The Sporting News named the San Antonio Spurs the team of the decade last September. Such a designation is not surprising when you consider that across the past 10 years the Spurs have won three titles and never failed to win at least 65% of their regular season games.
On Sunday, though, the Spurs lost to the Toronto Raptors. This defeat dropped San Antonio’s record in 2009-10 to 20-12, or a 62.5% winning percentage. Such a mark projects to only 50 wins across the entire regular season. For some teams – such as the Raptors (who have never won 60% of their regular season games) – a 50 win season would be cause for celebration. But for the Spurs, the current mark suggests that years of championship contention have come to an end.
About a week ago, Johnny Ludden – at YahooSports! – argued that cracks have already appeared in San Antonio’s title foundations. Looking at the standings today we see that currently four teams have posted a higher winning percentage in the Western Conference. So some numbers suggest the party is ending in San Antonio. But that’s not the story told by all the numbers.
Let’s start with efficiency differential (offensive efficiency minus defensive efficiency). For the Spurs, this number is currently 6.6. This mark currently ranks 2nd in the Western Conference (the Lakers mark is 7.7) and is consistent with a team that will win around 57 or 58 regular season games. Yes, the Spurs differential suggest this team is second best in the West and one of the top five teams in the NBA.
To put this differential in further perspective, here is what the Spurs have done recently with respect to this measure:
2008-09: 4.1
2007-08: 5.3
2006-07: 9.1 (won NBA title)
2005-06: 7.5
2004-05: 8.5 (won NBA title)
2003-04: 7.0
2002-03: 5.8 (won NBA title)
2001-02: 6.7
2000-01: 8.5
1999-00: 6.3
The Spurs current mark tops what they did the past two seasons, and even bests what San Antonio did on its way to an NBA championship in 2003. So this current Spurs team is not the best we have seen across the past decade, but it’s pretty good.
When we move from efficiency differential to Wins Produced we can see who is responsible for this outcome.
Table One: The San Antonio Spurs after 32 games in 2009-10
Not surprisingly – as Table One indicates – Tim Duncan once again lead this team in Wins Produced. Duncan, though, is not a one-man team. Of the Spurs 57 projected wins, about 28 can be tied to the play of Manu Ginobili, DeJuan Blair, Keith Bogans, and Matt Bonner. As noted a few days ago, DeJuan Blair was clearly a steal in the NBA draft. It’s possible that Blair will lead all rookies in WP48 [Wins Produced per 48 minutes] this season, just as Blair led all players taken out of college in per-minute production last year.
Once we move past this quintet, though, the Spurs are only projected to receive 10 more wins from the rest of the roster. “The rest” includes both Antonio McDyess and Tony Parker. As Table One notes, these two players have declined the most from what we saw last season.
For McDyess the decline might be tied to age. McDyess was drafted in 1995 by the Clippers and now is 35 years old. As noted before, age will ultimately reduce the productivity of all basketball players. And this means that McDyess – who is obviously not getting younger – may not improve as the season progresses. The Parker story, though, may be different. Ludden notes that Parker has not been healthy this year. If Parker’s health improves, the Spurs may be even better as the team approaches the playoffs.
And this means the Spurs might be very serious title contenders when this season ends. Again, the impact of age can’t be avoided forever. Two of the key contributors this season – Duncan and Ginobili – are well past 30 years of age. So the Spurs title window is definitely closing sometime in the future. But it’s possible – despite the team’s current place in the standings – that the window will stay open for the 2009-10 season.
– DJ
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Introducing PAWSmin — and a Defense of Box Score Statistics
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.
Wages of Wins Journal 
bags fly free
January 6, 2010
can you explain why in calculating wins score you subtract free throw attempts? I thought getting to the line was a good thing.
bags fly free
January 6, 2010
also as a pistons fans I’m sad to see mcdyess with such a decline from last year, either he gave all he had last season and has to steam left or hes still getting used to new spurs team.
dberri
January 6, 2010
Free throw attempts are a good thing. According to Win Score, break-even free throw percentage is 50%. In calculating Wins Produced it is even lower. So for most players, the more you go to the line the more you are helping your team win games. Maybe a post on this subject would be a good idea.
bags fly free
January 6, 2010
right but you are subtracting FTA thus lowering wins score so it appeared to me that FTA is a bad thing which of course its not
ilikeflowers
January 6, 2010
bags,
A FTA (or FGA) is a bad thing if you don’t score points well enough while using them. You have to look at points scored from FT as well as FTA to determine if they are good things for the player in question to use. For a hypothetical player who misses all of their shots FTA and FGA are opportunities to score lost and are indeed bad things.
Leon
January 6, 2010
The key thing is the minutes, those starters and key players on the Spurs are playing less minutes than their rivals, partly due to the strength and depth but also injuries. If Parker and McDeyss can produce what they have in the past, and that’s not unreasonable to ask, then that core of 6 to 8 players looks like one of the most productive in the league, and come the post season will be a strong one to handle.
brgulker
January 6, 2010
That would be very helpful!
