The below article shows how top Prem clubs are utilising statistical measurements more than ever before. Excellent read. It also talks about how the science of statistical measurements is evolving and growing in sophistication. http://www.ft.com/cms/s/2/9471db52-97bb-11e0-9c37-00144feab49a.html#
http://rsssf.com/tablesb/braz3-04.html has a list of all the results. Wikipedia doesn't have a page for that league and season.
Rev, Thanks for the link. I read an article about 3 Yrs ago mocking man city for using 'data nerds'. Its great not only to see man city gain some success but also recognition for their efforts. The next step is for some tangible stats to come out of this research.It is difficult for a person like me to fine hard peripheral data. Just something like completed passes is difficult to find for top leagues let alone secondary ones like the A-League or Russia. I have found some good sites for team table data but haven't found solid player data. If anyone has any links for historical player data I could use em'. Once again great thread Jj
http://avoidingthedrop.com/2011/07/19/how-does-sabermetrics-work-in-soccer/ is titled "How Does Sabermetrics Work in Soccer?"
English ARE somewhat into stats, just read a box score for a cricket match. If anything it contains more stats than the baseball box score.
Does a "chances created" metric explain Liverpool's moves? http://online.wsj.com/article/SB10001424053111904006104576502484064400802.html?mod=googlenews_wsj
Nice read. I believe for attacking players, chances created are such an important statistic we don't hear about enough.
You'll never get what you are looking for stat nerds. Companies are in agreements with leagues and clubs, and part of the allure of the information is that fact that it is not readily available. On top of that, compliation of statistics compared to American sports is much more difficult considering how many actions there are and how little play stops. And just to rebut Jjerg, soccer statistics have been popular within the sport for sometime now (over a decade in some clubs cases). They just don't talk about it as much, and as I said the information is proprietary. Also on the contrary, it is highly centralized which is how they are able to keep this information behind closed doors.
I think LFC are most definitely using data to help guide their selection of transfer targets. They are using their own performance data to help them though. Things like players being relatively robust and not having recent injury problems as well as being early 20's are also extremely important factors.
Just saw this thread and forum but I have been interested in this field for a while now. I have even written a few pieces on it here; One on Modric http://itsaballnotabomb.wordpress.com/2011/08/23/why-everyone-is-trying-to-sign-luka-modric/ Tottenham's midfield http://itsaballnotabomb.wordpress.c...have-the-best-midfield-in-the-premier-league/ And how Lampard and Gerrard are getting worse http://itsaballnotabomb.wordpress.com/2011/08/26/gerrard-lampard-and-how-theyre-both-getting-worse/
I couldn't read through everything. Has anyone mentioned the book Soccernomics, yet? Equivalent of Moneyball for soccer. Additionally, FIFA has released a slide show, I'll try and find it of relevant statistics. Like 70% of goals are scored in open play, 90% of goals scored occur from sequences consisting of seven passes or less, most goals in the men's pro game are scored in the last ten minutes of each half, the ball is in play 60 out of every 90 minute match, the team that wins the most 1v1s tend to win the game, a lot of good stuff.
Does anyone know if these four areas have meanwhile been further developed, and if so, what they are now?
Here is a breakdown of some new ways of to view a soccer match through stats. I started watching soccer 10 years ago, and though I love the game sometimes quite often my appreciation is for the aesthetics, without an idea of how productive a player was on the pitch except a for a feeling of how I remember his play. This post is will describe some potential soccer statistics and mention from qualitative assumptions how they can be measured quantitatively. The statistics I put forth are based on simple observations that even someone not knowledgeable of the game could learn quickly through the explanations of this post. As stated in previous posts soccer lags behind many other sports in using statistics to provide an explanation of how well a player performed. Many believe statistics has no place in soccer, I believe they do. For the casual American sports fan, in baseball you can measure someones productivity by Batting Average, Home Runs, RBI's, Stolen Bases, Hits, Singles, Doubles, and Triples for field players; and measure a pitcher based on their wins/losses record, ERA, and Strikeouts. With these type of stats we aren't trying to find percentages that correlate to future run or win, as Saber-metrics will state. These stats give a casual or hardcore fan who isn't a numbers person a common sense measurement of a players performance during a game or season. The stats and their measurements I am proposing give a casual or hardcore soccer fan a common sense measurement as to a players performance. Of the baseball statistics I mentioned above, some of them dont correlate with a players performance (wins/losses record) or a future win (stolen bases). But most