Thoughts on the 30-second shot clock

After experimenting with the 30-second shot clock in the non-Big-Dance postseason tournaments, it looks like NCAA Men's Basketball is going to change the rule next year:

Men's basketball is likely heading toward reducing its shot clock from 35 to 30 seconds, NCAA rules committee chairman Rick Byrd told ESPN.com on Monday.

Why the change? Andy Katz's summary was spot-on, I think:  "The trend is for this committee to do something, and that's the easiest thing for them to change."

Simply put, a lot of people harp on college basketball for being too slow, and this will presumably add a few possessions per game, which will likely increase total scoring a little bit -- assuming that having less time to score doesn't decrease offensive efficiency.

Here are some musings on the possible outcomes of this rule change.

Best-Case Scenario

Teams very rarely use the full 35-seconds of clock anyhow, so the change to 30-seconds has no effect on offensive efficiency. In fact, it increases offensive efficiency because:

1) Many teams that do use the full 35 seconds are actually just standing around for 20 seconds to rest so that they can play better defense. (Dick Bennett's Wisconsin teams sometimes did this.) With less time to rest on offense, these defense-first teams will be worse on defense, which will increase offensive efficiency.

2) Many teams respond to the compressed shot clock by running full-court press, then fall back into a zone defense -- not with the hope of forcing turnovers, but with the aim of preventing teams from getting into their offense until little time remains on the shot clock. But most teams are pretty bad at this, and capable college guards shred the soft press leading to transition buckets and open shots, with just a few more turnovers. The game becomes more run and gun, more exciting, higher scoring, and even more efficient on offense.

Worst-Case Scenario

Smart coaches quickly adjust to this pro-defense rule change, and the inconsequential increase in possessions per game is more than offset by a drop in offensive efficiency. Teams like Louisville, West Virginia, and Texas (with Shaka Smart at the helm) utilize their stifling full-court press to eliminate even the possibility of running of offense, forcing not only turnovers but frequent shot-clock prayers. More and more teams adopt this style of play, leading to benches filled with long but not particularly skilled players. Without freedom of movement, half-court offense in 12 seconds becomes impossible, and games in the 40s become common.

Most Likely Scenario

Nothing much changes. The shot clock rarely comes into play as it is, and shaving off five seconds is inconsequential. The extra few possessions don't change efficiency, much, and total scoring changes 0-3 points per game per team. The college basketball haters continue to hate with their hateful hate, same as it ever was.

For what it's worth, here were the final results of the 30-second shot clock experiment this last postseason -- keep in mind, though, that this is a small sample that in total amounts to a typical weekend's worth of games during the regular season:

.@nuclearbdgr All post-season pace & efficiency numbers: pic.twitter.com/vP8JKsl8t5
— Bart Torvik (@totally_t_bomb) April 8, 2015

New Rules

Here is a chart showing the offensive efficiency during Big Ten play since 2003 (courtesy of Kenpom.com):

The average from 2003 to 2013 was 102.9; this year it is 102.8.

In other words, the new rules seem to have stopped working. Probably because they're no longer enforced. 

More parity

I noticed that there seemed to be fewer blow outs than normal as I looked over the top 25 scores. I didn't know if this was just my imagination or if there was something to it, so I had my co-blogger Bart pull some data for me from Kenpom. Turns out I am not imagining things.

Here are the stats on games between top 30 and sub 200 Kenpom teams from this year and from last year through the same point in the season:

	2014	2013	Change
t30/s200 Margin	26.7	31.3	-14.7%
t30/s200 Blow	71%	79%
t30/s200 Close	19%	5%

Bart pointed out to me that:

So there is some support for your impression. Last year the average margin in these games was 31.3, and this year it's 26.7. There have also been fewer blowouts (20 pts or more) and almost 4 times as many close games (10 pts or less).

We both had some ideas about what may be causing this, and the new rules seem to be the most likely culprit. The new rules should help kids get off shots without being hand checked. They could also force more collapsing to the lane to help since guys just can maul drivers with their hands, which could lead to more open perimeter shots. I think there will eventually be more teams playing zone too.

Teams that recruit players who are phenomenal athletes but not great shooters could end up being "punished" as these players' once-dominating defense is now marginalized. I feel like there are probably more upper-level teams that recruit these types of athletes, and more small/crappy schools that just bring in shooters. This might lead to more upsets, or at least closer games amongst the haves and have nots until teams adjust recruiting or the rules are adjusted. 

Bart added:

Another way in which the new rules might help the Davids is that you tend to see quick undersized guards at the mid majors. In other words, they are just as good at basketball, and speedy, but don't get recruited to the major conference teams because they can't stand up to the beating and can't get their shot off over big defenders. Now we might see these little guys just going around the bigger slower defenders. 

We'll have to look at this more as the season goes on.

StatWatch™: Update on the new rules

We've played a full week of college basketball now (not including this Friday's games), for a total of about 319 games between D1 teams. I've been keeping track of how the new rules have been affecting scoring as compared to the same time-period last year, and I will continue to do so until I get bored with it.

