Home Unfiltered Articles Players Baseball Prospectus
Basketball Prospectus home
Click here to log in Click here for forgotten password Click here to subscribe

Click here for Important Basketball Prospectus Premium Information!

<< Previous Article
A Deserving Number One (02/11)
Next Article >>
Bracket Building (02/12)

February 12, 2009
Ranking Greatness
Evaluating the Value of Peak Performances

by Bradford Doolittle


On Christmas Day, 1984, Knicks captain Bernard King gave fans in Madison Square Garden a heartfelt gift, scoring 40 points in the first half against the Nets. He went on to wrap a nice little bow around the present, adding 20 more points in the second half, eliciting a stream of the "Bulls-eye!" playcalls that were the signature of legendary broadcaster Jim Karvalis. The end result was a Garden-record 60 points. Oh yeah, and the Knicks lost.

Seem impossible? Consider that on March 14, 1963, Guy Rodgers of the San Francisco Warriors racked up 28 assists against the St. Louis Hawks, tying Bob Cousy's four-year-old NBA record. The result? The Warriors lost by five points.

Here's one more--on December 13, 1961, the Lakers' Elgin Baylor scored 52 points, one of 14 times in his career that he surpassed 50 points in a game. This time, however, he added 25 rebounds and 10 assists, one of six 50-point game/triple-double combos in NBA history. The Lakers managed to win the game, but had to go to overtime before outlasting St. Louis 137-136. They won on a Baylor shot with three seconds left in overtime.

You can probably see where this is headed. Great individual performances mark the high points of almost any NBA season. These days, the regular season runs 1,230 games, and the pecking order of the respective conferences has generally been established by December. Sure, there are great games and teams are always jockeying for position. But as those team races are gradually unfolding, it's the great individual performances that keep us riveted.

Last week in the NBA was marked by a cluster of amazing displays. Kobe Bryant got the week started by breaking King's Garden record, scoring 61 points in the Lakers' win over the Knicks on February 2. That was a season high for the league this season, the fifth time Bryant has topped 60 points in a game and the 24th time he's gone over 50. Not to be outdone, Cleveland's LeBron James put on a show for Garden fans just two nights later, scoring 52 points, grabbing nine rebounds and handing out 11 assists in the Cavs' relatively easy win over the Knicks.

For a couple of days, we thought James' performance was even more spectacular than that. James grabbed a missed three-point shot by New York's Chris Duhon on the game's final play, securing what was thought at the time to be his 10th rebound. Two days later, a band of box-score auditors decided that one of Ben Wallace's famous back-tip rebounds had been mistakenly credited to James during the game. Thus James was stripped of the triple-double and the list of players with 50 points and a triple-double stands frozen at six.

Two days later, on February 8, Phoenix point guard Steve Nash capped the remarkable week by handing out 21 assists during the Suns' win in Detroit. The 21 dimes were the second-most in the league this season, and marked the fifth time in Nash's career that he's gone over 20 assists in a game.

So, a 61-point game, a temporary 52-point triple-double, and a 21-assist performance. The three spectacular performances by three of the NBA's biggest stars coming so close together serves as a perfect time to consider the ramifications of such a series of events. Are they a sign of the coming apocalypse? Is it evidence that the sinking economy is ready to rebound, and not in a Moses Malone sort of way? Is it a sign that the planet Vulcan is preparing to make first contact? Or, perhaps more germane to the subject, what do these performances really mean, both in the context of basketball history and in the practical sense of helping a team win a game?

Rare, Medium, or Well-Done?

Let's assume for the sake of this exercise that the NBA's statistical overlords had not revised James' rebound total against the Knicks. What can we learn about these feats? First of all, the games were yet another example of how numbers can be as descriptive as the words of the greatest writer or the images of the greatest photographer. When you think of Kobe Bryant, you think of one of the NBA's all-time great scorers, a player who can score inside and out, in the post or beyond the arc, on the break or from the foul line. Bryant is one of the game's best-ever shot makers, so it's rather convenient that he also happens to enjoy shooting a great deal. That he scored 61 points and took 31 shots fulfills every preconception that we have about Bryant as a player. Kobe is in actuality a wonderful all-around player, but on that night he didn't grab a rebound and handed out just three assists despite being the predominant ballhandler on a team that made 44 field goals. When we think of Kobe, we think of games like this.

