Archives

Categories

Football Match Result Database Project: The roadmap and the call

This is the fifth and final part of my week-long presentation of the Football Match Result Database project. Here are the previous articles: Part I: Selected screenshots Part II: Motivation, design and implementation histories Part III: Database objective, design, and features Part IV: Code release Now that my codebase for the FMRD has been made public, this will be the last post I make on the project on the blog. There will still be development, of course, and if there is a major release it will be announced here, but I will confine discussion to the project to the project… Read more ›

Football Match Result Database Project: Release Day!

This is the fourth part of my week-long presentation of the Football Match Result Database project. Here are the other articles in the series: Part I: Selected screenshots Part II: Motivation, design and implementation histories Part III: Database objective, design, and features Part V: Roadmap and call for contributors This is the moment that you — and I — have been waiting for. It’s Release Day. I have submitted an announcement to Freshmeat, and if approved, you should see that announcement in a few hours. (And here is the Freshmeat announcement.) Below is the information you need to know: The… Read more ›

Premier League Sharpshooters in 2010/11: Blues Shot The Most And The Least, While Spurs Shot Best

It’s that time of year again when I start going through my data from last year to make sure everything is up to date and accurate before the new season begins. In the (tedious) process of coding and cleaning data sets, I try to have a little fun and be… Read more ›

Football Match Result Database Project: Philosophy and functionality

This is the third part of my week-long presentation of the Football Match Result Database project. Here are the other articles in the series: Part I: Selected screenshots Part II: Motivation, design and implementation histories Part IV: Code release Part V: Roadmap and call for contributors Yesterday I wrote a long article on the origins of the FMRD and the various turns I took in developing the database design and its data-entry application. In this article I will discuss the functionality of the design and application: its purpose, its main features, and the philosophical rationale for the design and implementation… Read more ›

CONCACAF Club/League Coefficients: 2011-07-26 version

The 2011-12 CONCACAF Champions League has kicked off tonight, and it is time to publish the preseason set of club and country coefficients. For those new to the site and unaware of how the coefficients work, here is an explanation of my methodology. These are coefficients calculated over the last five seasons, inclusive. First, here is the league coefficient list (top ten). The numbers for Puerto Rico and Canada are misleading since they don’t have national leagues of their own that feed directly into CONCACAF. Puerto Rico has the PRSL, but the Islanders don’t come from there. Canada has the… Read more ›

Which Goals Matter? The High Value of Transition Play in the Premier League

While some goals are certainly more beautiful than others, do some actually matter more than others? I’ve been trying get a bead on this question in part by comparing the relative value of goals created in different ways. My earlier analysis of&nb… Read more ›

Football Match Result Database Project: A not-so-brief history

This is the second part of my week-long presentation of the Football Match Result Database project. Here are the other articles: Part I: Selected screenshots Part III: Database objective, design, and features Part IV: Code release Part V: Roadmap and call for contributors I’ve looked back at my previous posts on the subject and I realize that I have been writing about this project for almost two years! There are new visitors to this site all the time, so I think it would be good for my readers and me to revisit the origins for this project and the reasons… Read more ›

Why We Shouldn’t Put Much Value in Assists

Last week I wrote a post on why shots on goal are a misleading statistic. In keeping with the analysis of the problems with some commonly kept statistics in football, I decided to look at assists. 
If you think about it, assists are highly misleading. Simply playing with good players boosts your assist total. Similar to shots on goal, not all assists are the same. There are the assists where a player makes a short pass in the midfield that leads to a teammate dribbling through all the opposing defenders and finishing, and the assists where a player makes a beautiful cross where their teammate simply has to tap the ball in the open net. These obviously shouldn’t be counted as the same value to the team, yet they are. Hell, I could probably record an assist eventually in the EPL if I played for one of the top teams (OK, maybe an exaggeration but you get the point.)
First, let’s look at the assists data for all the teams in the EPL league. As the graph below shows, as the point value of a team increases (basically, the better the team is) the assist total also generally increases. This is no surprise. We would expect better teams to score more goals and thus have more assist totals.
Basically what this means is that the assist statistic should favor players on better teams. Players on better teams play with better teammates and should therefore have more opportunities for assists. Below is a screenshot from the EPL website of the players with the top 20 assist totals.
9 players from top 5 clubs are in the top 20 for assist totals. No players from bottom 3 clubs are in the top 20, with the exception of Blackpool’s Charlie Adam who was just signed by Liverpool. It’s easy to see assists totals are higher for players on better clubs.
A better statistic that is not influenced by the quality of your teammates are chances created. A chance created is defined as a pass that leads to a shot. These are obviously not as dependent on your teammates and give a more fair and true assessment of how much of a playmaker that player is for their team. 
The next time a club is looking to sign a player based solely on their assists totals, they should take a more in depth look. Assists can tell an inaccurate, or at the least biased, story.

Read more ›

Categories
Uncategorized