Soccermetrics Science Of Soccer Statistics Sheets - Sports Betting

Laref.win

Sports Betting

Soccermetrics Science Of Soccer Statistics Sheets

Category: Predictions

Description

Soccer player ratings are confusing, erratic, useless - and great

Soccer player ratings are confusing, erratic, useless … and great. Lionel Messi Goes to 11 Soccer's player ratings are confusing, erratic, and useless. More sports should have them.

The U.S. men's soccer team lost 1-0 to Costa Rica last Friday, but the result masked some strong individual play from the Americans as the players tried to implement new coach Jurgen Klinsmann's ball-control strategy. But which players stood out? It depends whom you ask. ESPN's Jeff Carlisle rated Jozy Altidore one of the team's top performers on the night, giving him a 7 on soccer's 1-to-10 player-rating scale and noting his "excellent link play." SI.com's Steve Davis gave Altidore his lowest grade of the night, a 4, writing that the powerful forward never came close to scoring.

For Altidore to earn such vastly different ratings, you'd think he gave a truly confusing performance—perhaps he dribbled Maradona-like through the Costa Rican defense only to Ronny Rosenthal the ball off the crossbar. Or maybe he was an asset in the attacking run of play but a liability on defense and set pieces. But no—Altidore played a typically Altidorean game in which he often got to the right place at the right time but lacked the touch to finish off his chances.

Jozy Altidore's conflicting player ratings have less to do with the player than with the scale. The game of rating players from 1 to 10 is a subjective mess with no statistical value—a poor stand-in for real statistics in a game desperate for them. Despite its flaws, the scale seems to be growing in popularity.

Player ratings first appeared in the late 1970s in England. To stand out from the glut of soccer magazines in the U.K., newcomer Match hired a London-based news agency called Hayters to provide in-depth statistics for every British match, all the way down to the Scottish Second Division. Each writer was tasked with compiling the Match Facts, which included an Entertainment Rating for each game, as well as the Player Ratings, in which the starters and substitutes were given a score between 1 and 10.

Match's readership took off, and the publication soon brought Match Facts in-house. The mag also began awarding a Matchman of the Month award to the player in each division with the highest average player rating, presenting the trophies at stadiums around the country. It was a great marketing tool, says then-editor Mel Bagnall, and the subjectivity of the ratings only added to their appeal. "Match [was] the currency for debate in playgrounds, on terraces and in the pubs," he says.

Other soccer rags, including Shoot and Football Weekly, soon started offering similar athlete appraisals. Over the ensuing decades, the practice has gone worldwide. Nowadays, every soccer game of consequence—and many of no consequence—is followed by sheets of 1-to-10 player ratings from reporters and bloggers.

These ratings wouldn't be so popular if soccer fans had more stats to bat around than goals and assists. While a few advanced statistics are slowly rolling out, soccermetrics are way behind those of baseball, basketball, and even (American) football. John Godfrey of the New York Times' Goal blog says that the current stats are both limited and misleading: "A striker might score two goals in a match, but one of them could be a sloppy deflected shot off a corner kick and the other could be through a penalty kick that another player earned." In that case, Godfrey says, the player ratings are a good way to add some context, and to give the players' teammates the plaudits they deserve.

While the raters don't usually come to a consensus, they do agree when someone turns in a spectacularly bad performance. The analysts from the New York Times, Washington Post, ESPN.com, and SI.com all said that Jonathan Bornstein, the much-maligned left back, was the worst player on the field during the United States' Gold Cup loss to Mexico. Breaking with the unspoken tradition of giving all players a 3 or higher simply for competing, several analysts gave him a 2. Godfrey gave him a 1.

As Match's Bagnall hinted, controversial ratings are good for business, and that's especially true in an era when pageviews are directly related to profitability. "I get many more emails commenting on my grades than I do about all the more in-depth stuff I write put together," says ESPN.com's Leander Schaerlaeckens. "Which is probably the point." The soccer player ratings are the equivalent of NFL power rankings and the Heisman experts' poll—an ultimately meaningless exercise that exists mostly to rile up fans.

