Conclusions from Blades 3-2 Preston
That was a pretty weird week for United. We lost a game we deserved something from, got a point from one we probably deserved to lose, then almost threw away a game we absolutely deserved to win before rescuing ourselves a few minutes later.
Still, I’m officially declaring that it is no longer too early to look at the league table, and what’s this? We’re 4th and 2 points off top, with no team having won more games than us. Here’s 10 conclusions from this weekend’s dramatic win over Preston.
1. The key events
For a long time in this game, a 3-2 scoreline felt extremely unlikely. Nonetheless, United started in a more purposeful, faster-paced manner than the lacklustre performance against Birmingham midweek. David McGoldrick had several headed efforts – one goalbound that hit a defender – and eventually helped to create the opening goal. Kieron Freeman slid an excellent through-ball for the striker, and his effort was pushed into the path of Billy Sharp who had an easy tap-in for 1-0.
United doubled the lead not long after half-time, with Chris Basham flicking Oli Norwood’s freekick into the top corner. That looked like being that, with Preston offering nothing and actually looking worse in the second half, and McGoldrick had a good chance to end the contest but put his shot wide when through on goal.
Preston brought on the dangerous Daniel Johnson and, virtually from nowhere, got themselves back in the game as Darnell Fisher’s cross found Callum Robinson. Minutes later, with United trying to change shape to regain control in midfield, Andrew Hughes found space on the left, and Johnson turned in his cross via Jack O’Connell’s legs.
Preston immediately went back into contain mode and United took advantage: Enda Stevens surged into the box and pulled the ball back for McGoldrick to net his first goal at Bramall Lane, and a deserved three points for the Blades.
2. We can be critical and happy at the same time
Something I’ve seen in response to this game is the tension between “I’m happy that United won” and “I’m annoyed that we allowed Preston back into it”. I think it’s perfectly possible for both things to be true, and is a pretty accurate summary of my feelings.
Did we play well? Yes, for 75 minutes or so. Did we deserve to win? Absolutely. Did we almost throw it away against a team who are bottom of the league, and is it therefore fair to question what on Earth happened there? Yes, of course it is, and I’m sure Wilder is doing exactly that right now.
Two points off top is a fantastic start to the season, but unless you look at what went badly and try to fix it, you’ll never get better.
3. The stats behind the game
A few things to call out here – I’ve deliberately omitted the fact we both had 3 shots on target because it suggests the game was kind of even (it wasn’t). United’s 11 shots inside the box shows our attacking dominance, even when we weren’t firing on all cylinders: for context, we had 3 against Birmingham, 7 against Bristol City, and (this may surprise you) 4 against Villa. Bolton are the only other team we’ve created chances against so regularly as we did v Preston, where we managed a whopping 14 shots inside the box.
I’m taking my Expected Goals data from Experimental 361 in Jay’s absence, and this again reflects that we were much the better team. What’s more, our xG in that walloping of Villa was a mere 0.9 vs Villa’s 0.5: Saturday was one of our better performances of the season in terms of chance creation. Which makes perfect sense when you think about it: none of the goals against Villa were as easy a chance as Sharp’s opener, McGoldrick’s winner, or arguably McGoldrick’s miss when in on goal.
4. Preston’s attitude isn’t helping them
Preston aren’t the worst team. Okay, fine, if you’re just looking at the league table, I suppose they literally are. But their underlying numbers are those of a midtable side. Jay is currently having a well-deserved holiday, but his Expected Points table after six games had Preston as the 10th-best team, just behind Leeds.
They have some good players: Callum Robinson was a standout for them on Saturday, and Alan Browne, Daniel Johnson and Ben Pearson are all very competent at this level. But good grief, their approach to this game was desperate.
Time-wasting is nothing new, but Preston weren’t even trying to disguise that their gameplan from the first whistle was to take as much time out of the game as possible.
Added to that, they showed a complete lack of ambition in and out of possession, with every player behind the ball and essentially conceding 70% of the pitch to United. Before they got their goal – more on that shortly – their only opportunities came from us giving the ball away close to our box. Even at 1-0, they stayed extremely negative. Given that United haven’t exactly been rock-solid defensively so far this season – only two clean sheets from nine games now – it was a somewhat baffling approach from Alex Neil’s team.
5. Changing shape – so what happened?
I’m still slightly amazed that Preston got back into this – but so they did. At some point midway through the second half, we stopped playing with the same intensity and made some poor decisions in possession, and also allowed Preston to get further up the pitch under less pressure. Wilder tried to address this by bringing on Marvin Johnson, moving Basham into midfield and going with a 4-5-1 to try and stop them getting into our third so easily, but it just didn’t work and suddenly it was 2-2.
I’ve pulled a 20-minute sample from 65 minutes to 85 minutes in the game, which is about when we started to lose some grip on the game up to conceding the two goals. In this 20 minutes we completed 39 passes to Preston’s 112. Of those 112 for the visitors, 55 were in the attacking third (for context, in the other 70 minutes Preston completed only another 22 passes in the final third). Those 112 passes were completed at an 84% rate compared to 72% in the game as a whole, showing that we weren’t putting the same pressure on the ball.
