WordsOfWisdom Posted August 5, 2021 Share Posted August 5, 2021 (edited) Since so many people struggle with this, I thought I'd start a TUTORIAL thread on prorating sports statistics! In this thread we will go through the math involved in the extrapolation of data. We'll explore forecasting, and how to factor in age, injuries, and other performance factors. We will look at real world player examples, and crunch the numbers. Edited August 5, 2021 by WordsOfWisdom Quote Link to comment Share on other sites More sharing options...
WordsOfWisdom Posted August 5, 2021 Author Share Posted August 5, 2021 Let's get started! LESSON 1: First let's examine how player performance changes with age: Source: A New Look at Aging Curves for NHL Skaters (part 1) https://hockey-graphs.com/2017/03/23/a-new-look-at-aging-curves-for-nhl-skaters-part-1/ As you can see, players in the NHL peak in their early to mid 20's and then performance begins to gradually decline, falling noticeably at age 37+. Any forecast that takes data from a player's prime and extrapolates it outward to the end of their career will almost certainly be wrong. The player will clearly underperform according to the estimate. Quote Link to comment Share on other sites More sharing options...
J0e Th0rnton Posted August 5, 2021 Share Posted August 5, 2021 55 minutes ago, WordsOfWisdom said: Since so many people struggle with this, I thought I'd start a TUTORIAL thread on prorating sports statistics! In this thread we will go through the math involved in the extrapolation of data. We'll explore forecasting, and how to factor in age, injuries, and other performance factors. We will look at real world player examples, and crunch the numbers. Most people here know how to pro rate statistics. We just don't care for the Bullsh t of it all. Quote Link to comment Share on other sites More sharing options...
J0e Th0rnton Posted August 5, 2021 Share Posted August 5, 2021 37 minutes ago, WordsOfWisdom said: Let's get started! LESSON 1: First let's examine how player performance changes with age: Source: A New Look at Aging Curves for NHL Skaters (part 1) https://hockey-graphs.com/2017/03/23/a-new-look-at-aging-curves-for-nhl-skaters-part-1/ As you can see, players in the NHL peak in their early to mid 20's and then performance begins to gradually decline, falling noticeably at age 37+. Any forecast that takes data from a player's prime and extrapolates it outward to the end of their career will almost certainly be wrong. The player will clearly underperform according to the estimate. That graph certainly only applies to Modern Players. Bobby Orr with modern knee surgeries would have played another 8 to 10 years. So would a lot of 70's and 80's stars. Back then the steep cliff was usually around 32 years old. 1 Quote Link to comment Share on other sites More sharing options...
JR Ewing Posted August 5, 2021 Share Posted August 5, 2021 @WordsOfWisdom Are you talking about pro-rating or forecasting? Quote Link to comment Share on other sites More sharing options...
WordsOfWisdom Posted August 5, 2021 Author Share Posted August 5, 2021 12 minutes ago, JR Ewing said: @WordsOfWisdom Are you talking about pro-rating or forecasting? We can cover both topics. Quote Link to comment Share on other sites More sharing options...
radoran Posted August 6, 2021 Share Posted August 6, 2021 I don't dispute that players tend to decline over time. But there's a reason the graph is an average and that's that there are a not insignificant number of players above that line and a significant number of players below it. There are players who remain productive. 2 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.