Tagged: Portugal

World Cup 2014 – Group Stages Completed; Who’s leading the PWP Pack?

No more draws… you’re out you’re out.  The true brutality of the game begins; if you’re faint of heart and don’t want to know how well the USA stacks up against the rest of the World in Possession with Purpose don’t read on.

I’ll lightly touch on my Attacking PWP to set the stage and then the reality of the Defending PWP and finally – the Composite PWP – it aint pretty if you fancy the United States…

To begin…

APWP INDEX GROUP STAGES COMPLETED

APWP INDEX GROUP STAGES COMPLETED

I walked through some major details on APWP in my last post so I won’t tarry here too long… a couple of things that stand out to me…

Only six teams fall below the pack of green bars up top – not a complete match but good enough when considering the ‘end state’ of PWP – come close to matching the League Table ‘without’ tracking wins, draws, and losses.

From an attacking standpoint there’s pretty solid evidence to support the USA being in a “group of death”; they ended up with the worst APWP in their group yet got through.

Not to be outdone though – there are the Greeks – they too finished lower than Colombia, Ivory Coast, and Japan.

Is the difference between this Index output and Results in the Group Stages a measurement of luck?

I don’t know but the outputs from the Index seem pretty compelling after just three games.

Now for the Defending PWP Index…

DPWP INDEX GROUP STAGES COMPLETED

DPWP INDEX GROUP STAGES COMPLETED

In short – the DPWP Index looks to have been much more accurate than the APWP Index; correctly ranking the top teams with just four exceptions.

For me that continues to reinforce that Defending (preventing the opponent from scoring) has more overall value than just scoring.

So how about some info behind the Index number; here’s the details on the differences between teams that advanced and teams that didn’t.

Opponent Possession: (PWP data point)

  • Teams not making the round of 16 who were in the top ten were Spain, Japan, Italy, Ivory Coast and Bosnia-Herzegovina.
  • Teams making the round of 16 who were in the bottom ten were Netherlands, United States, Greece, Algeria, Costa Rica, and Colombia.
  • Bottom line here; any approach with respect to possession can work provided the Head Coach has the right mix of players to execute that approach.

Opponent Passing Accuracy Entire Pitch: (PWP data point)

  • Opponents of the United States were the most accurate passers in the World Cup so far (87.33% accurate); perhaps another piece of objective evidence supporting how talented that Group was?
  • Other teams who faced opponents with high levels of passing accuracy, that made the round of 16, were Greece, Netherlands, Germany, Colombia, and Costa Rica.
  • The teams making the round of 16 that played against opponents with lower passing accuracy included Chile (lowest opponent passing accuracy – 76%), Brazil, Argentina, France, and Belgium.
  • Those teams “not” making the round of 16. that played against opponents with lower passing accuracy. included Japan, Spain, England, Ivory Coast and Ghana.

Opponent Passing Accuracy within the Final Third: (Supplemental PWP data point)

  • Opponents of the United States were also the most accurate passers in the Final Third (76.33%); perhaps??? another piece of objective evidence supporting how talented that Group was?
  • Greece remains a bed-fellow in this category as well – opponents of Greece were also 76.33% accurate within the Greece defending Final Third.
  • Both teams were the worst in this category; and were the only two teams, in the worst top ten, to make the round of 16.
  • Might this be a good “team” indicator on how effective those team defenses were in communicating and executing their roles?  Or was it good luck or great goalkeeping?
  • On the flip side Spain, England and Ivory Coast faced opponents with the lowest averages of passing accuracy in the defending Final Third.
  • Those three teams were also in the top ten ‘best’ for this category; and the only three teams in that ‘best ten’ that didn’t make the round of 16.
  • Might this also be a good “team” indicator on how poorly those team defenses communicated and executed their roles?  Or was it bad luck or bad goalkeeping?
  • I’m not sure about the answers, to those questions, but it certainly might be a good place to start as England and Spain lick their wounds and prepare for Euro2016…