This is a hard one for me to fathom. In theory, moving to a midfield five was a logical move. In practice, it seemed – if anything – to make things worse.
It could be that the players just aren’t as familiar with the roles in this kind of system. Basham and Duffy were bypassed in the lead-up to the first goal, where no-one tracks Robinson’s run from midfield into the box. The second goal comes from the area that Freeman usually occupies, but he was playing tighter in a back four at this point. Fleck, Norwood and Duffy were completely ineffective and barely touched the ball during this 20-minute spell: Fleck attempted 6 passes, Norwood 9, Duffy 6. We simply couldn’t clear our lines against a team that had had zero ambition up to that point.
What concerns me is that this isn’t particularly new this season: on numerous times we’ve tried to change to cope defensively, and it’s felt like we’ve got worse (Birmingham, Swansea, Bristol City – although to be fair, we did change to good effect against Villa) and lost more control of the game.
United have a very particular style under Wilder, and one of the main criticisms last season was that we stuck rigidly to that attacking style and it probably cost us points in tight games, or in away games at better teams (Wolves, Fulham, Cardiff). Equally, as Wilder himself noted after the game, we don’t have the energy to play with the same attacking intensity for 90 minutes. And he would know.
We do have to be able to adapt and change our mentality in games on the fly, if we hope to be anything more than an entertaining midtable team – the kind of team that sticks four past Wednesday but ships five to Fulham. It’s great that we’ve recognised that, but Plan B is definitely still a work in progress.
6. David McGoaldrick
After the toothlessness of our attack against Birmingham – and our strikers received hardly any service, but still mustered a mere 1 blocked shot between them – this was a very welcome return to the team for McGoldrick.
I praised him extensively after the Villa game so I’ll not repeat myself, but even though he’s missed some chances in the last couple of games, the fact he’s getting on the end of them is still a positive (especially when he does end up finishing one, as he did in the 87th minute). The timing of his runs for the first goal and then the second-half one-on-one was excellent, and an example of intelligence compensating for his lack of pace.
Speaking of which, and this is pretty anal in a game with five goals, but I loved his situational awareness with just seconds remaining in the first half: on the halfway line with two Preston men on him, he decided to avoid the risk of losing the ball and starting a counter, and simply turn round and play an easy ball all the way back to Henderson. No danger, 1-0 at half-time.
Like Sharp, McGoldrick is a player with some flaws, but also plenty of strengths playing in a system that suits him down to the ground.
7. Yup, we’re definitely good at set-pieces
After a brief interlude midweek where Norwood couldn’t hit a barn door, normal service resumed with another set-piece goal from the Blades. That makes it four goals from set plays already this season; last season we scored nine in total.
To put that into context, according to Whoscored we were 20th for set-piece goals last season. Nine games into this season, only Derby, Bristol City and Wednesday have scored more. It’s such a massive upgrade to our game, particularly against teams that play as defensively as Preston did. And remember, we didn’t have Norwood until game four.
Nice header as well, wasn’t it?
8. Rotation: frustrating but justified?
Wilder made five changes for the Birmingham game – Fleck came back into the team, but both wingbacks and both strikers were rested. It’s pretty safe to say that it was to our detriment in that game (although, to be clear, our first-choice midfield three all had stinkers), however I think Saturday showed why rotation will be a benefit in the long run.
All four returning players – Sharp, McGoldrick, Freeman and Stevens – had a big impact on the game. Both strikers scored, Stevens registered a brilliant assist, and Freeman’s through-ball created the opening goal. It’s not that these four had played poorly against Bristol City and looked like they were ready for a rest; more that there’s 46 games in the season and all four get through a lot of work in an average game.
I’ll cite Stevens here as the best example. It’s not an understatement to say that Stevens’ performances definitely declined as the season went on – but perhaps not surprising given he played 45 out of the 46 games, bombing up and down the wing.
Enda has been excellent in the last four or five games and was one of our best players on Saturday: he created 3 chances (most out of all Blades), completed 25 passes in final third (3rd-most of all players), completed 4 out of 10 crosses (most of all players) and had 3 interceptions (most for Blades). He consistently got close to the byline before delivering his crosses, as shown below (Opta count his assist as a pass rather than a cross so it’s not included here, but is probably the best example of them all).
Giving a key player the odd game off here and there might not be ideal in the short-term, but could be beneficial if we’re still in the hunt come March and April.
9. Compared to last season
Shout-out to Blades fan and SLO Gary Ogden for pointing this out on Twitter, but it’s interesting to compare our results this season with how we did in the same fixtures last season (throwing out the Swansea game, as we didn’t play them last season):
That’s a 9 point differential over what we picked up in the same fixtures last season – the only game we’ve done worse in is Bristol City.
I don’t know how meaningful this is – maybe it shows that we’ve improved our approach to these games, learning lessons from last season? – but at the very least, it’s nice to see us getting some revenge on a few annoying losses in 2017/18.
10. Finally - I couldn’t resist
Happy belated birthday, Chris Wilder.
What were you doing this time last year?