The Hidden Source

For those that remember the run up to the 1994 World Cup, there were various attempts at scaremongering in the British press. Suggesting that because it was being held in the United States, commercial pressures would dictate that the game be divided in to quarters along with several other silly notions which would shake the foundations of the game. For various reasons the majority of the scaremongering failed to come to fruition, however, that World Cup did see something emerge that wasn’t really in the British consciousness at the time. That something was the assist.

Assists have always been in the game, it’s an essential part of scoring a goal, however, around the mid-nineties they started to be classified and assigned to players and ever since a fascination has grown up around them. They are now a popular benchmark for a player’s attacking potency. Let’s be clear though, this didn’t become important then, it was already important, we just didn’t have the data and we probably called them ‘set ups’ or something akin to that. Today you can peruse all kinds of data about the game, but for many, the rise of football data as we know it began to form in the public consciousness with the addition of the assist in the mid-nineties.

Where is this all leading to? Well, this article will explore a little more about the assist, but more importantly explore what we don’t see. We know players assist goals, but not every goal is actually assigned an assist. It’s important to know who is assisting a goal for a team as you can potentially target the most creative players in a team that way.

Another way to view an assist is to call it a goal source of which it’s one of many. The reason for this is that it’s important to acknowledge that goals do come from other sources. It is this concept of ‘goal sources’ which will be explored here. Why explore it? Well, it’s tapping in to the unknown and it may or may not allow us to understand more about the game and the teams which participate in the English Premier League.

The data used here has come directly from Opta and on the 20th January after all clubs had completed twenty one games in the English Premier League. At this stage, the time of the season hasn’t been factored in; however, it’s acknowledged that during a full season these figures may level out or change in some other unspecified way. There may also be other ways to explain this which may render this article inert.

Up to date there have been 588 goals scored and 401 have been credited with an assist to an attacking player. This means that there are 187 goals scored where an attacking player hasn’t been credited with an assist which equates to 32% of the goals scored in the Premier League this season.

This appears to be a fairly significant chunk of the goals scored and leaves a void in the sources of goals when trying to understand where they come from. We know that David Silva has made 10 assists for Manchester City, but something or someone else is also creating goals without it being entirely visible. Before further allowances are made with the data, a swift table has been put together below so that you can look at the teams, their goals, number of assists and percentage of goals credited with an assist. They are ranked according to total number of assists.

What you see immediately is what you’d expect to see. Those teams with the most assists are the ‘better’ teams such as Manchester City and Manchester United, however, venture further down and you start seeing that teams are jumble up a little and the likes of Blackburn jump up the table due to the number of goals they’ve scored. The table below is instead ranked on the percentage of goals that are credited with an assist. Now look what happens.

Here you can see that Chelsea and Arsenal have pushed the Manchester clubs off the top and more interestingly Stoke City jump up the table. This ranking is determined as that which shows the percentage of goals which that team has assisted. The higher the figure the less scope there is for other factors to be the source of their goals, whilst the teams at the bottom of the table have greater shares of their goals coming from sources other than directly from their own players. However, before marching headlong in to a sticky mess it might be time to just check a few things, namely the definition of an assist, as the perceptive reader will have already ‘cottoned on’ to some details that are missing.

Opta have kindly provided data for this article and are also responsible for the dataset over at so it’s important to understand how they define an assist as this could potentially shape the outcomes of this article.

“The final pass or pass-cum-shot leading to the recipient of the ball scoring a goal.”

It can be assumed since we don’t see otherwise that only the attacking team are ever credited with an assist. It could also be assumed that penalties wouldn’t accrue an assist credit for an attacking player. There are other reasons that may lead to the attacking team not being credited and here’s a little run down of why that might be.

  • The team may have scored direct from a free kick
  • A rebound
  • A misplaced pass from an opponent
  • A goal created by the scorer.

These may not be exhaustive, so it would be dangerous to make too many sweeping judgements without understanding them all fully. What can be said is that the figures in the tables above should be levelled out a little to take in to account those goals that couldn’t have been assisted such as own goals, penalties and goals direct from free kicks. The table below does exactly that and then ranks the team based on their assist (net assist) percentage.

The first thing to note is that the number of goals has been reduced to a net figure taking in to account only those goals where an assist could be assigned to an attacking player. In total there are 504 net goals, once own goals, penalties and goals from direct free kicks have been removed. This leaves a shortfall of 103 goals that have not been credited and therefore come from other sources.

