By Rapids Rabbi
When a player has the ball at their foot, or moves into space anticipating the ball or to create a new shape or dynamic for the offense, everything is poetry and creativity. It is a splash of cerulean or azure on canvas, mottled with some ocre, to create a luxurious panorama. Step-over, in-step cut, long touch, feint, outside-of-the-boot pass. These are the actions of a craftsman, making artistic choices that will sculpt something not yet in existence, a goal, into being. It is creation from nothing. It is majestic and awe-inspiring. It is why Neymar is worth, literally, a half a billion dollars.
Defense is not nearly as awe inspiring, or as poetic. It is less art than science. Because it is reactive, instead of creative, it relies on training, and focus, and boring, boring practice. Watch game film on your opponent. Learn his or her mannerisms and go-to moves. Learn his tendencies and which foot he prefers. Watch his hips as he moves - don’t buy the flicks and feints and bobbles.
And, most importantly, do not make a mistake.
Attackers make mistakes all the time. They lose the ball in possession. They clip an errant pass. They shoot it miles wide, or high. Cristiano Ronaldo, arguably the best player in modern football right now, has played in 39 matches. He has 41 goals, or an unbelievably prolific 1.05 goals per game. He takes 6.8 shots per game. That means he fails to score on 85% of his shots. Nobody ever raises this point when discussing Ronaldo, because a 15% success rate is pretty spectacular for an attacker.
That’s not an option for defenders. They need to be pretty much flawless 99% of the time in their defending, because anything less and they’ll be replaced by someone stronger, faster, more precise, or more reactive.
I really, really enjoyed covering the 2016 Colorado Rapids. To watch the science of a defense, operating in perfect synchronization, frustrating their opponents time and time again, and frequently emerging victorious; it was a sight to see. Also, as a pundit, it gave me great pleasure to have the simple job of saying, game after game, that the defense was flawless; that nobody blew an assignment, that the Rapids held the line. I took very little joy in 2014, or 2015, or 2017, reviewing a game and calling attention to the errors of the players. When the team defends well, they get credit as a unit. When the team defends poorly, it typically will get pinned on the specific mistakes of just one or two players. It sucks to write. But a blown coverage resulting in a goal is as plain as the nose on my face. If I didn’t say something, somebody else would.
This whole prologue is, of course, due to the messy reality of Saturday’s 1-1 draw away to FC Dallas. After getting a 1-0 lead on a 62nd minute goal from Joe Mason, the Rapids were in control. They just needed to defend that lead for 28 minutes and walk away with all 3 points. But they couldn’t do it, yielding a goal in the 89th minute to Cristian Colman. It was the third time in Colorado’s four MLS games this season that they have conceding a game-tying or game-winning goal. So we need to spend a little time figuring out (sigh) who and what has been responsible for the late game defensive troubles of the Colorado Rapids.
Late Game Defending Part 1: TacticsFor starters, the teams late game tactics should be scrutinized. From the 75th minute onward, the Rapids kicked long passes or cleared the heck out of everything in the final third. While Colorado had 23 clearances from the 1st to the 70th minute, they had an insane 21 clearances after the 70th minute.
Moreover, from the 75th minute onward the team completed 16 passes of 37 total passes, gifting the ball back to Dallas 21 times. Often, those turnovers occured in just seconds. Colorado was out-possessed 80 to 20 in the final 15 minutes.
I understand the logic of kicking it long: make the other team chase it, make them tired, take time off the clock, make them reset and bring it back into your third, and you get to organize your defense with all 11 men behind the ball. But the other team of course can anticipate this tactic and adjust: they don’t need to defend as much and can put more men into the attack. They get a free walk into the final third. And they get more bites at the apple, because they have more possessions and more chances. In the final 15 minutes, Dallas squeezed off 8 shots. In the first 75 minutes, they had only 9 shots.
This 'bunker and pray' tactic didn’t work here for Colorado. This tactic probably doesn’t work, game over game, in the long run, for most managers. If you give the other team more chances, and you put no pressure on their defense, you’re asking for trouble. So one obvious reason for the Rapids late game struggles is tactical: they are putting themselves into a ‘last stand at the Alamo’ situation. The Alamo, for those of you non-history buffs, did not work out well for the defense.
Late Game Defending Part 2: Individual BreakdownsBut it isn’t just tactics. You can decide to go ultra-defensive to hold a lead if your defenders stay alert, and organized, and are physically up to the task. This was not what happened with Colorado.
