Atomic Events and PGAStream

Fri, 08 Apr 2016 00:00:00 +0000

Cal Newport, in his book So Good They Can’t Ignore You, uses narratives from successful careers to argue, among other things, that “to maximize your chances of success, you should deploy small, concrete experiments that return concrete feedback”. Iterating quickly on these small bets can lead to inspiration, increase your chances for success and open up new possibilities. One recent wager of mine, BirdieBot, tweets every birdie (or better) for the current PGATour event. I built this with no specific purpose in mind, just wanted to exercise some interesting frameworks and tools (like my own DataCombinator). One concrete set of feedback I noticed right away is the audience it soon found: Daily Fantasy Sports (DFS) players. However, I noticed the follower count would tend to fluctuate wildly, spiking up and then going almost immediately back down. The reason became quite obvious: the bot, like Twitter, is a firehose of information (I don’t even follow the bot). So, the logical next step (adjacent possible time) would be, how could I filter BirdieBot to only show the players and scores I cared about?

This leads us to bet #2, PGAStream. This looks very similar to Twitter, yet its stream is a record of every hole completed from every player in the current PGATour event (alongside the leaderboard). The novelty here is that you can filter the stream by player and by score type. Now the DFS players can create a stream with their stable of golfers. Or fans who want to see the best and worst holes can filter scores to only show the extreme outliers. Or maybe there’s other applications I haven’t thought of, especially as I add more filter criteria and track more events of interest like hot/cold streaks or if a golfer is on 59 watch.

The bigger point here is that users should be able to customize their data consumption down to an atomic level. The approach should be similar to Excel, which has a few simple concepts yet can act as a horizontal platform solving an entire universe of present and future problems. As consumers of information, however, we get little to no control over how we digest data. It’s a very passive experience, we’re constrained by software’s functions which usually only benefit the software itself (think sharing or retweeting on a social network). I want to build and use software that empowers users to have complete control over their data at a very granular level. PGAStream is a small start, I’m hoping to extend this idea to other realms (economic indicators, stock quotes/alerts) such that all of our media consumption can be customized.

Tech side note: I figured I’d expand my horizons a bit and learn some new tech in the process, so I built it using Python/Django. Having spent a lot of time in the Play/Scala world, I do appreciate the “batteries included” approach in Python and Django. I certainly get a better sense of the pros and cons of a different language by actually using it rather than reading someone’s blog post, so I’d encourage developers to constantly experiment with other languages. Learning a new language/framework can help revisit your assumptions, understand design decisions better, and maybe spark interest in a new way of thinking.

You’re welcome DFS folks, let me know if you have any feedback!

2014 PGA Championship Preview

Wed, 06 Aug 2014 00:00:00 +0000

The 2014 PGA Championship at Valhalla Golf Club starts tomorrow amid several interesting storylines. Will Tiger be healthy enough to play (and if so, will he tell us before our fantasy draft starts)? Will Rory continue his dominance and retain his World Number 1 status, or will Adam Scott parlay his strong majors performance and regain the top spot? Or will a past champion like Padraig Harrington or Y.E. Yang reclaim Glory’s Last Shot?

As an administrator (and participant) of a Golf Majors pool, I like to know the world rankings and betting odds for the field to help inform the draft and facilitate automated picks. For this last major, I also wanted to know how the field has performed in this year’s prior majors. Here’s how they’ve fared in 2014 (must have played in at least one major).

