Stats in the Wild

recent posts

about

  • On soccer’s declining home field advantage

    statsbylopez's avatarStatsbyLopez

    As part of their final assignment in my statistics and sports class, students were tasked with looking at the home advantage in the English Premier League (EPL). In some recent and related work, James Curley and Oliver Roeder found that, by 2014, an EPL home advantage had reached an all time low.

    Interestingly, that low reached new depths in 2016.

    Home teams have won 40.8% of games this past year, pending this weekend’s final contests. If that mark stands, it would be the lowest in EPL/English Division 1 history, one which dates back to 1888.

    Here’s a chart, similar to the one that James and Oliver produced. Overall home team win percentage in each year is shown in black, draw percentage in red, and away win percentage in green. The grey region reflects our uncertainty in the trend curve.

    EPL

    As we knew there’d be, it’s a fairly big drop in win percentage, from roughly 60% to 45% across about 120 seasons…

    View original post 729 more words

  • The NFL draft – where we stand in 2016

    statsbylopez's avatarStatsbyLopez

    Another NFL draft has come and gone, and with it has come the predictable displays of unyielding optimism, stale and arguably race-based generalizations of player skill, and, as a relative newcomer in 2016, lazy misuse of the term analytics.

    In following along this spring, it became clear that what is mainstream knowledge among researchers is far from it in the national media. This despite a decent amount of both academic and non-academic research into the topic.

    For those new to the scene, or even for a few veterans who may have missed an article or two along the way, I decided to write a quick review of what’s out there. Note that many of the following points are related to one another.

    1. Top draft picks are overvalued. 

    In an efficient market, the value of picks traded between teams would be equivalent.  That is, if Pick X was traded for Picks Y and Z…

    View original post 1,025 more words

  • New paper about Random Walks and Edge Sampling

    Respondent Driven Sampling is awesome.

  • JSM2016 Data Art Show

    JSM 2016 will host a Data Art show for the first time at this year’s meeting in Chicago.  If you are interested in submitting a piece, please follow the instructions here.   Everyone who is interested should submit something for acceptance.  And remember “I’m not an artist” is a bad excuse.

    If you have any questions to can email artjsm2016@gmail.com

    Cheers.

     

  • Cinderella Plots

    Cinderella2016CinderellaPlot2004-2016

    Cheers.

  • One Shining MGF: So, so, so, so, lucky.  

    So lucky.

    Cheers.

  • Cinderella Plot: 2001 – 2016

    I originally made this plot for the 2012 Stat Geek Idol contest run by Team Rankings, and they are still cool.  The original article describing them can be found here. Below I’ve updated the plots for games through March 19th.  (For games that haven’t been played yet this year, I’ve assumed the higher seed wins for the sake of the plot.)

    Cheers.

    Cinderella2016

    CinderellaPlot2004-2016.png

  • The ASA’s statement on p-values: context, process, and purpose

    “The ASA’s statement on p-values: context, process, and purpose”

    Cheers.

     

  • NCAA Tournament Field – Who Should be in?

    Here are my current picks for teams that should be in the tournament.  I’m NOT projecting the field, I’m modeling teams I think should be in.

    Cheers.

     

    Last 4 in: Syracuse, Colorado, Michigan, Florida St

    First 4 out: Providence, Kansas St, Creighton, Tulsa

    Next 4 out: BYU, G Washington, Oregon St, Georgia Tech

     

    1 seeds: Kansas*, Villanova*, Michigan State*, Virginia*

    2 seeds: North Carolina, West Virginia, Oklahoma, Miami (FL)

    3 seeds: Xavier, Louisville, Kentucky*, Duke

    4 seeds: Oregon*, Purdue, SMU*, Arizona

    5 seeds: Indiana, Texas A & M, Utah, Maryland

    6 seeds: Iowa State, Iowa, Vanderbilt, California

    7 seeds: Wichita St*, Baylor, Texas, Seton Hall

    8 seeds: Gonzaga*, Butler, St. Mary’s, Pittsburgh

    9 seeds: Wisconsin, Cincinnati, Notre Dame, St. Josephs

    10 seeds: VCU*, UConn, USC, Texas Tech

    11 seeds: Valparaiso*, Florida, South Carolina, Dayton

     

     

    12 seeds: (Play-In: Syracuse/Florida St), (Play-In: Colorado/Michigan), San Diego St*, Arkansas – Little Rock*

     

    13 seeds: Yale, South Dakota St*, Monmouth*, Hawaii*

     

    14 seeds: Stephen F. Austin*, Chattanooga*, Stony Brook*, UAB*

     

    15 seeds: Ohio*, Grand Canyon*, Weber St*, UNC Asheville*

     

    16 seeds: N Florida*,Texas Southern*, (Play-In:Lehigh*/Austin Peay), (Play-In: Wagner*/Norfolk St*)

     

    * – Indicates projected conference champion, Bold – Indicates actual conference champions

     

     

     

     

     

  • Talk at Columbia Business School

    People are idiots.

    statsbylopez's avatarStatsbyLopez

    I spoke this morning about prediction algorithms and NCAA basketball for Mark Brodie’s class at Columbia’s Business School. Also speaking was Ed Kaplan of Operations Research fame.

    Here are my slides, much of which was built from joint work with Greg in the Kaggle contest. There are also extensions to other things in life that I like, such as upset pools, game theory, and extensions/extrapolations to sports analytics as a whole.

    View original post