Monday, February 7, 2011

Inside The Numbers: A Change At The Top

Round One of the Ivy Heavyweight Battle went to the home team, as Princeton downed Harvard 65-61 on Friday night.

It took 50 minutes and a couple of bizarre events, but the Crimson managed to rebound from the loss at Jadwin by defeating Penn at The Palestra in double overtime 83-82. The end result pretty much matched expectations - sweep for Princeton, splits for Harvard and Penn - but the ride was quite the wild one.

We'll take a look at how those results affect the Ivy race and give you some next-level stats in a moment, but first, a public service announcement from the Please-God-Everyone-Notice-Jack-Eggleston-Is-Extremely-Talented Fund.


With all apologies to Kareem Maddox and Keith Wright, the best big man in Ivy play thus far has been Penn's Eggleston. Through four games, he has posted a 74.0 eFG% on a 27.3% shot load, an 11% assist rate with only a 15% turnover rate and has pulled down 29% of his team's defensive boards.

His last three games have been monster offensive showings - a combined 28-for-41 from the field and 8-for-12 from three. He has turned the ball over a total of two times in three games and chipped in with four assists and five offensive boards in that span.

After taking home the Ivy Co-Player of the Week a week ago, how he got passed over for Ivy Player of the Week with an even better showing is a complete mystery. Yes, Brown's Sean McGonagill dropped 39 on Columbia in a ridiculously efficient performance, but his 4-for-16 shooting, four turnover stat line in a home loss to Cornell the following evening should have disqualified him immediately. Eggleston might have averaged three fewer points per 40 minutes than McGonagill, but he shot 75 eFG% to McGonagill's 61 and turned the ball over once to McGonagill's eight.

Eggleston has always shown flashes of this type of ability but has struggled to string games together consecutively. As of now, with 10 games remaining in his senior season, he is the front runner for the Ivy Player of the Year award. It will be interesting to see if he can hold it together down the stretch and take home the hardware in his final year with the Quakers.


Princeton           23             7           11             3
Harvard           22             6           11             3
Yale           16           12             9             5
Penn           14           14             8             6
Columbia           16           12             7             7
Cornell             9           19             5             9
Brown           11           17             4           10
Dartmouth             6           22             2           12

After winning the first game at home, the Tigers slipped past the Crimson in expected wins, though both still round to 11. The leaders have maintained the separation from the field that they had last week, though Yale has crept a little closer in third place by defeating Cornell and Columbia over the weekend.

Despite Penn's hot start, the projections have the Quakers cooling off as they take to the road for seven of the final 10 games. At 6.6 expected wins, the Lions are teetering between the chase pack of Yale and Penn and the bottom dwellers Brown, Cornell and Dartmouth. Columbia could get back in the race with a sweep of the Quakers and Tigers this weekend or be pushed further toward the basement if they get swept by the pair of upper division foes.


1. Princeton (105.8)
2. Harvard (105.0)
3. Penn (103.7)
4. Columbia (102.1)
5. Cornell (100.2)
6. Brown (94.9)
7. Yale (93.5)
8. Dartmouth (81.9)

Aside from Yale, which (Spoiler Alert!!!) plays stout defense, the Ivy League standings pretty much follow the offensive efficiency ratings. That makes sense given that few of the league's defenses are strong enough to shut down a good offensive attack, giving the advantage to teams with more offensive talent.

Princeton has led the way with efficient scoring from Ian Hummer, Kareem Maddox and Douglas Davis, while its leader in points produced per game, Dan Mavraides, has struggled mightily with turnovers in posting just a 98 offensive rating through the first four league contests.

Harvard has five players producing in double figures in points produced per game - with only Brandyn Curry at a below average efficiency rating (97) - but the bad news is that beyond the top five, only two players (Laurent Rivard and Matt Brown) are seeing more than 15 percent of team minutes and only one is producing more than a couple points a game (Rivard - 7.1).