For one, Edgar Castillo, who was fantastic all game, was left out on his own with Miguel Barrios on the wing at the conclusion of this game, and that nearly ended in disaster. Watch here as Barrios chew up the left back, requiring emergency defending from Jack Price and Johan Blombergthat all was nearly catastrophic.
To watch the gif of Barrios clowning Castillo and the Rapids defense, click here. To return to this article, just click the ‘back button on your browser.’
Castillo makes three miscues on this play: first he gets turned inside-out by Barrios, next his clearance is right back, into Barrios, and last he doesn’t follow Barrios, hoping somebody else will pick him up. But the guy who does is Blomberg, who trucks the little dude over, nearly resulting in a penalty. This moment in the 82nd minute was a crisis, averted, but it portended very badly for what the defense would do down the stretch. I’ll mention Castillo again when we talk about game management and subs later on.
And then, there’s Deklan Wynne. Wynne has been called out here severaltimes for not being defensively up to par. In this match, Wynne is squarely on the hook for the game-tying goal.
To watch a gif of the game-tying goal, click here. To return to this article, just click the ‘back button on your browser.’
Mauro Diaz lofts an unbelievably perfect over-the-top pass to Cristian Colman, who finishes with his head. But Wynne is slow and late to react and ignorant of Colman breaking in behind him, and it results in a goal. Kip Colvey, debuting at right wing back, is guilty as well. He’s jogging back on defense and lets Colman go to Wynne without supporting his fellow Zealander, although there isn’t anybody else for Colvey to be guarding. The two of them blew this play, and got burned.
It is all the more frustrating because only 7 minutes earlier, Wynne and Colvey saw the exact same play, sniffed it out appropriately, and defended it without issue. That what I was talking about in my opening - it takes a 89 minutes and 55 seconds to construct a winner. A disappointment is the result of a 5 seconds lapse in concentration. Rapids announcer Marcelo Balboa was so certain in the Rapids ability to defend this kind of play, he said this on the broadcast: “And that’s perfect, that’s what Colorado wants. You’ve got the big guys (Wilson and Sjoberg) up top. They’ve got Colman, who’s not that big. Urruti who’s not that big. They put the ball into the box, especially over the middle? That fits perfectly into what colorado wants.” Marcelo’s right: the big guys can clear this ball all day long. Wynne, at 5’10”, and Colvey, also 5’10”, do not.
Sub selectionThere are also some questions after this match as to how Anthony Hudson is managing subs. Early in the CCL ‘pre-season’, AH chose to let the starters go a full 90, and didn’t sub at all, which one could ask questions about.
In this match, with a 1-0 lead, Hudson made three changes: Niki Jackson for Joe Mason at 77’, Dillon Serna for Nana Boateng at 80’, and Micheal Azira for Dominique Badji at 90’.
Jackson was a great addition because he ran around like a madman, harassing the ball-carrier at midfield while Badji was looking a bit too tired to give chase. Serna for Boateng is more of a puzzler. Serna’s an attacker - he can buzz and harass, but there were two more defensive choices in Sam Hamilton and Micheal Azira to go with here. And Boateng looked ok at this point, while it was his midfield partner Johan Blomberg who looked pretty well spent (my notes say ‘Blomberg was trash for the final 15). And then, finally, Hudson brought on Azira to close the door. Only, he did it after the game-tying goal. That’s pretty much the definition of closing the barn door after the horses have gotten out.
Also, subbing off strikers when your wingers have been covering a tremendous amount of ground all game and you’re clinging to a 1-0 lead strikes me as odd. Castillo and Colvey are flagging at the end of this one; I think no amount of fitness training would allow fullbacks in a 5-3-2 to perform past the 75th minute at the level you need them to perform
I have a feeling the Rapids head coach is still mentally in the phase of pushing players physically in the early season to build up their fitness. It is possible that, long term, it’s a good strategy. In the short term, it may have cost this tem two points on Saturday.
The offense under Anthony Hudson: is it better?I see a lot of folks on the twitters saying how happy they are that, even with the Rapids defensive struggles, the offense is much improved. They like the approach, and they like the attack. And, sure, the Rapids are scoring goals, so these folks have a good point. Colorado has 7 goals through their first four games. At this point in 2017, the Rapids had 4 goals. In 2016 after four games, Colorado had 3 goals. In 2015, Pablo Mastroeni’s first season, they had zero goals after four games - they wouldn’t get off the schneid until game five, when they curb stomped Dallas in Frisco for a four goal explosion (Badji had his first MLS goal that day). So it is true that the Rapids are scoring more this year.