For soccer followers, the player ratings are just confusing. SI.com's Davis consistently gives Altidore lower scores than ESPN.com's bloggers do, and recently has started to rate most players lower. He gave Altidore a 3 and midfielder Jose Torres a 4 for their performances against Belgium on Tuesday. ESPN.com's Schaerlaeckens gave them a 5 and a 7.5, respectively.

While part of it may be that Davis thinks that Altidore is a worse player than Schaerlaeckens does—and a comparison of their ratings over the past couple of years gives that impression—it's also true that different raters have different standards. "I, for one, think players should be graded within the context of their own ability and surroundings," Schaerlaeckens says. "Others seem to take a wider approach."

That simple 1-to-10 scale, then, has to do a lot. A rater must evaluate a player's performance in a particular game. He also must compare that performance to that of his teammates, weigh it against how the player did in previous games, and place it somewhere in the context of the worldwide level of play. If you're judging both Lionel Messi and Jonathan Bornstein on the same point system, you run out of room for intraplayer nuance pretty quickly. The result is a mess of precise but entirely useless numbers with no internal or external consistency.

While the player ratings are statistically worthless, they are vehicles for useful information. Besides the simple grade, each player is the subject of a bit of evaluative exposition. The best commentators pay attention to the player's role throughout the pitch, noting off-the-ball marking and runs, and give readers fresh perspectives that they couldn't have gotten elsewhere. Jack Bell of the New York Times, for example, noted that winger Robbie Rogers didn't just fail to convert any touches into scoring opportunities against Belgium—he never got into position to receive any passes at all.

We're missing that level of individual analysis in American football, where valuable players are ignored because they don't produce easily understandable statistics. I never hear about the play of Broncos right guard Chris Kuper, even though the quality of his blocks could be the key to a Denver victory or defeat. A tip for pageview-chasing NFL writers: Start doing postgame player ratings. Fans won't be able to stop themselves from clicking, and they'll debate the scores until it's time to start the next game. And who knows, they might even learn something.

Slate is published by The Slate Group, a Graham Holdings Company. All contents © 2017 The Slate Group LLC. All rights reserved.

Video

Other articles

Introducing the Football Match Result and Summary Database, Soccermetrics Research, LLC

Soccermetrics Research, LLC Soccer from First Principles Navigation Introducing the Football Match Result and Summary Database

The ‘lite’ dataset released by the MCFC Analytics project last weekend is a massive dump of almost 200 summary in-match metrics for every player who participated in every match of the 2011-12 English Premier League — more than 10,000 records and 2 million data cells. We set out to convert that dataset into something more structured and amenable to our analytics software. The result is the FMRD-Summary database.

The FMRD-Summary database schema models data that are less granular than the Football Match Event Database, but more detailed than the Football Match Result Database. It captures the much of the same type of data that you would capture with the Football Match Event Database, with the exception of the Venue and Venue History tables. The main feature is that it captures in-match statistical summary data down to very fine levels of detail, in particular field position, body part, and type of open or set play. The schema is built for those who track summary statistics in football matches as well as historical match data.

The current schema only models data from league competitions, but extensions to capture data from knockout and/or hybrid competitions can be added if enough people request it. We have added a Seasons table to simplify the Competitions table a bit (that change will eventually make it to the other FMRD schemas).

Besides the tables that are inherited from the Football Match Result Database schema, there are 33 tables that capture in-match statistics. Yes, it does sound like a lot and it does get quite involved, but it’s easier to deal with than 200+ columns on a sheet. We group these tables further into nine categories:

The schema exists — coded in SQLite — and lives on the Soccermetrics fmrd-summary repository at GitHub.

Over the weekend I’ll describe the database tables and their fields in a series of three posts (three categories each). I’ll set up a wiki on the repository and copy my descriptions over there.