When looking at the ranking above Arsenal retain their position as the club with the highest percentage of goals coming from their own players, but surprisingly Stoke City jump up to second place who like Arsenal have only have one goal scored where the source cannot be accounted for. At the bottom of the table reside Wigan who’ve assisted only half of their goals this season and ‘other sources’ have assisted the other half of their goals scored. This leads on to the final table which lists the number of goals each team has scored from ‘other sources’, called unassisted goals.

What you can see now is that Wolves have had the most goals from other sources, followed by Manchester City. However, the latter have a high assisted goal percentage due to their league high of goals scored. Then there’s a batch of teams with seven goals including Spurs and Norwich City. On average each team have scored five goals this season without the source being credited to their player. In total, 18% of goals scored in the English Premier League have come from other unspecified sources.

In terms of the game in general, this may have some significance or it may have none at all. There may well be a major oversight here and these goals from other sources are easily identified and explained. However, it appears that they remain unseen, people focus on the headline, the David Silva assist count and marvel at his creative abilities. The same for other players who’ve assisted goals, however, in truth the attacking team may have scored the goal but haven’t been responsible for the creation of the goal. Potentially, this may have implications when viewing a team and the way that they score their goals.

The upshot of this is that if a team isn’t responsible for their goals, then who or what is?

Could this be an indication of a team’s luck i.e. rebounding shots in to the path of a player who scores, a misplaced pass from the opposition. A team may in effect force these situations and look to seize upon opposition errors or it may be a case of their opposition having poor performances contributing to their own downfall.

What it perhaps suggests is that those teams who have the most assists as a percentage of their goals may be those teams that control their own outcomes better than others? It’s interesting to see that the two teams at the top of the assist percentage ranking are Arsenal and Stoke. These two are often cited as polar opposites in terms of playing style; however, they may be closer in their approaches than first appears. Arsenal have the accusation (unfairly you may say) that they try to walk the ball in to the net, whilst Stoke are accused (unfairly you may say) of being route one long ball merchants. However, these are their ways of giving them the control over their scoring opportunities, Arsenal want to pass the ball till the optimum moment of scoring a goal presents itself, whilst Stoke are doing just the same. Each approach is configured to their players, their style.

At the other end of the table is Wigan Athletic and applying the same logic to them as with the teams at the top of the assist percentage table then they are the team with less control over their destiny. Only 50% of their goals scored come from their own players, perhaps they press the opposition high up the pitch and steal the ball before scoring, perhaps they pounce on more bobbles off legs of opposing defenders. Whatever the source of their goals, they are some way away from being self sufficient. Should they intercept before goals then there’s clearly a strategic reason for this, but should the ‘rub of the green’ be putting the ball to their goal scorers then they are riding on the crest of a wave that will crash. When it crashes they will need to start creating more goals or the goals will dry up.

All of the above may also apply to those other teams at the bottom such as Wolves. What is surprising is that a team such as Blackburn who are fourth from bottom of the league table rank tenth in assisted goal percentage. Their defence may be holding them back in winning games perhaps, but perhaps they are more in control of what they are doing when trying to score. Perhaps this may be an indicator that things may turn around for them sooner or later or perhaps it is not.

Hopefully this has been enlightening and allows a better awareness of goal sources. However, before rounding up this article here’s a small point to note. Should a team be in receipt of goals given to them by the opposition then perhaps it might be interesting to see which players are responsible for those errors and for those to become public knowledge. Everyone who watches football will be aware of players committing gaffes but how many do certain players commit each season and how do they compare as a goal source to the top assisting players?

In conclusion, the only way to round this up is to just state that it’s important to try and understand the unseen elements of football, yes assists are a good indicator of where goals come from, but they aren’t the be all and end all. Perhaps there is some truth in clubs being in control of their destiny and this gives and insight in to how to understand that, perhaps not. Perhaps there’s a perfectly reasonable explanation for all of this. Certainly you could note down every instance that created a goal and share that. Perhaps that is what needs to be done; it would certainly make for interesting reading. Should you have your own thoughts on this subject feel free to comment below and if you’ve made it this far, thanks for reading.

Special thanks must go to Opta for providing the critical pieces of data which plugged some gaps and to who have made tons of statistics available to browse on their excellent website.

This post was written by Tangerine Dreaming. You can follow him on Twitter via @Tanger_Dreaming or visit his website

3 thoughts on “The Hidden Source

  1. I am curious if what you are measuring here correlates closely to chance conversion efficiency. Stoke and Arsenal aim to create highly efficient situations and it is unsurprising that they have the least amount of unassisted goals.

    Have you looked at unassisted goals against? That could be interesting.

  2. Pingback: Messi vs. Ronaldo: a goal-scoring comparison « 5 Added Minutes

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s