But let's look a little more closely. First, Badji’s three goal game might want to be qualified with a small asterix as a little fluky. But second, a better measure of whether the offense is clicking than goals is chance creation - whether the Rapids are getting more quality chances on goal this year than in previous years. The sample size of just four games makes it hard to really say definitely whether the Rapids are ‘much better’ offensively - things like home field advantage, quality of opponent, and luck have an outsized effect in such a small pool of data. Those things will even out in 6 to 10 more games. But chance creation, or the possibility of a good chance on goal, happens 15 to 30 times a match. Are the Rapids creating better chances this year than in the past?
The Rapids expected goals for (xGF) in 2018 is currently 4.9, which is on the low side among MLS teams (SKC leads with 9.9 xGF, Seattle [!] has the fewest expected goals with 4.4). Through four games in 2017, the Rapids had an xG of 5.96. Through four games in 2016, the Rapids had an xG of 4.59. This means the Rapids aren’t creating more offense than they used to; it’s roughly the same as in the past. Colorado is finishing better this year than in past years. That’s good! It indicates that the offense might be better. But it does not indicate that the offense is more dangerous or creative. The higher 2018 xG is also worrying: finishing is a fickle thing that grows and shrinks and generally stays near the mean. Badji and Joe Mason might be at the start of an unstoppable season in front of goal. Or he could be hot early, with an expectation that he can’t possibly keep bury If you really want to understand this, check out Harrison Crowe’s weekly column ‘Lowered Expectations’ on AmericanSoccerAnalysis.com. Crowe looks at five high-probably goals that didn’t happen, and one low probability one that did, each week, with gifs. It’s highly entertaining and educational.
…
The Rapids potentially have three players coming in in the next two weeks that will make them more dangerous and more attacking: Shkelzen Gashi, Stefan Aigner, and Yannick Boli. The first two are recovering from early season knocks, fitness issues, and illness. Boli is apparently ready to go for this week’s match against TFC. Let’s hope adding a few more artists to the field will allow the Rapids to paint a more luxurious canvas, offensively. The defense, meanwhile, will be back in the laboratory trying to get the science of defending right.
When a player has the ball at their foot, or moves into space anticipating the ball or to create a new shape or dynamic for the offense, everything is poetry and creativity. It is a splash of cerulean or azure on canvas, mottled with some ocre, to create a luxurious panorama. Step-over, in-step cut, long touch, feint, outside-of-the-boot pass. These are the actions of a craftsman, making artistic choices that will sculpt something not yet in existence, a goal, into being. It is creation from nothing. It is majestic and awe-inspiring. It is why Neymar is worth, literally, a half a billion dollars.
Defense is not nearly as awe inspiring, or as poetic. It is less art than science. Because it is reactive, instead of creative, it relies on training, and focus, and boring, boring practice. Watch game film on your opponent. Learn his or her mannerisms and go-to moves. Learn his tendencies and which foot he prefers. Watch his hips as he moves - don’t buy the flicks and feints and bobbles.
And, most importantly, do not make a mistake.
Attackers make mistakes all the time. They lose the ball in possession. They clip an errant pass. They shoot it miles wide, or high. Cristiano Ronaldo, arguably the best player in modern football right now, has played in 39 matches. He has 41 goals, or an unbelievably prolific 1.05 goals per game. He takes 6.8 shots per game. That means he fails to score on 85% of his shots. Nobody ever raises this point when discussing Ronaldo, because a 15% success rate is pretty spectacular for an attacker.
That’s not an option for defenders. They need to be pretty much flawless 99% of the time in their defending, because anything less and they’ll be replaced by someone stronger, faster, more precise, or more reactive.
I really, really enjoyed covering the 2016 Colorado Rapids. To watch the science of a defense, operating in perfect synchronization, frustrating their opponents time and time again, and frequently emerging victorious; it was a sight to see. Also, as a pundit, it gave me great pleasure to have the simple job of saying, game after game, that the defense was flawless; that nobody blew an assignment, that the Rapids held the line. I took very little joy in 2014, or 2015, or 2017, reviewing a game and calling attention to the errors of the players. When the team defends well, they get credit as a unit. When the team defends poorly, it typically will get pinned on the specific mistakes of just one or two players. It sucks to write. But a blown coverage resulting in a goal is as plain as the nose on my face. If I didn’t say something, somebody else would.