LeBron James, meanwhile, is a physical freak, with the size and strength of a tall NFL tight end to go with as much speed, quickness and agility as any player in the league. In short, he can do it all, and the evidence of that often manifests statistically, as it did that night against the Knicks. Steve Nash's game in Detroit was similarly telling. We think of him as a playmaker, one who lives to create opportunities for others. He took 12 shots to go with his 21 assists, making five of them. If Nash is the starter for an offense, Bryant is the finisher, and James is the handyman that keeps it all going. Each of the three performances was unique, but just how rare was each?

If James had been allowed to keep his 10th rebound (can't Big Ben just will it to him or something?), his feat would have obviously been the rarest. Through February 9, there have been 51,518 games played in NBA history. Since there have been six 50-point/triple-double combos, then that's one every 8,586 games. There are more games played nowadays, with 30 teams in the league, but in a historical sense we're talking about a once-a-decade accomplishment. However, no one has surpassed 50 points, 10 rebounds, and 10 assists in the same game in over 34 years, not since Kareem Abdul-Jabbar did it on January 2, 1975. Of course, James didn't accomplish the feat, either. Alas.

Which feat was rarer between Nash and Bryant, though? There have been 51 games in league history in which a player has scored 61 points or more in a game, a list compiled by 15 different players. That's 0.1 percent of all games played in NBA history. A little more than twice that--105, to be exact--have reached 21 assists in a game, or 0.2 percent of all games. So it's safe to say that, historically, it's awfully rare for a player to reach either of those totals in a game but it's about twice as "easy" to hit the mark in assists.

We do need to consider some league context here, though. The game's pace, among other factors, is wildly different that it was in the 1960s, when Wilt Chamberlain surpassed the 50-point mark 104 times by himself, and famously averaged over 50 during the course of his amazing 1961-62 season. Here's a breakdown of 61-point and 21-assist games by decade:

Decade   61      % of    21      % of
        Points  Games  Assists  Games
1940s     1     0.11%     0     0.00%
1950s     2     0.06%     2     0.06%
1960s    32     0.67%    12     0.25%
1970s     4     0.04%    16     0.15%
1980s     2     0.02%    40     0.42%
1990s     4     0.04%    25     0.23%
2000s     6     0.05%    10     0.09%
Total    51     0.10%   105     0.20%

Putting assists into context is a tricky business because most long-time NBA observers will tell you that the de facto definition of the assist has changed over the years, and today's passers have it a little easier. This chart might suggest that is true but, really, that is a topic for another day. Still, it's important to remember that we're talking about comparing a purely objective statistic (scoring) with a comparatively subjective one (passing). Nevertheless, we can see that if you remove the 1960s from the equation, as many a neo-conservative would love to do, it's actually been almost five times "easier" to cop a 21-assist game as compared to a 61-point game.

So Bryant's performance last week was the rarer, and the more impressive, of the two games. But if not for the spiteful eyes of the NBA's box score police, LeBron would have trumped them both.

True Value and Other Topics

Of course, rare doesn't necessarily mean valuable. The one year I won a fantasy baseball league was a rarity, but its practical value was squat. In a basketball sense, we know that production has value. In the case of our standout trio of last week, however, which one was of the most value?

Most, if not all, methods of establishing a value to overall player performance in basketball involve assigning some sort of point value to each statistical category. There have been many such systems and, really, mine is not that much different. John Hollinger's PER statistic derives the value of each category based on the value of a possession. Martin Manley's 20-year-old efficiency formula simply assigned a value of one to each category. In both of those cases, you can simplify the final result of the formulas by saying you are subtracting the good stuff from the bad stuff.

The method used in my system, called NBAPET, tries to separate production from efficiency, them combines them to determine the point and, ultimately, the win value of a player's offensive and defensive contributions. I don't subtract the "bad stuff" like missed shots, turnovers, etc., but instead attempt to look at positive production in the context of possessions used. I don't believe in negative production. So we're going to look at the three games in question from this perspective. We'll answer two questions about each player's line: how many points did it create, and how many possessions were burned to create those points?

