Clutch hitting revisited

Slides:



Advertisements
Similar presentations
Baseball Statistics By Krishna Hajari Faraz Hyder William Walker.
Advertisements

Markov Chains in Baseball
Is There Really Racism Among MLB Umpires? Revisiting the Hamermesh Study Phil Birnbaum
Linear Transformation and Statistical Estimation and the Law of Large Numbers Target Goal: I can describe the effects of transforming a random variable.
March 2012 All data and design concepts property of Baseball Info Solutions. No part of this material or the accompanying data may be reproduced or distributed.
Moneyball in the Classroom Using Baseball to Teach Statistics Josh Tabor Canyon del Oro High School
Chapter 8 Standardized Scores and Normal Distributions
Random Numbers and Simulation  Generating truly random numbers is not possible Programs have been developed to generate pseudo-random numbers Programs.
Baseball Batting Averages A part to whole of hits versus at bats.
Chapter 8: Introduction to Probability. Probability measures the likelihood, or the chance, or the degree of certainty that some event will happen. The.
(but only when they are hitting!) David Vincent. Since 1876, 1,214 pitchers have hit 3,790 homers in the major leagues (264,773 total home runs). 567.
My Baseball survey by Angel Aguila
Simulations with Binomials Mean and S.D. of Binomials Section
Progress Report #2 By Adam Rothstein and Jesse Cox.
Comparing Counts Chi Square Tests Independence.
Chapter 11 Analysis of Variance
The statistics behind the game
Random Numbers and Simulation
Chapter 8: Estimating with Confidence
Chapter 8: Estimating with Confidence
Power and p-values Benjamin Neale March 10th, 2016
Illusions of Patterns and Patterns of Illusion
The Baltimore Orioles, Relationship of Wins and Loses, Batting Average, Earned Run Average, and Errors Stalanic Anu, Matthew Beeman, Jonathon Chudoba,
Chapter 7 Exploring Measures of Variability
PCB 3043L - General Ecology Data Analysis.
Sampling Distributions
Chapter 25 Comparing Counts.
Understanding Standards Event Higher Statistics Award
Vocabulary and Context
Unit 6 Probability.
Displaying and Describing Categorical Data
Using the Empirical Rule
Chapter 11 Goodness-of-Fit and Contingency Tables
Software Development Project Success Survey 2007
The Normal Probability Distribution Summary
Bellwork Suppose that the average blood pressures of patients in a hospital follow a normal distribution with a mean of 108 and a standard deviation of.
The Binomial Distribution
Act utilitarianism, partiality and integrity
Techniques for Data Analysis Event Study
SEE SOMETHING, SAY SOMETHING
Do Pitchers Try Harder for Their 20th Win?
The Math of Baseball Will Cranford 11/1/2018.
3.13 Probability Concepts.
Adam Smith on Trade From An Inquiry in the Nature and Causes of the Wealth of Nations (1776) “It is a maxim of every prudent mast of a family, never to.
Significance Tests: The Basics
Sampling Distribution of a Sample Mean
Paired Samples and Blocks
Least-Squares Regression
Chapter 5: Probability: What are the Chances?
Chapter 8: Estimating with Confidence
Pixel Non-Uniformity Study
Chapter 26 Comparing Counts.
The Quality Control Function
Adam Smith on Trade “It is a maxim of every prudent mast of a family, never to attempt to make at home what will cost him more to make than to buy.
Chapter 8: Estimating with Confidence
Chapter 8: Estimating with Confidence
Chapter 8: Estimating with Confidence
The Essentials of 2-Level Design of Experiments Part I: The Essentials of Full Factorial Designs Developed by Don Edwards, John Grego and James Lynch.
Chapter 8: Estimating with Confidence
Chapter 26 Comparing Counts Copyright © 2009 Pearson Education, Inc.
Exploring Numerical Data
Chapter 8: Estimating with Confidence
Chapter 8: Estimating with Confidence
Chapter 26 Comparing Counts.
Chapter 8: Estimating with Confidence
Chapter 8: Estimating with Confidence
Chapter 8: Estimating with Confidence
Chapter 8: Estimating with Confidence
Chapter 8: Estimating with Confidence
UNBREAKABLE RECORDS (I think.).
Presentation transcript:

Clutch hitting revisited Pete Palmer and Dick Cramer

Why another study? Persistence of a controversy Bill James argues “Clutchness” is objectively non-existent vs. Human perceptions of pressure Bill James argues Clutchness might be obscured by the “fog” of random variation Might this “fog” be overcome by clever grouping of players? How do pressure situations affect batting generally?

Summary of Our Findings Game situation does not significantly affect average batting performance The “fog” of statistical variation is much thicker than almost anyone appreciates The variation in career “clutchness” among the 897 players with >3000 BFP’s between 1957 and 2007 seems random Ortiz and Mark Grace are tied for ~80th and ~100th rankings among the 897

All ML Hitters Under Pressure OPS when: “Pressure” Definition (who) Tense Other Tensest 10% of BFP (best 897) .779 .771 Elias “late close” 15% (all) .704 .715 Tense Situations are different: More intentional walks Better pitchers More pinch hitters

David Ortiz’s Clutch Performances by Season 2005 Win Value 2006 10 2004 2003 2007 2000 2002 2001 1998 Linear Weight Runs 100

Comparing the “Fog” to the Clutch “Results” Width => Fog Density: calculated several ways (probability theory, simulation). All agree. “Fog” Scott Fletcher “Results” Richard Hidalgo The other 895 players who: had 3000+ plate appearances between 1957 and 2007

“Fog” Density: Starting Points Many “pressure definitions” considered All, weighted by “pressure” 10% highest “pressure”, vs. other 90% Elias “late and close” (15%), vs. other 85% First 100 batting appearances of player Individual AB’s critical => noisy win average Example: Adam Dunn on 6/30/2006 tensest 2% BFP == easiest 35% BFP

Most and least “clutch” players Nellie Fox Don Lock Dave May Pat Meares Minnie Minoso Desi Relaford Jose Uribe Sandy Alomar Earl Williams Damian Miller Mike Lieberthal Michael Barrett Frank Thomas Dick Schofield Chris Sabo Manny Ramirez

Other “pressure” non-effects Clustering of “consecutive seasons”? (e.g., Ortiz 2005-2006). Overall, r2 for “clutchness” = .002 Overall, r2 for OPS = .43 The first 100 BFP’s of a career? Performance distributions by pressure situations compared to performance distributions by game date

Conclusions Yes, the fog that probability theory demands and empirical observation confirms is thick. But why believe in something whose existence can objectively be demonstrated to be unprovable? Especially since a “clutch” hitter must actually be someone who doesn’t always perform at his best!