I suspect that McDyess’s numbers will pick up. Dyess always started slow in Detroit (or at least that’s my observation). I doubt he ever gets back to .200, but I’ll be shocked if he’s not at least average by season’s end.
Oren
January 6, 2010
“right but you are subtracting FTA thus lowering wins score so it appeared to me that FTA is a bad thing which of course its not.”
Have you heard of the Hack-a-Shaq strategy? Basically, teams would foul Shaq since he’d miss his FTs. You need to be bad at the FT line for this to work, but if you’re a 40% shooter, then this strategy becomes viable.
FTAs are a bad thing if a player is unable to successfully convert them. If I foul you and you miss both of your FTs, then that’s one less possession that your team has. Thing is that it’s more likely to hit two FTs then to hit a FG.
Tommy_Grand
January 6, 2010
Lets say I go to the line 10 times and score 7 points.
You go to the line 13 times and score 7 points.
Arguably, I am the more efficient (*not nec. more productive) player even though (or rather because) you have more FTA. But if we both drive the lane 14 times, and both create 7 points, that’s different.
Man of Steele
January 6, 2010
Dr. Berri,
I’ve always wondered about the way FTs are valued in Win Score. Although the frequently cited critique about the cutoff points for shooting efficiency in PER and NBA Efficiency is certainly valid, I wonder if perhaps the same critique is valid with regard to FTs. If a player hits 51% of his FTs, this is a positive, although the league average is significantly higher. That seems strange – it seems that a player should have to shoot above the league average to help the team win. A post on this topic would be very interesting.
TBall
January 6, 2010
Tommy,
If you both drive the lane 14 times and both create 7 points, neither of you are doing your teams any favors.
TBall
January 6, 2010
Here’s the win score formula, which is simpler than and shares characteristics with WP:
Points + Rebounds + Steals + ½Assists + ½Blocked Shots – Field Goal Attempts – Turnovers – ½Free Throw Attempts – ½Personal Fouls
If your sole contribution to a game is 2 points on 1-2 shooting from the field, you have a win score of 0. If you have 2 points from 2-4 shooting from the line, you have a win score of 0.
MoS – Arguably a player should shoot better from the line, but there is a value in being able to get to the line and an implication that 2 points were achieved in two possessions making 1/2 FG comparable to 2/4 FT. The and1’s complicate things (2 points on 1/2 FG and 0/1 FT is a -0.5 WS), and over the course of a season probably means a player has to shoot ~55% from the line to ‘break even’.
At the root, though, the key is points per possession (which tend toward 1 point/possession) not shooting percentages relative to the league average. The criticism of PER and NBA Eff. is that players can be inefficient (convert less than 1 point/possession) and still be favored for volume shooting.
kevin
January 6, 2010
You know, the whole area of fouls, fouling, free throws and the like are a vastly neglected area of statistical research, which is dumbfounding, since coaches basxically game manage around the foul situation.
There are good and bad fouls, but nobody bothers to distinguish and enuimerate them. there are offensive and defensive fouls, but nobody bother to distinguish or enumerate them, at least on an individual basis. Minutes lost due to foul trouble are not counted. There souyld be some way of measuring how soon teams gets to the foul limit/quarter, or allows the other team to get there.
For such a crucial area of strategy and win outcome, foul statistics are pretty poorly kept. some enterprising person could learn a lot just by going to the online game logs and counting stuff like I outlined above.
Austin
January 6, 2010
The thing about the Spurs this season that this post neglects is the Spurs’ competition and who they’ve won against. I’ve seen some statistics that a) their strength of schedule is weak and b) they win consistently against lesser teams but struggle against playoff-caliber teams.
So while I expect the Spurs to win 50 games, if the playoffs started today I’d be surprised if they won more than a single series.
Robert O'Malley
January 6, 2010
With McDyess there is also the possibility of diminishing returns. With Blair now in the picture and Duncan improving so much, that is going to take available stats away from McDyess. Or at least that’s my theory.
Robert O'Malley
January 6, 2010
In addition to his aging
Italian Stallion
January 6, 2010
I would argue that drawing fouls has value over and above what you do on the FT line.
Suppose a player is GREAT at drawing fouls but not so good at making FTs. Let’s say he makes 50% (break even). At first glance it seems like there’s not much value there.
But drawing all those fouls puts the other team in the penalty quicker. That in turn also gets teammates to the FT line quicker where they typically will be better than 50% and more productive than usual.
It also gets opposing players in foul trouble (typically good ones) and causes them to either play more cautiously on the defensive end or get benched.
I think we clearly also need a stat for drawing offensive fouls. Drawing an offensive foul has a lot of value.
Italian Stallion
January 6, 2010
Kevin,
I think you are making a very good point about good and bad fouls. I mentioned that once also.
If a guy has a clear path to the basket and you foul him and put him on the FT line, statistically it counts against you as a personal foul. But clearly a foul like that can have a positive value if the player would have dunked the ball and got a certain 2 points if you didn’t do that. You reduced the probable production on that possession by sending him to the line.
