Sabermetric Revolution Notes PDF

Preview

Summary

Benjamin...

Description

SABERMETRIC REVOLUTION CHAPTER 1  Central tenet of this unconventional philosophy is that teams pay too much attention to hitter’s BA and not enough attention on OBP (PAGE 2)  Identified undervalued players to assembled winning team for cheap (PAGE 2)  Sabermetric wisdom is itself limited by inadequate consideration of how threat of steal affects pitcher’s concentration, selection of pitches, and stamina (PAGE 5)  Level of play is too low and uneven to be able to make much out of performance statistics in intercollegiate baseball (PAGE 12)  Term “sabermetrics,” advances sophistication of statistical analysis and help significantly to spread its practice (PAGE 13)  Specific application of sabermetrics in order to identify market inefficiencies (PAGE 17)  High payroll does not guarantee success (PAGE 20)  Game of statistics has begun to run away with game of baseball. It’s not a sport anymore, it’s a multiplication table with baselines (PAGE 22)

CHAPTER 2  Sabermetric movement has been accessibility of research conducted by those who are not writing for academic audience (PAGE 24)  Working on analytics full time: (PAGE 26) - Teams with at least two people working mostly on analytics = 19 - Teams with one person who likely spend most their time on analytics = 2 - Teams with one person who spend part of their time on analytics = 2 - Teams with no apparent analytical presence = 2  Motivation for sabermetric self-education grew out of desire to overcome what is perceived to be own shortcomings in traditional scouting (PAGE 27)  In and outside of baseball industry were identified misfits relegated to fringes of established power structure (page 28)  Website allowed anyone to quickly answer questions that would otherwise require extensive baseball card (page 30)  Database allowed sabermetricians to compute any statistic for any major league history and sort results (PAGE 30)

 Pace of sabermetric activity and potential for its assimilation swiftly increased (PAGE 31)  Sabermetric audiences: (PAGE 32) 1. Lay audience that wants to know results of study without detail 2. Technical audience that wants to verify details  What was happening to capitalism should have happened to baseball: technical man with analytical magic should have risen to prominence in baseball management (PAGE 33)  Other resources of data include: (PAGE 37) o Track muzzle velocity and trajectory of batted balls coming off bat o Measuring magnitude and direction of how far pitcher missed spot o Radar technology to measure spin rate and flight time of pitching  More data is not always better data. What we seek is relevant data (PAGE 37)

CHAPTER 3  Strong relationship between number of runs that team scores and allows over course of season, and number of games that they win (PAGE 39)

 Cumulative number of runs scored and allowed over course of season, with no information about distribution of how those runs are scored in any game (Page 39)  Expected winning percentage: based on known ration of runs scored to runs allowed (PAGE 40)  Only quantity that really matters when evaluating team’s offense is number of runs that they score (Page 41)  Offense divided into  hitting and baserunning (PAGE 41)  Statisticians use measurement called correlation coefficient to describe strength of linear relationship (PAGE 44)  Relationship between run scored and OBP is more closely linear than it is for batting average, and correlation coefficient is higher (PAGE 45)  More recent sabermetric hitting statistics are quantifiably better at estimating number of runs that team will score than older statistics such as batting average (PAGE 47)  Randomness = Unpredictable (PAGE 48)  It is hopeless to try to predict outcome of random process, but what one can do is attempt to understand distribution of that randomness, and conversely focus









one’s effort on predicting skills accurately (PAGE 48) Notion of rehabilitee helps statisticians distinguish between signal (skills) and noise (chance) present in measurement (Page 49) Technique for quantifying notion of reliability is to examine autocorrelation of statistic which is to measure how statistic changes with respect to previous instances of itself (PAGE 49) Regression to the Mean: longstanding statistical phenomenon that governs behavior of random variables  stands to reduce error in future predictions (PAGE 55) Developing dual notion of accuracy and reliability (PAGE 56)

CHAPTER 4  “Defense wins championships” is as prevalent in baseball as it is in other sports (PAGE 57)  Pitcher’s record depended heavily on factors outside pitcher’s control, era was still considered to be a reflection of quality of pitcher’s performance (PAGE 58)  Pitchers should be judged by what happens when the ball is not put into play

