BIG BLUE FANS FOR
02 How Does Shooting at Home and Away From Home
For many years, I have arbitrarily adopted a 6-point home court advantage (HCA) for my handicapping of UK basketball games. When UK played at Rupp, I added this HCA to its ANE (neutral court value) and conversely, when UK played on the opponent's home court, I added this HCA to the opponent's ANE. When UK played at a neutral venue, I ignored the HCA impact.
During the summer of 2017, Ken Pomeroy has published several items at his website, http://kenpom.com , regarding how he views HCA for all college teams. Most recently, he wrote the following in his blog post on the subject.
So it's pretty clear that fouls drive home-court advantage more than any other thing recorded in the box score. 1 Another piece of evidence that foul bias is an important source of home-court advantage is the identical trend in home-court advantage and home-foul advantage 2 over the past decade-and-a-half. ( https://kenpom.com/blog/how-to-measure-site-specific-home-court-advantage-part-two/ )
I do not have access to the broad ranging database he possesses for all teams, but I do have a considerable database regarding UK teams and UK opponents over the general time span referenced by Pomeroy, i.e. the past decade-and-a-half. I have summarized this data in the following table.
FIGURE 1: Shooting and Defense Against Shooting Data
There are many factors that affect how well or how poorly any team shoots in an particular game, but the database on UK spans 18 seasons and over 700 basketball games. Approximate ½ of these games occurred at Rupp, ¼ of them on opponent courts, and about ¼ at neutral locations. Over the course of so many games and seasons, I would expect the vagaries of outlier games that occur at both extremes of the scale would even out, yielding reliable measures about how the UK team has performed at home, away, and at neutral venues.
In his most recent blog post on the HCA issue, Pomeroy opines at length about how salient game issues such as officiating differences and the ability of teams to make steals impact the HCA. It occurred to me that the real answer about why a HCA exists should be revealed by the most fundamental of measures, shooting percentage (those acts that produce scores).
UK has scored better than its opponents over the study period, whether at Rupp or away from Rupp. At Rupp, UK has outscored all opponents by an average 17.8 ppg. At neutral locations, the scoring differential drops to only 6.8 ppg, and for away games, the average scoring margin falls to only 4.4 ppg.
UK has shot the ball better than its opponents over the study period whether at Rupp or away from Rupp, and the average difference in overall shooting percentage has been 7.7% (UK shooting Opponent Shooting). However, the shooting gap does vary by the venue type. At Rupp, UK has shot the ball at a 10.6% higher rate than its opponents but on the road and at neutral locations, this gap has been 4.6% and 5.6%, respectively. This trend holds for 2 point shooting, 3 point shooting, and free throw shooting for all games played, regardless of venue. The venue specific relationship for 2 point and 3 points shooting mirrors the overall values, with a higher gap at Rupp, and a slightly higher gap for neutral venues than for away venues. However, while UK has shot free throws at a slightly higher rate than its collective opponents (+0.9%), the venue specific values reveal an interesting variation. The free throw shooting gap for home, away, and neutral venues have been 0.8%, 1.5%, and 0.6%, respectively.
The shooting gaps generate average game point differentials. For 2-point shooting, UK has enjoyed a +7.3 ppg advantage, while the 3-point shooting gap only provides about 0.9 ppg advantage. Free throw shooting has provided UK an advantage of about 0.4 ppg for all games. This is an overall point advantage of 8.6 ppg for UK over its opponents in the over 700 basketball games over the last 18 seasons. The shooting gaps, filtered based on venue, provide overall point advantages for UK at home, away, and at neutral locations of 12.3, 5.4, and 5.9 ppg, respectively.
Based on this data, it appears that the advantage that UK has gained from its shooting, and defense against shooting when playing at Rupp is about 6.42 points as compared to UK's performance levels at neutral locations, and UK has performed about 0.52 ppg better at neutral locations than on opponents' floor.
I admit that the venue may impact other aspects of the game, e.g. steals, turnovers, assists, blocked shots, or fouls called against either team. However, it is my opinion that these factors, whether considered individually or collectively, would make a relatively minor impact as compared to the impact of shooting effectiveness. I disagree with Pomeroy's conclusion that fouls drive home-court advantage more than any other thing recorded in the box score.
For example, turnovers also vary by venue. Over all UK has enjoyed an average NET advantage on turnovers of about 0.76 per game, but that average has been 2.37, -0.85, and -0.21 turnover per game at home, away, and at neutral locations, respectively. The point impact, using average efficiency values has been about 5.37 points at Rupp, and about 0.94 points at neutral locations. When UK plays on an opponent's court, the impact of turnover differential contributes essentially no net (-0.04) points to either team.
By adding the shooting percentage gap to the turnover differential, they account for 10.86 points per game of the 10.97 point per game actual average scoring margin. The other secondary factors should have a negligible effect on the HCA.
Thus far, the discussion has not applied any strength of opponent factor. In 2016-17, the most recent season, the strength of UK's opponents was 0.121 points per possession overall, and 0.069 at home, 0.169 away, and 0.153 at neutral venues. The factor affects the raw scoring margins for the various venue filters. For this season only, the adjusted margins are about 4.9 points overall, 12.5 points at home, 2.7 points at neutral locations, and -6.7 points for away games. This produces a HCA value of about 9.79 points for the season as compared to neutral court games.
I conclude that the overall advantage that UK has enjoyed while playing at Rupp rather than neutral locations over the past 18 seasons has been slightly less than 11 points per game, and this advantage is gained by enhanced shooting at Rupp and the greater turnover margin at Rupp.
Submitted by Richard Cheeks
Submitted by Richard Cheeks