Analysis: Can Lammens save Man Utd from goalkeeper conundrum?

Ryan Benson of Opta Analyst explores the data behind the decisions as Manchester United move through another transition in goal, with Andre Onana, Altay Bayindir and now Senne Lammens all involved.

When Erik ten Hag arrived as head coach in 2022, Manchester United were looking to implement a more specialist approach in build-up, playing out from the back with the goalkeeper and centre-backs.

That brand of football wasn’t a good fit for David De Gea and it led to Man Utd needing a different sort of goalkeeper – Andre Onana, signed from Internazionale in July 2023.

But two years after he replaced De Gea, Onana has departed on a season-long loan to Turkish side Trabzonspor.

Why has Onana fallen out of favour?

There is an argument Onana’s strengths weren’t maximised at Man Utd and the 'keeper himself might claim that to be the case.

For instance, the assumption following his arrival was that he’d potentially make Man Utd an entirely different proposition when it came to building from the back, but they actually became more direct.

The average speed at which they progressed up the pitch with their open-play passing sequences was 1.8 metres per second in 2022/23, going up to 1.9 metres per second in 2023/24.

Similarly, their passes per sequence decreased from 3.7 to 3.6; while obviously only slight differences, they don’t exactly suggest Man Utd became significantly more intricate as a result of Onana’s signing.

One potential rebuttal to this could be that Onana’s greater confidence with the ball naturally led to more mid-length passes out from the goalkeeper and therefore allowed Man Utd to get upfield quicker and more efficiently.

The fact the team’s best ball-playing defender – Lisandro Martinez – featured prominently in the same line-up as De Gea but missed most of Onana’s first season also isn’t completely irrelevant. 

Both are reasonable points, but the fact is, from the same number of minutes played (3,420), Onana registered 428 long open-play passes in 2023/24 compared to De Gea’s 321 the season before.

Now, it’s worth acknowledging that a long pass isn’t necessarily an aimless punt up the pitch.

Similarly, whether that pass qualifies as accurate or not is highly influenced by the defenders that team is up against and the attacker the pass was intended for, so a pinch of salt should be taken with any direct comparison between players in this respect.

But De Gea actually had a better success rate with long open-play passes in 2022/23 (38.9 per cent) than Onana did in 2023/24 (36.4 per cent).

The fact Onana created three chances to De Gea’s zero probably supports the widely accepted view that the former is better and more deliberate in possession than the latter.

But it’s difficult to make the case that Onana’s abilities on the ball were in any way transformative.

Whether that’s down to him or his manager(s)’s instructions isn’t for us to say.

Onana didn’t go long quite as often in 2024/25, though it was such a dismal season all round for Man Utd that their goalkeeper’s passing habits almost feel irrelevant.

Onana’s error record

While there is some evidence Man Utd didn’t make the most of Onana’s passing ability, there is also data to suggest he made too many mistakes.

Since joining the club in July 2023, Onana’s nine errors leading to goals across all competitions is the joint-most of any goalkeeper from the top five European leagues.

While it is relevant that he’s also played more minutes (9,300) over that period than all but one other keeper (Jan Oblak – 9,660 minutes), the frequency of his mistakes still doesn’t make for great reading.

Among current Premier League goalkeepers to have played at least 900 minutes over the same time, only six have made an error leading to a goal at a greater frequency than Onana (every 1,033 minutes on average).

There’s even an argument the data cuts Onana some slack, too. For instance, in Man Utd's shock EFL Cup exit at Grimsby Town, it was only their hosts’ second goal that was counted as coming from an error by Onana, though he was arguably also culpable for the first as he was beaten at his near post.

Will it be Bayindir in goal - or new signing Lammens? 

Despite the fact Onana may be about to leave (at least until the end of the season), a sense of uncertainty continues to hang over Man Utd's goalkeeping situation.

Man Utd allowing Onana to go on loan would suggest they have confidence in Altay Bayindir and Senne Lammens, their 23-year-old new signing from Royal Antwerp, but most unofficial briefings from the club have indicated Lammens is seen as one for the future.

And although it’s possible Bayindir retains the faith of the coaches, he hasn’t inspired confidence among supporters.

In fact, Bayindir is one of the six current Premier League goalkeepers who’ve made goal-ending errors at a greater frequency than Onana. None of the others commits mistakes as frequently as Bayindir's one every 430 minutes.

Mins per error leading to goal since Aug 2023

Corners and crosses: How Man Utd 'keepers compare

Corners have been highlighted as an area of susceptibility for Onana and Bayindir, and Lammens' statistics in that regard compare favourably.

Lammens recorded twice as many claims (12) as the first contact from corners in Belgium's Pro League last season as Onana (six) managed in the Premier League.

In fact, 63.2 per cent of Lammens’ first contacts at corners were successful claims of the ball, a greater proportion than any other Pro League keeper to get first contact more than four times. Meanwhile, only 31.6 per cent of Onana’s first contacts were claims.

It’s early days this season, of course, but Bayindir has yet to claim a corner delivery (admittedly only five goalkeepers have in the Premier League) and he’s got first contact on just one – that accounts for 7.1 per cent (1/14).

While we must bear in mind that we’re only looking at small sample sizes here, only two goalkeepers have a lower percentage - Brentford's Caoimhin Kelleher (5.9 per cent – 1/17) and Crystal Palace's Dean Henderson (4.2 per cent – 1/24).

We can actually go a little deeper here, looking at how they’ve commanded their respective boxes in open play with respect to facing crosses.

In Premier League matches since the start of last season, Man Utd's opponents have put 369 open-play crosses into the box while Onana was in goal – he’s successfully claimed 13 of them, which is 3.5 per cent.

Man Utd's opponents have delivered 106 open-play crosses into the area while Bayindir was in goal – he claimed five of them, or 4.7 per cent.

Over the same period, Antwerp’s opponents put 466 open-play crosses into their penalty area – Lammens successfully claimed 42 of them, which equates to 9.0 per cent. So, not only is the young Belgian used to dealing with these situations, he has a considerably better claim rate.

We are of course talking about two very different leagues here; there are numerous factors that could be influencing the numbers such as the height and physicality of forwards, the average quality of the crosses in question, and even tactics used by attacking teams to put goalkeepers off.

But Lammens could only play against and contend with what was in front of him. On paper, his top-line numbers offer encouragement when shown side by side with those of Onana and Bayindir.

Visit Opta Analyst for more features on the Premier League