Championship Week 7: Reading and Charlton – brothers in fancystats

Last time around I looked at Middlesbrough and Brighton, two teams seemingly cursed by the football gods. So this time let’s look at those teams apparently in the Almighty’s good books - Reading and Charlton.
As we can see from the early season real life table, both teams are sitting nicely in the top six on 13 points – one win away from Norwich in second.
But there’s a clue even in this table that these positions might not be fully reflective of the clubs’ performances already.

Position	Team	Played	Won	Drawn	Lost	Goals For	Goals Against	Goal Difference	Points
1	Nott'm Forest	7	5	2	0	17	7	10	17
2	Norwich	7	5	1	1	15	5	10	16
3	Watford	7	5	0	2	14	7	7	15
4	Wolves	7	4	2	1	7	3	4	14
5	Charlton	7	3	4	0	11	8	3	13
6	Reading	7	4	1	2	11	10	1	13
7	Derby	7	3	3	1	13	8	5	12
8	Sheffield Weds	7	3	3	1	7	4	3	12
9	Middlesbrough	7	4	0	3	9	7	2	12
10	Ipswich	7	3	2	2	9	6	3	11
11	Brentford	7	3	2	2	9	9	0	11
12	Millwall	7	3	1	3	8	8	0	10
13	Blackburn	7	3	1	3	13	14	-1	10
14	Leeds	7	3	1	3	7	10	-3	10
15	Bournemouth	7	2	2	3	11	10	1	8
16	Wigan	7	2	2	3	9	8	1	8
17	Cardiff	7	2	2	3	8	9	-1	8
18	Brighton	7	2	1	4	8	10	-2	7
19	Rotherham	7	2	1	4	5	9	-4	7
20	Birmingham	7	1	3	3	5	12	-7	6
21	Bolton	7	1	2	4	7	12	-5	5
22	Huddersfield	7	1	2	4	7	15	-8	5
23	Blackpool	7	0	1	6	3	10	-7	1
24	Fulham	7	0	1	6	6	18	-12	1

Charlton have a goal difference of +3, while Reading’s is just +1.
Derby (+5), Sheffield Wednesday (+3) and Ipswich (+3) all have goal differences equal to or better than Charlton’s, while teams as far down as Bournemouth and Wigan (in 15th and 16th respectively) are equal to Reading’s score.
This certainly isn’t definite proof that these teams are currently sitting above performance levels, but it gives a hint that we should be scrutinising them.
So let’s do that and have a look at the deeper stats.

Position	Team	Shots for total	Shots against total	Corsi/TSR	Shooting % For	Save %	PDO
1	Nott'm Forest	99	85	0.538	50.02	76.66	126.67
2	Norwich	125	84	0.598	34.88	80.77	115.65
3	Watford	101	111	0.476	40	76.67	116.67
4	Wolves	100	83	0.546	29.16	83.34	112.5
5	Charlton	70	99	0.414	45.83	73.33	119.15
6	Reading	74	97	0.433	52.39	69.7	122.09
7	Derby	94	105	0.472	35.13	80.49	115.62
8	Sheffield Weds	102	82	0.554	23.34	79.99	103.33
9	Middlesbrough	120	66	0.645	29.04	61.11	90.15
10	Ipswich	104	96	0.52	29.99	80.64	110.63
11	Brentford	90	92	0.495	29.03	70.97	99.99
12	Millwall	102	92	0.526	23.53	69.22	92.76
13	Blackburn	98	105	0.483	34.21	65.86	100.06
14	Leeds	58	116	0.333	38.87	70.59	109.46
15	Bournemouth	107	88	0.549	29.72	62.96	92.68
16	Wigan	81	75	0.519	45.01	71.41	116.42
17	Cardiff	73	112	0.395	29.62	62.5	92.13
18	Brighton	110	71	0.608	25.01	52.37	77.38
19	Rotherham	90	87	0.509	18.51	65.39	83.91
20	Birmingham	90	104	0.464	22.73	65.72	88.44
21	Bolton	93	130	0.417	25.92	70.73	96.65
22	Huddersfield	108	103	0.512	17.5	58.32	75.82
23	Blackpool	85	92	0.48	15	61.53	76.53
24	Fulham	97	96	0.503	22.21	49.98	72.2

Well we can see that the indications from the goal difference are right in this case.
And it’s remarkable just how similar these teams are: both are well under 50% shot share (Corsi/TSR) and both have high PDO (luck) scores driven largely by sky high shooting %s - first and third in the league.
And both are wildly unsustainable in the long term one would expect.
Going down to the raw data it is amazing how similar they are – Charlton have taken 70 shots, conceded 99, with Reading taking 74 and allowing 97.
If we want to see which of these team’s success is more sustainable, we need to find a bigger difference between them than the odd shot here or there.
Well three points of PDO late in the season would be a more significant margin later on in the season, but in these early days it can be as little as the difference of one goal scored, or one extra save made.
However, the shots on target tell a different story.

