Breaking down last season by shot selection: Part 1

One aspect about college basketball that I've become interested in is the analysis of different kinds of 2 point field goals.  Despite the obvious differences between layups, dunks, and jump shots, too often shot selection analysis is confined to a simple two point field goal vs three point field goal argument.  This is true even for the people doing more sophisticated quantitative analysis out there, such as the good folks at Basketball Prospectus.  (There are a few rare exceptions: before the season started Ken Pomeroy wrote a quick piece about shooting accuracy as a function of distance from the basket which Tru posted as a fan shot.)

The reason for this simplified approach is certainly understandable: box scores make the process of dividing shots into two pointers and three pointers easy.  When we're talking about 340+ college basketball programs, "ease" is definitely a virtue.  However, nearly every college game played these days comes with reliable play-by-play data that does a better job of categorizing two point shots, thus making a more detailed breakdown of shot types possible.

With this in mind, I went back over UK's season and looked at the various shots the Cats and their opponents attempted to gain a better understanding of how the offense and defense executed.  In part 1 I'll take a team-wide view of UK and their opponents and in part 2 I'll look at the Cat's three best offensive players (at least in terms of ppg) and how well they performed at each kind of shot.

First a quick word about compiling these results.  I started out by using the play-by-play data at Statsheet.  In addition to play-by-play, Statsheet summarizes makes and misses for each shot type - a real time saver.  Unfortunately, the site did not have data for several games (all of them away from Rupp) so I had to look elsewhere for 9 games.  There are 4 games that come from CBS which also provides shot summaries for some (but not all) games and 5 games that I had to do the hard way using play-by-play data from the UK Athletics website.  I won't go into detail - mostly because it's of no interest - but there are a few discrepancies with the CBS data.  It only affects a total of maybe 12 shots though (UK and opponents combined) so I'm not going to sweat it.  The Statsheet and UKA data is completely reliable.

Two point shots can generally be divided into the three categories I mentioned in the opening paragraph: layups, dunks, and jumpshots.  In addition, we can make a fourth category for tip-ins, but as you'll see those shots are pretty rare.  They are differentiated in the data though so I've included them as their own category (CBS lumps them in with layups and dunks though).  Below is the summary of all shots taken by UK and their opponents, divided into these categories and separated into makes and misses.  I've included three pointers and free throws just for completness.

Shot Type Opponents UK
Made 3 Point Shot 256 202
Missed 3 Point Shot 463 371
Made Dunk 101 154
Missed Dunk 6 8
Made 2 Point Jumper 224 236
Missed 2 Point Jumper 554 438
Made Layup 247 318
Missed Layup 248 182
Made Tip In 35 24
Missed Tip In 14 8
Made Free Throw 452 601
Missed Free Throw 214 176


As you can see, UK had a decided advantage inside taking and making more dunks than their opponents and making more layups with about the same number of attempts.  Made two point jumpers were also about even, but UK took significantly fewer shots than did their opponents.  The numbers make a lot of sense - UK had a big advantage inside with Patterson and Stevenson and the shot selection reflects that, with UK taking more shots around the rim than did their opponents.  This also resulted in the Cats getting to the line about 120 more times than the opposition.  Other teams took far more jump shots - both two point and three point varieties - than did the Cats and we can attribute a good portion of this difference to the shot blocking skills of Patterson and Stevenson forcing shots away from the basket.

Those are the numbers, here are the shooting percentages for each type of shot.

Shot Type Opponents UK
3 Point Shot 35.61% 35.25%
Dunk 94.39% 95.06%
2 Point Jumper 28.79% 35.01%
Layup 49.90% 63.60%
Tip-In  71.43% 75.00%
Free Throw 67.87% 77.35%


It should come as no surprise that dunks have the highest percentage, followed by layups, with jump shots bringing up the rear.  What might come as a surprise is the actual shooting percentage on two point jump shots, particularly when compared to three point jump shots.  Kentucky shot just as well from inside the arc as it did from outside when it came to jumpers.  This surprised me until I went back and took another look at the Pomeroy article I linked to above.

A key piece of information that we lack is what the average NCAA Division 1 team shoots on dunks, layups, and jumpers.  Getting this information is possible, but would require looking at the play-by-play of every game for every team - a daunting task that would certainly require a script of some kind.  However, I think we can make a pretty good approximation about two point jump shots by using the afore mentioned Pomeroy article. 

If you haven't already, click on over to the article and look at the graph.  Go ahead, I'll wait.

Finished?  Excellent.

Pomeroy charts not just the shooting percentages by distance, but also the average number of field goal attempts in a game from that distance.  Using this information, we can get a weighted average for shooting percentages over a range of 5 feet to 20 feet from the basket and use this number as a proxy for what an average team shoots on two point jumpers.  The distances work as anything closer than 5 feet could potentially be a dunk or layup and anything beyond 20 feet is likely to be a three point shot.  Doing this gives us an estimated shooting percentage of 35.60% for the average D1 team.

Taking that estimate and comparing it to the chart above shows that UK was basically an average team last year when it came to shooting jump shots.  On the other hand, UK's opponents had a terrible time against the Cat's defense shooting well below what an average team would be expected to do.  You'll also note that other teams shot below 50% on layups.  I'm not sure what the average shooting percentage on a layup is, but I'd wager it's a lot better than 50%.  Last year UK had a real edge inside the arc, a combination of getting 50% more dunks than their opponents and making like difficult on layups and jumpers.  If that continues next year - and there's no reason to think it won't - we could be looking at an exceptional defensive squad.

That ends part 1.  In part 2 I'll look at how Meeks, Patterson, and Stevenson performed in each of these categories.  There are some things there that I think will surprise some people.


In This FanPost