**Best Majors Performers in 2014**
Player	The Masters	U.S. Open	British Open
Rickie Fowler	T5	T2	T2
Adam Scott	T14	T9	T5
Jim Furyk	T14	T12	4
Rory McIlroy	T8	T23	1
Jimmy Walker	T8	T9	T26
Justin Rose	T14	T12	T23
Jordan Spieth	T2	T17	T36
Henrik Stenson	T14	T4	T39
Chris Kirk	T14	T28	T19
Matt Kuchar	T5	T12	T54
Jason Day	T21	T4	T58
Francesco Molinari	50	T23	T15
Louis Oosthuizen	25	T40	T36
Martin Kaymer	T31	1	70
Brandt Snedeker	T37	T9	T58
Bill Haas	T21	T35	T51
Kevin Stadler	T8	T63	T39
Stewart Cink	T14	T54	T47
Gary Woodland	T26	T52	T39
Keegan Bradley	CUT	T4	T19
Thomas Bjorn	T8		T26
Graeme McDowell	CUT	T28	T9
Victor Dubuisson	CUT	T28	T9
Sergio Garcia	CUT	T35	T2
Ian Poulter	T21	T17	CUT
Stephen Gallacher	T34	CUT	T15
Phil Mickelson	CUT	T28	T23
Steve Stricker	T31	T21
Brendon Todd		T17	T39
Hunter Mahan	T26	CUT	T32
Billy Horschel	T37	T23	CUT
Ryan Moore	CUT	T48	T12
Brendon de Jonge	T37	T28	CUT
John Senden	T8	CUT	T58
Kevin Na		T12	T54
Darren Clarke	T44	CUT	T26
Brooks Koepka		T4	T67
Hideki Matsuyama	CUT	T35	T39
Thongchai Jaidee	T37	CUT	T39
Zach Johnson	CUT	T40	T47
Russell Henley	T31	T60	CUT
Kevin Streelman	T42	CUT	T54
Paul Casey		T56	T47
Thorbjorn Olesen	T44		T64
Bubba Watson	1	CUT	CUT
Jonas Blixt	T2	CUT	CUT
Erik Compton		T2	CUT
Miguel Angel Jimenez	4	CUT	CUT
Marc Leishman	CUT		T5
Charl Schwartzel	CUT	CUT	T7
Lee Westwood	7	CUT	CUT
Edoardo Molinari			T7
Shane Lowry		CUT	T9
Robert Karlsson			T12
Jamie Donaldson	T14	CUT	CUT
J.B. Holmes		T17	CUT
George Coetzee			18
Angel Cabrera	CUT	CUT	T19
Matteo Manassero	CUT		T19
Chris Wood			T23
Gonzalo Fernandez-Castano	T26	CUT	CUT
Joost Luiten	T26	CUT	CUT
Ben Martin			T26
Brian Harman			T26
Steven Bowditch	T26
Kenny Perry		T28
David Hearn			T32
K.J. Choi	T34		CUT
Patrick Reed	CUT	T35	CUT
Ernie Els	CUT	T35	CUT
Branden Grace	CUT		T36
Vijay Singh	T37
Marc Warren			T39
Koumei Oda			T39
Nick Watney	T44	CUT	CUT
Webb Simpson	CUT	T45	CUT
Danny Willett		T45	CUT
Harris English	CUT	T48	CUT
Jason Dufner	CUT	CUT	T51
Tom Watson	CUT		T51
Seung Yul Noh		T52
Matt Jones	CUT	CUT	T54
Ryan Palmer		CUT	T58
Luke Donald	CUT	CUT	T64
Boo Weekley	CUT	66	CUT
Charley Hoffman			T67
Tiger Woods			69
Matt Every	CUT	CUT	71
Roberto Castro	CUT	CUT	CUT
Y.E. Yang	CUT	CUT	CUT
Graham Delaet	CUT	CUT	CUT
Pablo Larrazabal		CUT	CUT
Bernd Wiesberger		CUT	CUT
Hyung-Sung Kim		CUT	CUT
Scott Stallings	CUT		CUT
Ross Fisher			CUT
Padraig Harrington			CUT
Mikko Ilonen			CUT
Richard Sterne			CUT
Geoff Ogilvy		CUT
Shawn Stefani			CUT
Chesson Hadley			CUT
Brian Stuard		CUT
Chris Stroud			CUT
Cameron Tringale			CUT
Fredrik Jacobson			CUT
Kiradech Aphibarnrat			CUT
Rafael Cabrera-Bello			CUT
Anirban Lahiri			CUT
Tim Clark	CUT
George McNeill			CUT
Tommy Fleetwood			CUT
Brendan Steele			CUT
John Daly			CUT
Ryo Ishikawa			CUT

No big surprises here. Rickie Fowler has been sniffing at a major championship all year, falling short to two dominant performances by Kaymer and McIlory. Some other folks have been quietly consistent, like Jimmy Walker and Chris Kirk (haven’t heard their names on the Golf Channel all week). I was surprised to see normally consistent players turn in poor performances for the year (looking at you Jason Dufner and Angel Cabrera).

Hope this information helps your golf handicapping/gambling purposes! If you have any other suggestions for stats/blog posts please let me know!

Joe Conley Tagged pgatour

Atomic Events and PGAStream

2014 PGA Championship Preview