and still lose. You can be a good team at three factors and poor at the other and only end up with a mediocre team. Identifying the factors that are generally important then helps in identifying a strategy for constructing a successful team.

Specifically, at the NBA level, the general ranking of importance of the Four Factors is

  1. Shooting percentage (10)
  2. Turnovers per possession (5-6)
  3. Offensive rebounding percentage (4-5)
  4. Getting to the foul line (2-3)

The number after each of the factors is an approximate weight on the factor, on a scale from one to ten, that indicates how important it is to winning a game. So, with shooting being most important, turnovers and offensive rebounds end up about half as important. Getting to the line is half again as important.

These weights were determined through a program I call "Roboscout." Roboscout has a number of advantages on human scouts. First of all, he gets to "watch" a lot more games, as many as I can get numbers for. Second, he doesn't get distracted by cheerleaders or dance teams when they stand right in front of him, as has happened to many human scouts (it really is a tough job). Roboscout is not mature enough to be distracted by female roboscouts either, in case you're interested. Roboscout is willing to put his analysis in terms of numbers and projections that can be supported. Plus, Roboscout works cheap - I take 95% and he gets 5% of the deal. The downside of Roboscout is that he doesn't see the whole game. He doesn't see how players run a play or what is happening off the ball. He can't suggest how to stop a team's fast break, though he can tell you whether it is important to do so. One last downside is that Roboscout isn't much fun at parties.

Having these weights on the Four Factors suggests how to build a team. Get players who can put the ball in the basket and, very importantly, players who can keep opponents from putting the ball in the basket. Those are the first things to consider in constructing a solid team. At the NBA level, shooting and stopping shooting are two very different skills and knowing which players to go after first becomes important. This relates to the scarcity of talent. At different times in NBA history, one type of player has been more difficult to obtain than the other. In the 1980's, offense won - so getting shooters was more important. In the 1990's, defense won and obtaining players who could shut down opposing shooters became more important. But note that improving shooting or shutting down shooting - whichever a team chooses - is more important than controlling turnovers or getting rebounds or getting to the line (or their defensive counterparts). Keeping this in mind while starting a team from scratch or while building a team from its current components will make for a shorter road to success.

Roboscout and Game Strategy

The previous section highlighted general team-building strategy that comes from an understanding of the Four Factors. But a team can also help develop a game-plan for a future opponent based on Roboscout's analysis of the opponent's Four Factors. Yes, Roboscout has seen the entire league, but can limit his discussion to a specific team if I tell him to (which is why I get 95% of the cut). By applying the analysis to that opponent, Roboscout may see variations in the team's weights from the league's that suggest strategy against that team. For instance, it may show that the weights on turnovers and offensive rebounds go down to 3 for an opponent, with the weight on foul line appearances dropping to a 1. That would be a team of a lot of shooters and a good strategy for that team would be to play tight defense on the shooters, not working as hard on the defensive glass.

As a specific example, Roboscout will look at the 2004 Dallas Mavericks, whose array of international players in combination with a coach who tinkers a lot, gives a broad spectrum of types of games to do the analysis. Teams that don't alter strategy much can also be analyzed because opponents will vary their strategy against them, but Mavs' coach Don Nelson is so well-known for playing small lineups, going to strange matchups, or playing gimmick defenses, that he provides a pretty useful array of styles to evaluate.

Letting Roboscout work on the overall Mavericks team (not focusing on either offense or defense), we see that their weights do change from the league weights. In particular, with shooting still a 10, offensive rebounding comes in fairly high at 4.9, turnovers come in quite low at 3.9, and getting to the line is a meager 2.2. This implies that shooting is the most important thing with the Mavericks, with nothing else even half as important. Forcing them into more turnovers or worrying about committing more turnovers against the Mavs - those don't make as large a difference as against the rest of the league. Those are the two most important messages from the overall weight analysis.

Roboscout can look even more specifically at the Mavs' offense and defense individually to refine these a bit. For instance, looking at offense and defense separately shows that forcing the Mavs into turnovers is less significant than preventing your own turnovers. Steve Nash, Dirk Nowitzki, Michael Finley, Antoine Walker, and Antawn Jamison are good enough at controlling the ball through a great balance in their offense that trying to force turnovers out of them just doesn't help much. But being careless with the ball on your offensive end, while still not huge, is more likely to help the Mavs. Also, in looking at the Mavericks' offense, Roboscout's weights show that going after your own missed shots (increasing your offensive rebounding percentage) helps your offense against the Mavs nicely, but it also helps their offense - they like to release off of missed shots for easy shots. In terms of magnitude, though, it is still a good strategy to pound the offensive boards against the Mavericks. The Mavs may convert their breakouts off missed shots to an extra point per game, but allowing more offensive boards (note that their weight of 4.9 was toward the high end of league weights) costs them three to five points per game. In looking at the Mavs a bit last year, Roboscout found this pattern then as well. Given that the Mavs do play a lot of zone, which is known to be more prone to giving up offensive rebounds, this makes a lot of sense.

Taking a step back, note that these strategies add up to a few points per game. Talent is a big factor in basketball and the Mavs have built a team with a lot of talent. That talent is going to get them high shooting percentages and low turnover rates, two of the most important Four Factors. They have built their team the right way. But teams that recognize their strategic strengths and weaknesses against the Mavs can only increase their odds against them. And increasing your odds is what wins in basketball.