Without getting into the hairy math of how these values were created, I look at each of the box score categories as a fractional value of the actual points scored in a game. A made field goal, for example, is worth 1.449 points in this season's context. Offensive rebounds are worth 0.33 points. Assists are worth 0.655, and so on. If you apply the various point values to the categories at a league-wide level, the sum will equal the actual number of points scored in the league. That's NBAPET in a nutshell and it tracks these numbers both for players on the offensive end, but also for the players they cover on the defensive end, with the hope of determining just how many points, and wins, a player's total contribution is worth.

Let's start with Kobe. Against the Knicks, Bryant's 61 points gets carved up a little bit because he didn't contribute much in categories other than scoring: no rebounds, three assists, no steals, and one block. L.A. scored 126 points in the game, and by the time you finish redistributing those points amongst the different box score categories, Bryant gets credit for 49 points. He burned up 41 possessions to get those points, a fairly efficient performance. Meanwhile, his various Knicks counterparts created 14 points in 16 possessions. I'd call that a win for Bryant.

To determine Bryant's net value in the game, NBAPET has two more steps. First, it calculates Bryant's net production as his points created (49) minus that of his counterparts (14), for a final net production value of 35. Next, we adjust for efficiency, which is based on the league-wide possession value of about 1.09 points this season. Bryant was 4.0 points better than average with his offensive efficiency while he saved 3.2 points with his defense, giving him an efficiency score of 7.2. The production score is averaged with the efficiency score, yielding a +21.0 net value (or gRATE as I call it) for Bryant against the Knicks. In short, you can say that Bryant's presence on the court for the Lakers that night was worth +21 net points for his team, which is one of the better game scores for a player this season.

Without going through the entire process again, let's run through the comparable figures for James and Nash. James' case is kind of murky since, on the defensive end, he gets hit with Al Harrington's 39 points on 16-of-24 shooting night, to go with 13 rebounds. James created 48 points on 43 possessions, which means that his well-rounded line was a smidge less than Bryant's scoring burst. However, fair or not, by the time you factor in the defense that night, James was worth +2.1 points for the Cavs that night. His memorable performance was great, but if the defensive component is reasonable, always a question mark in basketball analysis, Bryant actually did much more to help the Lakers win.

Nash's production rating is considerably less than those of Bryant or James, simply because an assist in this system is not worth as much as a made basket, though it is worth more than a rebound. Nash created 26 points for the Suns in Detroit, but he used only 14 possessions to do so, giving him the best efficiency score of the trio. However, Nash spent a lot of time chasing around Allen Iverson that night. All in all, his net value to the Suns was 11.3 points.

You can sum this up like as such:

Player  Production   Value
Bryant    49        +21.0
James     48        + 2.1
Nash      26        +11.3

There is no substitute for putting the ball into the hoop. Bryant was the most productive of the three players in the three games in question, and did the most to help his team win.

Scarcity and Player Types

Comparing three players based on one game only tells us so much. The value of a system isn't in what it tells you about a single game, but in what it tells you about a season or a career. LeBron James, Kobe Bryant and Steve Nash have very different talents, but they all add wins to the bottom lines of their teams. Using a similar method to the box score analysis we've already run through, NBAPET says that James' performance is on pace to be worth 17.5 wins to the Cavaliers this season. That leads the NBA. Bryant is next, at 15.8. Nash, having a down year, is 20th at 9.5.

In the three years previous to this one, James had a total of 45.1 wins for Cleveland, Bryant had 38.1 wins for the Lakers, and Nash had 31.4 wins for the Suns. James and Bryant are one-two on the league-wide list, and Nash is seventh. All three are extremely valuable players who lead their respective teams' success, though they go about it in different ways. In the aggregate, James' all-around game seems to trump Bryant's scoring ability, while Nash just doesn't put the ball in the hoop often enough to be at their level, a statistical observation that I think most observers would agree with. You can extrapolate that out to career value. Great assist men just aren't quite as valuable as great scorers that also have well-rounded games.