1. Yale (94.8)
2. Penn (96.1)
3. Harvard (97.0)
4. Princeton (99.0)
5. Dartmouth (101.2)
6. Columbia (102.2)
7. Brown (104.9)
8. Cornell (108.5)

Among the leaders in defensive efficiency all year, the Bulldogs have only gained ground on the rest of the pack in Ivy play. With Greg Mangano swatting eight percent of opposing shots and rebounding 27 percent of opponents' misses, Yale has been able to bolster its weak perimeter defensive with the knowledge that it has a stopper if an opponent drives the lane.

The Quakers have been the biggest surprise during league play, boasting the second-best defensive efficiency, primarily driven by Penn's amazing work on the defensive boards (78 percent). The Quakers have also managed to hold opponents to 49.3% eFG shooting over the past four contests.

Harvard and Princeton are right behind Penn in third and fourth. The Crimson is benefiting from its good defensive rebounding (71 percent), strong shooting defense (45 eFG%) and ability to keep opponents off the line (20% FTA/FGA rate). The Tigers are putting opponents on the line more than Harvard (39% FTA/FGA rate), but have a similar shooting defense (46 eFG%) and defensive rebounding rate (74 percent).


1. Harvard (8.0)
2. Penn (7.6)
3. Princeton (6.9)
4. Columbia (0.0)
5. Yale (-1.3)
6. Cornell (-8.4)
7. Brown (-10.0)
8. Dartmouth (-19.4)

This one might be a surprise. Numbers two and three are leading the way as the Crimson and Quakers come out ahead of the Tigers. It's still early (i.e. the differences between the top three are insignificant right now), but it would take a lot of luck for the season-ending, third-place efficiency margin team to win the league. Over time, playing closer games than the two teams in front of you will catch up with you in the win-loss column.

It should come as no surprise that the efficiency margin standings mirror the league standings, but one has to be intrigued by how poorly the Bulldogs fare in this metric. At the final horn, Yale has played everyone close - no games decided by more than eight - and the efficiency model is less than impressed. While six games does not a strong sample size make, often when an efficiency margin and record are discordant, the latter makes a move toward the former, not vice versa.

95% CONFIDENCE INTERVALS (Mean Ivy Wins, 95% High, 95% Low)

1. Princeton (11.1, 14, 9)
2. Harvard (10.7, 13, 8)
3. Yale (8.6, 11, 6)
4. Penn (7.9, 11, 5)
5. Columbia (6.6, 9, 4)
6. Cornell (4.8, 7, 3)
7. Brown (4.1, 6, 2)
8. Dartmouth (2.2, 4, 0)

Columbia getting swept this past weekend has really forged a divide between the upper and lower division that we haven't seen all season. The Lions have reached the easier part of their schedule, but it might be too late, as the 3-3 start has left them too much ground to make up.

With Princeton's win over Harvard, the Tigers have taken the lead in the projected win chart, almost a half win higher than the Crimson. Likewise, Princeton is a 50.9 percent favorite to win the league (either during the regular season or in a playoff), with Harvard at 40.3 percent. Yale (5.1 percent) and Penn (3.5 percent) round out the top four. In terms of winning at least a share of the title, the odds are Princeton (64.6 percent), Harvard (51.2 percent), Yale (6.5 percent) and Penn (4.4 percent).

Given that both of the top teams are on pace for an 11-3 finish, the odds of a playoff have actually increased a little bit, rising to 24.2 percent this week.


1 Princeton 6 1
2 Harvard 7 1
3 Yale 5 3
4 Penn 4 3
5 Columbia 4 4
6 Cornell 2 6
7 Brown 2 6
8 Dartmouth 2 6

NOTE: Teams are ordered by expected winning percentage, not rounded record. Thus, Princeton and its 0.888 expected winning percentage is ranked higher than Harvard and its 0.839 expected winning percentage, even though the Crimson would be a half-game ahead at the rounded 7-1.

No comments:

Post a Comment