Sounder At Heart: An SB Nation Community

Navigation: Jump to content areas:


Pro Quality. Fan Perspective.
Login-facebook
New Blog: The Nova Blog for Villanova Fans!

On Formations and Fullbacks

Or: Why James Riley Was Left Unprotected

You'll probably need some understanding of the formation aspect of soccer strategy before reading. Wikipedia is a good resource, and if you've watched many Sounders games you'll have picked up quite a lot of information as well. Sorry in advance for the English spelling.

Four at the back, one or two at the front. While there exist a huge variety of midfield combinations a Major League Soccer team can run with, the formations are typically bracketed by these five or six players. I imagine at this point several of you are already readying the comments that say I've completely ignored 3-5-2, which indeed I have. Before getting to the heart of the post, let me explain why.

The Dangers of 3-5-2

3-5-2 is a formation which takes the two wide defenders in a four-man back line (the full-backs) and pushes them into midfield, which reinforcing the defence with an additional central defender (centre-back). This has the effect of providing an additional defender to cover the two central strikers in the standard 4-4-2 whilst simultaneously overloading the midfield, allowing advantages in two key areas of the field. The formation came about as a direct response to 4-4-2 pulling the old-style wingers down the field and slightly inside. Why ignore a system designed to combat the most common formation on the planet, then? The answer is simple. Modern teams are perfectly capable of making changes to their shape, and within the 3-5-2 lie the seeds of its own demise. Withdrawing one forward from 4-4-2 into the central midfield turns the formation into a 4-5-1, leaving one of 3-5-2's central defenders standing around essentially doing nothing, almost turning the game into 11 vs. 10. So while we accept the possibility of a 3-5-2 appearing, its shortcomings mean that we don't expect it to encounter it too often. This is a formation whose heyday seems to have already passed, although we may see a form of it arise with one centre-back and two wing-backs, which would be functionally completely different (and defensively suicidal, one would only see this against teams playing something like 4-6-0, which nobody does. Yet).

Why Not Three Up Front?

4-3-3 is a formation you'll also see talked about which seems to turn my first sentence on its head (perhaps I should have written a more accurate first sentence!), but in most cases the attackers are staggered in a way which means that the two side players of the three in the front are only slightly advanced from the outside midfielders in 4-4-2, with a centre forward withdrawn in favour of a play-making central midfielder. It's rare that the outside forwards lie so far upfield that they could be considered bona-fide strikers, and there are no three players in MLS capable of providing the service that such a formation requires. Spain can pull it off fairly effectively against weaker teams, but they can field Fabregas, Iniesta, and Xavi, which might be the best set of midfielders who've ever lived, and it's still not that useful against their elite opponents.  So let's also neglect the possibility of seeing three 'real' strikers running rampant.

I hope at this point I've convinced you of the merits of my first sentence. If not, feel free to ignore the rest of the post!

Star-divide

The Full-back/Centre-back Divide

Defenders are not all cut from the same cloth, mainly because there are normally only one or two true forwards to deal with. Centre-backs are typically tall and strong in order to deal with the aerial threat posed by opposing centre-forwards. Extremely good pace is a plus but not an explicit requirement, and good distribution is also overlooked in favour of hoofing the ball out of dangerous areas of the field as hard and fast as humanly possible. So although a centre-back has to be a good player in his own right, they tend to be stronger than fast, good in the air, and uncompromising in tackles and clearances.

Full-backs (when they are deployed a little further upfield they become wing-backs, but for our purposes the two designations are interchangeable) are a different story. Their main opponents are not the central strikers of the world. Rather than forwards, they deal with the outside midfielders, whose main goal in life tends to be cutting inside and shooting or cutting outside and crossing to the strikers. These players - I'm going to refer to them as wingers, but this isn't technically true as wingers are the outside forwards in 4-2-4 - need to be fast, clever, and have good deliveries. Defending against them requires different skills. Aerial prowess isn't as critical as it is in the centre-backs, although it's useful for defending crosses coming in from the opposite flank. Decent speed suddenly becomes vital as wingers are very often the fastest players on the pitch. Clearing the ball may not be as important as sensibly distributing it to one's midfielders.

Most of all, though, the difference between the outside and inside defenders lies in their attacking responsibilities.

Attacking From Defence

Centre-backs are utlised as attacking options fairly frequently, but this generally is due to their heading ability being useful in set-pieces such as corners. Other than very specific situations, we do not expect to see the central defenders in the opposing half. Full-backs, on the other hand, have become very attacking in recent years. Why?

Imagine 4-4-2 vs. 4-4-2. Who are the only players with space directly in front of them, with no immediate opponent? The full-backs. If you are having trouble, the picture below might help.

Figure 1: 4-4-2 vs. 4-4-2. The outside defenders are the only players with unoccupied space immediately in front of them.

In the past fifteen years, the full-back position has possibly undergone the most drastic tactical shift in the sport. Full-back is no longer a static position, waiting for wingers to make their move and counteracting their attacks. Modern full-backs are expected to fully contribute to the attack, providing overlapping runs when a winger cuts inside and opens space at the touchline and crossing when in advanced positions. While it's rare to see these players with a knack for goal, they are so common in the immediate build up to goals that it's totally unremarkable that they record assists. Many have said that the wing-back is the most important position in attack, and I'm not particularly inclined to disagree with them. Having a wing-back who cannot go forward is the equivalent of shooting yourself in the foot.

