Editor’s note: this article was originally published at Sons of Penn on February 28, 2016, and is being used with permission.
So you want to make the playoffs in the Eastern Conference, eh?
Here’s what the Philadelphia Union must do to make the playoffs within the framework of statistical analysis of the Eastern Conference.
The 2015 Eastern Conference table
The teams in bold went to the playoffs; the teams in italics did not. What do you notice right away? What are some characteristics playoff teams show that non-playoff teams do not?
- Teams who went to the playoffs in the East had a positive or nearly-positive goal differential. Teams that went to the playoffs had an aggregate GD of +27. Teams that didn’t had an aggregate GD of -47. Playoff teams simply scored more or nearly more goals throughout the season than they gave up. If you do that, you win games. There’s a couple ways to do it; you can prevent teams from scoring and score the occasional goal yourself, like DC (only 45 goals allowed, with 43 goals for) or you can do it through sheer offensive firepower like Toronto (58 goals allowed, but 58 goals scored, for an even goal differential). Speaking of winning games…
- Every team that made the playoffs in the East won at least 14 matches. Again, common sense, but wins are worth three points and draws are one; that’s a big swing. It doesn’t matter if you win 4-2 or 1-0, winning 14 matches gives you 42 points off the top, with 20 matches to factor in for possible draws. Only one non-playoff team (Orlando City) cracked the 42 point barrier, and that was with wins and draws combined.
So score more goals than you give up for the season, and you’ll probably win at least 14 matches and make the playoffs. It sounds simple, but as Sun Tzu once said, “one may know how to conquer without being able to do it”. The Union scored 42 goals and gave up 55 last year for a -13 goal differential, second-worst in the East. That’s a mighty big difference. How do you close that gap? How do you get more points out of each game? Coach Jim Curtin believes you start with defense.
In a series with Philly Soccer Page (which is well worth the read), Curtin talks about how defense is very important, and he cites some stats:
“Interesting stat, and it’s an obvious stat when you hear it: In games where teams kept a clean sheet last year, they averaged 2.5 points,” Curtin told PSP. “That’s common sense. You don’t give up a goal, more likely to win. The only other thing that can happen is a 0-0 draw.
“In games where you give up a goal, the average is 1.2 points. The focus has to be not conceding first. We’re going to score, other teams are going to score, but, simply put, keeping a zero is the easiest way to get points at the end of the day. It’s an obvious one, but it’s one that jumps out at you.”
Okay, there’s a lot to unpack there.
- A clean sheet means more points. An Eastern Conference team recorded a clean sheet 72 times last year, or 21.1% of matches. That’s a little more than 1 in 5, or an average of about 7 clean sheets a year per team.
- Giving up a goal means less points. An Eastern Conference team conceded at least one goal 78.9% of the time last year.
- The focus is on not conceding first.
The third point is particularly interesting. When it comes to winning a match, does it matter who scores first?
We charted all 34 matches for all 10 Eastern Conference teams by who scored first, the final result, and the average points gained in various situations (such as scoring first or going up two goals). The following is what we found.
First, let’s look at points garnered per game in which a particular team scored first, versus the points gained when they conceded first.
There’s a lot to be said for getting off to a good start
Right away you can see a couple things. Overall, there is a significant discrepancy between the points gained in a game where a team scores the initial goal (PPG/INIT) versus points gained per game in which a team concedes the initial goal (PPG/VS INIT). Does scoring first entail success? Teams who scored first averaged 2.14 points gained from those matches; teams who conceded first averaged 0.70 points from those matches. Something else to consider: every team (RBNY, Columbus, Montreal, and New England) that scored first at least half of the time (INIT column) in the 34-match slate made the playoffs.
“But Jay” you say, “of course the points gained in matches in which a team scores first will be much higher due to the games in which the opposition doesn’t score!” Quite right! So let’s take a look points per game in which a team scores first but then concedes at least one goal. Is scoring first still important if you eventually concede?
Teams who scored first but eventually conceded still did better per game.
The answer in this case is yes. The first entry (INIT/CONC) is the number of games a team scored first minus the games they scored first and kept a clean sheet. The second (I/C PPG) is their points per game in those situations. In the Eastern Conference last year, teams who scored first but eventually conceded in a match still averaged about a point more (1.68 vs 0.70) versus a match in which they conceded first.
So it’s important to get the first. Scoring first lets you dictate the game; everything from the opposition’s play style (defensive clubs will have to open up to find an equalizer) to the substitution patterns for both teams relies on the score. Getting that first goal is a big advantage.
We’ve seen the effect of scoring first, both in keeping a clean sheet and scoring first but then conceding, and we’ve seen that scoring first allows a team to gain more points on average per match. But how good is a one goal lead? What about that second goal? How important is it to push the margin to two?
