Let’s forget Fulham’s 2015-16 forgettable season


(by timmy; apologies for mislabeling when Kit was fired/Joka hired on a few of these images)

The 2015-16 Season can be told in 3 chapters: Kit, Interim, Joka. Throughout these chapters was the same bad stats, poor defense, inconsistent offense, and otherwise anonymous cast of characters.

Kit was kept on as manager because he initially did extremely well taking over for Magath’s Reign of Terror. Things regressed heavily toward the end, as we crawled toward the finish, but we felt most of that could be attributed to the poor squad composition and PTSD from Felix. I dare you to visit a match report from late 2014-05 and not laugh at the starting XI’s. That wasn’t a team, it was a casserole.

This season was supposed to be different. Summer signings like Tom Cairney, Ben Pringle, Jamie O’Hara, and failed attempt at Lewis Dunk showed the club’s movement toward analytics and the process, over big names well past their prime .

A full offseason of Kit’s philosophy would undo whatever damage and upheaval the past 18 months had left.




But, things never really took off. Inconsistency reigned supreme. The defense kept shipping goals. The offense made up for this with an astronomically high conversion rates, but most knew it couldn’t last. Soon it became evident Kit simply wasn’t the right man, and probably shouldn’t have been kept on over the summer.

There was an interim period where under the guise of Alan Curbishley(?!?) and Peter Grant, furtive gestures were made to Steve Clarke (got fired anyway!), Nigel Pearson (wanted his assistants still at Leicester to come with; good thing they stayed put!), before finally convincing Slaviša Jokanović to leave Israel for London.

Joka corrected the defense, but the that offensive regression we worried about came with a vengeance. A league-high 36% Shooting Percentage under Kit would tumble to a league-average of 29%. DangerZone shots for would also decrease from 36% to 32%, with about a half-less shot in the DZ per game (when you’re only attempting 4 to begin with while giving up 6, that’s a big drop).

SoT Share

Rolling Sh%.png

Shooting% For

Three wins in a row for the first time since 2012 (was that the year?) would be immediately followed with the wheels falling off the bus. 13 goals against in final 5 games. Opponents shots surpassing 19 on 4 of the final 6 games.

The task at hand, already appearing daunting, requires more work than most suspected. Unless Joka can work magic this summer, a long stay in the Championship beckons. The pieces are there, but several of the large ones must be sent away in exchange for additional parts.

dz fordz against

On Saturday this not particularly fun or noteworthy season ends.

Let’s analyze Slavisa after just 7 games!

(by timbo)

As I mentioned yesterday, Fulham’s TSR has increased whereas the Shooting % has decreased in Slavisa Jokanovic’s 7 games in charge. This is a very small sample size, and barely half of what Kit’s is, but it’s still illustrative of the changes he’s made.

Below are four graphs. The first set is Fulham’s TSR and then PDO under Kit, the interim period (weren’t like 3 people in charge then, including Alan Curbishley?), and Slavisa.

The second set is Fulham’s Shots on Target Share (i.e. our % of a match’s total shots on target; if we have 3 and our opponent has 7, we have 30% SoT Share) and Shooting % (how many shots on target result in goals). There’s also our Danger Zone Shots For % and Danger Zone Shots Against %.


Notice the TSR uptick under Slavisa? Despite our overall bad TSR metric (see yesterday’s post), we’re actually trending upwards, ever so slightly.

The defense, Kit’s achilles heel, is actually improving ever so slightly. We’re giving up 1 less shot in the Danger Zone per game than before (6.3 to 5.4). That might not sound like much, but our offense was carrying this team in a very unsustainable way.

And the second set of images:

Shooting % & DZ %.png

Holy crap our shooting has nosedived, especially in the Danger Zone. This may contradict what I wrote about our TSR improving, but that’s why it’s helpful to look at these things in many different ways.

Part of this is because:

  • We’ve played some very good opponents since Slavisa took over. Hull are currently in an automatic promotion spot. Derby and Sheffield Wednesday are currently in the playoffs. Wolves were a top-10 team in mid-January. Blackburn, despite having the then-tied-for-worst-offense, has a top 6 GA defense.
  • He’s kept the same formation between matches just once (Kit changed formations once in his 16 games in charge).
  • If you count his first game as zero, he’s averaged 2.6 changes to the starting lineup between each match (Kit averaged 1.5).
  • Ross McCormack missed a match where we had but 9 shots total, 2 on target (remember, this is only a 7 game sample size).
  • Just watching the team you see a new offense: whereas before the offense was an open, exciting/frantic give-and-go system that too often exposed the defense; now we’re possessing the ball and playing with a high line.
    • On average we’re still shooting the same amount of times, but far less on target and much less in the Danger Zone. This isn’t surprising when you introduce an entirely new offense half way through the season.