This whole prologue is, of course, due to the messy reality of Saturday’s 1-1 draw away to FC Dallas. After getting a 1-0 lead on a 62nd minute goal from Joe Mason, the Rapids were in control. They just needed to defend that lead for 28 minutes and walk away with all 3 points. But they couldn’t do it, yielding a goal in the 89th minute to Cristian Colman. It was the third time in Colorado’s four MLS games this season that they have conceding a game-tying or game-winning goal. So we need to spend a little time figuring out (sigh) who and what has been responsible for the late game defensive troubles of the Colorado Rapids.
Late Game Defending Part 1: TacticsFor starters, the teams late game tactics should be scrutinized. From the 75th minute onward, the Rapids kicked long passes or cleared the heck out of everything in the final third. While Colorado had 23 clearances from the 1st to the 70th minute, they had an insane 21 clearances after the 70th minute.
Moreover, from the 75th minute onward the team completed 16 passes of 37 total passes, gifting the ball back to Dallas 21 times. Often, those turnovers occured in just seconds. Colorado was out-possessed 80 to 20 in the final 15 minutes.
I understand the logic of kicking it long: make the other team chase it, make them tired, take time off the clock, make them reset and bring it back into your third, and you get to organize your defense with all 11 men behind the ball. But the other team of course can anticipate this tactic and adjust: they don’t need to defend as much and can put more men into the attack. They get a free walk into the final third. And they get more bites at the apple, because they have more possessions and more chances. In the final 15 minutes, Dallas squeezed off 8 shots. In the first 75 minutes, they had only 9 shots.
This 'bunker and pray' tactic didn’t work here for Colorado. This tactic probably doesn’t work, game over game, in the long run, for most managers. If you give the other team more chances, and you put no pressure on their defense, you’re asking for trouble. So one obvious reason for the Rapids late game struggles is tactical: they are putting themselves into a ‘last stand at the Alamo’ situation. The Alamo, for those of you non-history buffs, did not work out well for the defense.
Late Game Defending Part 2: Individual BreakdownsBut it isn’t just tactics. You can decide to go ultra-defensive to hold a lead if your defenders stay alert, and organized, and are physically up to the task. This was not what happened with Colorado.
For one, Edgar Castillo, who was fantastic all game, was left out on his own with Miguel Barrios on the wing at the conclusion of this game, and that nearly ended in disaster. Watch here as Barrios chew up the left back, requiring emergency defending from Jack Price and Johan Blombergthat all was nearly catastrophic.
To watch the gif of Barrios clowning Castillo and the Rapids defense, click here. To return to this article, just click the ‘back button on your browser.’
Castillo makes three miscues on this play: first he gets turned inside-out by Barrios, next his clearance is right back, into Barrios, and last he doesn’t follow Barrios, hoping somebody else will pick him up. But the guy who does is Blomberg, who trucks the little dude over, nearly resulting in a penalty. This moment in the 82nd minute was a crisis, averted, but it portended very badly for what the defense would do down the stretch. I’ll mention Castillo again when we talk about game management and subs later on.
And then, there’s Deklan Wynne. Wynne has been called out here severaltimes for not being defensively up to par. In this match, Wynne is squarely on the hook for the game-tying goal.
To watch a gif of the game-tying goal, click here. To return to this article, just click the ‘back button on your browser.’
Mauro Diaz lofts an unbelievably perfect over-the-top pass to Cristian Colman, who finishes with his head. But Wynne is slow and late to react and ignorant of Colman breaking in behind him, and it results in a goal. Kip Colvey, debuting at right wing back, is guilty as well. He’s jogging back on defense and lets Colman go to Wynne without supporting his fellow Zealander, although there isn’t anybody else for Colvey to be guarding. The two of them blew this play, and got burned.
It is all the more frustrating because only 7 minutes earlier, Wynne and Colvey saw the exact same play, sniffed it out appropriately, and defended it without issue. That what I was talking about in my opening - it takes a 89 minutes and 55 seconds to construct a winner. A disappointment is the result of a 5 seconds lapse in concentration. Rapids announcer Marcelo Balboa was so certain in the Rapids ability to defend this kind of play, he said this on the broadcast: “And that’s perfect, that’s what Colorado wants. You’ve got the big guys (Wilson and Sjoberg) up top. They’ve got Colman, who’s not that big. Urruti who’s not that big. They put the ball into the box, especially over the middle? That fits perfectly into what colorado wants.” Marcelo’s right: the big guys can clear this ball all day long. Wynne, at 5’10”, and Colvey, also 5’10”, do not.