 



     

against them  by number of strikeouts recorded and number of walks and homeruns they yield (PAGE 59) Public evaluates fielding through basic statistics: assists, errors, and putouts (PAGE 64) Errors tend to be subjective because it relies on judgement of home scorer  human being whose being objectivity, not including visual acuity, was called into question (PAGE 64) Most important question is not “how often do you turn a ball into an out, given that it is hit to you?” but rather “how often do you turn a ball into an out?” (PAGE 65) (RF) = Range Factor (PAGE 65) (DER) = Defensive Efficiency Rating (PAGE 65) Player’s range encompasses more than side to side but also leaps a player can make (PAGE 70) (WAR) = wins above replacement (PAGE 72) Presence of clutchness imbues certain players with ability to raise their game at most critical justice (PAGE 78) Major categories: (PAGE 82) 1. Defense 2. Offense

3. Pitching 4. Developmental strategies 5. Economics 6. Strategic decisions 7. Tactical decisions  Sabermetricians access to critical information about pitcher’s repertoire, work ethic, mental discipline, health, etc. (PAGE 83)

CHAPTER 5  Baseball lends itself to use of statistical analysis to evaluate player performance because it is easier to isolate productivity of individual players (PAGE 85)  Embrace notion that they could improve team Performance by more sophisticated application of statistical analysis (PAGE 88)  Per = Player Evaluation Rating (PAGE 92)  When understanding true efficiency of player, it is necessary to know cost AS WELL AS BENEFIT OF ACTIONS (PAGE 92)  RELIABLE PERFORMANCE METRIC SHOULD HAVE AT least two qualities: (PAGE 96) 1. Strong relationship with wins 2. Predictive  Predictive ability comes from year-to-year consistency (PAGE 96)

 Not simple to put together winning team – no matter how perceptive your statistical analysis might be (PAGE 99)  When using analytics, people often make decisions with little to no experience (page 100)  Want objective and subjective to line up (page 101)  Keep information of any area of market that you find that may be undervalued (PAGE 101)  Tape is going to be 1st choice since people must look good on film (PAGE 101)

CHAPTER 6  Quantitative analysis directly affects sabermetric concerns of player performance and team strategy (PAGE 102)  Conflicts among owners over revenue sharing frequently spill out into labor conflicts and other inefficiencies (Page 104)  System is supposed to result in payroll compression which is supposed to promote greater performance balance (page 104)  Computations depended on basic systems: (Page 111)

1. Straight pool plan: tax rate is equal for all teams 2. Split pool plan: tax top teams & distribute to bottom teams  Primary distribution mechanism continued to be straight pool system (page 113)  Economic theory teaches that incentive matters (page 114)  Designing systems & implementing are separate matters but latter always confronts reality of political divisions and constraints (page 114)

CHAPTER 7  Sabermetricians measure performance  seeking to inform through new metrics and analysis what produces wins and profits (page 115)  Factors for individual effects measurements of DER (Page 117)  Identifying more useful metrics and engendering strong team performance resides in team’s development system and its coaches (PAGE 118)  Baseball is infused with chance and uncertainty (page 118)  2 Step Analysis: (page 119) 1. Attempt to identify intensity of sabermetric practice



  



 

2. How this intensity is correlated with team performance One of most fundamental of sabermetric insights is that walks are almost as good as singles; more precisely, both walks and singles share essential and important characteristic that they do not use up one of three outs per inning (PAGE 121) DER encompasses team’s fielders’ ability to reach balls in play and to make necessary throws (PAGE 123) Fip = Fielding Independent pitching (PAGE 123) Metrics included in Index of saberintensity: (PAGE 124) 1. OnBase 2. Sac bunt 3. Brun  stolen bases 4. Der  defensive efficiency rating 5. Fip  fielding independent pitching 6. Iso  isolated power Philosophy lies in relationship between value of certain skills in producing wins an value of these skills in baseball’s labor market (PAGE 124) Estimating impact of payroll on win percentage (PAGE 131) Breakdown of player’s characteristics: (page 134)

1. work ethic 2. concentration and focus 3. competitiveness and selfconfidence 4. ability to manage stress 5. humility...