But James Riley is a good attacking player, and I promised that I'd try to shed some light on why he was left unprotected. So what gives?

On One Dimensional Full-Backs

Although I question that Riley really deserves 'good' attached to his attacking performances, that's a subject that will have to be revisited once I get my teeth into more advanced metrics. My belief (unvalidated as of yet) is that he gives the ball away far too often in midfield, which leads to dangerous fast counterattacks by the opposing team, but many intelligent observers think he's more than adequate as an offensive threat, so I'll leave it at that for the time being and instead focus on defensive frailties, which is something more or less everyone has noticed about his game.

Simply put, James Riley is an attacking player playing defence. He's more or less totally incapable of shutting down wing play on his side, and has his central defenders and goalkeeper to thank for covering up his mistakes. Riley does not close players down - not because he doesn't know that pressuring the opposition is a good idea, but because he's not really very good at tackling, and letting a winger run straight by you is probably a worse plan than keeping them in front of you. I don't think anyone will argue that Riley's strength lies in going forward, while his defensive play is a weakness, and by leaving him off the protected list, the Sounders seem to agree with me.

In the previous sections, I have described why a full-back who cannot attack is a liability in soccer, which seems to explain the absence of Zach Scott far better than it answers any questions about Riley. The key here lies in how teams counter good attacking full-backs, which is something Fabio Capello demonstrated in Zagreb last year. Jonathan Wilson, whose wonderful book Inverting the Pyramid is the source of much of my knowledge on formations and tactics, explains it far better than I could:

England's 4-1 victory in Croatia... can be seen as a tale of winger and full-back. [Left-back] Danijel Pranjic had looked excellent in the [2008 European Championship], his overlapping runs allowing Ivan Rakitic to drift infield off the left flank on to his favoured right foot. Faced with the pace of Theo Walcott, though, he never had the confidence to abandon his man and surge forward, which had the effect both of stymieing Croatia as an attacking force, and of exposing his own defensive shortcomings. That Walcott scored a hat-trick underlined the point, but came almost a bonus alongside his primary role of disrupting Croatia's left flank. [Article]

As an attacking threat, Riley is completely neutered by a good left winger. He then becomes a pure defender, forced backwards by both the skill of his opposition and his own defensive ineptitude. Without the requisite defensive talents, he is essentially unable to break out of his own territory, because he's forced to sit much tighter on his direct opponent. And not even that works very well, due to his habit of dropping away whenever they get the ball. So while a non-attacking full-back is a liability against any team, a good team can push Riley completely out of the game simply by calling his defensive game into question more often.

One imagines the Sounders know this, and that they are prepared to lose him to the Union in the draft. A more well-rounded player at right-back would improve the team (and I'm not touching the disciplinary issues here either). At the end of the day, James Riley is a shiny piece who can be a critical part of the Sounder attack on good days. But on bad days, his liabilities collapse the team's attacking outlet from the right side. And playoff games? They're probably going to be bad days for Riley. I wouldn't be too concerned if he ends up in Philadelphia.

FanPosts only represent the opinions of the poster, not of Sounder at Heart.

4 recs  |  Comment 61 comments |

Story-email Email Printer Print

Comments

Display:

Riley - It works both ways

Great post Graham. I am not quite so down on Riley as it seems you are here though I would tend to agree that he is not a shut-down fullback, to borrow a metaphor from throwball.

Offensively, Riley was more integral to the right-sided attack than you allow. Going simply on my memory, Riley was primarily responsible for service down the right side. Whether we were deploying Ljungberg, Montero, Jaqua or a rotation at RM, all of those three would tend to tuck inside and only pop out to the wing on a diagonal run. Riley was our only consistent wide threat and his service of the ball from the wing was the best on the team. I don’t have the metrics to look at his number of crosses nor the percentage that found targets but his topline stat of 4 assists on the season is equal to that of Zakuani and better than any other midfielder besides Ljungberg.

The one thing I will also say on the tactics side is that while a good wing can expose an offensive wing-back’s liabilities, it works both ways. In the instance you cite in the Croatia game, I saw it as a tactical mistake for Croatia to hold Pranjic into a defensive posture to counteract Walcott. As you point out, Pranjic was never going to be able to hold up to the pressure applied by Walcott, who played almost like a third forward than a true wide mid. Capello put in Brown at RB and he pretty much stayed there the whole game. If Croatia released Pranjic down that left side, it could have put a lot more pressure on Walcott to track back to help and forced him into a more defensive role. Now maybe Croatia was simply out-gunned and could not have held up in midfield enough to retain possession to get the ball into offensive space for Pranjic (and the left-sided mid I can’t remember), but by letting Walcott play offense was a mistake.

For our situation, putting Riley out there means that opposing teams can’t let their LM adopt a purely offensive posture as it means that they will often be exposed to a 2v1 on the right side with Riley combining with Montero, Jaqua, or Ljundberg. It happened frequently though and we put in at least 4 goals because of it.

by brokejumper on Nov 25, 2009 8:02 AM PST reply actions   0 recs

You're right in that attacking the wide midfield can work for a full-back as well

But generally you need to have the defensive ability to cover when things go completely pear shaped, and I don’t think Riley has that ability.