Position	Team	Corsi/TSR	shots on target for total	shots on target against total	PDO	Shots on target share
1	Nott'm Forest	0.538	34	30	126.67	0.531
2	Norwich	0.598	43	26	115.65	0.623
3	Watford	0.476	35	30	116.67	0.538
4	Wolves	0.546	24	18	112.5	0.571
5	Charlton	0.414	24	30	119.15	0.445
6	Reading	0.433	21	33	122.09	0.389
7	Derby	0.472	37	41	115.62	0.474
8	Sheffield Weds	0.554	30	20	103.33	0.6
9	Middlesbrough	0.645	31	18	90.15	0.633
10	Ipswich	0.52	30	31	110.63	0.492
11	Brentford	0.495	31	31	99.99	0.5
12	Millwall	0.526	34	26	92.76	0.567
13	Blackburn	0.483	38	41	100.06	0.481
14	Leeds	0.333	18	34	109.46	0.346
15	Bournemouth	0.549	37	27	92.68	0.578
16	Wigan	0.519	20	28	116.42	0.417
17	Cardiff	0.395	27	24	92.13	0.529
18	Brighton	0.608	32	21	77.38	0.604
19	Rotherham	0.509	27	26	83.91	0.509
20	Birmingham	0.464	22	35	88.44	0.386
21	Bolton	0.417	27	41	96.65	0.397
22	Huddersfield	0.512	40	36	75.82	0.526
23	Blackpool	0.48	20	26	76.53	0.435
24	Fulham	0.503	27	36	72.2	0.429

Both teams have a variation of about 0.04 between their shot share and shots on target share scores – only in opposite directions, producing a 0.075 difference in Charlton’s favour.
Sure that might not seem like much, but continued at the same rate, that would mean an extra 7.5 shots on target per 100 taken.
And if we extrapolate that at league average shooting rates (31%) – we can see that Charlton would expect to score around 2.3 more goals than Reading per 100 shots taken.
Over the course of a season, that can certainly prove the difference between being in the playoffs or just outside, or in the relegation zone or not.
However the big take away from Charlton and Reading is that neither team has pretty numbers right now.
Sure, score effects will likely be playing their part, but a near 50% shooting level just isn’t sustainable, and sooner or later the PDO score will cool off.
Whether this pair can adapt when they do will be the important question, or it could be a very bumpy ride.

Ouch

Speaking of bumpy rides, I’m really beginning to feel very sorry for Felix Magath at Fulham.
Despite scoring three goals away at the league leaders he still gets burned by PDO with Forest scoring on all five shots on target.
Even Ross McCormack’s first league goals for the club couldn’t help him out.
It really is painful to watch at the moment. I’ve not had time to look at the goals Fulham conceded to see what was going on there, but it seems pretty harsh to ship five goals on five shots on target, even if one was a penalty.
I’m quite frankly amazed that Fulham owner/chairman Shahid Khan has stuck by Magath so long. Either he appreciates the team isn’t getting outplayed (a 50.3% shot share shows that) and expects the luck (72.2 PDO) to turn around at some point, or he’s simply not ready to write-off his investment in Magath (personally and financially) yet.
Either way I tip my hat to him for waiting it out.

[UPDATE: Apologies but by the time I published this post Khan had actually sacked Magath. It's been an incredibly hectic week for me and I didn't get a chance to double check it before publishing. As I wrote, I was surprised Magath lasted as long as he did in this age of short-term managers, but would have liked to see him given a while longer. So who will be the lucky receipient of Fulham's PDO rebound (when it does)?]

Goodbye

In contrast Ole Gunnar Solskjaer’s departure from Cardiff was probably fairly deserved through both statistical analysis and on the pitch performances.
More on this later, but the Bluebirds had been constantly outplayed all season with only a highly inflated share of the shots on target and slightly elevated PDO keeping them above water.
Once the PDO started to fall and the share of all shots (Corsi/TSR) continued dropping it really was time for him to go.
And for those who watched the final few games of Solskjaer’s reign, they will have been all that was needed to seal his fate.