Bryan Mbeumo (£7.9m) is a bit like Data Dribble.
Putting out the best football analytics content week in, week out. No slowing down.
Jokes are still suboptimal, but hey, you can’t have everything.
Look, I caved to the current meta of Arsenal defence, DEFCON defenders, and several sensible shield picks, but I’ll never stop searching for old-school goal-scoring midfielders.
Dual threats? Even better.
We’re still in the phase of the season where you need to be extra careful with data, because several match-up/game-state-driven outliers can still skew it significantly. And even though Manchester United are guilty of one of the biggest outlier games out there (that Ipswich one…), all charges against Bryan Mbeumo should be dropped.
That man is doing something in almost every game.
Despite Bruno Fernandes’ (£11.7m) creative output falling off a cliff.
Despite the team creating just 1.12 xG per game since Gameweek 4. You might be asking why I’m manipulating sample sizes just 9 GWs in, but those opening 3 GWs were basically still pre-season. Remember what Fulham did. Look at what they’re doing now. Also, the summer transfer window was still open.
Wait for the magic I will perform from Gameweek 12 onwards.
Mbeumo’s predicted points according to the FPL Copilot model
Anyway, another telling number is 0.08 non-penalty xG per shot created.
That’s over the whole sample size, and it’s the lowest in the Premier League, along with the Spurs and Coventry. You probably won't find a more telling line in this article.
But to make it more granular:
53% of their shots originated outside the box, which is one of the league’s biggest shares
Only 36% of their non-blocked shots find the target (3rd worst)
38% of their shots are blocked (2nd worst)
3% shot conversion rate, which is… the worst. For comparison, Brighton are on 17%
And amidst this mess, there’s Bryan Mbeumo.
This season so far, he averages 3 shots, 0.47 non-penalty xG, and 0.48 xGOT, along with 0.16 xG per shot, with most numbers well above the 90th percentile for his position.
Out of his 27 shots, only 7 were blocked, and only two of those blocked attempts were attempts from central areas. He’s great at creating separation from defenders, and lately, only about 1 in 10 shots gets blocked.
The good news is that his movement in the box and advanced areas is amazing this season. He regularly receives the ball close to the goal and makes things happen. Advanced VAEP models measuring how his touches increase the team’s chances to score/not concede make that pretty clear.
I mentioned his consistency.
Hull: 4 shots, 0.50 npxG, 0.23 xA
Ipswich: 5 shots, 1.76 npxG, 0.21 xA. Yes, there was this one:
Everton: 2 shots, 0.05 npxG, 0.03 xA
City: 3 shots, 0.49 npxG, 0.08 xA
Fulham: 5 shots, 0.27 npxG, 0.25 xA
Spurs: 2 shots, 0.24 npxG, 0.10 xA
Mbeumo: 3 shots, 0.65 npxG
Bournemouth: 2 shots, 0.26 npxG, 0.04 xA
Chelsea 🟥: 1 shot, 0.02 npxG, 0.33 xA
Just one truly poor game in the opening 9 gameweeks, with solid numbers despite playing for a team ranked 14th in xPoints, and producing all the numbers I discussed earlier.
Losing set pieces lately is slightly annoying, but he always plays 90 minutes, which is certainly nice.
Another from the series:
Now, the situation is still chaotic at Old Trafford, and United will face Aston Villa, Liverpool, and Brentford before playing Newcastle, Coventry, and Crystal Palace.
The Villa game is anyone’s guess – expect a fairly even contest. Playing Liverpool on the Anfield Road will be hell for United, provided they keep performing the same way they do right now. You know I’m a big admirer of the Reds’ defence and discussed it several times in my previous articles.
Brentford at home after that isn’t too bad, especially for Mbeumo.
💡 But because that run is disrupted by the international break between GWs 10 and 11, the Liverpool clash is annoying, and this is a super turbulent period for United, I think waiting until Gameweek 13 and monitoring what happens to Mbeumo’s final product in those games is a pretty nice move in isolation.
The Cameroonian who seems to have an endless physical tank at the peak age of 27 is exactly a pick for the Christmas congestion.
And if there’s a new-manager bounce, even better, as it might put Bruno Fernandes in a better mood and stop him constantly waving his arms around on the pitch.
Top 10 for projected points for the upcoming 8-GW period according to the Solio Analytics model
They play every 3 days when the good runs begin, so he might get fewer minutes in some games, but he’s still one of the more comfortable picks for that period.
At halftime of the AS Roma game, Carrick swapped him with Cunha, who then played up top, so let’s see if Cunha starts there in Gameweek 10. While I don’t think Mbeumo on the flank would be a worse asset, if that happens, that new dynamic is another thing to watch in the coming weeks.
What to read next
Which DEF will haul next? And is Gibbs-White finished as a pick?
Data Dribble is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
GW10 Thought Process: Jan Paul van Hecke special
Data Dribble is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
Join the Data Dribble premium community.
This newsletter is 100% reader-supported. There’s free content every week, but becoming premium unlocks the full edge:
2 deep dives per week
Weekly Q&A
Discord community
Prize mini-league, pre-season content, Subscribers’ chat and other perks
… for the price of one pint.