The Ivy League is unique among Division I conferences for its conference schedule, which forces every team to play games on both Friday and Saturday nights for six consecutive weeks. If you wonder what kind of toll that takes on players, listen to literally any postgame press conference — without fail, at least one player or coach will bring up the Ancient Eight’s unique schedule, usually alongside the words “grueling” or “grind.”
“Saturday nights are always the toughest games … That’s when you have to bring it,” Harvard guard Brandyn Curry said after the Crimson’s loss to Yale last season. Indeed, many of the biggest upsets over the last four years have happened on the latter half of a back-to-back: Penn over Harvard in 2012 and 2013; Yale at Princeton last season and Dartmouth over Princeton this year; Yale over Harvard in 2011 and now 2014.
But are Ivy Saturdays any different on the court than Ivy Fridays, or even Ivy Tuesdays? Here’s how back-to-back capstones compared to all other intra-league games from 2003-13 (all data from KenPom.com):
1. The pace slows down. On average, the pace of each game slows down by about three-quarters of a possession per team on Ivy Saturdays. This change isn’t really enough to be noticeable to the naked eye, but it fits the stereotype — when playing a second game in 26 hours, it makes sense that teams would slow things down a bit.
2. Offensive efficiency increases. One might think that Ivy Saturdays would be defense-dominated affairs, with shooters and penetrators battling tired legs; but actually, offensive efficiency has been slightly higher on back-to-back Saturdays than in all other games. Once again, this is a small effect in the grand scheme of things (about 1.2 points per 100 possessions), though it’s worth mentioning that it’s twice as strong in games since 2010.
3. Home-court advantage increases. This is probably the strongest and most interesting finding — home teams outscore visitors by about 3.0 points per game (.047 pts/poss) in back-to-back Saturday games, compared to 2.0 points per game (.032 pts/poss) in all other games. This should also make sense; if you have to play two games in consecutive days, better to rest in your own dorm room than on a late-night bus trip. (Note that this is not a new finding — Michael James of @ivybball has a Saturday-specific home-court adjustment in his prediction model, which I believe shows approximately the same effect.)