Sub selectionThere are also some questions after this match as to how Anthony Hudson is managing subs. Early in the CCL ‘pre-season’, AH chose to let the starters go a full 90, and didn’t sub at all, which one could ask questions about.
In this match, with a 1-0 lead, Hudson made three changes: Niki Jackson for Joe Mason at 77’, Dillon Serna for Nana Boateng at 80’, and Micheal Azira for Dominique Badji at 90’.
Jackson was a great addition because he ran around like a madman, harassing the ball-carrier at midfield while Badji was looking a bit too tired to give chase. Serna for Boateng is more of a puzzler. Serna’s an attacker - he can buzz and harass, but there were two more defensive choices in Sam Hamilton and Micheal Azira to go with here. And Boateng looked ok at this point, while it was his midfield partner Johan Blomberg who looked pretty well spent (my notes say ‘Blomberg was trash for the final 15). And then, finally, Hudson brought on Azira to close the door. Only, he did it after the game-tying goal. That’s pretty much the definition of closing the barn door after the horses have gotten out.
Also, subbing off strikers when your wingers have been covering a tremendous amount of ground all game and you’re clinging to a 1-0 lead strikes me as odd. Castillo and Colvey are flagging at the end of this one; I think no amount of fitness training would allow fullbacks in a 5-3-2 to perform past the 75th minute at the level you need them to perform
I have a feeling the Rapids head coach is still mentally in the phase of pushing players physically in the early season to build up their fitness. It is possible that, long term, it’s a good strategy. In the short term, it may have cost this tem two points on Saturday.
The offense under Anthony Hudson: is it better?I see a lot of folks on the twitters saying how happy they are that, even with the Rapids defensive struggles, the offense is much improved. They like the approach, and they like the attack. And, sure, the Rapids are scoring goals, so these folks have a good point. Colorado has 7 goals through their first four games. At this point in 2017, the Rapids had 4 goals. In 2016 after four games, Colorado had 3 goals. In 2015, Pablo Mastroeni’s first season, they had zero goals after four games - they wouldn’t get off the schneid until game five, when they curb stomped Dallas in Frisco for a four goal explosion (Badji had his first MLS goal that day). So it is true that the Rapids are scoring more this year.
But let's look a little more closely. First, Badji’s three goal game might want to be qualified with a small asterix as a little fluky. But second, a better measure of whether the offense is clicking than goals is chance creation - whether the Rapids are getting more quality chances on goal this year than in previous years. The sample size of just four games makes it hard to really say definitely whether the Rapids are ‘much better’ offensively - things like home field advantage, quality of opponent, and luck have an outsized effect in such a small pool of data. Those things will even out in 6 to 10 more games. But chance creation, or the possibility of a good chance on goal, happens 15 to 30 times a match. Are the Rapids creating better chances this year than in the past?
The Rapids expected goals for (xGF) in 2018 is currently 4.9, which is on the low side among MLS teams (SKC leads with 9.9 xGF, Seattle [!] has the fewest expected goals with 4.4). Through four games in 2017, the Rapids had an xG of 5.96. Through four games in 2016, the Rapids had an xG of 4.59. This means the Rapids aren’t creating more offense than they used to; it’s roughly the same as in the past. Colorado is finishing better this year than in past years. That’s good! It indicates that the offense might be better. But it does not indicate that the offense is more dangerous or creative. The higher 2018 xG is also worrying: finishing is a fickle thing that grows and shrinks and generally stays near the mean. Badji and Joe Mason might be at the start of an unstoppable season in front of goal. Or he could be hot early, with an expectation that he can’t possibly keep bury If you really want to understand this, check out Harrison Crowe’s weekly column ‘Lowered Expectations’ on AmericanSoccerAnalysis.com. Crowe looks at five high-probably goals that didn’t happen, and one low probability one that did, each week, with gifs. It’s highly entertaining and educational.
…
The Rapids potentially have three players coming in in the next two weeks that will make them more dangerous and more attacking: Shkelzen Gashi, Stefan Aigner, and Yannick Boli. The first two are recovering from early season knocks, fitness issues, and illness. Boli is apparently ready to go for this week’s match against TFC. Let’s hope adding a few more artists to the field will allow the Rapids to paint a more luxurious canvas, offensively. The defense, meanwhile, will be back in the laboratory trying to get the science of defending right.