He might actually be better off playing right wing than anything else.

by Graham on Nov 25, 2009 8:10 AM PST up reply actions   0 recs

Another note:

I agree that Riley was integral to our right wing play, and he’s certainly a good crosser, but his buildup play frustrates me. But again, I won’t argue with you if you say he’s a good attacking player, as it simply becomes one person’s belief vs. another with not a whole lot of evidence to tip the balances. I tried to assume that he was indeed a good attacking player when talking about why good attacker+shoddy defender is not really a good combination at his position.

by Graham on Nov 25, 2009 8:14 AM PST up reply actions   0 recs

I think he is better defensively than you give him credit for

But, I also think that his skillset is replaceable. I cannot argue with any of the players the Sounders protected and likely would have protected basically the same players as well. The Sounders have talent and when you have talent and can protect only eleven players, someone is going to get left out.

For as offensive as the Sounders played (and I think that perhaps the Sounders tried to play offensive soccer more than any other team), they were along with Houston the best defensive team in MLS.

You are not that good defensively by the GK and CB covering for him.

But, as I said at the top, he is replaceable.

by Coug1990 on Nov 25, 2009 10:07 AM PST up reply actions   0 recs

I have to disagree with the idea that the Riley is good at defencesimply because the Sounders didn't give up many goals

It completely fails to isolate him from his teammates, it doesn’t allow for missed chances by the opposition, and it’s essentially results-based analysis, which is a good way to measure actual talent incorrectly.

Until we have better statistics, I have to go with my eyes.

by Graham on Nov 25, 2009 10:11 AM PST up reply actions   0 recs

In any sport team sport, it doesn't matter which one

any player can be isolated and be taken advantage of. If Riley was so bad, smarter coaches than the both of us would have taken advantage of him. Statistics can hide some things, but they don’t lie.

Well, I watched every game as well and I never saw him as a problem and eyes do lie.

by Coug1990 on Nov 25, 2009 10:19 AM PST up reply actions   0 recs

Inadequete statistics provide in adequete truths

And for MLS, the publically available stats are a best a partial image of individual talent. Team stats at least correlate at about .80 which is decent but not good. (Graham I will show you/world what I mean – IIRC I used Shots on Goals For/Against to PPM)

But for individual judgement I think currently the best stat is +/-, but in Riley’s case he spent so much time on the pitch the image is colored more by team performance, especially since the 1 stinker the Sounders played was without Riley for much of the game.

I am not a Supporter
I am not a Fan
I am a Sounder
Sounder At Heart

by Sounder At Heart on Nov 25, 2009 10:27 AM PST up reply actions   0 recs

Look, I am not arguing that he is an allstar by any means

But from reading Graham’s article, he makes it sound like Riley is awful. He is not awful. There are things in his skillset that adds value to the Sounder offensively and defensively. If that were not the case, it would have been Riley that was replaced from the back line earlier in the year, not Scott.

Yes, I agree with you that inadequate statistics do provide inadequate truths. But, so far neither you, nor Graham, has shown any of us any statistic that backs up his premise.

No matter how basic my statistic was, I at least used one to back up an argument.

Now, I love statistics to make arguments and it is what attracted me to your site way back when. It is also why I love baseball and think the “Fielding Bible” award that voters strongly use statistics to make a decision versus the “Gold Glove” award that voters only use their eyes is a highly superior award.

I also love the +/- as I was very familiar with that in regards to basketball. So, in those terms, I don’t understand your argument in how it relates to Riley.

There is a reason that Riley got more minutes than Brown or Scott last year; Sigi thought he could help the team win more than those two. There is a reason that Riley played the second most minutes on the team; Sigi thought he was worth being on the field.

Do you discount Jacqua’s accomplishment’s or Montero’s just because they spent so much time on the pitch? Does their accomplishments diminish by team performance? You can make that about every single player. It is a circular argument.

Yes, welcome Graham, I have enjoyed reading you on LL.

There is a reason that it is better to look at a season’s worth of statistics than any one game.

by Coug1990 on Nov 25, 2009 11:24 AM PST up reply actions   0 recs

In baseball, statistics have progressed to the point where I am comfortable using them in most circumstances

(Although I’m still a scouts/stats guy rather than pure stats).

In soccer, things are much more immature. But since I don’t have any concrete evidence to back my opinion up, you’re more than welcome to disagree with me. I just probably won’t accept the very basic stats we have at our disposal now as real evidence one way or another.

by Graham on Nov 25, 2009 11:30 AM PST up reply actions   0 recs

That is a fair point and one that I agree with entirely

I also think that baseball statistics will continue to evolve for the better, which in itself is exciting.

I believe that this country is so statistically sports oriented more than possibly the rest of the world, that I think the US might possible/eventually lead the world in statistical analysis in regards to soccer and that could help improve American soccer teams in the long run.

by Coug1990 on Nov 25, 2009 11:39 AM PST up reply actions   0 recs

looking forward to better stats

This is a pretty subjective argument because of our lack of stats. I am more in the Coug1990 camp than the Graham camp. It’s not that Riley is great at defense — but I think he is better than what is implied here.

The trouble is that we are all just “thinking” at this point.

