The record of the 2012 NexGen Team surely ranks as one of most surprising and impressive accomplishments in recent ultimate. A team composed of 15 college all-stars posted a record of 11-4-1 against the top Open teams across the country. How did this happen?
November 14, 2012 by Sean Childers in Analysis, Featured with 8 comments
Editor’s note: When we first announced Ultiworld back in July, we discussed our intentions of developing innovative statistics that could launch new opportunities for analysis and understanding. We are proud to announce the first phase of that work. This article is the first in an extended series about this summer’s NexGen tour, examined through the lens of thousands of data points gathered and analyzed over months of work.
The record of the 2012 NexGen Team (and, to a lesser extent, the inaugural 2011 edition) surely ranks as one of most surprising and impressive accomplishments in recent ultimate. To briefly recap, a team composed of 15 college all-stars defeated ten top Open teams around the country, including current champion Austin Doublewide, 2011 champion San Francisco Revolver, and perennial favorite Boston Ironside. Not only did the NexGen team post an 11-4-1 record, but they won by greater margins than they lost by — compiling two blowout wins and just one blow out loss. Their other losses were by three points or fewer.
How did this happen? Elite open teams roster 27 players, most of whom have played years of ultimate together. Almost all were college stars in their own right but have since added significant club experience. Club teams practice together, develop offensive plays, and design defensive sets. According to the conventional ultimate wisdom, all of this is supposed to matter.
Our project looks into that how: we want to dissect the NexGen performance at a statistical level, using a level of detail and specification that hasn’t yet been employed in ultimate analysis. Using a prototype iPad application, we recorded information about every single pass, possession, and point in 12 NexGen games. Player-specific statistics will be launching on Ultiworld over the next month, answering questions like which players on the team were most likely to handle and who was the most efficient cutter. Team statistics will also be available. But we also processed the raw data through statistical packages and looked at correlations, regressions, and other statistical tests.
In addition to analyzing the NexGen performance, we hope to kick start a conversation in the ultimate community that goes beyond the simple “Goals-Assists-Ds” narrative and pushes the ball forward on understanding what makes good individual and team play possible. So accompanying the release of statistical information on Ultiworld will be articles giving a deeper look at what we’ve seen. Part of these articles will question and evaluate the current conventional wisdom.
NexGen Won Games Because Of Their Defense
One sentiment in the ultimate community is that elite teams win games because they have elite offensive units that can hold serve and turn the disc less than their opponents. Casually watching NexGen play could easily reinforce this notion: the all-stars throw with ease, rarely drop the disc, and routinely make athletic plays in the endzone.
But our data suggests that it was NexGen’s defense, even more than their offense, that contributed to their success. It’s easy to see this side of the story too: they make some huge plays on the defensive side of the disc. But we think there might be something to it than that. Essentially, in games where NexGen’s defense performed well, NexGen was more likely to win — even if objective metrics and evaluations of their offense suggest that they were playing below their average. I will dive into the statistical analytics later, but offer one anecdotal example now to help ground the problem.
One of the worst NexGen offensive performances was their game against Chicago’s Machine. But this was also their largest blowout win. To be sure, there are many possible explanations for this (some of which we discuss later), but it’s interesting to keep in the back of your mind.
Why Does the Offense/Defense Distinction Matter?
The relative importance between the two sides of the disc matters at both an individual play level and for developing team strategy as a whole. At the individual level, there is a limit to the amount of physical energy (endurance and effort) that one can expend per point, and perhaps a limit to the amount of mental energy (mental focus and concentration) as well.
To be sure, both limits can be increased through proper conditioning and practice, but limits will still exist and players may be forced to choose between them. This phenomenon has been noted by basketball commentators, who suggest that players may make tradeoffs between the amount of defensive effort they expend and the amount of usage or efficiency they are able to invoke on offense. Furthermore, players may face choices in-game between offensive moves with defensive implications. A second deep cut against a zone, for example, may be a good idea on the offensive side but leave the team short a defender in case of a turn.
Finally, a decision may involve a risk-reward tradeoff. The simplest example is a layout bid that would result in no mark if unsuccessful. That decision should be made in light of the reward involved, which may be different depending on whether one sees Ultimate as a game where good, turn-generating defense is impossible to play consistently.
From a team perspective, the captains have to choose whom to play and strategies to employ. Players specialize in different areas and the top seven offensive players will differ from the top seven defensive players. The simple solution of placing the former on offense and the latter on defense is an obvious fallacy: there could be a significant increase in expected value by playing someone who is slightly worse at defense but much better at offense over an elite yet one-dimensional defender, even on a D Line. The relative importance of offense compared to defense for any given team is going to matter at a fundamental level when they choose who and how to play.
Finally, it’s fun and interesting to evaluate the old adage “defense wins championships.” At least with regards to basketball, it has been shown that a great defensive team is a more likely to win than a great offensive team. We are a long ways away and will need a lot more data before we can answer questions that definitely, but we hope to at least start the conversation.
Stay tuned all week for the continued rollout of our project on the statistics of NexGen. The deepest debt is owed to Ian Guerin, whose countless hours of work developed the program. It would also not have been possible without immense help from Wesley Cronk and Kahyee Fong, who coded the games. Jeremy Weiss also offered thoughtful discussion and comments on early drafts of the piece and project. And big thanks to NexGen for making such great videos!