of them illustrate the an artistry of America's past time through the speed of Ricky Henderson stealing any base at his will, or Pete Roses eye for hits, Tony Gwyn amazing batting average which led to his getting on base always, Deon Sanders getting triples when others would only single, etc. These statistics aim is to help people who are not soccer fans understand The Game and It's Tapestry through a medium they understand, statistics. I will first mentions the statistics being measures and the assumptions behind their measurement, while finishing up this post with their units of measure. These statistics are segregated by 1) Offensive Stats; 2) Defensive Stats; 3) Goalkeeping stats. The offensive stats are categorized by Ball Movement, Possession, and Striking the Ball. Offensive Stats: Ball Movement Stats- Cross, Passes, Runs, Assists Possession Stats- Dribble, Trap (Ball Control) Striking Stats- Goals, Shots Defensive Stats: Tackles, Unsuccessful Tackle, Interception Goalkeeping Stats: Saves, Interceptions Assumptions regarding the stats Unit of Measure: Cross- an elevated pass that is aimed for the 18 yard box. Only recorded when taken in middle 1/3 or attacking 1/3. Passes- only recorded when pass completed to the middle 1/3 or attacking 1/3 of pitch. Passes completed within the attacking 1/3 are worth more points. (I decided to exclude passes made to defending 1/3 of field because they bare significance or artistry to a players performance on the field). Run- is a player running with the ball at least a 1/3 of the pitch. Runs that extend to 2/3 of pitch are worth more points. Assist- last pass before a goal. Dribble- touches on the ball that beat defender(s). The amount of points given is based on area of pitch: middle 1/3, attacking 1/3. Dribble in attacking 1/3 worth more points. Trap- when a player receives a pass successfully. The amount of points given is based on area of the pitch: middle 1/3, attacking 1/3. Trap in attacking 1/3 worth more points. Goal- It's nice when the title can also be the description. Tackles- the points given is based on the area of the pitch. middle 1/3, or defensive 1/3 & attacking 1/3 (these last two areas are given equal points and more than middle 1/3 because they have a greater effect on the game and their position on the field, i.e bocanegra tackling an Italian forward or Jozy tackling within the att 18yd box). Interception (for field players)- points given are based on area of pitch. Def 1/3 > Mid 1/3 > Att 1/3. Fouls- points are given passed on area of pitch. Def 1/3 > Mid 1/3 > Att 1/3. Saves (goalkeeper)- Another description from the title. Interception (goalkeepers)- categorized based on distance to goal. 1) Within 18 yards; 2) Outside 18 yards (each category illustrates a different act in goalkeeping). The Statistics Units of Measurement: Cross: 1 Pass: 1,2,3 Run: 1,2,3 Dribble: 1,2 Trap: 1,2 Goal: 1 Shot: 1 Tackle: 1,2 Interception (field player): 1,2 Fouls: 1,2,3 Saves (goalkeeper): 1,2 Interception within 18 yards (goalkeeper): 1 Interception outside 18 yards (goalkeeper): 1 These points of each stat are not used to be cumulative with different stats. They differentiate same actions based on relation to importance of individual performance.
Hey guys, while doing some research on this kind of stuff I came across this blog from a company that does statistical analysis for teams. Has quite a few interesting articles on statistics applied soccer. Thought some of you might be interested. http://blog.statdna.com/
I'm reading Soccernomics now. I was really surprised at the Chapter that concluded that 92% of EPL's teams win/loss/tie record correlates to the team salary. Isn't that saying that despite the lack of stats, club management are fairly adept at identifying the best soccer talent? I would be interested to know how salary correlates with winning % in major league baseball. I suspect it would be far less then 92% (but I'm not sure). Moneyball was about exploiting inefficiencies in the market to allow a low payroll team to compete with high payroll teams. Do you think that kind of inefficientcy exist in the EPL?
No. It could mean they're all just as bad at identifying talent. No, because apart from youth development, EPL teams bring in more players than the 'eleven' they line up every week. The 92% would be more interesting if it would correlate with the salaries of the players that actually play every week. Yes, of course it does.
I think you are spot on now but I think Soccernomics was written in 2008 and they were analyzing a period of some period before that (20 years?...I need to check). I'm not sure they had a lot of statistics to base decisions on during that time frame.
This doesn't make any sense. Unless you're talking about negative correlation which is obviously not the case as the richer teams tend to win more often. Edit - Oh, I see what you mean. You think salaries are priced based on features X throughout the EPL and features Y are not being priced despite being more important. Is that it? If features Y are important, I would still expect much less correlation. If they are not that important, then the market is being efficient. I wonder how often it happens that the team with the highest salaries in the starting XI is not also the team with the highest salaries in the rest of the squad. Especially given the much deeper squad depth that rich teams tend to have.