I too think the US Soccer community will drive the development of this work. I’m sure you are aware of Soccernomics — which starts pushing the envelope on this conversation. We also have the dedicated work of Dave locally and Nate Silver nationally. Plus, I know that Billy Bean has taken an interest in this from a front office perspective. All good signs.

by zeagle on Nov 25, 2009 12:48 PM PST up reply actions   0 recs

Statistics will never become definitive in soccer

Whereas baseball is the most statistic friendly sport possible, with dozens of discrete events that can be counted and tabulated, soccer is just too fluid to break down into statistics beyond a certain level. An example was the goal we suffered against KC at Qwest. It’s a goal against, so it counts against the defensive stats, but the player who was actually closest to Arnaud and who could have potentially stopped him from shooting was Nate Jaqua. Other than Dragavon, no member of the defense was involved in the play that gave up the goal.

On the other hand, how do you quantify pressure where the player doesn’t take, but rushes a pass that turns over the ball? The opposing player takes a statistical hit for the bad pass, but what statistic does that go into for the defender? Bad passes forced? What if it doesn’t force a bad pass, but forces the opponent to pass back and the other team to try another avenue of attack? Back passes forced?

by CarlosT on Nov 25, 2009 12:48 PM PST up reply actions   0 recs

I understand your argument

However, some day some smart person will figure out a way to come up with some fairly reliable statistics in soccer.

Before the wright brothers thought of humans actually flying, most would have said it was a stupid thought.

Before someone actually flew or sailed around the world, there were thousands of people who could give great and valid reason’s why it never could happen.

Before Bill James had written his first baseball article, BA and RBI’s were just about the only measures that anyone had for an offensive player and most people thought that it was enough.

Now, James is a dinosaur and many people have taken his work to to higher levels than he ever could imagine until today there are statistics in it’s infancy that measures defense like UZR.

I understand your reasoning Carlos, but when you say never, there is always some smart guy somewhere that will prove us wrong.

by Coug1990 on Nov 25, 2009 1:30 PM PST up reply actions   0 recs

I can amend to "Highly Unlikely"

And I admit that sports statistics have never been a passion for me, but statistics in this country are helped by the fact that American sports are very statistics-friendly, and baseball seems like it was invented to generate statistics.

But if I were going to go look for models of sports statistics that might be useful in soccer, I’d check hockey and basketball. Like I said, sports statistics aren’t a passion of mine, but hockey and basketball are more fluid than baseball or football. What statistics are available in those sports?

I do have to dispute your first two analogies though. Human flight was something people had been working on for centuries before the Wright brothers. Leonardo Da Vinci had a sketch of a flying machine in one of his notebooks and balloons and dirigibles had been around a long while before the flight at Kitty Hawk. The achievement of the Wrights was winning the race for powered heavier than air flight, not the idea of flight itself.

Similarly, with circumnavigation, after Europeans rediscovered the fact that the world was a spheroid, it was more a matter of who would dare do it than was it possible. Eventually someone did, in the form of Juan Sebastián Elcano, who took over as captain after Magellan died in the Phillipines.

by CarlosT on Nov 25, 2009 1:59 PM PST up reply actions   0 recs

basketball is making amazing advances

good point, CarlosT. american football may not be as analagous as basketball.

but basketball too has made amazing advancements in statistical analysis.

here’s a little wikipedia on it:
http://en.wikipedia.org/wiki/APBRmetrics

here is one of the better writers/thinkers on the subject: http://basketball-statistics.com

even fluid sports like basketball and soccer can be analyzed statistically. it is just a matter of time.

Dave has actually done a lot here at Sounder at Heart to advance this type of analysis. and i suspect we will here more on the subject from Graham.

by zeagle on Nov 25, 2009 4:57 PM PST up reply actions   0 recs

Again I understand your reluctance

But, the examples that you gave to counteract my argument are the worlds visionaries, the exceptions to the normal thought. The majority of the world’s population would not have agreed with them.

So again, I understand your reluctance to agree with the notion that someday some smart person will come up with some relevant statistical analysis. However, it could just as equally mean that you do not have a progressive vision (and I do not mean any disrespect) in this regard.

Now, it took almost 400 years from Da Vinci’s death to the Wright brother’s successful flight. Flight didn’t happen overnight and revolutionizing soccer statistical analysis will not occur overnight.

But, if it were ever to happen, it starts with one person, one visionary to go against the norm.

by Coug1990 on Nov 25, 2009 7:15 PM PST up reply actions   0 recs

No offense taken

I think you’re reading my comments more strongly than I intend if you think I’m saying that no relevant statistical analysis is possible in soccer. And you’re right that Dave has done some great stuff with statistics. I’m actually hoping to work with him next season to produce not only season-long stats, but stats seen through smaller windows through the season, so we can see if there are changes in performance that might be hidden by a longer term analysis. A quantification of “form”, basically. What I’m saying is statistics won’t be able to get as deep as baseball, football, or maybe even basketball, because of the nature of the game itself.