Slavisa was given half a season to try different things to see what will work for next year. He’s clearly attempting that, and some metrics are taking a slight hit.

It will be fascinating to see where these numbers, and the team’s results, go in the remaining 15 matches this season.


Championship Advanced Stats

(by timbo)

I’ve been keeping advanced stats for Fulham all season long, but I’m happy to announce that I’ve finally figured out how to tally up data for the entire league. Here are the fancy numbers, via WhoScored:

# Team TSR SoT Share Shooting% Save % PDO
1 Hull 0.584 0.621 31.1% 79.3% 1.105
2 Middlesbrough 0.533 0.563 28.9% 84.8% 1.137
3 Brighton 0.545 0.546 29.2% 73.7% 1.029
4 Burnley 0.450 0.510 36.4% 79.5% 1.159
5 Sheffield Wed 0.508 0.555 37.1% 68.9% 1.060
6 Derby 0.543 0.554 31.0% 74.0% 1.050
7 Birmingham 0.476 0.486 30.4% 77.3% 1.077
8 Ipswich 0.515 0.542 28.0% 67.8% 0.957
9 Cardiff 0.475 0.502 30.2% 73.6% 1.038
10 Notts Forest 0.525 0.533 21.2% 78.9% 1.001
11 Preston 0.490 0.507 26.4% 72.9% 0.993
12 Wolves 0.471 0.451 32.3% 72.2% 1.044
13 Brentford 0.472 0.465 30.2% 70.0% 1.002
14 QPR 0.518 0.518 28.5% 68.6% 0.971
15 Reading 0.618 0.592 23.9% 65.3% 0.892
16 Leeds 0.479 0.430 29.6% 73.8% 1.034
17 Huddersfield 0.542 0.523 30.9% 64.5% 0.954
18 Blackburn 0.529 0.578 22.2% 70.7% 0.929
19 Fulham 0.457 0.463 34.1% 66.9% 1.009
20 MK Dons 0.452 0.415 25.0% 70.2% 0.952
21 Bristol City 0.474 0.504 22.3% 59.7% 0.820
22 Rotherham 0.492 0.470 28.3% 61.5% 0.899
23 Charlton 0.386 0.435 21.7% 64.1% 0.858
24 Bolton 0.481 0.452 24.4% 65.1% 0.895

Unfortunately I don’t have a sorting function handy (feel free to copy/paste into your own excel or google spreadsheet), but here’s where Fulham are overall within each metric, with notes in parenthesis:

21st overall in TSR (this means we’re getting outshot a lot, and it’s happening more often than the rest of the league. This is not news.)

19th overall in Shots on Target Share (we’re giving up less shots on target than we’re giving up in total, but 18 other teams are both shooting and conceding on target better than we are.)

3rd overall in Shooting % (our offense is scoring a lot of goals despite not attempting many shots on target. This is probably due to the system we were running under Kit, but that is material for another post. That said Experimental361 dubbed us “Languidly Clinical”—so I guess the ghost of Berbatov still lives with us!)

18th overall is Save % (our Save %, which is essentially ‘Goals Against’ divided by ‘Shots on Target Against’, is where it is because our defense is so atrocious. If you look at the current table you’ll see only Rotherham & Charlton have conceded more goals. The fact this metric is not worse is, in my opinion, a credit to the play of Andy Lonergan.)

11th overall in PDO (this is hard stat to compare each team with, but as the mean PDO is 1, we’re basically where we should be in the table because of our, in the context of of our defense, astronomical Shooting %. Yep, things could be worse.)

One silver lining is that our TSR has been increasing steadily since Slavisa Jokanovic took over. However, our Shooting % has nose dived—presumably because he’s running an entirely different tactical setup.

I’ll write another update this week discussing his first seven games in charge, especially comparing things to Kit and whoever was in charge during the interim period.

Why Kit Had to Go

(by timbo)

What to say about Kit Symons that hasn’t already been said? We got our immediate bump when he replaced Magath, but have essentially plateaued for the past calendar year.

Our offense is on fire and something to behold when fully functioning. The one-two passing and fluidity were such a reprieve from the nebulous Magath and static Jol era. And it worked: currently we lead the league in goals scored and shooting percentage. Our PDO is second to Boro.