Percentage of Opponent Passes within the Final Third: (PWP data point)

  • The easiest teams to penetrate against, so far, have been Colombia, Greece, and the United States.
  • All three have seen their opponents penetrate their defending third more than 28% of the time given total possession of the ball.
  • Those three, plus Switzerland, also made the round of 16, all the other teams in the worst ten, for this category, are going home.
  • In looking at the teams with the least amount of penetration per possession we have France leading the pack at just 13.98%; with Netherlands next at 15.96%.
  • What is interesting about Holland is that their opponents possessed the ball (overall) third most (60.95%).
  • Truly amazing that with over 60% of possession their opponents penetrated just 16% of the time – can you say high pressure that was extremely well organized?
  • As for those who didn’t advance; Spain, England, Australia and Bosnia-Herzegovina were in the top ten for being stingy; the rest advanced.

Percentage of Opponent Shots Taken per Completed Pass in the Final Third: (PWP data point)

  • Algeria was the top team in preventing shots taken, per completed pass, in their own defending third (11.75%); next up was England, Iran, Australia, Russia.
  • Intriguing that six of the top ten teams in this category (Iran, Australia, Costa Rica, Netherlands, Greece, and the USA) were also six of the top ten in yielding possession and facing high passing accuracy numbers by their opponents.
  • Is that an indicator of a ‘solid’ team defensive approach within the defending Final Third (particularly the 18 yard box)?
  • I think so; another supporting indicator might be Blocked Shots; more to follow on that…
  • On the other end of the spectrum, France opponents took shots 33% of the time they completed a pass within the Final Third.
  • In other words, with just ~40% of the possession, the opponents of France were quick to take their chances… perhaps that’s an indicator that they weren’t given a lot of time and/or space?  Or is it simply down to impatience?
  • Others not yielding many shots taken, per penetration, were Chile, Argentina, Japan, Uruguay, Croatia, Brazil, Nigeria, Korea and Ecuador.

Shots Taken: (Supplemental PWP data point)

  • Spain and England faced the fewest shots taken of any teams in the World Cup.
  • Spain opponents averaged 8.33 shots per game and England’s averaged 8.67 shots per game – yet both failed to advance?
  • When they got exposed, they got exposed big time.
  • I’m not sure there is a way to quantify mental lapses but a good indicator to me that the balance of players in the back-four, for a team, is not good, is when they have high levels of possession in attack, high level of passing accuracy moving forward yet face few shots taken.
  • I talked about that in my previous post on APWP; perhaps??? this is another supporting indicator that helps point out that both England and Spain didn’t test themselves and push the fine line far enough between brilliance and boring.
  • Put another way perhaps???
  • Might this also reaffirm, that at least for Spain and England, the goals scored against were more influential in them losing than the goals scored for in winning?
  • On the flip side – the United States and Ecuador both faced over 18 shots taken per game…
  • So the United States not only faced opponents with high amounts of possession, high levels of passing accuracy, and high levels of penetration – they also faced the most shots taken – yet they advanced!
  • Is that great goalkeeping or good luck?  I think I’ve asked that question about the Americans before…
  • But before moving on – both Colombia and Greece were also in the top ten for shots faced – all the others with high shots faced did not advance.

Opponent Shots on Goal per Shot Taken: (PWP data point)