Schemas are nice to have, but what we really want is a database created with these rules. So I will present some code that starts the process of loading the database with the MCFC Analytics data. I do want this to be an open source project, so contributions are very much welcome.

About Howard Hamilton Follow Soccermetrics

Follow us on the following social networks:

Soccermetrics on Twitter

.@RacingClub 1-0 @catigreoficial. At last Racing's most expensive buy comes through, but still vulnerable in back. @DataFactoryLA @argsaf 9 hours ago

.@Velez 1-0 @CANOBoficial. Back on track for Vélez after two losses. Newell's need more shot creation from open play. @DataFactoryLA @argsaf 9 hours ago

.@CARCoficial 1-3 @AAAJoficial. Argentinos opportunistic, three consecutive Bicho wins, Central nervous/imprecise. @DataFactoryLA @argsaf 9 hours ago

.@clublanus 2-1 @clubaunion. Unión converted once from a set-piece in the box, but not enough vs surging Granate. @DataFactoryLA @argsaf 1 day ago

.@AAAJoficial 1-0 @ChacaOficial. Held over from Matchday 1 - Argentinos do just enough in Clásico de los Ascendidos. @DataFactoryLA @argsaf 1 day ago

Happy to be moving on to club football. 1 day ago

CONMEBOL has the best World Cup qualifiers and it's not even close. 1 week ago

A sports analytics conference at Carnegie Mellon on 28 October cmusportsanalytics.com/conference/ 2 weeks ago

Mejores arqueros luego de Fecha 5: Agustín Rossi (-0,75 GC con respeto de lo esperado), Sebastián Torrico (-0,50), Marcos Díaz (-0,23) 2 weeks ago

.@catigreoficial 1-1 @CARPoficial. Parece que River dejó su poder ofensivo en aquél partido contra Wilstermann. @DataFactoryLA @argsaf 2 weeks ago

.@BocaJrsOficial 1-0 @ChacaOficial. Estuvo lejos de sus rendimientos recientes, pero igual Boca logró la victoria. @DataFactoryLA @argsaf 2 weeks ago

Latest Posts Blog Archives

Soccermetrics Research, LLC © 2017. All Rights Reserved.

Goalscoring variances at close of MLS regular season, Soccermetrics Research, LLC

Soccermetrics Research, LLC Soccer from First Principles Navigation Goalscoring variances at close of MLS regular season

Three months ago I wrote an article on the goalscoring characteristics of champion and relegated teams, and showed that champion clubs in general have very tight and consistent defenses.  In particular, the goals allowed variance — the spread of the individual numbers about the average goals allowed — tends to be less than 1.0 for the top teams, which means that the standard deviation is less than 1.0 goals/game. (Standard deviation is the square root of variance.)

Now that the 2011 MLS regular season is over, it might be useful to plot a similar chart of team goalscoring statistics. 

Below are two plots — the first shows a scatter plot of the mean and variance of goals scored by MLS teams, the second shows a similar plot for goals allowed.  I've tagged each point with a three-letter code for a MLS team, and the key follows the figures.

(Key: CHI = Chicago Fire, CHV = Chivas USA, CLO = Colorado Rapids, CLM = Columbus Crew, DCU = DC United, FCD =  FC Dallas, HOU = Houston Dynamo, LAG = LA Galaxy, NER = New England Revolution, NYR = New York Red Bulls, PHI = Philadelphia Union, PDX = Portland Timbers, RSL = Real Salt Lake, SJE = San José Earthquakes, SEA = Seattle Sounders FC, SKC = Sporting Kansas City, TFC = Toronto FC, VAN = Vancouver Whitecaps FC.)

There are several characteristics of the playoff teams that are borne out in the charts.  The playoff sides generally score between 1.2-1.5 goals per game, but their offensive goal variances vary widely. Houston scored more than two goals in just four matches, while Chicago had a large number of matches in which they scored one goal and five in which they scored three (none in which they scored four or more).  Seattle were the big outlier as they showed themselves equally capable of being free scoring and unable to find the back of the net.  DC United were another outlier among non-playoff teams; their average goals scored is higher than LA's and exceeded by just three other teams (Sporting KC, NY Red Bulls, Seattle).