Yet this type of offense would be Kit’s undoing, as the defense would be exposed whenever our midfielders moved upfield. It was also due for a major regression (those numbers weren’t going to keep up with a defense this bad).

Speaking of, stating our defense is bad is now akin to saying iTunes sucks. It’s been so bad for so long we’ve sort of just accepted it.

Ted Knutson, creator of of StatsBomb.com and now analytics honcho at Brentford, came up with a theory that relegation candidates often average 16 shots against per game.

Although this is presumably for the EPL, and things in the Championship are slightly different, but if we still apply that metric Fulham were averaging about 14 shots against per game this season. Last season was over 15.

Additionally, Fulham’s Danger Zone Against percentage (this being shots that are essentially point blank) this season is 42%. Last season, when I last calculated it in early April, it was 41.9%.

Despite all the changes in personnel, there hasn’t been much systemic improvement with our defense.

So Kit had to go. It’s a shame as he seemed like a genuinely nice person. And one thing becomes evident when looking at the numbers: if we wanted to remain a mid-level Championship club, Kit would have been a fine manager.

But we don’t want to be mid-level Championship side, let alone remain in it.

We want to be in the Premier League.

Those final 15 minutes vs Charlton

(by timbo)

Sometimes leads are blown because the lead was perilous to begin with, a result of some fluky behavior and good fortune, and it catches up with the team as some sort of metaphysical reckoning.

Other times it’s bad luck or a momentary lapse in judgement, a result of some cruel joke by Loki.

And then there are other times when you’re two goals up with 10 minutes plus stoppages to go. Then but fifteen minutes later,  you are no longer two goals ahead but instead level with your opponents. And it’s because you made a Gamble when you should have probably just, oh, I don’t play cards but you know what I mean.

In the first 80 minutes of the game, this was Charlton’s shot chart:


In the final 15, it was this:final15

Numerically, they had slightly more shots in the first 80 minutes. But only 4 of the 12 were in the Danger Zone, and most were speculative long rangers.

In the final 15 minutes, 7 of 9 (not the Star Trek character), nearly 80%, of Charlton’s shots were in the Danger Zone, basically point blank.

Here are their shots in this span, logged by minute:

81(g), 82, 85, 86, 88, 88, 90, 90, 96(g)

I don’t normally like to speculate on substitutions, but Cauley Woodrow is worth looking at. Kit gambled, and it almost paid off after clearly not paying off and then eventually did not pay off at all. In the 88th minute The Gamble could have very easily could have iced the game for us had an unfortunate deflection not nicked his shot wide. But, the shot didn’t go in. There was no goal and no ice and no two goal lead with minutes remaining.

Up to that point Charlton had gotten a goal back and attempted 6 shots, 4 within the DZ. After The Gamble they’d attempt 3 more, all in the DZ, and one would score.

A second defensive substitution was not made until Mattila came on in the 90th, about ten minutes in to Charlton’s blitz of point-blank shots.

Fulham 2015-16 Stats

Timbo here!

Here’s a link to a google spreadsheet I will be updating with data from each match (or, I hope to keep updating): Fulham Data.


  • It’s still very early, so there’s been a lot of wacky variance within and between each match (see: the PDO in our opener, our Sv% at Hull and then vs Huddersfield…)
  • Our PDO is about at league average, so we’re not lucky nor unlucky.
  • Despite getting, for all intents and purposes, a new back line our defense is still abysmal. Nearly 50% of our opponents shots are in the ‘Danger Zone’ (80% of Sheffield Wednesday’s were in the DZ!), and we’ve been outshot in all but two matches.
  • Our offense is much better, with marked improvements in Shooting%, DZ Shots For Total, and DZ Shots For %.
  • It’s still very early, so there’s been a lot of wacky variance within and between each match.

Once data for the entire league becomes available I’ll post that so we can see where we stand in comparison to others.

Final Data Dump of 2014-15 Season

Well the season is long in the books, and Rich has been doing some wonderful analysis so far. I just wanted to add the final data dump of the season so we can look back, laugh, cry, and whatever else we so desire.