After the first weekend, overall scoring was about about 6.9%. Now it's up even more—9.1%. The increase is even large in certain subsets of games: close games (10 points or less), games that top 10 in Ken Pomeroy's "FanMatch" (that is, games expected to be competitive and quality basketball), and games between teams ranked in the top 100 of the Kenpom ratings:

D1 v D1	2014	2013	Change
All Scoring	73.5	67.4	9.1%
Close Games	74.4	66.2	12.4%
Top 10 FM	75.3	67.2	12.1%
Top 10 Close	76.1	66.9	13.8%
Top 100	76.1	68.0	11.9%
#OT	15	18

(Hey, look at that, I finally figured out how to embed an excel table into a blog post all pretty-like.)

Of course, for each of these subsets there is a small number problem, so the difference could just be a mirage. It's fun to speculate, though. I speculate that scoring is up in the games between good teams because the games between good teams are on television, so the officials are especially motivated to make a show of enforcing the new rules. 

That the increase also seems to be accelerating could be a sign that the rule changes are actually working to open up the offense, rather than just creating more free throws. But we shall see.

Scoring is way up ... so far

The biggest sub-plot in college basketball this season is the new rules preventing the use of hands on defense, and favoring the offense on the block/charge call. So far scoring is up, and in games we actually care about it is way up.

After three days of play, and 163 games involving D-1 teams playing other D-1 teams, scoring is up by about 4.7 points per team as compared to the first three days of last year, an increase of about 6.9%. Kevin Pauga breaks down the additional scoring and finds that it amounts to about one bucket and three FTs per team. In other words, so far about 60% of the scoring increase is due to more fouls, and 40% of it is perhaps due to freer play.

Other than a relatively small sample size, one problem with the data so far is that the opening weekend of college basketball season has a lot of mismatches, which lead to a lot of blowouts. One might think that the new rules wouldn't affect the blowouts much, or at least might affect them in weird ways.

So I looked at some subsets at the data. First, I excluded all games involving a non-D1 team. That made just a slight difference, but it weeded out a bunch of probably worthless data.

Second, I looked at games decided by ten points or less. In those games, scoring has increased from 66.6 points per team last year to 74.2 points per team this year. That's an 11.5% increase, which is huge. (The result is essentially the same if you account for overtime minutes.)

Third, I wanted to look at games between relatively good, relatively evenly matched teams. To determine this I used only the top ten games in Ken Pomeroy's daily Fan Match feature. The downside of this is that it decreases the sample to just 30 games each year so far. The upside is that we can be more confident that the samples are similar. Also, these are the games we really care about. The result: in those games, scoring has increased from 68 to 76.1 points per team—an 11.9% increase. Assuming Pauga's ratio holds true in those games, that's about three more buckets and ten more made free throws per game so far.

Fourth, I looked at the games that were both close and expected to be competitive. In those games, scoring increased from 66.8 to 77.1 points per team—a 15.4% increase.

Given the sample sizes, I'm not ready to declare these increases Torvistically significant, but it does seem like the people who think we need even more scoring in college basketball might be getting their wish this year.

More on NCAA rules, sorry I just can't do baseball yet

The NCAA rules committee came out with their recommendations and they did not include the change to the 35 second clock. This pretty much means that the clock will remain at 35 seconds for another season. The proposed changes are outlined in Eamonn Brennan's column.

The best suggestion in the article is not from the NCAA committee, but from Brennan himself regarding timeouts. Basketball has too many. There are already official timeouts at each 4 minute mark, do we really need many more?

The rule change to 5 timeouts a game instead of 3 per half was sort of a move forward, but it hasn't helped. Since teams lose a timeout at halftime if they don't use it, teams often use one at the end of the first half to set up a play rather than lose it at halftime. This is totally unnecessary and slows down the game.

Teams also don't need to have 4 timeouts in the 2nd half. They already get one every 4 minutes, why do we need to have a half of basketball with potentially 8 team timeouts plus the 16, 12, 8 and 4 minute timeouts.

Brennan brought this up because the NCAA committee proposed to expand the ability of refs to review plays to the last 2 minutes which will slow down games even more. I'm OK with the extra review as long as there is an NFL like time limit on the review, but there should be fewer timeouts.

In the NBA there are 82 games and a need to make that huge number of games as competitive as possible so they entertain. The NBA allows a team to move the ball to half court after a time out to get an advantage in end game situations and allow teams a better chance to come back. NCAA games don't allow the movement of the inbound, so the only real advantage is to set up a defense after a made basket. I don't think getting rid of timeouts will lead to a ton fewer steals on inbound plays, as teams usually just foul anyway. Just think if teams only had 1 or 2 timeouts at the end of games. Fouling slows down the game enough, do we need all the time outs?

I also think the foul out timeout needs to go. Why should a team get to huddle and talk to the coach just because a guy fouls out. If the coach needs 30 seconds to make a decision about a sub, why does he need to huddle with his players to do so.