On defense, Los Angeles Galaxy were head of the class with 17 clean sheets, but their goals allowed variance crept above one thanks to two matches in which they allowed three goals and one in which they allowed four.  Real Salt Lake's defense is just as good as LA's (14 clean sheets, 10 matches allowing one goal), but they allowed more than two goals in ten matches, equal with Philadelphia and Seattle.  Houston had the lowest goals allowed variance in the league thanks to allowing one goal in 17 matches, but their inability to keep a clean sheet inflated their goals allowed average.  DC United's defense proved to be a liability to the team's playoff hopes, and Toronto FC's defense was wildly inconsistent (they got on enough of a run to advance to the CONCACAF Champions League quarterfinals, though).

I know that the MLS Cup Playoffs — and all postseason tournaments — are randomizing events in a league season.  But if goal defense characteristics are the same indicators of playoff success as they are of league success during the regular season, I'd expect a Galaxy-Union MLS Cup final.  However, I am intrigued by the upcoming RSL-Seattle series and the impact of a Sporting Kansas City side that is much improved from the one that opened its new stadium in June.

About Howard Hamilton Follow Soccermetrics

Follow us on the following social networks:

Soccermetrics on Twitter

.@RacingClub 1-0 @catigreoficial. At last Racing's most expensive buy comes through, but still vulnerable in back. @DataFactoryLA @argsaf 9 hours ago

.@Velez 1-0 @CANOBoficial. Back on track for Vélez after two losses. Newell's need more shot creation from open play. @DataFactoryLA @argsaf 9 hours ago

.@CARCoficial 1-3 @AAAJoficial. Argentinos opportunistic, three consecutive Bicho wins, Central nervous/imprecise. @DataFactoryLA @argsaf 9 hours ago

.@clublanus 2-1 @clubaunion. Unión converted once from a set-piece in the box, but not enough vs surging Granate. @DataFactoryLA @argsaf 1 day ago

.@AAAJoficial 1-0 @ChacaOficial. Held over from Matchday 1 - Argentinos do just enough in Clásico de los Ascendidos. @DataFactoryLA @argsaf 1 day ago

Happy to be moving on to club football. 1 day ago

CONMEBOL has the best World Cup qualifiers and it's not even close. 1 week ago

A sports analytics conference at Carnegie Mellon on 28 October cmusportsanalytics.com/conference/ 2 weeks ago

Mejores arqueros luego de Fecha 5: Agustín Rossi (-0,75 GC con respeto de lo esperado), Sebastián Torrico (-0,50), Marcos Díaz (-0,23) 2 weeks ago

.@catigreoficial 1-1 @CARPoficial. Parece que River dejó su poder ofensivo en aquél partido contra Wilstermann. @DataFactoryLA @argsaf 2 weeks ago

.@BocaJrsOficial 1-0 @ChacaOficial. Estuvo lejos de sus rendimientos recientes, pero igual Boca logró la victoria. @DataFactoryLA @argsaf 2 weeks ago

Latest Posts Blog Archives

Soccermetrics Research, LLC © 2017. All Rights Reserved.

APItools, Soccer Metrics API

Soccer Metrics API

Connecting the world's data to the world's game.

About Soccer Metrics API

Rich data and sophisticated analytics on the world's football competitions, to power Soccermetrics' research projects, applications, and website.

Important Links
  • Developer Portal https://developer.soccermetrics.net/
  • Documentation http://soccermetrics.github.io/connect-api/
  • Get a Key https://developer.soccermetrics.net/login
  • Developer Console https://developer.soccermetrics.net/docs
  • Support https://developer.soccermetrics.net/contact-us
  • Forum https://developer.soccermetrics.net/forum
  • Twitter https://twitter.com/sccrmetricsapi
Universal Middleware CORS 1.0.0 by 3scale

Cross Origin Resource Sharing allows you to use API directly from the browser

Cache 1.0.0 by 3scale

Caches responses so you don't have to care about rate limiting!