As for the historical examples, I was actually just nitpicking your interpretation of history independently of anything else. My thrust there was that we remember the names in the history books who finally succeed (and in Magellan’s case, not even the guy who succeeded). But before those guys were all the people who failed again and again. Now if you’re talking about laypeople with no knowledge of the subject, then yes of course those people wouldn’t believe what was possible. But they’re ignorant, so that’s to be expected.

by CarlosT on Nov 25, 2009 7:54 PM PST up reply actions   0 recs

Good, I am glad you weren't offended

However, I am offended that you were nitpicking my interpretation of history. Just kidding. You were taking it more literal that I was intending, which is probably my fault because I didn’t make myself clear. My point and I think you understood this from your most recent post was that there are the few, the idea people who believe something can be done and then the many, the rest of the world who need to see something done before they believe it can be done.

By definition, “ignorant” is the correct word, but it has negative connotations so I think a better word is uninformed or unenlightened. There is a saying that “you don’t know what you don’t know.”

Regarding statistical analysis, I guess where we differ is just in our perspectives. Since I grew up following baseball, basketball and football, the statistical data was natural for me.

Now, the so-called sabermetrics has gone so much further than anyone but those true visionaries could have foreseen (and I am not a visionary in this regard).

What is currently being accomplished is because of technology that is available today that didn’t exist in the past.

Computers, video and other technologies have merged to create data outcomes that weren’t conceivable a few years ago.

In baseball, when looking at pitching, the speed of a ball can be measured, the movement of the ball can be measured, the location of the ball can be tracked, etc. What the computer can do is merge this all together into some meaningful and graphical data.

How do you measure defense? If a batter gets three at bats and gets one hit, that is easily measurable (although in and of itself not a good measurement).

If a centerfielder catches two flyballs and the rightfielder catches none during a game. Can you make a conclusion that the CF is a better defender than the RF? No, it could just mean that the CF had two flyballs hit directly at him and the RF had none.

However, with technology someone was able to figure a way to process information and come up with a defensive metric.

In my very rudimentary explanation of baseball statistics, I don’t even scratch the surface of what is being accomplished.

During the half-time of MLS Cup, ESPN showed a color-coded graphic that tracked Beckham and a RSL player (I forget which one) on the pitch.

Now, this is just the beginning. In the future a computer program will be able to track all twenty field players independent of each other and in relation to the ball. Then someone will be able to figure some other use that none of any of us could have conceived.

It could be that it is tracked how well a player makes himself available in space. Someone can design a measurement for a player taking free kicks is based on the velocity of his kicked ball, the bend of the ball and the accuracy of the kick to rate the best players taking a free kick.

There a several other areas in soccer that I think someone could possibly think of a way to measure, but this post is way to long as it is.

So to end, I do not pretend to have all the answers and I could be wrong. But, I could just as easily be correct as well. Do we both have 400 years to wait around and find out?

by Coug1990 on Nov 26, 2009 10:01 AM PST up reply actions   0 recs

A few things

It is important to note that for the EPL at least much of that data exists right now in a closed form, for competitive reasons. I think this actually hampers things. Open data is the main reason why MLB has advanced so much in the intellectual community.

2 – Stats will never replace the eye and mind of a trained observer. They supplement the knowledge that we can from watching and loving the game. It is not a replacement. Statistics will only ever reveal probability, not certainty, and part of what makes sports so exciting is the improbable – the upsets, the run of South Korea to a World Cup semi-final, Iraq winning the Asian Cup title, the USA making the Confeds final.

3 – The battles between the stats/saber guys in baseball became ridiculous at times, with many insults bandied about by both sides. Hopefully for soccer it doesn’t have to get there.

4 – Sounder at Heart will always welcome all opinions in this debate. In fact, though I look to stats I do not have the background that Graham does; and though I look to tactics and technical knowledge Carlos T is stronger than I at that. What the 3 of us will look to be doing is giving you not just the best sight covering the Sounders, Cascadia and Democracy in Sports, but hopefully one of the best covering the beautiful game. In 2010 I promise more interviews, more insight, more stats, more tactical diagrams, more editorials. 2009 was the baseline for the club and this site.

It is time for us all to get better.

I am not a Supporter
I am not a Fan
I am a Sounder
Sounder At Heart

by Sounder At Heart on Nov 26, 2009 10:16 AM PST up reply actions   0 recs

Good post

1) Good point about the EPL being closed. In the long run, there are usually work around’s or the EPL can change their way of thinking. With the outrageous contracts that many players have, it is in the owners best interest to evolve to something like some MLB teams have with replacement level players.

I understand anything that will happen will be at a snail’s pace and possibly not in our lifetime.

2) I agreee. Stats are a tool, that is all. Some MLB teams use it a lot and other’s not so much. The best teams try to merge the two in it’s decision making.

3) I hope that doesn’t happen either. The thing that get’s me is that many cannot admit when they are wrong. I read sites where if people disagree, they are ridiculed and ostracized. They fail to realize that sometimes people come up with a better way and what once was thought the gold standard becomes obsolete.

4) Thanks for your thoughts. I know that Carlos has forgotten more about soccer than I have ever known. You have the best Sounder statistical based site and I look forward to reading more from Graham.

by Coug1990 on Nov 26, 2009 11:00 AM PST up reply actions   0 recs

I don't disagree that statistics can and will be formulated

I doubt either of us will be around in 400 years, but you can never say never, right :) I think the graphic you’re referring to tracked Beckerman on the RSL side.

My point is whatever the state of technology, statistical analysis will be able to go farther in American sports, especially baseball, because of the structure of those games provides dozens of discrete events that can be tracked and tabulated, whereas in soccer you have to go in and make your own distinctions and that by definition makes the statistics more subjective, which is what statistics are trying to get away from in the first place.

