2010 Retrospective: Point Efficiency

June 22, 2010 - Pretoria, South Africa - epa02216461 USA soccer head coach Bob Bradley prepares balls for his players during a training session at Eesterust Stadium in Pretoria, South Africa, 22 June 2010. USA will face Algeria in their Group C preliminary round match of the FIFA 2010 Soccer World Cup on 23 June in Pretoria.

Over the long club season players have the luxury of going in and out of form, having hot spells and droughts, and even weathering the occasional negative off-the-field press. When it comes to a national team, however, there is no such luxury afforded – every precious minute is like a miniature sales pitch in which you must convince the coach, and a nation, that you merit being chosen over the interminable number of players lined up behind you.

It would seem, then, that efficiency is key in proving your case. Specifically in the case of forwards and midfielders, when a player is on the pitch, is he at maximum production in terms of goals and assists? Today FFG will scour the USMNT 2010 stat book and analyze the ‘point efficiency’ (PE) of every forward and midfielder who made an appearance in 2010.

Methodology

How do we calculate this stat, you ask? Well, our formula was designed so that even [insert the name of your least favorite soccer thug] could calculate it after gleefully bashing his head against a goalpost. We took each player’s total minutes and divided that number by total points produced (goals plus assists). This yields a number, in minutes, that represents the average amount of time a player takes to produce a point. So, obviously, the lower the number the better.

Yes, this is a highly imperfect statistic. We acknowledge this. A holding midfielder, for instance, doesn’t have as many opportunities to produce points as a striker, and shouldn’t be expected to do as well in this measure.

Plain dumb luck is a factor as well. The 80th minute substitute who knocks in a ‘gimme’ will have an extremely good PE compared to the forward who played the first 80, even if that forward scored. Oh well. In a way that’s exactly what we’re talking about here. It’s been argued time and again that a good player “creates his own luck,” and PE might be a good measure of that.

Without further delay, here are the 2010 PE scores for USMT forwards and midfielders (players with a PE of zero, ie Robbie Findley, are not listed–ouch!).

PlayerPos.Min.GoalsAssistsPointsPE
Diskerud, MikkelM1201112
Agudelo, JuanF3010130
Ching, BrianF9011245
Buddle, EdsonF20821369
Davis, BradM7501175
Gomez, HerculezF16420282
Donovan, LandonF706336118
Beasley, DaMarcusM126011180
Jones, JermaineM180011180
Dempsey, ClintM696213232
Kljestan, SachaM240101240
Altidore, JozyF750213250
Holden, StuartM286011286
Edu, MauriceM558101558
Bradley, MichaelM930101930

Analysis

Odd/Worrying Trends

So what are the biggest insights and surprises? Well, Juan Agudelo and Mikkel Diskerud top the chart with low PE’s. Remember, this is a measure of efficiency and, regardless of opponent, match status, etc. this DOES show how well these players used their time on the pitch. And check out Mr. Buddle’s number. FFG may have to retract some of its harsh commentaries on HIS play. Has Bob Bradley been crunching PE’s?

That being said, some of these numbers are frightening. Jozy Altidore, our “premier” forward, was extremely inefficient in 2010. I could forgive a 250 from a defender who played 750 minutes, but from a striker?

The biggest shocker, however, appears to be Michael Bradley’s score. He’s normally the more attack-minded of USA’s central midfield pairing, so you’d expect him to produce points regularly.

But that’s not the way Bob Bradley’s 4-4-2 works. The central midfielders function as facilitators. One drops towards the center backs to collect the ball. The other inches just a bit higher to serve as a base of operations in the opposing team’s half. Rarely do you find them charging towards the box. Michael Bradley really didn’t do anything wrong in generating a low PE.

Additionally, a center midfielder can make his mark without scoring. For instance, Xavi, widely considered Spain’s best player, generated just one assist and didn’t score any goals during Spain’s World Cup run.

Star Performance

Now let’s take a look at USA’s superstars, Landon Donovan and Clint Dempsey. Donovan, considering his high total minutes, scores reasonably well at 118. As a starter who usually goes the full 90 minutes, this means we can expect a goal or assist out of him every other game. For a coach that’s a pretty comforting number, especially since Donovan played midfield most of the year.

Clint Dempsey’s PE of 232 (a point every third game) is not quite as encouraging. While that may seem low compared to Donovan’s – particularly since Dempsey played forward more often than Landycakes – Clint did have two goals ‘stolen’ from him. He had a goal incorrectly called back for offside against Algeria and Donovan took the PK Dempsey earned against Ghana. Counting those two missed opportunities Dempsey’s score is a more respectable 139.

And, again, because Dempsey and Donovan spend so much of their time in the midfield, creating a goal every other game is pretty good.

Now do you see why Altidore’s 250 is so worrying?

What do you think, gringos? Is point efficiency a useful statistic?

5 Comments Post a Comment
  1. Blake Owen says:

    Good stuff, Ben.

    Donovan and Dempsey were both excellent throughout the year. I think PE certainly shows their quality.

    In January, it’ll be nice to see if Diskerud and Agudelo can contribute on another goal or two against Chile.

  2. Bradley's Receding Hairline says:

    I think it shows overall offensive efficiency. Donovan’s numbers seem absurdly high for someone who plays on the wing in a 4-4-2.

  3. John says:

    I wonder what a hockey-style plus/minus would show. That’s a figure that would also be applicable to the defensive players.

  4. Dave says:

    I think the PE is a starting point for conversation (thus I enjoyed the post), but it also demonstrates the relative lack of useful statistics for soccer. Ideally, I would like a Goals above Replacement (or GAR) similar to baseball’s wins above replacement (WAR). or perhaps an efficiency measure similar to football’s (i.e., NFL) approximate value, described on the web at Pro-Football Reference.

    However, we currently lack the underlying data to determine average results for a particular pass or position, etc. I’m hoping that the MLS partnership with Opta will change that, but at first glance, it seems that the best data is reserved for the editors/sportswriters and not the public.

    • Ben Koch says:

      Yes, FFG would love access to that “reserved” statistical data. Hopefully those of us interested in the intellectual and numbers side of the game will help evolve some measures like those NHL/NFL ones mentioned above.

Leave a Reply to Dave




Powered by WP Hashcash