Measure X-Ratelimit-Remaining 1.0.0 by 3scale

Allows measure custom metrics.

Accept-Encoding = Identity 1.0.0 by 3scale

Sets the `Accept-Encoding` header to `identity`. This deactivates gzipped responses. This middleware is inserted by default in all new services.

add-args 1.0.0 by 3scale

adds a paramter

Add Keys 1.0.0 by 3scale

Add authentication key as a header.

Response Time Alert 1.0.0 by 3scale

Sends an email in case a call takes more than expected

response-size-metric 1.0.0 by 3scale

Adds a metric called 'size', with the response size. This metric can be used in the analytics tab to show graphics.

xml-to-json 1.0.0 by 3scale

Transforms an xml response into a JSON response

Privacy 1.0.0 by 3scale

All-purpose middleware. Customize to adjust it to your app's privacy concerns

404 Alert 1.0.0 by 3scale

Sends an email in case a call returns a 404

HTTP Cache 1.0.0 by Ludek Berecka

Cache server responses using HTTP semantis like Etags and Last-Modified.

500 GitHub issue 1.0.0 by Giovanni Cappellotto

Open a Github issue when an error occurs.

500 Webhook 1.0.0 by Giovanni Cappellotto

Send a POST request to URL when an error occurs

Slow request GitHub issue 1.0.0 by Giovanni Cappellotto

Open a GitHub issue when a request takes too much time

500 Bitbucket issue 1.0.0 by Giovanni Cappellotto

Open a Bitbucket issue when an error occurs

Related APIs (tags: sports, football, analytics) WorldCup API

World Cup' in JSON.

ChurnBee API

Secret weapon for SaaS startup founders.

Ducksboard API

All of your data. In one place.

Google APIs

Google Apps Application APIs

Keen IO API

Custom analytics shouldn't be a pain in the backend.

Mixpanel API

The most advanced analytics platform ever for mobile and the web.

Soccermetrics Soccer Data And Analytics API Is The Future Of Global Football, Today, Sports Techie

Soccermetrics Soccer Data And Analytics API Is The Future Of Global Football, Today

July 13, 2014 , Blog One comment

Soccermetrics Soccer Data And Analytics API Is The Future Of Global Football, Today – Sports Techie blog

Soccermetrics Soccer Data And Analytics API Is The Future Of Global Football, Today

Today is the 2014 FIFA World Cup Finals between heavyweights Germany and Argentine, two of the best teams going into the tournament that comes as no surprise to the sports analytics world, Bloomberg Sports and many of the billions of fans worldwide. When trying to get all of the key soccer stats and info, a new API, powered by 3SCALE, is now available from Soccermetrics for media organizations, sports teams, software developers and soccer leagues. Soccermetrics’ Howard H. Hamilton, Ph.D., and Guillaume of 3Scale, spoke to the Sports Techie community blog from here in Atlanta and in Barcelona, Spain, about how Soccermetrics soccer data and analytics was born to ponder problems in applied mathematics and statistics that have applications toward a better understanding of soccer.

When trying to get all of the key soccer stats and info, a new API, powered by 3SCALE, is now available from Soccermetrics for media organizations, sports teams, software developers and soccer leagues.

Soccer has been the last of the major sports to witness a data revolution over the last ten years for a number of technical and cultural reasons.

The Soccermetrics Connect API, is a sports modeling and analytics layer on top of in-match data sources at varying levels of complexity. The soccer API delivers advanced match analysis of data in an accessible form and makes it easier for end-users to create their own customized analysis tools on football data. Data has already disrupted industry and now it’s disrupting sport.

These applications range from a macro level (team performance, comparative performance of national leagues) to a micro level (player performance and evaluation, influence of formations/strategies, indicators of future success).