  • Remember that Colombia were in the top ten for shots taken by their Opponent…
  • Well that higher amount of Shots Taken did not translate to a higher amount of Shots on Goal – they were 4th best in the fewest Shots on Goal versus Shots Taken.
  • And a good reason why is they had the highest average in Blocked Shots of their opponent; 6.33 to be exact.
  • Brazil lead all teams in the fewest Shots on Goal per Shots Taken by their opponent; they were also third best in blocking their opponent shots.
  • In looking at the top ten; only Bosnia-Herzegovina and Croatia where in the top ten that didn’t make the round of 16.
  • And of the top 16 teams in this category, only Korea is in that pack who didn’t qualify for the next round.
  • A pretty strong single indicator, with the exception of Algeria and Switzerland, who were in the bottom five for this category.
  • Five goals against to France certainly didn’t help the cause for Switzerland.
  • As for the United States, Netherlands, and Greece?
  • Those three teams, where the opponent had high numbers in possession and passing accuracy, saw all three in the top 15 (defensively) for this category.
  • With the United States offering up 4 blocked shots per game and Greece averaging 4.33 shots blocked per game.
  • Netherlands, who had one of the 4th lowest shots taken against, was 11th best in minimizing shots on goal per shot taken; their blocked shots were only 2.33 per game (midway in the pack).

Opponent Shots on Goal: (Supplement PWP data point)

  • Only one team in the top ten, who faced the fewest Shots on Goal did not advance, England.
  • They faced the 3rd fewest Shots on Goal while also seeing 35.26% of their opponents Shots on Goal net goals.
  • A worthy note here is that England only averaged one Opponent Blocked shot per game – and ironically enough Spain was worst – averaging just .33 Opponent Blocked Shots per game.
  • The two surprise teams kicked out of the World Cup were the same two teams with the lowest amount of average Opponent Blocked Shots.
  • Other teams who moved on that had high Shots on Goal Against were Nigeria, Switzerland, Algeria and the United States.
  • If Blocked shots has value as a supporting indicator then Nigeria, Switzerland and Algeria are more likely to lose their next game than the United States.
  • Nigeria averaged 2.67 blocked shots per game, Switzerland averaged 2.33, while Algeria averaged 1.33.
  • The USA averaged 4 blocked shots per game – sign of a swarming defense that really focuses on protecting the 18 yard box.
  • All told, the rest of the teams in the top ten in preventing shots on goal were Brazil, France, Costa Rica, Belgium, Netherlands, Germany, Mexico, Uruguay, and Argentina.

Opponent Goals Scored per Shots on Goal (PWP data point)

  • Two teams in the top ten for this team performance indicator didn’t advance, Italy and Ecuador.
  • The top team with the lowest Goals Scored Against versus Shots on Goal was Nigeria; 7.69%.
  • Others following in the top five were Costa Rica, Germany, Mexico, and Colombia.
  • Both Greece and the United States did well here; they finished 13th and 14th respectively.  Netherlands, a team who gave up quite a bit of possession, was 12th best.
  • The teams with the worst ratio were led by Australia (56.11%) followed by Bosnia-Herzegovina, England, Hondurus and Japan.
  • Brazil was actually sixth worst and Belgium was 13th worst.
  • Might that be a worthy indicator where Chile may “upset” Brazil – or, given the Index information would it really be an upset?
  • How about the United States taking on Belgium?
  • The United States are good at blocking shots – while Belgium seems more inclined to yield space ‘within’ their 18 yard box.   Does that translate to an ‘upset’?

Goals Scored Against (Supplemental PWP data point).

  • Most seem to agree that one of the single greatest indicators is goals against; in looking at the top ten Goals Scored Against Switzerland lead the pack at 2.00 (per game) for all teams that are advancing; and yes that is a bit dodgy seeing as they gave up five goals to France – but it is what it is…
  • Mexico, Belgium and Costa Rica all lead the pack in fewest Goals Against (.33) per game.
  • The only team, not advancing, in the top ten not for fewest Goals Against is Italy (1.00) per game.
  • As for Greece and the United States?
  • Both finished on 1.33 Goals Against per game.
  • Overall, nine of the top ten teams in fewest Goals Against advanced.  And only one team, in the top ten for most Goals Against, advanced; Switzerland – against France.
  • Uruguay was the other team who matched the United States and Greece at 1.33 per game.

