Fulham’s record when scoring, and not scoring. Also just how bad are our defenders?

Thanks to Rich for graciously letting me guest post.

Of all the issues plaguing Fulham this season, two have stood out the most for me. First is Fulham’s knack for getting a result when scoring 2 or more goals (and lack of results when scoring 1 or less), second is which defenders have been privy to the various shellackings we’ve received so far.

Fulham’s inability to get a result when scoring less than 2 goals is no secret considering our terrible defense. Chances are we’re not scoring a lot is because we’re chasing games, and it’s often more difficult to score when a team is having to chase. Additionally, something I’ve been saying all season is that a club of our aptitude can’t expect to get a result when we’re not scoring. We could under Roy, we can’t now.

Our record when scoring less than 2 goals:

Team Pld GF<2 W D L F A GD GFA GAA   PpG Pts
Fulham 20 15 2 1 12 9 35 -26 0.6 2.33   0.46 7

For context, the league average for this is .811 PpG; and 9.45 Pts. So we’re 16th in Total Pts, but third worst (Swansea and WestHam are T-1st) in PpG. Taking a step back, we’re T-13th in amount of games when scoring 1 or less goals, but our terrible PpG shows our inability to “keep it close” and “grind out” draws: in 2008/09 we had 10 draws when scoring less than 2; in 2010/11 we had 12. This season we have 1.

Now here’s our record when we score more than 1 goal:

Team Pld GF>1 W D L F A GD GFA GAA   PpG Pts
Fulham 20 5 4 0 1 12 7 5 2.4 1.4   2.4 12

Again for context, the League average is 2.4 Ppg and 12 Pts. Right on par considering we’re also T-13th in this category. This stat also shows how terribly our defense has been regressing: in 2008/09 our GAA when scoring more than 1 was .75; under Sparky it was 1.08; even Jol’s first season was a .75(!) Last season was an atrocious 1.66, highest since the Coleman days.

So basically if we score more than 1 goal, there’s like an 80% chance we’ll win. If we score less than 2 goals that drops to 15%. (Now this is when the likes of StatsBomb and James Grayson stumble upon the site and rightfully lambaste me for my terrible arithmetic…)

Second is the Plus−minus (+/-) of the club’s Defenders. This stat is widely used in hockey, and even stretching to sports like Basketball and Ultimate Frisbee. It’s a bit incongruent in footy because of the numerous variables that don’t exist in the former sports (main being the lack of timeouts/substitutions, which can create wide variances). But there have been moments this season when we’ve thought to ourselves “Boy, Richardson is always playing when we’re losing a lot” or “Actually, Senderos might be better than we think”. Both are actually true.

Player Appearances +/- +/- pG
Briggs 2 0 0.0
Hangeland 8 -4 -0.5
Reither 16 -6 -0.4
Senderos 11 -6 -0.5
Riise 9 -12 -1.3
Zverotic 4 -14 -3.5
Richardson 15 -15 -1.0
Hughes 13 -15 -1.2
Amorebieta 14 -18 -1.3

Players like Amorebieta, Hughes, and Richardson have been on the field when the opposition is scoring a lot. Hangeland, Reither, and Senderos haven’t. Again, this is quite imperfect considering how awful the team was up until about a month ago, and the symbiotic nature of the game: defense is really only as good as the midfield, which is only as good as the forwards. Also, if we don’t ever see Zverotic again this might be why.

At some point I’ll crunch the numbers for our midfield, and our defenders of past seasons. But this is still quite telling: Hangeland and Senderos need to get healthy ASAP; and boy oh boy do we need a good Left Back.

5 thoughts on “Fulham’s record when scoring, and not scoring. Also just how bad are our defenders?

  1. Interesting stuff. But I think the defence stats at the end are skewed unfairly against Richardson by the Hull result. He was playing midfield in that 6-0 defeat. (I am assuming that game is one of the 15 games you include – if not ignore my comment!). So perhaps it should have been -9 over 14 games instead?

    Lucky that Riether wasn’t playing against Hull since then his +/- score would have plummeted too (down to -12 the same as Riise) and we would be saying we would need a right back as well. I am discounting the fact that he would have made that much difference to the scoreline.

    1. Good catch, but he substituted in for Karagounis at the 74 minute mark. So he only got a -1 for that game; but I suppose he could easily count as a MF in that instance.

      Should have elaborated above, but I went through each game and didn’t count the overall score in relation to whether or not a player appeared; the plus or minus only gets tallied when Fulham or the Opponent score *when the Player was actually on the field*. So take the City match: both Senderos and Amore a -1, as Senderos left when it was 1-0 City, and Amore was on when we scored 2 and conceded 3.

  2. Interesting stuff. I’ve been thinking that it would be neat to try to come up with useful +/- numbers based on shots for/against (since there’s a bit more data than goals, and it still correlates with success). I’ve also wanted to do things like look at how certain midfield combinations or certain fullback/wing combinations affect short +/-, to see if it produced anything useful.

  3. I’d be very wary of using +/- to evaluate footballers. Despite its ubiquity in hockey it’s considered a flawed stat that is generally worthless unless you’re talking about a very large sample size, and 16 games is not that.

    1. That is true–it’s fairly flawed, which is why I added the disclaimers. You are correct in saying 16 games isn’t a big sample size, but I’d argue that conceding a league-leading 46 goals against is plenty of data to see who’s doing what.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s