Biographical and demographic data of players, referees, and managers are captured. Venue data includes geographic data — not just city and country, but also latitude, longitude, and altitude — as well as capacity, field temperatures and field characteristics. Match data encompass not just data pertinent to league matches, but to group and knockout stage matches as well.

The Soccermetrics Connect API permits distribution of rich data at all levels of complexity in a football match, whether a club match or one involving national teams. The core of the API is its access to match event data, from the major events such as goals, penalties, bookings, and substitutions, to summary statistics and the micro events that make a match happen — every touch and every movement of ball and player. The API makes it easier for end-users to create their own customized analysis tools on sports data. Averages, expected values, shot ratios, Pythagoreans, adjusted plus/minus are some of the metrics available. Soccermetrics Connect is a 100% Unicode, 100% JSON-based REST API.

Because of this sport business trend, the pioneering sports technology Company unveiled a paid version of the API that end users will find packed with historical, statistical, and touch-by-touch data and analytics for every match of the World Cup in Brazil and more.

I asked Hamilton about localization of the API for other languages other than English. He responded, “Most people will access the API through what’s called a client or a library of functions that communicate with the API. We’ve already built one client in the Python programming language (here is the link), and will build others shortly. All of our clients are open-source, so if someone wants to create a localized version of the API client they can do so and we would publicize that on our site.”

The Soccermetrics Connect API is available at https://developer.soccermetrics.net, where you can find links to the API documentation and obtain your own API key.

Soccermetrics Founder, CEO – Howard H. Hamilton, Ph.D.

Howard is the founder and CEO of Soccermetrics Research, LLC. He earned a Bachelor’s degree from Georgia Tech here in Atlanta, and both a M.S. and Ph.D. degrees from Stanford University, all in Aerospace Engineering (otherwise known as Aeronautics/Astronautics in Palo Alto). Howard shared that he attended the 2010 MIT Sloan Sports Analytics Conference expecting that sport analytics could explode and left thoroughly convinced the field was mature enough to ramp up his Soccermetrics vision. By 2011, Howard felt the discipline of soccer analytics was truly a game-changer. He grew the Soccermetrics knowledge base through Newsletter and Podcasts, and now has over 5,000 twitter followers @Soccermetrics including us. Howard searched for API management options and found 3Scale.

API scalability is what 3Scale does and they were looking for strategic partners who valued their assets of allowing operators access and control of traffic via API’s. Their infrastructure gives their 430 customers in 13 different Industries, ranging from the Fortune 500 to startups, rapid access to researchable products hosted over the Internet in the clouds.

The Soccermetrics Connect API permits distribution of rich data at all levels of complexity in a football match, whether a club match or one involving national teams

Sports Techie, Within a few hours after the World Cup Final match ends, Soccermetrics will have posted data specific to the game many of you who are involved with sports teams, researching and entertainment, will all find valuable, I know I do.

When I was the Director of Sales for Sports 2000 in 1997/98, Colonel Jim Thomes developed the pioneering GridIron 2000 together with the Super Bowl 48 Champions, Seattle Seahawks. The G.I. 2000 was an expert system that ran on Window 95 and was used for football game planning. We eventually scaled out the model for a US Soccer project together with their Coaching Development Director, Bobby Howe.

Many of the GI2000 sports analytic components of the software have not only become standard mode of operations for the NFL, but I have witnessed a trickle down to MLB, NBA and NFL, and now, futbol, with FIFA National teams, the Premier League and MLS.

Today, however, is a new soccer Big Data dawn because Soccermetrics launched a new soccer API, powered by 3Scale, to create sports tech history as will the winner of the 2014 FIFA World Cup.

I will see ya when I see ya, THE Sports Techie @THESportsTechie – http://twitter.com/THESportsTechie

Sports Techie Social Media Networks

Related Posts:

One comment to Soccermetrics Soccer Data And Analytics API Is The Future Of Global Football, Today

These guys are dreaming if they think you can charge $12k a month for data that can easily be scraped and put into a poisson model readily available open source on the net…

© Sports Betting 2018