For example, you can create a decision tree for a standard play in baseball. First the pitcher has to commit to the pitch, instead of throwing to a base. Next the batter either swings or not. If he swings and makes contact, then the defense springs into action. Assuming it’s not a home run, a defensive player will need to retrieve the ball and the ball needs to get to an infielder to attempt the out and so on and so forth. There are a lot of different ways a play can happen, but there are discrete events and decisions that can be tracked and evaluated and no subjective judgments are required to tease them out.

In contrast, you really can’t generate a very detailed decision tree for a soccer play. Your first challenge is deciding what a “play” is in the first place. If a team advances part way up the right flank then passes back to the keeper then attacks down the middle, gets stymied but retains possession, passes back again and then attacks down the left and scores, where is the play in that? Is it the whole sequence? Is it the attack that finally generated the score? How many passes back go you go? Does it matter how focused the passes are? What if in the middle a player from the other team steals the ball but then loses possession again after a few seconds? Is that a new play or a continuation?

The problem is that you’re either going to have to set some arbitrary standard of what a “play” is or you’re going to have to make a case-by-case decision. And that’s just the first of many subjective judgments you have to make. Another example is a player takes a high-arcing free kick from a long distance and it goes over the keeper’s head and in. The question is was it a shot or a cross that took a lucky turn. Arguments for a shot would be that it was on target and went in without any other player touching it. Arguments for a cross would be that a shot wouldn’t have a high-arcing trajectory because it makes it too easy for other players to track but you want your own players to be able to track a cross. This is exactly the debate over Ronaldinho’s free kick goal against England in the 2002 World Cup. He says it was a shot, but he might say that in either case, so that’s not definitive.

So, to boil it down, statistics in soccer will always have a much greater level of subjectivity and therefore will not reach the same level of objective analysis that you can get in baseball. Technology can provide us with ever more fine-grained data about the game, but we can’t get away from having to make those subjective judgments.

Finally, one of the things I’m really enjoying about being on the network is the depth and breadth of the discussions that follow from posts. I like the back and forth and having to develop arguments and refine one’s ideas, and yes, admitting I’m wrong if I am. I don’t think I am in this case, but I also think that Coug1990 don’t actually disagree, so I don’t think he’s wrong either.

by CarlosT on Nov 26, 2009 11:53 AM PST up reply actions   0 recs

Here's why I think that some day a decent framework will emerge for soccer statistical analysis:

Most of us here have played, I imagine, and any who’s played a lot knows that vision and intelligence with the ball can cover a multitude of sins. Good players have a number of choices of what to do on and off the ball, and the fact that they take certain decisions imply that there must be some sort of logical structure inherent in the game. If there wasn’t, there’d be no decision to make!

So however complicated and fluid the game is, at the end of the day there’s an optimal way to play it. We’ll never get to a perfect understanding of what should be done in every situation vs. what a player actually did, but the fact that it’s there means we can chase it, and someone will get close enough on the analytical side to make reasonable conclusions about player value. And I think ‘reasonable conclusions’ aren’t really that far down the line.

by Graham on Nov 27, 2009 10:01 AM PST up reply actions   0 recs

they've done it in American Football

this same argument was made against statistical analysis in american football. most of the barriers have been overcome — and today there is an amazing (and dizzying) array of statistical analysis being done on a similarly complex team sport. John

it will be done. and it is being worked on by some very smart folks.

by zeagle on Nov 25, 2009 4:50 PM PST up reply actions   0 recs

stay on your feet, man

If Riley was as weak defensively as this entry suggests, I suspect he wouldn’t have been given the playing time he was by Sigi. For all the talk of us creating an attacking style of soccer, I really think Sigi believes in a more balanced approach and at the end of the year we had our entire midlfield behind the ball when defending (that’s where I think a lot of our good defensive stats came from).

One thing I noticed with Riley is he has this habit of every so often leaving his feet when tackling. I’m fine with it if the ball is a little ahead of the attacker and you’re slide tackling, but if the attacker has the ball at his feet and you’re an outside back, why take such a risky approach? You got the touchline at your disposal, just jockey him out to that. It’s a habit I think that is fixable with good coaching.

I would concur with others he is replaceable and losing him wouldn’t be as devastating as some fans are making it out to be. Though I also have this funny feeling he is not going to get selected.

by PeterJH on Nov 25, 2009 11:11 AM PST up reply actions   0 recs

I think most of our good defensive stats came from Alonso, Hurtardo, and Keller, but right now it's impossible to judge

The touchline thing drives me absolutely insane too. There is nothing wrong with putting the ball out and regrouping the defence.

by Graham on Nov 25, 2009 11:15 AM PST up reply actions   0 recs

Part of the playing time came because we really had no better options.

Our reserves at the fullback position were downright frightening to watch.

by Sec 108 on Nov 25, 2009 12:55 PM PST up reply actions   0 recs

I'm still not convinced...

