If you watched the FC Dallas vs. D.C. United game on MLS Live, you got stuck with the D.C. commentary. At first their blatant homerism and misreading of the game was comical. But by the end, the constant insistence that D.C. United was playing a "winnable game" grew wearisome. Here are three things that the D.C. commentary team said that were just plain wrong.
D.C.'s Argentines are just as good as FC Dallas' Argentines
Dave Johnson: "[Mauro Diaz] is developing a partnership with fellow Argentine Max Urruti who was very successful as a super-sub with the Portland Timbers but is now in the starting eleven for FC Dallas."Bruce Murray: "What’s very interesting is that D.C. United has almost the same dynamic duo in [Fabian] Espindola and [Luciano] Acosta. So, two Argentines for us and two for them."
Is the partnership of Fabian Espindola and Luciano Acosta equivalent to the partnership of Mauro Diaz and Maximiliano Urruti?
Let’s look at the Dallas duo first. Diaz and Urruti have played 347 minutes together this year. In the season opener against Philadelphia, Maxi scored from a Mauro pass. Diaz also assisted a Fabian Castillo goal. Against Montreal, the two goals in the game were scored by Diaz and Urruti. Finally, against D.C., Urruti’s goal came from Diaz playing the ball out wide to Castillo, who crossed the ball into the path of Urruti. Diaz also assisted one of Michael Barrios’ goals. Other than the unassisted Barrios goal, every FC Dallas goal this season has involved Mauro Diaz, Maxi Urruti, or both.
As for United, Fabian Espindola has one goal this season, in a game that he entered in the 60th minute to replace Luciano Acosta. Espindola and Acosta combined have created fewer chances than Diaz. This competition is not even close, even when you adjust for Acosta and Espindola having played fewer minutes. I don’t think many people other than Bruce Murray think that Acosta and Espindola stack up to Diaz and Urruti.
The team with more possession should win the game
Half-timeDave Johnson: "Something to ponder – why DC United is trailing FC Dallas 2-0… What we’re talking about here: 62% of the possession for D.C. United. That’s terrific."
Bruce Murray: "FC Dallas has not shown ANYTHING – other than the two goals – but not in terms of the run-of-play."
Does possession win soccer games? I plotted points vs. possession for each MLS team in 2013, 2014, and 2015 to see if there was a correlation.
In 2013, there was a weak correlation between possession and points. See that team at the bottom that had 53% possession but only finished with 16 points? That's D.C. United! Dave Johnson and Bruce Murray should've learned in 2013 that having the majority of possession doesn't guarantee results! If D.C. is treated as an outlier and removed, the R squared becomes 0.47, showing much higher correlation between points and possession.
By 2015, we can see that there is almost no correlation between points and possession. Possession accounts for 4% of the variance in where teams finish on the table. In MLS, possession has become less and less important over the past 3 years. In 2015, the team with 60 points and 55% of possession is the New York Red Bulls, while the team with 60 points and 48% of possession is FC Dallas, showing that there are multiple ways to be successful in today's MLS.
Attempting to predict the winner of a game based on possession percentage does not take into account team tactics. Dave Johnson and Bruce Murray (and Ben Olsen, it seems) fundamentally misunderstand FC Dallas’ desire to concede possession, absorb pressure, and then explode on the counter. In FC Dallas’ 2-0 win vs. Philadelphia Union, Dallas had 48.6% of possession. In the 0-5 loss vs. Houston Dynamo, Dallas had 54.3% of possession. FC Dallas doesn’t really want the ball until the opposing team has moved up the field and there is space in behind defenders (kudos to Owen Coyle for grasping this concept).
54th minute (after third goal)
Dave Johnson: "Well that theory about United and the better possession is… not holding up as well anymore."
Well said, Dave.
3 goals down in the 66th minute is still a winnable game
Regardless of how many goals his team conceded without scoring, Bruce Murray continually thought that D.C. United could win the game. What is the likelihood that United could come back to win when he made those comments?
To find out, I consulted an Outcome Probability Calculator, which gives the chance of a win, loss, or draw for a club depending on the venue (home or away), the minute, and the goal difference in the game. I used Ford Bohrmann's model, found at www.soccerstatistically.com, which includes data from over 4000 EPL games.
Start of Second HalfBruce Murray: "There's no question. This is a game that is winnable for D.C. United."
In the 46th minute, a team at home losing by 2 goals has a less than 3% chance to win and less than 10% to draw. In 87.23% of games in that scenario, the home team lost.
66th minute - After PK is awarded and before it is takenBruce Murray: "Again, I keep harping on it. I'll harp on it. This is a winnable game!"
In the 66th minute, a home team losing by 3 goals is doomed, with no chance to win and a 99.18% likelihood of a loss. However, approximately 77% of penalty kicks are scored. Murray was probably assuming that Espindola would score. At 2 goals down in the 66th minute, the home team has a less than 1% chance to win, a 7.4% chance to draw, and a 91.86% probability of losing. Bruce Murray probably picks the horse with 100-1 odds to win the Preakness every year.