Tuesday, January 29, 2013

NCAA Quality And Talent Are Down... Right?

We all know that there's been a talent drain in the NCAA, and the result has been reduced quality and reduced scoring. It's getting to the point of a crisis. Eric Sharp wrote in the Detroit Free Press about the embarrassing play in the most recent Final Four in an article called "Revise rules to help woeful shooting":
But the three Final Four games were a microcosm of the entire season's ills: Nobody could shoot the ball. Shooting percentages overall plummeted to depths not seen since the days when the dunk was prohibited. Everyone gasped in horror when only twice during Final Four games did a team shoot better than 40 percent....  
The college game's apologists [like Dick Vitale] point out that improved defenses have made it harder for players to get open shots. That's ridiculous. The open shots are there. The good shooters aren't. We saw enough air balls heaved from open three-point shots that they could have inflated the Goodyear blimp.

In a piece for the Atlanta Journal-Constitution called "Amid Game's Flash, Scoring Slows", Tony Barnhart suggests some ways to improve the dramatic reduction in scoring and shooting percentages, about which:
The guardians of the game are more than embarrassed. With television ratings down for this tournament, they are worried that the game could be in a decline and are wondering if anything can be done about it.
Barnhart suggests that we consider reducing the shot clock from 35 seconds to keep inferior teams from taking the air out of the ball. We need to get the "grabbing and holding and banging" out of the game, because it's no longer fun for anybody, even the players. But of course, the elephant in the room is that the talent just isn't there. As tv ratings decline, the reality is that it's all about:
Talent drain. Early defections to the NBA are starting to take their toll on the quality of play. The ACC, for example, should have had a senior class of 29 this season. Instead, there were 12 seniors playing meaningful roles on [an ACC team].
Steve Wieberg wrote in the USA Today about the impact of reduced quality of play on television ratings and crowds. Attendance is down for the fourth time in five years and CBS NCAA Tournament ratings are down almost 40% since 1992-93. And everybody knows the reason why:
The college game takes a double hit to its star power and, less noticeably, to its quality of play.
[Mike Krzyzewski warns] "The game keeps getting hit, over and over. People don't look for ways to improve things. All of a sudden, you can have a castle that looks like it's falling apart."
As Stephen A Smith wrote in the Philadelphia Inquirer about the exodus of talent from the NCAA game: "The current version of Hoop Dreams is turning into a nightmare for college basketball." Terence Moore of the Atlanta Journal-Constitution sees an even more dire situation, in an article titled "Collegiate early exodus for NBA bankrupts programs":
Mediocrity will get you far these days in college hoops
This actually is an epidemic with no cure in sight, and everybody knows the problem. Many of this season's All-America players are in the NBA. Most are doing well only in their bank accounts. 
So the epidemic will continue, with more quality players skipping either some or all of their years of college eligibility to create future Utahs, Clemsons and New Mexicos to bore the rest of us.
In Sports Illustrated, Jack McCallum rued the one-and-done NCAA player in an article called "Going, Going, Gone; The growing exodus of star underclassmen to the NBA is ripping the heart out of the college game":
The sad truth is that one-year plans aren't even news anymore. And while there are coaches, such as Georgetown's Thompson, who can talk about education and not sound like complete hypocrites, there are few, if any, who will turn down a player even if he's almost certain to leave early.
As Rick Pitino added in that same article: "Quite frankly, I think college basketball is in serious trouble." McCallum repeated the stat about the dramatic reduction in NCAA Tournament television ratings on CBS.


You may or may not have realized by now that all of those articles were written last century. Those articles cited were all written between 1996 and 1999. Of course, scoring is down. But why? Let's look at some statistics. Below are plots featuring the average data from all Division I games going back more than 60 seasons. The year on the x-axis refers to the calendar year that the season ends (so "1950" means 1949-50). All of the data is normalized to 1980, which is a nice middle ground that also happens to be near the beginning of the "modern era" of college basketball. Go watch college basketball from 1950 - it looks like a rec league. Note also that I have labeled the implementation of the shot clock (originally 45 seconds), the 35 second shot clock, and the implementation of the three-pointer. You can click to embiggen:

The first thing that you notice here is that while scoring is down over the past couple of decades, it's really not down that much. We're talking about approximately 4-5% over the past 30 years. The much bigger changes happened in the earlier years, when basketball as a sport was just different than it is now. Also, note that the implementation of the shot clock (and then reducing it to 35 seconds) did not have a significant impact on the game at all.