Final thoughts on DPWP:

  • The DPWP Index is not intended to be a predictability model; not with Goals Against included; but all told the Index looks very reasonable after just three games – far exceeding my initial expectations.
  • The Correlation to the sum of points earned (R2) is -.7988.
  • The Correlation of Opponent Goals Scored to sum of points earned is -.7366
  • The Correlation of DPWP to Opponent Goals Scored is .7994
  • All told the Correlation of DPWP to Points is the best Correlation.

In closing… Composite PWP:

CPWP INDEX GROUP STAGES COMPLETED

CPWP INDEX GROUP STAGES COMPLETED

Bottom line here is that with just three games played the CPWP Index shows just two teams outside the ‘bell curve’.

Pretty reasonable – and while many may poo-poo Costa Rica belonging in the upper echelon they finished in the top 7 for four of the six team defending performance indicators; while facing opponents who averaged 57.58% of the ball while also completing 82.67% of their passes.

As for the United States, even when removing that late goal by Portugal in the Index analysis, the CPWP for the United States would still be in the negative (-.3120) instead of (-.3596).  I.E. 6th worst and not 5th worst; that goal did impact the results table but really didn’t impact the Indices of PWP.

In thinking about the next round…

These Indices are not predictability indices, with Goals Scored and Goals Against included they can’t be; but… it does provide a great litmus test for showing which teams (and their overall performance) are on form and ‘what form’ / ‘style’ those teams might be playing to.

Given that, there’s a pretty reasonable chance that Germany beats Algeria, France beats Nigeria, Costa Rica beats Greece, Argentina beats Switzerland, and Colombia beats Uruguay.

Toss ups (and indeed what I think will be really great games) are Chile v Brazil, the Netherlands v Mexico and Belgium v the United States.

Chile can win against Brazil given their better than average defending (and) attacking PWP compared to Brazil; in other words Chile are showing themselves to be in better form.

The Dutch have been masters at the counter-attack and are very efficient in preventing Goals Scored Against; that will be a very dangerous game for Mexico!

With respect to the United States?

They have given time and space but still seem to hold on – it’s a tactic oft used by teams who aren’t quite on the same cutting edge as others – they just simply found the right mix to advance; can that continue?

And lest it’s forgotten – when it comes to defending the 18 yard box, no other team was more effective given the volume of traffic by the opponent!

Best, Chris

 

Taking on Portugal… USA – Sunday…

There is quite a bit of information to consider in how difficult this game will be – I’m not sure I’ll scratch the itch about everything but here’s some information I think has relevance for those who intend to watch the game this Sunday.

To begin…

Possession:

The correlation of possession percentage to average points taken in the World cup after yesterday is (R2) .37 – in other words there is no real correlation between possessing the ball and winning – or for that matter drawing.

And maintaining possession wasn’t the issue against Ghana – it was lack of passing accuracy – and that led to poor possession.  So as others may ponder or offer that the US needs to add another midfielder in lieu of Altidore they may want to reconsider.  Perhaps if put this way it may be better understood…

The USA accuracy in passing against Ghana was 73% – fourth lowest of all teams in the World Cup.  Said differently; given the poor level of passing accuracy it is ACTUALLY better for the USMNT not to pass the ball more in the midfield – as more passes in the midfield – with a poor trend in passing accuracy – means more opportunity for the opponent to get a quick counter-attack that is more likely to result in a goal scored against.

A bit cynical but perhaps more accurate?!?

Formations:

The formation that averaged the highest amount of possession percentage so far was the odd one labeled a 4-2-1-3; run by Croatia in their 4-nil thrashing of Cameroon.

Next up in the formation scheme with the highest possession percentage was the 5-3-2; which when in attack usually takes the form of a 3-5-2.  Both Mexico and Argentina have been very effective in that approach.