Teams who pushed the margin to two goals averaged 2.77 points gained from those matches.
Teams who had a one-goal lead at any point (UP 1) in a match averaged 2.22 points for that match (U1 PPG). Teams who had a two-goal lead (U2+) at any point in a match averaged 2.77 points gained from that match (U2+ PPG). That two-goal lead doesn’t guarantee a win, but as it turns out, it comes pretty close.
Here’s the averages in various conditions in one table.
PPG in various conditions
Note that teams gained more points from matches in which they gained a two-goal lead than kept a clean sheet; for a team who focuses on defense (such as the Union) it may however be easier to get a clean sheet than take a two-goal lead.
Let’s take what we’ve learned and apply to what the Union must do to be successful:
- Score first at least half the time (17 matches). The Union were actually not that far off from doing so last year; they scored first 15 times, good for fifth overall in the East. That’s three more times than Toronto FC and two more times than DC United, both playoff teams.
- Find a second goal. Every playoff team in the East averaged at least a point in matches in which they conceded a goal. Every non-playoff team was below a point a match. With respect to Jim Curtin, it’s just not possible to win all your matches 1-0. We’ve already discussed the benefits of scoring a second goal and taking a 2-0 lead. They need to be able to win the 2-1 games too.
- Turn leads into wins. Scoring first wasn’t a big problem last year. No, the two big problems last year were the defense and not scoring a second goal. The Union played 18 games in which they lead at any point last year; they allowed equalizers in half of those matches. Only NYC (58.8 percent) and Chicago (64.7 percent) had a higher percentage of matches in which they once lead but then lost the lead. Leads need to turn into wins, not draws or losses. Speaking of which…
- Handle adversity better. The Union folded like an accordion last year when teams equalized against them. Out of the nine games last year in which they allowed an equalizer, they got six points total. That comes out to 0.67 per game, and that’s good for dead last in the East (average is 1.18 points in such a situation). In games in which they allow a goal (and in games they don’t allow a goal too, for that matter), they must find a second tally. Scoring one goal a game all year won’t cut it.
- Fight back. Furthermore, the Union went down a goal in 23 of their 34 matches last year. Only the lowly Fire went down in more matches (27). They got nine points from those 23 matches. That’s 0.39 points per match in which they were losing at any point, which is (you guessed it!) last in the East (the average was 0.62). To state it another way, the Union were not good at keeping a lead and not good at coming back from a deficit. It’s hard to make the playoffs like that. They need to get better at doing both.
The Union had a below average offensive output last year (42 goals for, average was 49.7) and allowed more goals than the average team (55, average was 51.7). Is the midfield improved enough that they can score enough themselves along with setting up Sapong to score six more goals than last year? Is it possible the defense is improved enough to allow six fewer goals than last year, with Blake in net and a seasoned vet as his backup, and with a reconstructed back line? It’s unlikely, but those changes would put them at a -1 goal differential and in the running for a playoff spot.
A Possible Playoff Path
Let’s figure out a path to the playoffs for the Union now, using what we know. (Yes, that’s a trolley that just went by and yes, you have entered the Neighborhood of Make-Believe. Just roll with it.)
Hypotheticals: the Union have improved to the conference norm in both points garnered for games in which they score first as well as points gained on average when conceding first. All aboard the trolley.
First, we have to figure out how many points from games they need. Going back through 2013, every single team in the East who got to 49 points would have, in the current format, gone to the playoffs. So let’s set our target at around 49.
We’ve established that scoring first is good for your point total; a team who scores first gains 2.14 points on average from that match. We also saw that the four teams who scored first at least half the time went to the playoffs. So let’s score first half the time.
17 1-0 leads multiplied by 2.14 points per match = 36.38 points from those 17 matches. Let’s assume for the moment those 17 1-0 leads turn into 11 wins, 3 draws, and 3 losses, for 36 points.
That leaves 17 times when the other team scores first, which carries with it a 0.70 average of points gained.
17 1-0 deficits multiplied by 0.70 points per match = 11.9 points from those 17 matches. Rounded up to 12, that’s 3 wins, 3 draws, and 11 losses in those circumstances.
Added up, that’s a 14-14-6 record, good for 48 points. Change a one-goal loss to a draw and you have 49, and you’re in the playoffs. Again, it sounds easy, but this is assuming the Union will both find a little more offense and concede a bit fewer goals, which would conceivably make any team better. Still, this is one potential path to the playoffs.
The Union’s success this year rests on their ability to score first and hold onto leads, as well as their ability to equalize when they go down. If they can do those two things, they’ll be in the playoff picture. If they reprise the familiar practices of losing leads at a high rate and being unable to equalize once they go down, 2016 will look much like 2015.