I think this is the kind of thing (good and bad fouls) that players instinctively calculate while in action (some better than others).
benamery21
January 6, 2010
MIP?
Breaking news:
Sun’s Channing Frye has team highest 18.5 Win Score in Phoenix defeat of 11th ranked Houston (.571). Channing’s projected season WP continues to rise as the 26 y.o. ‘center’ grows better at the new perimeter based offensive role he plays, and starts to incorporate defensive improvements into his game.
Frye currently (as of his 36th game tonight) leads the NBA in 3 point field goals made this season (92), while ranked 10th in 3pt percentage. Prior to this season, Frye had averaged just five 3pointers per season (1 per 278 minutes) across 4 years in the pros.
When evaluated as a center, Frye’s current 7.34 projected season WP (based on average minutes played and PAWSmin over 36 games), leaves him below the 15th best center’s mark in the league last year, but about the 65th mark overall in the league, and if evaluated as a power forward his 8.2 projected WP would place him 7th among PF and 49th overall.
This is a 15.5 WP season improvement from what WOW would have predicted for this many minutes played based on last season’s WP48.
It is a 5.0 WP season improvement over Frye’s best ever season.
MIP?
Man of Steele
January 7, 2010
Tball,
I see what you’re saying, but isn’t the baseline of 1 pt. per possession based on the league average? I thought Dr. Berri had said somewhere on this blog that the league average was 1 pt./poss. (and I think Dean Oliver as well).
mrparker
January 7, 2010
The reason fta attempts are subtracted is the same reason fga are subtracted. It is a gauge of efficiency. Efficiencies are often written in fractions. Fractions can usually be expressed in linear equations. 1apples-1oranges=0 can also be expressed as 1apple/1orange.
*The league average is 1.066 per posession. The league average is .267 ft/fga. The league average ft percentage is .757. The league average efg% is .496 or .992 pts/shot.
Italian Stallion
January 7, 2010
>>I see what you’re saying, but isn’t the baseline of 1 pt. per possession based on the league average? I thought Dr. Berri had said somewhere on this blog that the league average was 1 pt./poss. (and I think Dean Oliver as well).<<
I think that's clearly it. 50% = 1 point
The idea is not to measure FT efficiency relative to the league average. It's to measure the scoring efficiency of the possession. Most players are more efficient from the FT line than elsewhere. So if they get to the foul line more often it helps their overall scoring efficiency.
kevin
January 7, 2010
What Stallion says. Plus, getting fouled gets the other team in foul trouble, limits their defensive options and, ideally, gets their better players off the floor because of foul trouble. If linear weights were applied to free throws, the true impact would be much higher than 50%.
brgulker
January 7, 2010
That’s one of the reasons I like comparing players in terms of points per shot attempt. It helps to capture their contributions from the FT line, 2 pts attempts, and 3 pt. attempts.
Doing so for guys who do all three of those things efficiently shows how valuable their offensive contributions really are.
Ray
January 7, 2010
The Spurs are so weird. They have very few quality wins, yet they have a great point differential. It’s likely due to schedule, but you just wonder how they can be this inconsistent.
kevin
January 7, 2010
Injuries, Ray. It seems like every game, somebody’s limping around.
Man of Steele
January 7, 2010
Okay, so getting to the free throw line counts as a “victory” for the offense; that is to say, the offense wins the possession if they get to the free throw line because they get 2 (or 3) undefended shots to achieve the league average 1 pt. per possession (or 1.066).
I think I see what you all are saying, but this makes me think about differences between situations. In football, a 2 yard gain on 3rd and 2 is a lot different than a 2 yard gain on 3rd and 10. So even though the league average is roughly 1 pt. per possession, the league average on FTs is like 1.5 pts/poss. (if the league avg. FT& is 75%). So in reality making less than 3/4 of your free throws is a below average performance.
Anon
January 7, 2010
But suppose hypothetically the league free throw average was 90%, and the league fg% was 10%. If you were good at drawing fouls and getting to the line, but only shot 60% when you got there, it would still be better than shooting a regular shot even though you’re below average at converting free throws when you get to the line.
Tommy_Grand
January 8, 2010
“Okay, so getting to the free throw line counts as a “victory” for the offense;”
Usually. If they don’t make a shot (or gather in the rebound) it is not a victory. If they hit only one, it’s a below aver. offensive possession, but when you factor in adding a foul to some opposing player it’s probably a slight net positive.
benamery21
January 10, 2010
I think some of the newby confusion with respect to the “penalty” for shots attempted would be avoided if they thought about it in this mathmatically equivalent, more complicated way: Instead of -1 point for each 2 point or 3 point shot taken, or -0.5 for each free throw attempt–Just say +1 for each 2 pointer made, -1 for each 2 pointer missed, +2 for each 3 pointer made, -1 for each miss, +0.5 for each free throw made, -0.5 for each miss. It’s the same thing, and it reduces the perception that there is a penalty for taking shots or getting to the line. The penalty is for missing.