Week 5 77 65 0.542 18 24
Week 10 133 125 0.516 44 44
Week 15 194 206 0.485 68 65
Week 20 272 282 0.491 92 94
Week 25 324 372 0.466 106 118
Week 30 380 470 0.447 127 151
Week 35 451 538 0.456 143 177
Week 40 533 632 0.458 167 220
Week 46 601 712 0.458 193 246
Week 5 0.429 16.66 58.33 74.99
Week 10 0.5 27.27 54.55 81.82
Week 15 0.511 32.36 56.91 89.27
Week 20 0.495 32.61 58.51 91.12
Week 25 0.473 33.01 62.71 95.73
Week 30 0.457 33.85 64.91 98.76
Week 35 0.447 32.15 66.66 98.82
Week 40 0.432 30.54 67.27 97.81
Week 46 0.44 32.12 66.26 98.38

This last set contains 6 matches instead of 5, so the numbers may be skewed ever so slightly. But considering it’s the end of the season and we all know the story about the club, let’s at least look at each match individually:

Opponent Fulham Shots Fulham SoT Opp Shots Opp SoT
Charlton 13 6 8 1
Wigan 14 3 8 3
Rotherham 13 3 12 5
Blackpool 6 2 17 3
Boro 14 8 17 6
Norwich 8 4 19 8

Despite netting 9 points from these 6 matches, Fulham were outshot on target in all but 2: against Charlton, one of the worst shooting teams, and Boro, who were down to 10 men for 20mins (and they almost won that match). ‘Tis a funny game sometimes.

Now if we graph SOT Share, PDO, and TSR a few things stand out. First, our PDO under Kit impressively hovered in the mid to upper 90s for most of his reign. That basically means our table position was correct all along. Second, the Shots On Target Share increased in the final set after dipping consistently since November.

Finally, look at that TSR. Remember all the writings about our horrible defense? Well that’s why Fulham’s TSR was so atrocious, having a final ranking of 20th overall. Although it was actually quite high under Magath, it declined all season long despite flatlining a bit the last 10 matches of the season. The teams with a lower TSR all had higher PDOs, except for Blackpool who were the worst ever, meaning that we were really bad AND finished exactly where we deserved (unlike 12th placed Charlton who had the league lowest TSR).


If we sort all the stats for the entire Championship by Largest to Smallest/Most to Least, you’ll see Fulham posting some not-so-pretty numbers:

Statistic Final Ranking
SAVE % 19
PDO 17

I could probably spill more ink going into all this, but at this point … what’s the point? Fulham survived with an 11pt gap, the same difference between a top 10 finish. Changes should be coming in the summer. Kit seems like he may be sticking around, but who knows.

This entire campaign can probably be summarized as such: ¯\_(ツ)_/¯

Fulham 1-0 Blackpool; Fulham Survive And Yet…

Jesus H Blackpool is but 1 point away from setting an all-time low record for points and yet:


Blackpool. Shot. Nearly. Twenty. Times.

This (tragic) dumpster fire of a club is 22nd in total shots for and, thankfully for Fulham, 23rd in shots on target for. Their Shooting %, despite somehow better than Rotherham, is 22nd but not even 24% (i.e. only 1 of their 4 shots on target go in the net.)

As of last month, they barely had 130 DZ shots total in about 40 games. On Saturday they got 5.

And Fulham set out a fairly defensive lineup.

Which may explain whey they had six total shots. Against an opponent that is DEAD LAST in total shots AND shots on target against (Spoiler alert: Fulham also had two shots on target. TWO. But hey, 50% Shooting Percentage!)

Oh and looks like Fulham’s second leading scorer and shot taker was left on the bench, again.

Statistically speaking, this may match may have crossed the rubicon for me. There’s no tenable reason to keep Kit as manager after the Norwich match. The defense is bad, the offense is petering, results are being achieved when they shouldn’t be, and the team selection is confounding.

Any of those in isolation would be acceptable. Two of four? Eeeeep. All Four? Toodle-oo.

Bring Back Hugo … ?

Woah hey maybe it’s time for Hugo Rodallega to get some more minutes? The fella hasn’t started in last 7 games, and although his performances right before his benching were quite bad, our offense isn’t getting that much better without him.

Here are some shot figures, in a table (filtered so no less than 5 shots) and bar (no less than 3 shots) format.

fulham shooters table

fulham shooters bar

For comparison, here’s similar data from early March and late February.

You see Hugo is still our #2 shooter by an absurdly wide margin, despite being benched for the past month and a half.

Things can’t get much worse (okay yes they can) than reintroduce Hugo (and possibly give Kaca more minutes). As of yesterday our Shooting Percentage was 30.69. In the past 8 games where Hugo has barely featured, the Shooting% is 31.37. That’s not a great increase.

But rather than diversify our offense, his absence has resulted in Ross just shooting from distance, by himself, a lot.

I dunno. Just spit balling.

(Say is Adam Taggart still alive?)