Scott, Wahl, Sturgis (speaking of Wikipedia, take a peak), even Evans…

Riley played practically every minute he was available to play this past season. If he was as weak defensively as is being portrayed here, there would have been a lot more experimentation at right back. When in New England he also got quite a bit of playing time under Steve Nicol. There is no way Nicol would let anyone as weak defensively as is being portrayed here remotely near his back line.

by PeterJH on Nov 26, 2009 1:16 AM PST up reply actions   0 recs

ok then

How about providing some specific examples in games of what you are talking about?

That way we can go back to the game films and see this. It may very well be that some of us just didn’t notice this before and I’m certainly willing to learn.

by PeterJH on Nov 26, 2009 12:18 PM PST up reply actions   0 recs

you must have had something in mind to write all this

I just watched the first half of the USOC semi-final (http://www.soundersfc.com/media-library/Videos/Match-Replays/2009/090721-Houston-First-Half.aspx?svt=0)

I see Riley leave his feet in the 2nd minute, and around 16:45. Beyond that, I’m really not seeing much to be concerned about, and he does seem to blend into the attack very nicely. Positionally, he’s very sound.

You must have at least one incident in a game in mind that illustrates your concerns with Riley. One?

by PeterJH on Nov 28, 2009 8:05 PM PST up reply actions   0 recs

I'll go hunt something down if you absolutely must have a specific event cited

But I wouldn’t have posted if I had one event in mind, because that would be a ridiculously small sample to base a conclusion off, and would make for an excessively weak argument. I think Riley’s a bad defender because I’ve never seen him be a good defender.

by Graham on Nov 28, 2009 11:22 PM PST up reply actions   0 recs

come on!

One example is better than no examples.

“As an attacking threat, Riley is completely neutered by a good left winger. He then becomes a pure defender, forced backwards by both the skill of his opposition and his own defensive ineptitude. Without the requisite defensive talents, he is essentially unable to break out of his own territory, because he’s forced to sit much tighter on his direct opponent. And not even that works very well, due to his habit of dropping away whenever they get the ball. So while a non-attacking full-back is a liability against any team, a good team can push Riley completely out of the game simply by calling his defensive game into question more often.”

Is this just a theoretical MLS winger? Or has this theory been proven?

by PeterJH on Nov 29, 2009 12:23 AM PST up reply actions   0 recs

I have numerous examples, but since I have no idea what games they're from I can hardly tell you what they all are

As for that paragraph, Riley was completely shut down in the second playoff leg by Houston’s left winger.

by Graham on Nov 30, 2009 8:30 AM PST up reply actions   0 recs

LL?

Not sure what LL is. Where do I go?

The left mid for Houston is Brad Davis. When work/family gives me the time, I’ll be sure to take a closer look. Thanks.

by PeterJH on Nov 30, 2009 11:31 AM PST up reply actions   0 recs

LookoutLanding.com

Mariners site on the network here.

I am not a Supporter
I am not a Fan
I am a Sounder
Sounder At Heart

by Sounder At Heart on Nov 30, 2009 12:35 PM PST up reply actions   0 recs

I'm not seeing it

Well, I just finished watching the first half, and while there are a couple occasions when Riley did something that made me cringe, I didn’t see anything remotely like what you’re describing.

Houston’s left mid, Brad Davis, was practically invisible in that half. He certainly wasn’y any sort of speedy threat who caused Riley to hold back.

You might have been thinking of Oduro, but he’s a striker and as such was much more Huratdo’s mark than Riley’s.

I see Riley overlapping in attack on the right side in minutes 4, 8, 26, 28, 36. Considering Houston was the better side that half, this isn’t indicative of holding back.

The only time I saw Brad Davis do anything was in the 21 min, when he played a dangerous ball in while on the flank. In this case I thought Riley was jockeying too square to the touchline when Davis wasn’t square to him (if Davis was really dangerous he would have taken advantage and cut into the middle and caused real problems). In the 23’, there’s an instance where Oduro is cutting across the top of the box and Riley is the closest defender to him. Instead of closing Oduro down, it looks like Riley is trying to switch him off onto Hurtado, and Oduro took advantage and rang a shot off the post. Those were the two instances that made me cringe. I am happy to report he didn’t leave his feet once the first half (though we all know he made up for that with vengeance at the end of the game).

In 33’, 41’ and 44’, Riley is carping at the referee. So I would say I tend to agree with Dave’s assessment in a later post. What I saw tended to match what Dave was describing.

In terms of priorities, I really think a new RB is way down the list. We badly need to address free kicks (minus corners), control in central mid and an attacking physical presence up front much more desperately than we need to address Riley.

by PeterJH on Dec 4, 2009 12:09 AM PST up reply actions   0 recs

Just like to say.

Welcome Graham! I look forward to the quality that infuses Mariner blogs attaching to the quality of Sounder At Heart and making some sort of Voltron-like super statistically minded blog of awesomeness.

by Perrinbar on Nov 25, 2009 10:51 AM PST reply actions   0 recs

I should be clear

At this time Graham is not an “official” writer with Sounder at Heart. At this time… hopefully only because I haven’t asked.

I bumped this from a FanPost because quite frankly I woke up to this stunning bit of writing.

I am not a Supporter
I am not a Fan
I am a Sounder
Sounder At Heart

by Sounder At Heart on Nov 25, 2009 11:04 AM PST up reply actions   0 recs

I saw that.

and while he may or may not have the time, just having him here means we get that much better.

by Perrinbar on Nov 25, 2009 11:23 AM PST up reply actions   0 recs

Though he just said yes

So he’s listed on the masthead.

