Last month’s piece on the value of chance creation over the course of a season showed that there is some logic behind Liverpool’s decision to respectively sign and target Jordan Henderson and Stewart Downing.
Does the rationale extend to Charlie Adam, the Reds’ latest purchase? Although creating a healthy 64 chances in the league last season (Downing 85, Henderson 82), 45% of these came from set pieces (Downing 31%, Henderson 40%).
The original analysis focused on total chances created; that is those from both open and set plays. Unsurprisingly, the majority of chances created in the division last season were from open play.
The striking non-relationship between league position and total set play chances created is validated by a near-zero correlation when set play chances are instead plotted against total points collected in the season.
Modelling the relationship between set plays and points collected – whilst holding constant attacking ability amongst possession and defence-based variables – confirms that creating more chances through set pieces adds absolutely nothing to your points total over the course of the season.
Meanwhile, every additional chance created from open play significantly contributes to total points – as described in last month’s post.
The ineffectiveness of set pieces has been written about in detail by a number of bloggers, and these two pieces are certainly worth a read:
- Soccer by the Numbers: Why Wigan Were the Most Unusual Team of the Season: Goal Creation in the Premier League in 2010/11
- Bass Turned to Red: The Effectiveness (or not) of Corners
Teams converted on average 7.5% of all set piece chances in the league last season, compared to 9.9% of open play chances.
The numbers all suggest that signing Charlie Adam primarily for his set piece ability is misguided, as it adds limited value.
Yet everyone has seen Adam’s ability striking a dead ball; Tangerine Dreaming points out the specifics of his talent:
He is excellent at delivering set pieces. Wide free kicks are better delivered from wide on the right hand side and generally hits them just above head height swinging inwards. His free kick delivery from wide left have a tendency to be hit low towards feet and behind the defensive line, swinging away from goal. He generally takes the majority of his corners from the right side, in-swinging, although has a tendency to over hit the ball. His striking of the corner can be inconsistent with a scuffed low and running corner being the key fault. His goal against West Ham was scored in this fashion, but it wasn’t deliberate as his celebration would confirm.
The numbers I’ve analysed do not include goals from direct free kicks or penalties, situations that Adam equally enjoys. It may be that Adam creates ‘better’ chances from dead ball situations, though no such measure of chance quality exists.
Soccer by the Numbers also highlights Blackpool’s remarkable efficiency from corners last season, and whilst this cannot be entirely attributed to Adam, there’s no question he played an important role in this source of goal creation for the Seasiders.
Sir Alex Ferguson meanwhile felt that Adam’s corners were worth “£10 million”, though as the original recording of the interview shows, the comment was perhaps more light-hearted than it was later made out to be.
Charlie Adam may well create an abundance of chances from set pieces for Liverpool next season, but the numbers suggest this will do the side little good. That said, he has the ability with a dead ball to affect that great intangible: momentum. An inch-perfect corner onto the head of Andy Carroll in the 90th minute of an important game is not an inconceivable scenario, and it is those types of moments the numbers do not account for.
When it comes to set pieces, the question is do you sign a player for those moments, or what he may add on average over a whole season?
This post, along with other articles I’ve written, is available on 5 Added Minutes. You can also follow me @OmarChaudhuri.
[…] Data courtesy of EPLIndex. Post also available on anfieldindex.com. […]
FKs are a essential attribute, I think most people are overlooking other Adam’s qualities as his number of involviments per minute in a match and the kind of pass that he uses. He is a type of player that we don’t have, considering that Gerrard cannot play as a second QB (or CM in front of a CDM in a 4-2-3-1) anymore because of his age and decreasing physical conditions.
Adam maybe is not a WC player, but probably will help the team to be a WC team.
What an excellent comment. I totally agree with this. I believe Adam may not be a Xabi Alonso but may just add exactly what he did. It’s also what we have been missing. We will be very dangerous from set pieces now that we have Charlie Adam. Andy Carroll must be very pleased with this signing.
I agree. I want to make clear I’m not criticising the signing of Adam, merely pointing out the lack of value in set piece chance creation; this is only one facet of his game. For the purpose of this post I’m not terribly interested in what he contributes elsewhere (though obviously I’m still open to a discussion on his ability!).
[…] View the full story here: Anfield Index […]
Set pieces vs points is a fairly useless statistic, though so is chance creation vs points. Have a look instead at chances vs goals and you’ll see a far more useful correlation. Remember that the reason some teams don’t get higher placed in the league is down to conceded goals, and not just goals scored.
If you are looking at a players attacking statistics, you need to ignore the goals against or you are clouding the output somewhat. Having a lethal set piece delivery can have no (or very little) impact on the amount of goals a team may concede.
“Set pieces vs points is a fairly useless statistic, though so is chance creation vs points. Have a look instead at chances vs goals and you’ll see a far more useful correlation.”
I wouldn’t call it useless at all – if a team knows how many more chances it needs to create to gain x amount of points next season, it can sign players appropriately, as Liverpool have done. Chances vs goals would tell you something entirely different – useful perhaps, but entirely different – and not something I’m concerning myself with this post.
“Remember that the reason some teams don’t get higher placed in the league is down to conceded goals, and not just goals scored.”
I’m of course aware that chances created isn’t the only factor affecting points totals, which is why, as I stated in the post, I’ve allowed for defensive variables (i.e. goals conceded) in the model. This means I get a true, unbiased measure on the effect of chances created on points over the course of a season.
“If you are looking at a players attacking statistics, you need to ignore the goals against or you are clouding the output somewhat. Having a lethal set piece delivery can have no (or very little) impact on the amount of goals a team may concede.”
That’s totally irrelevant to the post. I’m not looking at the effect of chances created on how many goals you concede (not sure why anyone would), nor have I alluded to that relationship in any way. The only player’s attacking statistic I look at is chances created, I haven’t looked at anything else.
If you could clarify some of your criticisms I’d really appreciate it.
To make clear, I’m asking: what is the effect of (set piece) chances created on total points over the course of the season? For the purposes of this post, I’m really not interested in anything else, though of course I account for other factors that affect points in the model described.