A Statistical Look at Cardiff City’s squad in 2019/20 - Defence

League games only

Goalkeeper:
There has been much debate as to who is Cardiff City’s best goalkeeper. Neil Etheridge had the jersey for 2 seasons until his injury in the opening game of the season. Alex Smithies kept Etheridge out until a 4-2 defeat at West Brom. Etheridge would lose his place after the horror show at QPR.

Smithies tops the clean sheet chart with 37%, compared with 26% for Etheridge. You might say that’s down to our improved form under Neil Harris. Under Warnock, Etheridge had a marginally better goals conceded per game ratio (1.24 vs 1.3) yet only kept one clean sheet (in the 56 minutes he played against Wigan), compared with 3 kept by Smithies, at percentages of 18% and 30% respectively.

Under Harris, Etheridge conceded 1.7 goals a game and had a clean sheet percentage of 30%. Smithies conceded 0.9 goals a game and 40% of his starts resulted in clean sheets. Etheridge’s stats take a bit of a hit from the QPR game, but it appears that Harris’s decision to go with Smithies is the right one.

Interestingly, Smithies makes more saves from close range than Etheridge, who statistically is better at saving long range efforts. Whoscored.com suggests Smithies’ distribution is fractionally better.

Verdict: Smithies deserves the shirt next season

Centrebacks:
The season started with Sean Morrison and Aden Flint as the first-choice partnership. Morrison lost his place in November and December through injuries, but after the defeat at Loftus Road, Morrison and Curtis Nelson established themselves as the main centre back pairing.

Sean Morrison
Played: 35
Goals conceded: 42 (1.2 per game)
Clean sheets: 11 (31%)

Aden Flint
Played: 26
Goals conceded: 43 (1.69 per game)
Clean sheets: 11 (19%)

Curtis Nelson
Played: 33
Goals conceded: 41 (1.27 per game)
Clean sheets: 11 (33%)

It might be more interesting to look at partnerships.

Morrison/Flint
Played: 14.5
Goals conceded: 1.38 per game
Clean sheets: 21%
Points per game: 1.24

Flint/Nelson
Played: 10.5 (Bamba played as a back 3 for the first half against QPR)
Goals conceded: 1.33 per game
Clean sheets: 29%
Points per game: 1.81

Morrison/Nelson
Played: 20.5
Goals conceded: 0.88 per game
Clean sheets: 39%
Points per game: 1.83

It’s clear that the Morrison/Flint combination was the poorest. Funny how that seemed obvious to most fans but Warnock persisted with it. Flint is decent enough cover though at the amount he cost I think we could have expected more.

According to the data on whoscored, Flint was the defender most easily dribbled past. Nelson is less likely to tackle than Flint or Morrison. Morrison is our best interceptor and best at clearing the ball.

Verdict: Morrison and Nelson should continue next season. Flint is likely to be here unless he could be offloaded.

Fullbacks:

Joe Bennett played virtually the whole season, bar a 0-0 at home to Preston and the QPR game. Lee Peltier only missed one game before moving to West Brom. Dion Sanderson came on loan from Wolves while Leandro Bacuna has also filled in there.

With Sanderson we conceded 1 goal a game. With both Peltier and Bacuna, that rises to just over 1.3 a game.

It’s interesting to spot that full backs who attempt more tackles are more often dribbled past. Bacuna was our full back most likely to dive in and was dribbled past more often, followed by Bennett. Sanderson and Peltier didn’t make as many challenges and weren’t dribbled past as often.

Sanderson blocked more crosses per game than everyone, followed by Bennett. Bennett made more interceptions than the others. Peltier made more clearances. Sanderson and Bacuna committed most fouls on average.

Attacking is now a necessary part of a full back’s game. Bennett is joint second with Ralls behind Tomlin for assists and 3rd overall for key passes. None of our other full backs come close. Surprisingly Bacuna was bottom of that list when used as a right back.

Verdict: Bennett has the shirt and deserves to start next season. Peltier’s stats suggested he wasn’t the greatest of full backs, certainly not in an attacking sense. If Sanderson is brought in on loan, he’d be a useful acquisition, but a specialist right back signing would be preferred.

Whoscored.com's credibility takes a big hit if it only rates Smithies' distribution as fractionally better than Etheridge's, but, apart from that, I'd say the stats look to mirror most supporters' opinion when it comes to the players you talk about - in particular, the centre backs.

On the subject of right backs, I read the jacks are interested in Wales international and ex Sheffield United defender Keiron Freeman who is now out of contract - he might be worth a look at for us.

Thanks Eric. Perhaps us spectators should pick the team in future as we can see things that the manager can't!
I really cant place the Freeman fellow Bob mentions. Perhaps our manager has him on his list. Whatever, a right back, creative midfielder and a striker must be priority. I hope VT is able to fund 'one more push' as Wales Online puts it.
I cannot see, in the present circumstances much cash being spent in the Championship.

Quote:

---

Originally Posted by the other bob wilson

Whoscored.com's credibility takes a big hit if it only rates Smithies' distribution as fractionally better than Etheridge's, but, apart from that, I'd say the stats look to mirror most supporters' opinion when it comes to the players you talk about - in particular, the centre backs.

On the subject of right backs, I read the jacks are interested in Wales international and ex Sheffield United defender Keiron Freeman who is now out of contract - he might be worth a look at for us.

---

I don't know where whoscored get their stats from, but they've been consistent with Smithies' figures all season. I agree they seem odd and the question of where they come from is appropriate.

Take the 2 playoff legs. In the first game it is claimed Smithies made 27 passes, only 3 of them short. In the second leg he apparently made 31 passes, all long. I don't know what they classify as a long pass, but I don't recall him playing anything short at Craven Cottage, where of those 31 passes, 8 found a Cardiff player.

A little maths work shows that whoscored claim 20% of our passes in the first leg went long and 33% in the second leg went long. That seems about right to me.