Draft Day Analytics

The draft day analytics event at the 2018 Wharton Sports Business Summit consisted of a three-

person panel of industry experts, including Glenn DuPaul, the director of basketball analytics for

the Brooklyn Nets, Namita Nandakumar, a statistics analyst for the Philadelphia Eagles, and Stu

Siegel, the founder and CEO of Hockey Tech. Both Nandakumar and Siegel are graduates from

The Wharton School. The panel was moderated by Cade Massey, a professor in the Operations,

Information and Decisions department at The Wharton School. Massey is also a co-host of

“Wharton Moneyball” and the co-creator of the Massey-Peabody NFL Power Rankings in the

Wall Street Journal.

The discussion started with a comparison of the draft in the NHL, NBA, and NFL, as well as some

mention of the MLB, with Stu Siegel first introducing the NHL draft. What they pointed out as

the most notable aspect about the draft in hockey is how international many of the prospects

are. Although the NBA is becoming more international, for example recent lottery pick Luka

Doncic, a Slovenian who played in Europe for Real Madrid, for the most part the top prospects

in basketball and baseball play in the NCAA college system. This has a large impact on how

hockey scouting is conducted.

First of all, with prospects playing across the world, often in low-profile leagues, there are no

advanced analytics available on players. The prospects also very rarely play against each other,

adding another layer of difficulty to the scouting process. As a result, there is heavy emphasis

placed on two particular tournaments, the u18 and u20 World Championships. One final key

aspect of drafting in the NHL is the expectation of players. Whereas in the NFL a large number

of rookies are starters and in the NBA many rookies are key contributors, in the NHL prospects

are generally drafted with the intention of them only starting to play for the team years down

the line. The majority of players remain in their domestic leagues or are sent to a minor league,

only the very best prospects play immediately.

For the basketball prospects, the main stage to impress scouts is March Madness. There is a

large emphasis on players’ performance over the course of the tournament, and often a good

March Madness can drastically change a player’s draft stock. In terms of the draft itself, NBA

teams tend to take the best player available, with less regard for the positional need. In

contrast, NFL teams have to consider positional need heavily, and a major issue for teams to

overcome is deciding how to value each position.

Another difficulty for NFL teams is accurately predicting how a prospect will fare in the league.

The panel identified running backs as a particularly hard position to predict, and similarly

goalies in hockey. As a result, both positions are generally being picked later on in drafts,

although recently there have been some sensational talents at the running back position who

have been taken with high picks. The same can’t be said for goalies, who for the most part are

now almost exclusively taken in the second round or later.

Moving on from the draft, the panel went on to discuss the analytics themselves, and the

opportunity for improvement in the future. In the NBA, Glenn DuPaul shared that the statistics

available are already very strong. What he did say, however, was that the influence of analytics

could grow. The opposite is true for the NHL, as the nature of hockey as a constant game with

no real set plays means that there is less opportunity for incorporating analytics. Consequently,

very few statistics are actually kept by the NHL, and there is little desire for more, as Stu Siegel

explained. Meanwhile in the NFL there is great demand for analytics, not only from teams but

fans as well, especially with the development of the NFL’s Next Gen Stats. There is, however,

room for improvement in terms of understanding the data. Namita Nandakumar suggested that

there is a need to have a better understanding of how each statistic varies in importance

depending on the position.

The final topic of conversation for the panel was motion tracking, which is in its sixth year of

use in the NBA, and third year in the NFL. What was noted about motion tracking in the NBA is

that although it is widely used, it has not changed the way that teams evaluate players.

Widespread usage cannot be said about motion tracking in the NHL. Stu Siegel shared the story

of Hockey Tech’s investment into the technology, however, he went on to say that the

investment ultimately failed as there was no interest from any of the teams.

Overall the Draft Day Analytics panel was a fantastic event, and it was especially interesting to

have a representative of the NFL, NBA and NHL in attendance to give an insight into each of the

sports, as well to compare and contrast their different practices.

Finn Teoh

USBC Journal Writer

Class of 2022