Third in the overall formation scheme is the 4-2-3-1 (also the most popular formation).  And, oddly enough, the amount of possession percentage that teams have had running that formation gets as high as 70% (for Nigeria) and as low as 29.75% (for Iran)

Another oddity is that the team most successful in running the 4-2-3-1 is Columbia; they have six points in two games – and their average possession percentage for those two games is 46.84%; again indicating that the 4-2-3-1 is not indicative of a game where midfield possession has value.

And the team with the most possession, in the 4-2-3-1, is Nigeria (70% possession) and they could only muster a draw against another team also running the 4-2-3-1; Iran.
If anything – this ‘formation’ is highly deceptive and it really doesn’t indicate – or even hint at meaning that the team running it will have more possession in the midfield.

The Diamond 4-4-2:  

So far three teams have employed this ‘formation’ and both Uruguay (game #2) and the USA (game #1) won their games with less than 38% possession.

In both cases – both teams were playing against other teams known for possession-based play in the 4-2-3-1  (Ghana and England).

So what about Chile, the other team to employ a Diamond 4-4-2; they ran that formation in game one against Australia and took three points while having 66% of the possession.

As for Portugal – they ran a 4-3-3; which – if you follow soccer pretty closely – is a very close cousin to the 4-2-3-1; a primary difference for some folks is whether or not the Head Coach wants to advertise playing a single pivot central defending midfielder or a double pivot central defending midfield pairing.

Viewed either way it usually means there is one true forward on the team and any number of multiple players who can act as a #10 or #8.

Bottom line here is that the formation that is publicized, prior to a game, really has no bearing on what style of approach a team might take.

An approach – keep it simple…  

The less some players have to think about on the pitch – the better.

In other words instincts built up over time (repetitive training) suggest Klinsmann will run Dempsey up top with another striker/forward; whether it’s Wondolowski or someone else really doesn’t matter.

The key is keeping it simple; if the player can afford to think less about positional play that opens the player up more to think about creating and using spaces (knowing) that his partner will be near-by — and vice versa.

The Midfield…

In looking at the first game it seemed pretty clear that the lone attacking midfielder (really) was Michael Bradley.

Bedoya added value in attack but his presence was more about defending the midfield and supporting the back – four; recall that the real wide right pressure actually came from Fabian Johnson and Graham Zusi after Ghana had equalized.

And remember a wee bit ago – Klinsmann wanted ‘game changers’ to be available on his bench – Zusi was a game changer; not a soldier…  if you run the game changers through the normal run of the game they get tired and can’t add that value when others are tired…

Next up was the tandem of Jones and Beckerman – the good thing here was both players are used to playing narrow and both players could rotate into a double pivot or, individually, control a single pivot system – flexibility….

And with Beasley, out wide left, and Johnson, out wide right, the speed of those fullbacks allowed Jones and Beckerman to drop deeper to clog the middle when Ghana had the ball.

The challenge, in all that, was dealing with crosses – as expected – in that narrow formation – Ghana offered up 38 crosses; with none of them ending up being an assist.

So what about Portugal?

In their game against Germany they offered up 21 crosses – and like Ghana – none of them ended up being an assist.

They have a stud up top named Ronaldo; all hands on deck for this guy – but perhaps the greatest danger he offers is his ability to create space for others.

A tighter back-four with support from the Midfield should help minimize those open spaces; but if the USA commits too many players behind the ball they, then, minimize their counter-attacking opportunities.

And lest it’s forgotten, that first goal against Ghana really came from a quick penetration (when the Ghana defense wasn’t set) – exactly the type of scenario you look for in a counter-attack out of the Midfield…

In closing…

Klinsmann will go with what he thinks best suits the scenarios he wants to work from in beating Portugal.

However viewed it is likely he goes with simple, strong and steady to start the game and then, flash-and-dash with perhaps some panache, to finish the game, depending on score-line.

For me that means two strikers starting the game – perhaps Wondolowski, this time, to pair with Dempsey?

More to follow…

Best, Chris