Performance or Results?
If a team is failing to get good results, we assume that it’s under-performing. Equally, if a team is exceeding all its targets, we assume it’s in great shape. But is a team’s performance as simple as its results?
Expectations or results?
In football, comparisons are done between the team’s actual results and the results that were expected. For example, the category of expected goals scored is known as xG and the category of expected goals conceded is known as xG Conceded. These expected figures are calculated based on the quality and quantity of the chances to score.
At the end of every game, both teams are allocated expected figures for these and other categories. You can then compare the expected figures with the actual ones. Average figures are calculated over a season. These days, every professional football team is assessed against its expected performance.
The average figures get more and more meaningful as the season progresses, but even then form can quickly change, so averages based on a shorter more recent period might be more accurate, especially if key players have joined or left the club part way through the season.
Predicting results
Sometimes a team significantly exceeds its expected figures or significantly underachieves by comparison to them. That’s because what’s expected doesn’t always happen. We live in a VUCA World after all. When our environment is volatile, uncertain, complex and ambiguous, it’s no surprise that our ability to predict events is limited. And our ability to assess opportunities is equally flawed. We can use hindsight to analyse what’s already happened. However, we don’t have the gift of precognition, to predict the future.
Expected results
When assessing a team’s performance we obviously need to look at its results. We also need to look at its expected results. If a team has an expected goals average of xG 2.1, it means it is expected to score 2.1 goals per game. If it is scoring an average of 1.1 goals a game, it means it should be scoring one more goal every match than it is. It is therefore producing performances in which it is capable of scoring more. That means there may be better results to come if the reasons for that can be addressed. Seeing an improvement to getup to average 2.1 goals per game wouldn’t be a surprise.
If halfway into season, a team suddenly starts scoring 3.1 goals a game, despite having an average xG of 2.1, we know that those good times won’t continue indefinitely, unless there are reasons why the new norm is 3.1 goals per game. If nothing’s changed, the team seems to be overachieving and is likely to score fewer goals going forwards.
Studying just a team’s actual results, without the context of reviewing its expected results, limit the effectiveness of that analysis. We need to show what ‘should have happened’ when we assess what has actually happened. Many industries won’t have a published xG figure to compare their performance with, but there are ways to objectively assess expected performance prior to and after key events. You just need to find the best ones for your team.
Regression
Statistically, even when a team is really consistent, there is an always some random variation. There are always outliers to the norm. Extreme results can’t always be predicted. What can be predicted is that where there’s an extreme result, the next result is likely to be closer to the average. This is called regression to the mean.
If a team has a run where it is outscoring its expected goals, that is likely to come to an end. Its results overall will re-settle back to the average,. This is unless and until the average changes away from that figure, in which case results will still regress to the mean, it’s just that it will be the new average.
Performance
Studying expected results can therefore help to assess how successful a team is. That’s not the only consideration though. Measuring effort, engagement, collaboration, body language and feedback can also help to assess how well a team is performing. With performances having a direct link to results, it’s critically important to measure every contributing element to performance.
The aim is to increase a team’s performance so that its expected goals or equivalents increase. Increasing the opportunities for success will ensure that the actual results will improve.
Studying the reasons for the deviations from the expected figures can tell us which player actions and behaviours caused the team to underachieve. If there were key misses, they can be addressed differently next time. Those are the actions which need to change next time, to improve the results.