I am not a Supporter
I am not a Fan
I am a Sounder
Sounder At Heart

by Sounder At Heart on Nov 25, 2009 11:34 AM PST up reply actions   0 recs

Voltron complete!

by Perrinbar on Nov 25, 2009 11:47 AM PST up reply actions   0 recs

Who Then Replaces Riley?

Let’s assume your analysis regards Riley’s defects is correct – I’m not saying it isn’t correct, just that the numbers aren’t there. That still begs the question of 1) who takes over his position and 2) is filling Riley’s position a priority given other needs? I’m not sure it is.

Who on the team can fill Riley’s position? Wahl doesn’t seem to have any capability to attack (at least none that I saw). Marshall? Surely more of an enforcer than Riley, but does he have the legs and ability to be an attacking defender? Is there another player (Ianni?) who can be converted? Or do we need to bring some one on board?

And if bringing someone on board is the only solution, is Riley’s play so poor that replacing it should be a priority? I’m not sure. The ability to move the ball from the defense through the midfield to the forwards looked like an immense hole to me that must be filled if Jacqua/Montero are going to be successful. Too often – it seemed to me – the defense simply cleared the ball (generally to the other side) with little attempt to start a counter attack. Problem with the defenders or a problem with the midfield? Which needs to be addressed?

by snesbitt on Nov 25, 2009 11:11 AM PST reply actions   0 recs

Well you could have asked the same question at the start of the season with Zach Scott

Then Leo Gonzalez fell into our laps in the middle of the season. Clearly, we don’t have anyone on the roster capable of replacing Riley at right back, but it’s the offseason, and there will be players out there.

by Graham on Nov 25, 2009 11:17 AM PST up reply actions   0 recs

They picked Le Toux

Great discussion, but it’s all irrelevant now. The bastards took Le Toux.

by J Sep on Nov 25, 2009 12:14 PM PST reply actions   0 recs

Just because they didn't take Riley

doesn’t make it irrelevant. It means we’ll probably still be looking to upgrade at right back. I’ll miss LeToux, but beyond being a class-act he was getting a lot of money to come off the bench, or not come off the bench as it was at the end of the year.

by Perrinbar on Nov 25, 2009 12:25 PM PST up reply actions   0 recs

Sorry, I didn't mean it like that

I meant more that the point about Riley being likely the one to leave in the draft. I agree that the rest makes sense still.

by J Sep on Nov 25, 2009 12:28 PM PST up reply actions   0 recs

Comment withdrawn. I’m very interested to see what we do this offseason. If we can find another Leo, that would be awesome!

by Perrinbar on Nov 25, 2009 1:12 PM PST up reply actions   0 recs

Spot on, at least to my eyes

Riley really did seem like a wing player, which was good and bad – good in that he was able not just to draw a defender wide on the attacking end and spread the field a bit, but because he could actually drive and cross.
On the other hand, his defensive play seemed to consist of yielding ground until Hurtado double-teamed and forced a pass.
Like a number of players, there seems to be more in him than he showed in 2009; I could see him improving his footwork on defense and becoming a very good player (yay that we didn’t lose him).

by marc w on Nov 25, 2009 1:44 PM PST reply actions   0 recs

Comments For This Post Are Closed


User Tools

Sounder at Heart is a blog about the Seattle Sounders FC, with occasional forays into Democracy in Sports, Roster Management, Soccer Statistics and Life in Puget Sound.
Start posting on Sounder At Heart »

Join SB Nation and dive into communities focused on all your favorite teams.

Connect_with_facebook

FanPosts

Community blog posts and discussion.

Recent FanPosts

3302_1084868615194_1630752284_179446_4689125_n_small
Hey!
Tight_with_elephant_small
The future of "Sounders FC Weekly"...
Small
CBA deal not looking good
Sounder_4_small
Notes from the team - Feb 22, 2010
Small
Luis Gil to Seattle or RSL? Currently with Wizards, what is his value?
Small
Players are in no-win situation
Small
Season may be in jeopardy
Small
MLS is in no danger of turning into the NASL
Img_0349_small
Markus Kamp for Alliance Council
Enchantment_clark_060_small
Arlo White to succeed Kevin Calabro as Sounders' Play-by-Play announcer

+ New FanPost All FanPosts >

SBNation.com Recent Stories

David Beckham watches the Los Angeles Lakers against the Houston Rockets in an NBA basketball game, Sunday, Nov. 15, 2009, in Los Angeles. (AP Photo/Lori Shepler)

David Beckham Likely Ruptures Achilles' Tendon, Will Miss World Cup

Olympique Lyonnais players celebrate after scoring their first goal during the UEFA Champions League round of 16 2nd leg match between Real Madrid and Olympique Lyonnais at Estadio Santiago Bernabeu on March 10, 2010 in Madrid, Spain. (Photo by Denis Doyle/Getty Images) link

Real Madrid Out Of Champions League After 1-1 Draw With Lyon

wayne rooney man u link

Manchester United Rolls Past Milan For Spot In Champions League Quarter-Finals

More from SBNation.com >


Managers

Sounder_4_small Sounder At Heart

Editors

Small BayAreaRefugee

Authors

Small CarlosT

Hms_surprise_small Graham