Goals in soccer can be greatly influenced by luck or random moments such as near misses, deflected shots, errors, or referees. Expected goals measures the probability that a shot will result in a goal. Factors include angle and distance to goal, current score, header or not, during a counter attack, etc. This statistic can be used to more accurately determine the performance of a team. If the score was 0-0 but the keeper's had 20 saves each, that score could have been 3-3 or more on another day.

Expected Goals

High-level game-by-game statistics that aren't directly correlated to player but possibly team. Such as line-ups, attendance of stadium, goals, cards, substitutions, etc.

Game basic statistics

Data that is dethatched from players and it more indicative of a managers performance over seasons or a teams history.

League standings and results data

Data on a players athleticism such 40-yard dash time, vertical, etc. Also, their goals scored, assists, salary, etc. All this can be used to make player driven models to determine the performance of a team as a whole.

Player details & characteristics data

Gathering this data is where a lot of novel computer vision techniques come into play. Determining frame-by-frame basis of player from both teams, locations, ball locations, action details, tackles, passes, direction, speed is part of this effort.

Game event and tracking data

Its common for good teams to control the ball for an entire game but never score a goal. Possession models look to further quantify what kinds of possessions are actually valuable. A simple version is location-based possession value which tells you the probability of scoring based on where the ball is at that point in time. As you get closer to goal that probability increases.

https://fivethirtyeight.com/features/possession-is-the-puzzle-of-soccer-analytics-these-models-are-trying-to-solve-it/

Soccer possession models

Big data has become more and more prevalent in the sports world but for some sports like soccer there isn't as much information in the data. It's difficult to quantify the skill of a team or player outside of just goals and assists so creative metrics and techniques are used to measure their performance.

University of Michigan - Ann Arbor

The sports industry generates around 90 billion dollars in revenue each year. With a rapidly growing industry like sports, the need for analytics and data science is essential. What is the likelihood that team Lakers win the championship? How valuable are Nikola Jokic and Jamaal Murray to the Nuggets? Will Kevin Durant be as successful next year after tearing his ACL and skipping a year from the league? These are all questions GM's and Owners ask as some of these decisions can make or break their franchise. Looking back at decades of sports data to try to predict the future is crucial for this industry to continually boom in the coming decades.

Data Science In Sports

There is clear evidence that the NBA is taking advantage of Data Science wide range of uses. Such as designing winning strategies, predicting and avoiding player injuries, and scouting

Data analytics in the NBA

For example, players with the NBA’s Minnesota Timberwolves wear Fitbit health tracking watches during practice to collect data on heart rate, sleep and movement. The team analyzes this data to better understand when players are overexerting themselves and to create strategies to avoid exhaustion during games.

With this information, players may identify weak spots in their form so they can be more cautious when playing. Sport medicine employees may also analyze how they treated injuries to better understand the success rate. They may change treatment plans for certain injuries or players to try to speed healing.

Data Science working for Athlete Safety

Some of the most important data that teams analyze before a game include:

Opposing team player statistics, such as common plays or configurations and types of scoring.
Recentwins and losses and how individual player performance contributed to these games.
Game-day weather conditions and players’ experiences in these conditions.
Game statistics, including how many games they must win to make it to the playoffs or surpass previous records.
Professional sports teams work hard to gather relevant data to prepare for games.

Data science used to win

Fan Analysis
Sports is a business and the more engaged the fans are, the more profit organizations experience. By learning about data analytics in the online world, sports management teams can discover how and when fans are likely to attend events or buy merchandise.

Management analyzes social media patterns, attendance and merchandise sales to better understand what consumers want out of the game. This allows them to identify what’s important to fans, such as an engaging mascot or unique merchandise. Management can then be sure to provide these amenities to keep fans satisfied and coming back.

With fans in mind, management can develop marketing strategies and advertising campaigns that target these consumers. Data helps them to easily identify fans that are likely to engage with the team so they don’t spend advertising dollars on consumers who aren’t interested in their sport.

Data Science used to improve sales in sport

Talent scouts typically observed potential recruits practicing or playing competitively. However, data visualizations and analytics that include statistics on past performance are also a vital part of the scouting process. Professional scouts can’t always physically visit every promising collegiate player, so they rely on statistics to identify and prioritize who to visit and observe.

Data science for scouting in sport

	
Sports Analytics and Data Science: Winning the Game with Methods and Models
(https://play.google.com/store/books/details?id=7LXxCgAAQBAJ&gl=us&hl=en-US&source=productsearch&utm_source=HA_Desktop_US&utm_medium=SEM&utm_campaign=PLA&pcampaignid=MKT-FDR-na-us-1000189-Med-pla-bk-Evergreen-Jul1520-PLA-eBooks_Computers&gclid=Cj0KCQiAx9mABhD0ARIsAEfpavTZ1LqMrGoxA8pDI6MWeFmWyIOEOVmSUnALSBXWHXN_aZEmB683MZAaAj-NEALw_wcB&gclsrc=aw.ds)

Sports Analytics and Data Science: Winning the Game with Methods and Models

One of people's favorites things to do in the sports world is bet on the outcome of a game or players performance. In order for the sportsbook (entity offering the bets) to determine what the odds will be, they must do sophisticated data analytics that rely on data science techniques. On the contrary, nowadays more and more people are able to create their own models to predict the outcomes of games.

Sports betting

Soccer Analytics

Football is an interesting sport because there's numerous coaches and a lot of stoppages allowing analytics to play an important part in every facet of the game. Fantasy football has popularized the advanced metrics side of the NFL but there's useful analytics that expand beyond just player focused models. Much like soccer, there's an element of randomness that has to be factored in because of how sporadic and important one score can be so creating accurate models is a difficult process.

American Football Analytics

Statistics have always been apart of sports but for baseball they've become the most accurate and useful. Many modern metrics don't give a good understanding of what's actually happening on the field, just the broader result.

Baseball Analytics

Much like the others, hockey analytics has shifted over the years from caring solely about a teams scoring to a player's overall contribution.  It can help us see players that stand out and are doing more good than bad for their club and vice versa.

Learn Before

Related

Learn After