I'll stop bitching now.

Cha Cha Cha Changes

The Bucks are almost out of the playoffs, so I will soon have to turn my attention to the Brewers. I promise I'll start to post some baseball stuff, or at least comment on Brett's posts.
Until then I have one more post on NCAA basketball.

Bo Ryan had some interesting comments about the proposed NCAA rule changes to the shot clock and practice start. The full article is on Andy Katz's blog.

On starting practice early:

You will have 30 days between Oct. 4 and your first game, so they would be expanding the time to get your 30 workouts in so you can give guys more days off. You can spread it out a little bit better. (Wisconsin's Bo Ryan)

On shortening the shot clock to 30 seconds:

I'm OK with 30 seconds. [But] when they say go to 30, they'll think 'Oh man, the points will go up.' People are much more sophisticated on defense, bodies stronger. There is a lot more than the shot clock on why scoring is down. (Wisconsin's Bo Ryan)

The practice rules don't seem like a huge change. Practice would start 2-3 weeks early, but the amount of time coaches can spend with players would still be regulated. As Bo states, it would allow kids to have fewer back to back practices and give bodies more time to recover.

I imagine his response to the shot clock changes will surprise those who still believe UW to be a run the shot clock down and chuck it team. Bo is dead right that shortening the shot clock will increase possessions but may not necessarily significantly increase scoring. Contrary to what many may think, if anything I think the change may benefit UW. UW will still run the same offense and 5 less seconds won't force any big changes to their system. However teams that play good defense like UW could gain an even bigger advantage if teams don't have an extra 5 seconds to get a good shot.

The best quote in the article was from Steve Alford about the Cuse D:

I don't buy the argument that shooting will go up or scores. Seems college kids struggle enough knowing how to get good shots in 35 seconds. The shorter clock helps better defensive teams, though. Have fun going against the Syracuse zone with five less seconds. (UCLA's Steve Alford)

Can a player goaltend his own shot?

I saw one of the more spectacular plays I have ever seen in the KU vs MI game the other night. Kansas's Kevin Young did what would be considered a goaltend if he wasn't a freak of nature. On a fast break he went up for a lay up and lost the ball as he was going up. In the same play as the ball was traveling down he tiped the ball which then went into the basket. Since this was all in one crazy freak of nature motion, no ref even thinked to call a goaltend, but the shot left his hand and was going down when he touched it and directed it into the goal.

Look at it again and tell me this isn't a goaltend.

How much scoring is enough?

Some people claim college basketball is "broken." The basic criticism is that the number of possessions is down, and overall scoring is down. Both of these trends are long-term and continuing. We have more and more games in the 40s and 50s, and more and more outliers like Eastern Michigan's 42-25 win over Northern Illinois (A game in which NIU scored just 4 points in the first half.)

I frankly reject the premise that college basketball is too low scoring. Look at that epitome of ugly basketball, EMU's 42-25 win. That's a total of 67 points, and it was accomplished with 21 field goals and 20 free throws. So the ball went through the hoop 41 times.

There is no other sport where 41 goals scored is considered an abomination of offense. Even if you count just field goals, that's 21 field goals, or one every other minute of game play. Imagine if there were 21 scores in a football game. There's a name for that: arena football. Not a popular sport, and for good reason.

Looking at the box score, Eastern Michigan and NIU played a slow tempo game, with just 52 possessions. But Northern Illinois alone took 61 shots, making just 8 of them. Here's my question: are missed shots boring? I don't really think so. In sports, what you want to keep things interesting is legitimate opportunities to score. The NIU / EMU game had plenty of that: NIU put up 33 three-point shots (and made just one).  The point is, this isn't soccer, where entire games can be played without even legitimate scoring chances. Even in the worst college basketball game, there are an absolute abundance of scoring chances. If 96 shots on goal in 40 minutes of play is too "boring" for you, the problem is yours.

In fact, I've always thought the problem with basketball is that it is too high scoring. This is problem because individual scores are close to inconsequential to the outcome. For this reason, many casual sports fans will joke that they just turn on a basketball game with five minutes to go, because the first 35 minutes are pretty much meaningless. And there's a lot of truth in that. The sport absolutely does not need more scoring chances.

What's really going on here is a stylistic clash. There's an element in college basketball that thinks beautiful offense is running, jumping, and dunking. These people prefer high-possession games with lots of transition offense. There's another element in college basketball that thinks beautiful offense is passing, screening, cutting, and shooting. These people prefer half-court offense.

In my opinion, the beautiful thing about college basketball is that both styles can be successful. The issue is that the fast-break proponents seem to think it's unfair that the half-court teams can beat the fast-break teams. But that's just stupid. The athletes that are good at fast-break offense are often just different than the athletes that are good at half-court offense. And it so happens that a good coach can usually teach his players how to neutralize a team that just wants to run and jump and dunk. Unless you have truly elite, NBA-bound players, a good half-court offense is just way harder to stop than a transition offense. More and more coaches have figured that out. Is that a problem? I think not.