However, James, Bryant and Nash--and the few players like them--are extremely valuable for another reason, which is that their combination of skills is very hard to find. Even if you set the bar pretty low, you find how unique these players are. In the last 20 years, only Bryant, James and Michael Jordan have averaged at least 25 points, four rebounds, and four assists per game. That's pretty remarkable, when you think about it, because in any given box score a player with 25 points, four rebounds, and four assists isn't liable to garner many headlines. Of course, if you raise the bar just a little bit, to 25 points, five boards, and five assists, James stands alone over the last two decades. He is joined historically by Jordan, Jerry West, and Oscar Robertson.

Meanwhile, Nash is also historically unique, but in a different way. Over the current decade, only Nash and Chris Paul have hit 15 points, eight assists and three rebounds per game. Six players have hit those figures historically, but Nash isn't one of them, since his career got off to a slow start. Nash's career averages of about 14 points and eight assists a game have been matched by only nine other players in league history.

There is one final question in all of this: What, if anything, do the big numbers out trio posted last week tell us about their greatness? In other words, are the thresholds of 61 points or a 50-point/triple-double or 21 assists attainable only by the game's elite, or can any ham-and-egger rise up for one night?

In the case of James' almost triple-double, it's hard to say, because it's only happened six times by four players. Three of those players--Elgin Baylor, Wilt Chamberlain and Kareem Abdul-Jabbar--are Hall of Famers, worthy of being mentioned among the best 10 or so players to ever suit up. However the fourth player, Richie Guerin, is not in the Hall of Fame. He's a borderline case, but he's also an argument that says that once every half-century, a very good (but not great) player can hit the 50-point/triple-double jackpot. Either way, I wouldn't say that the achievement tells us that much more about the talents of a given player than we already knew.

Nash's 21-assist game is a small chapter in his career story. Consider the list of 11 players that have posted 25 assist games. Only two of those players are in the Hall of Fame: Bob Cousy and Isiah Thomas. Two others, John Stockton and Jason Kidd, are cinches to get there. Kevin Johnson is another borderline case. However, others on the list--Kevin Porter, Scott Skiles (the single-game record holder), Guy Rodgers, Geoff Huston, Nate McMillan, and Ernie DiGregorio--were all good players that simply had a big night. Porter and Rodgers had several huge assist games; neither is going anywhere near Springfield, Massachusetts.

However, Bryant's 61-point game is indicative of a truly elite player. Most thanks to Chamberlain, the 51 games of 61 points or better in league history can be accounted for by only 15 players. Here is the list:

David Thompson
Elgin Baylor
George Gervin
George Mikan
Jerry West
Joe Fulks
Pete Maravich
Rick Barry
Wilt Chamberlain
David Robinson
Shaquille O'Neal
Tracy McGrady
Karl Malone
Kobe Bryant
Michael Jordan

The first nine players on the list are in the Hall of Fame. The other six, with the possible exception of McGrady, are virtual locks to get there. A good player might get 21 assists in a game, but it takes a great player to get 61 points.

Bradford Doolittle is an author of Basketball Prospectus. You can contact Bradford by clicking here or click here to see Bradford's other articles.

0 comments have been left for this article.

<< Previous Article
A Deserving Number One (02/11)
Next Article >>
Bracket Building (02/12)

State of Basketball Prospectus: A Brief Anno...
Tuesday Truths: March-at-Last Edition
Easy Bubble Solver: The Triumphant Return
Premium Article Bubbles of their Own Making: Villanova, Temp...
Tuesday Truths: Crunch Time Edition

Bracket Building: Using Tempo-Free Stats

2009-02-27 - : Week of March 2, 2009
2009-02-24 - : Week of February 22, 2009
2009-02-17 - : Week of February 16, 2009
2009-02-12 - Ranking Greatness: Evaluating the Value of P...
2009-02-09 - : Week of February 9, 2009
2009-02-04 - Hanging Fifty: Half-a-Hundred Fun Facts
2009-02-03 - : Week of February 2, 2009

Basketball Prospectus Home  |  Terms of Service  |  Privacy Policy  |  Contact Us
Copyright © 1996-2017 Prospectus Entertainment Ventures, LLC.