![](https://omsstats.wpenginepowered.com/wp-content/uploads/2021/09/Header_Tracking_Article_Web-1000x380.png)
![](https://omsstats.wpenginepowered.com/wp-content/uploads/2019/05/icon-soccer-ball-16x16.png)
As performance analysis processes across elite professional football mature, analysts working for leading clubs are striving to identify tactically relevant insights and trends across multiple datasets, which in many cases are utilised in isolation of one another.
One such dataset has been domestic tracking data, available to all clubs across a league, which captures the positioning and movement of all 22 players on the field. This data, which has been captured for over 20 years, allows analysts to review the actions of every player against set KPIs, from both a performance and tactical perspective. Historically this has only been available in a raw format, resulting in limitations in relation to the identification of longer-term trends.
However in recent months a significant number of metrics, derived from tracking data, have been made available in a merged tracking and event dataset, which for the very first time can be interrogated within a front-end analysis tool, Stats Perform’s ProVision. This is opening up new avenues of performance analysis to provide further objective insights, in a matter of seconds, into strengths and weaknesses of opposition players.
In total, more than forty different filters relating to tracking can be utilised alongside event data, which cover the following three areas:
- Player location data: featuring average distance metrics and off the ball runs.
- Performance metrics: including extensive player distance and sprint data.
- Passing data: including pass velocity, total players bypassed by passes and detailed crossing data.
Using this information it is now possible to build reports, run on a daily or weekly basis, which provide another layer of context to event outputs.
Here we provide five examples of analysis which can now be undertaken using the combined dataset.
1 – Good and Bad Decision-Making From Crosses
One of the key benefits of the combined dataset is that for every ball event that takes place, we have an understanding of the location of every player on the pitch. This is particularly useful for analysing the crossing performance of players in wide positions.
One of the tracking filters in ProVision enables a user to determine how many players are in the opposition’s penalty area when a cross is delivered. This allows an analyst to identify players who regularly deliver crosses when there are very few teammates in the box, which can highlight bad decision-making in attacking areas.
Conversely, using the same principle an analyst can also use this filter to highlight teams who look to pack an opposition box when the ball is in a wide area, looking to get on the end of deliveries played in. This can be analysed on a play-by-play level, at game level and season level, with average numbers providing an insight into a team’s tendencies from open play crosses, compared to competition averages, from both sides of the pitch.
So as well as identifying the number of crosses and their outcome, we now have additional context to help attain what the key reasons are behind these outcomes.
2 – Progressive Passing By Central Defenders
When analysing the passing tendencies of defenders, establishing an accurate picture of the players who are effective at moving the ball out of defence and progressing it up the pitch successfully can influence a team’s approach without the ball.
Filtering from event data, such as the direction, start/end location and outcome of passes, can offer insights into how comfortable a defender is in possession. However by adding tracking, we can take this analysis further.
Using this data, an analyst can apply a filter for how many opposition players are bypassed by a pass, as well as the average number of players who are bypassed per pass. Other filters can also be applied to fit a club’s own definition, including narrowing down the number of passes to only include ground passes and the area where passes are played from (e.g the defensive third).
Using these insights, across multiple matches or against specific types of opposition, analysts can identify opposition players who score highly and compare their performance in this metric to other passing outputs, which can help identify central defenders who are effective at passing through the opposition and using Possession Value (PV), also establish which ones increase the probability of their team scoring through progressive passing. It can also help identify defenders who are not as effective in possession and who may be less comfortable being pressed, which leads in nicely to our next example.
3 – Passing Performance Under Pressure
A key benefit of tracking data is that an analyst can determine how close every player is on the field to the player in possession at any one time. This means that each player’s passing performance can be analysed, taking into account how they perform when they have opposition players in close proximity.
ProVision allows an analyst to add a tracking filter to passing information, where they can see how their numbers fluctuate based on how many opposition players are within two metres of the ball. If a player has at least two defensive players in close proximity, they are likely to be under greater pressure when attempting a pass – and these numbers can be used to compare the contrasting performance of players in a squad, and who is most likely to give the ball away when they are put under more pressure.
4 – Linking Work Rate to Effective Pressing
When working solely with event data, the PPDA metric (passes per defensive action), used in conjunction with high turnovers, has been a useful proxy to identify if a team looks press the opposition. A low PPDA total, across multiple matches, would indicate that a team strives to win the ball back quickly, whilst a high total suggests that a team is more likely to be sitting in a low block and allow the opposition to have the ball.
When we add physical tracking data, it is possible to establish correlations between insights derived from event data and the intensity of each player’s movement on the field. We can do this using two metrics available within the merged dataset: peak player speeds, which is measured in metres covered per second; and the total metres covered by each player at high speed.
As well as providing context as to which players cover the most ground in teams adopting an aggressive pressing approach, these outputs can be applied against various ProVison filters to establish how a player’s work rate and their team’s approach out-of-possession changes against different types of opposition, such as teams who look to play over the top of a press, compared to their season averages.
5 – Use of Personnel at Attacking Corners
During 2020/21, 20% of all non-penalty goals scored in the top five European leagues came from set pieces, a number which reinforces the importance of a team maximising their opportunities from corners and free kicks, as well as being able to defend them successfully.
Goals Scored From Corners: 2020-21 Season
League | Goals |
---|---|
Premier League | 58 |
Ligue 1 | 58 |
Seria A | 52 |
Bundesliga | 43 |
La Liga | 36 |
Whilst advanced metrics, such as Expected Goals, can offer insight into which teams create and concede high quality chances from corners, the availability of player location data adds another dimension to establishing how teams set up at these situations.
Using tracking filters, an analyst can identify the number of scenarios during the season where a team sets-up with a specific number of players in the penalty area, or inside the six-yard box. So for example, a club can identify the number of instances where a defending team has to deal with two attacking players standing in the six-yard area as a kick is being taken – and crucially how many chances they concede from these scenarios.
This can help identify potential areas of weakness when a team is defending set pieces, across multiple matches, which a team can use to inform their tactics in the lead-up to a game.
These are just five examples of how a merged domestic dataset, combining outputs from tracking and event data, can help with the identification of players who either excel, or potentially have an area of weakness, during recurring in-game scenarios.
All of the tracking filters in ProVision can also be added to an analyst’s own reports and customised stats, which opens up a wide range of different possibilities for analysis across different contexts.
To learn more about how ProVision can provide deeper insights into long-term performance trends, supporting your club’s pre and post-match processes, visit our Match Analysis page or contact us to find out if tracking data, from your domestic league, has been integrated into the platform.