NYC Tech Events - GarysGuide | The #1 Resource for NYC Tech

LOCATION

EVENT DETAILS

Fordham Schools of Business CDT and FBAS Present the Annual:MarchData CrunchMadness

The NCAA Tournament in March is fast approaching.Are you ready to fill out yourbrackets?
<H2>Sponsored by:Deloitte</H2>

What if you could use data analysis and machine learning techniques to optimize your picks and beat everyone in your pool? Thats what Fordham Universitys March Data CrunchMadness will aim to find out! The event is being supported by the Center for Digital Transformation at Fordham and the Fordham Business Analytics Society.

March Data CrunchMadness will be a fun, data-centric competition hosted by Fordham Universitys Business Analytics Society on 2016. Students will compete in groups of 4 to build a model to predict the NCAA "March Madness" Basketball Tournament results.

Current season data will be released soon. A poster session and event will be held on the end of Marchwhere judges will be on hand to rate the submissions on presentation, methodology, and results. Prizes will be awarded to the winners.
Further information will be released at the kickoff event on - February 8th
<UL>
<LI>Team Size: 4 Members</LI>
<LI>Sponsors: Deloitte</LI>
</UL>
Evaluation of Submissions
<UL>
<LI>3 parts (Results, Analytics & Insight, Presentation)</LI>
<LI>Results evaluated using LogLoss formula</LI>
</UL>
<DIV>
<DIV><IMG SRC="https://cdn.evbuc.com/eventlogos/133781087/evaluation-1.png" ALT=""></DIV>
<DIV> </DIV>
<DIV> </DIV>
</DIV>
<DIV> </DIV>
<DIV>Where:</DIV>
<UL>
<LI>n=number of games played</LI>
<LI> = predicted probability team 1 beats team 2</LI>
<LI>yi = outcome of each game</LI>
<LI>log() = natural (base e) logarithm</LI>
<LI>Goal is to minimize LogLoss</LI>
</UL>

Methodology Examples
 
<IMG SRC="https://cdn.evbuc.com/eventlogos/162894471/capturenewjpg.jpg" ALT="Methodology Example">
 
 
Poster Board will be Final Report
Format of Poster Board
Overview & Introduction
Hypothesis & Methodology: Data Selection, Analytics Explored, Data Mining Techniques
Analytics & Results: Results of Analytics, Results of Data Mining Techniques
Conclusions & Suggestions for Improvement: Performance of Model
 
Possible Analytics
What impact, if any, does distance from home campus and location of game have on outcome?Which efficiency metrics have most predictive power of outcome of game?
What predictive power does seeding have on predictions?
Is there additional data you can find to assist in the model creation?
 Glossary for Dataset
<UL>
<LI>game_id id of each game</LI>
<LI>season Years that the NCAA season spanned. The tournament takes place in the latter year</LI>
<LI>team1_id Team 1 id</LI>
<LI>team2_id Team 2 id</LI>
<LI>team1_score Final score of team 1</LI>
<LI>team2_score Final score of team2</LI>
<LI>numot Number of overtimes</LI>
<LI>team1_seed Team 1 seed in the tournament. There are 4 brackets of teams seeded 1 through 16.</LI>
<LI>team2_seed Team 2 seed in the tournament.</LI>
<LI>host_site The location of game</LI>
<LI>host_lat Latitude of location of game</LI>
<LI>host_long Longitude of location of game</LI>
</UL>
<DIV> </DIV>
The following stats are for both the team 1 and team 2:
<UL>
<LI>lat Latitude of teams home campus.</LI>
<LI>long Longitude of teams home campus.</LI>
<LI>teamname Team name</LI>
<LI>fg2pct Shooting percentage on 2 point field goals.</LI>
<LI>fg3pct Shooting percentage on 3 point field goals.</LI>
<LI>ftpct Shooting percentage on free throws.</LI>
<LI>blockpct - Blocked shots divided by opponents 2 point field goal attempts.</LI>
<LI>oppfg2pct Opponents shooting percentage on 2 point field goals.</LI>
<LI>oppfg3pct Opponents shooting percentage on 3 point field goals.</LI>
<LI>oppftpct Opponents shooting percentage on free throws.</LI>
<LI>oppblockpct Opponents blocked shots divided by 2 point field goal attempts.</LI>
<LI>f3grate Percentage of field goal attempts that are 3 point field goal attempts.</LI>
<LI>oppf3grate Percentage of opponents field goal attempts that are 3 point field goal attempts.</LI>
<LI>arate Assists divided by field goals made. Percentage of field goals that were preceded by an assist.</LI>
<LI>opparate Opponents Assists divided by opponents field goals made.</LI>
<LI>stlrate Steals divided by defensive possessions.</LI>
<LI>oppstlrate Opponents steals divided by opponents defensive possessions.</LI>
<LI>tempo - Estimated possessions using this formula: FGA-OR+TO+0.475xFTA. For each team, possessions are counted for the team and their opponents, and then averaged. A teams average tempo is total possessions divided by minutes.</LI>
<LI>adj_tempo An estimate of the tempo (possessions per 40 minutes) a team would have against the team that wants to play at an average D-I tempo.</LI>
<LI>oe Points scored per 100 offensive possessions.</LI>
<LI>adjoe An estimate of the offensive efficiency (points scored per 100 possessions) a team would have against the average D-I defense</LI>
<LI>de Points allowed per 100 defensive possessions.</LI>
<LI>adjde An estimate of the defensive efficiency (points allowed per 100 possessions) a team would have against the average D-I offense</LI>
<LI>ap_preseason The preseason AP Poll ranking of each team (top 25 only)</LI>
<LI>ap_final The final AP Poll ranking of each team (top 25 only)</LI>
<LI>coaches_preseason - The preseason Coaches Poll ranking of each team (top 25 only)</LI>
<LI>coaches_before_final The most recent Coaches Poll rankings before the final. The Coachs poll final poll is released after the tournament (making it not very valuable for predictions)</LI>
<LI>rpi_rating The RPI rating of each team</LI>
</UL>
 The following stats are coach data
<UL>
<LI>coach_id coachs id</LI>
<LI>coach_name coachs name</LI>
<LI>pt_school_ncaa Number of NCAA Tournament appearances at current school</LI>
<LI>pt_overall_ncaa Career overall number of NCAA Tournament appearances </LI>
<LI>pt_school_s16 Number of NCAA Sweet Sixteen appearances at current school</LI>
<LI>pt_overall_s16 Career overall number of NCAA Sweet Sixteen appearances</LI>
<LI>pt_school_ff Number of NCAA Final Four appearances at current school</LI>
<LI>pt_overall_ff Career overall number of NCAA Final Four appearances</LI>
<LI>pt_career_school_wins Number of wins at current school</LI>
<LI>pt_career_school_losses Number of losses at current school</LI>
<LI>pt_career_overall_losses Career overall number of losses </LI>
<LI>pt_team_season_wins Teams number of wins in this season</LI>
<LI>pt_team_season_losses Teams number of losses in this season </LI>
<LI>pt_coach_season_wins Coachs number of wins in this season</LI>
<LI>pt_coach_season_losses Coachs number of losses in this season</LI>
</UL>
Have questions about March Data Crunch Madness?
<A TARGET="_blank" CLASS="contact_organizer_link js-d-modal" HREF="https://www.eventbrite.com/e/march-data-crunch-madness-registration-15610896612#lightbox_contact">Contact Fordham Business Analytics Society</A>
<A TARGET="_blank" HREF="https://drive.google.com/a/fordham.edu/file/d/0B59ZSR_8Bny9cTBLRzRzNlA2Rmc/view" REL="nofollow"> </A>