The implementation of the three-pointer did increase scoring, and also increased free throw attempts. The former is obvious. The latter is a bit confusing - perhaps the three-pointer spread out defenses, which made it harder to give defensive help without fouling? But overall foul rates didn't increase? I'm not sure what to tell you. That might be an interesting topic to research some day.

The bottom chart is fascinating to me. It shows just how amazingly constant shooting has been for decades. Shooting back 60 years ago was terrible, but by the early 1970s it was essentially where it is now. The only change is FG%, which declines from the institution of the three-point shot until around 1995. It took close to a decade for teams to really learn how to use it properly. But since then? Shooting has basically been flat.

So what about efficiency data? Unfortunately, I can only find that back to the 1996-97 season. Here it is:

What you'll notice in the top plot is that scoring is down over the past 15 years, but it basically just tracks possessions. In fact, possessions have slowed more than points have decreased. And you can see that in the bottom plot, where PPP is actually very slightly up over the past 15 years (by around 1.3%).

What else do we see? Turnovers per game are decreasing even faster than possessions (meaning that the turnover rate has decreased). Furthermore, the assist-to-turnover rate has increased and offensive rebounding percentages have dropped.

So what has changed over the past few decades? The most dominant change is that the game is slower. There are fewer possessions per game. But the change isn't that much. Analysts saying that teams used to score 100 per game and now we have 43-42 games are anomaly hunting. The average game is only down over the past 30 years from around 70-71 points per game to around 68. There's always been large game-to-game and team-to-team variation, and it washes out over a sample of around 5,000 games per season.

The more important thing to me is that games are much cleaner. Teams are better at not turning the ball over. They are better protecting the defensive glass. And scoring efficiency is, ever so slightly, increasing.

So in other words, the common myth that you will read constantly from major publications and will hear constantly on television that scoring is significantly down because quality is down and the game is much more physical is not at all backed up by the statistics.

What the statistics tell us is that players are more cautious now. They take more time running their offense, and play a cleaner game. And it's been a steady trend for decades, and does not seem to have been impacted in any significant way by the increase in players going pro early.

Why is this happening? I can only speculate. There is more film study now, the attention paid to the sport is greater and the stakes are higher. I guess it makes sense that coaches would slow down and clean up the game a bit. But like I said, that's just speculation.

But the idea that the quality level is down significantly and that the game has gotten boring, which you have heard for decades and will hear for decades into the future, is simply not backed up by anything objective.


In researching this article I had originally wanted articles from the 1980s or early 1990s, but any article from that era complaining about reduced scoring or decreasing quality of play was tied up in discussions about the shot clock and the three-point line, and I didn't feel it was quite fair to quote reporters complaining about reduced scoring prior to those rule changes.

But while doing that research I came up with great quotes worth sharing to remind you that, in a sense, nothing ever changes:

In a February 4, 1979 Washington Post article called "Trend is to parity in college basketball play" by David Dupree:
The trend the last few seasons indicates that there never will be another UCLA-type dominance of college basketball.
In a February 13, 1982 United Press International article called "Scoring drought hits college basketball" by Richard Rosenblatt on scoring, which was its lowest point since 1952:
[Southern California Coach Stan Morrison]: "I think there is greater parity in basketball than ever. In the past, there would be obvious differences in two teams. Parity had made the weaker team not that much weaker. Now, the weaker team feels if it can control the tempo of the game it can win."
[Syracuse Coach Jim Boeheim]: "There are a little more zones than in the past and teams are more conscious of ball control. I don't think there is as much emphasis on the fast break. Running causes more mistakes and the tendency by coaches is to be more careful. You always have the underdog trying to control the ball, more so now, and it can be successful."
But everybody's favorite curmudgeon, Billy Packer, warns that a shot clock would be a terrible answer:
''It's a travesty to put a clock in,'' he says. ''It would put more pressure on recruiting and give the man without top talent no opportunity to play the game. If put in, it would be legislating a terrible game."
In an April 2, 1985 Associated Press article, Bob Baum quotes coaches worried that the implementation of a shot clock will ruin the game:
"I think you will see upsets are going to start to disappear. Simply, the rich are going to get richer and the poor are going to get poorer," [Oregon State coach Ralph] Miller said. "You will not see a team like a North Carolina State or a Villanova go through and win a national championships with a clock."
But even the shot clock didn't concern people as much as the three-point line. In a November 23, 1986 Washington Post article called "Open season, and masters of change will rule" by John Feinstein:
"It's stupid, horrible, awful," said N.C. State Coach Jim Valvano, echoing many colleagues. "The game doesn't need the thing. It's going to penalize teams that play the game the right way and it brings us closer to being like the pro game. Why would we want that? The rule will be gone after this year, guaranteed."
And the classic proof that old fogies complaining about statistical analysis is not confined to the "sabermetric era" is this classic article from a Sports Illustrated article title "The great numbers nonsense":

The greatest menace to big-time sports today is neither the shrinking gate nor TV, either in the free or paid version. It is a nonsense of numbers, the stupefying emphasis on meaningless statistics which is draining the color from competition, stifling the fans' spontaneity and distorting their appreciation of skills.
Sports statistics are meaningless because they do not measure the most important factor in the business—the resourcefulness that is the hallmark of a champion. The Braves compiled more hits than the Yankees in the last World Series, for all the good it did them. Or, for that matter, all the good it did the Yankees in 1957 to tally not only more hits but more runs too.
That article? It was published on November 24, 1958.


Anonymous said...

great read.

keep doing your thing, i'm sure lots of people really appreciate it

Jeff said...

Thanks. And yes, the response to this post has been almost entirely positive. I had a couple of people who clearly didn't get past the first three paragraphs and were angry at me for bashing the quality of the college game. But that's how it goes I suppose, haha.

Go Blue 1985 said...

Nice statistical analyis of the data. Offensive efficiency is slightly up.

Here is what I wonder aloud: What if the game was officiated the way
it was back in pre-1985 3-pt era?

I do think the game is officiated much differently than it was in 1985. I do no thave any data to back up my claim other than personal experience. There is so much banging inside now of days. Every post move has cpontact - every time. I rewatched the Michigan v Wisconsin game. I wish I counted inside possessions and how many shots had some contact. It seems like every post move is not only contested but the players are definately banged or arm barred or landed on after a wild block attempt.

What do the numbers say about inside offensive efficiency say when compared to pre 1980?

-Go Blue 1985

Jeff said...

That's an interesting question, though I'm not sure how to answer it since those stats aren't readily available.

I want to write a post at some point soon about the addition of the shot clock and the lowering of the shot clock, as it relates to a lot of people now wanting to lower the shot clock further. So if I do that then I will put a little bit more of a microscope on that era of basketball.

Pedro said...

Nice article, but I'm wondering if using averages for all of D1 basketball skew the charts a bit. There are a lot of good D1 programs now compared to back in the day. So could it be the bad teams are better and the good teams are worse? So the averages are about the same? I'd love to see these same numbers using only the top 25 teams for each year.

Jeff said...

I actually think that looking at just the Top 25 will go the other way (though I don't have the stats at my disposal at the moment). For one, the Top 25 included a much larger percentage of teams back in the day. So the average won/loss record of a Top 25 team has increased over the years, and I'd expect that things like shooting and efficiency have increased as well.

Also, if we look at tempo only, there's basically no correlation with how good teams are and how fast they play (link). So I don't think basketball is played significantly differently at the top or bottom of D-I.

C. Mont in Sapporo said...

First time on your site. Nice read! This is better than any article I've read on any those "professional" sport sites... ever. Properly researched and concisely written. You are obviously passionate about both your writing and basketball. Very refreshing. You've been bookmarked and I'll be back. Thanks.

Jeff said...

Thanks for the kind words.

Tara Tyler said...

This is really something!
I like your post,visit my homepage here:

coffee machine reviews