I meant top target for CM not overall, should have specified.
Are we a one man team? (And that man is Casemiro)
- Thread starter The Corinthian
- Start date
I meant top target for CM not overall, should have specified.
He turns 32 during next season. His minutes will need to be managed at some point, especially if our midweek games are Champions League ties instead of Europa League ones where we can afford to rotate in McFred: this would be suicidal for either our League or Champions League campaign.
I absolutely agree with this, the player we purchase should be one who is happy to play less than half of our minutes initially, but be ready to take the step up in a year or two's time. We've lost our succession planning since Fergie retired. Signing someone young to groom to be his replacement should be a priority, rather than waiting until he's announced his retirement. Not doing so is what has got Liverpool into the mess they're now in with their midfield.
Below is the fixtures and the points and expected points for every team in the league. For our home fixtures I've used the other team's away record and vice versa, giving a rough measure of fixture difficulty. This is marginally in favour of Casemiro when looking at the average but against him when looking at the median. I don't think there's much in it
Without Casemiro:
| Opponent | Home/Away | Opponent xPts per Game | Opponent Points per Game |
| City | A | 2.40 | 2.67 |
| Arsenal | A | 2.18 | 2.53 |
| Newcastle | A | 2.16 | 2.07 |
| Brighton | H | 1.85 | 1.67 |
| Arsenal | H | 1.85 | 2.25 |
| Brentford | A | 1.75 | 1.80 |
| Brentford | H | 1.38 | 1.00 |
| Leeds | A | 1.29 | 1.25 |
| Liverpool | H | 1.25 | 1.00 |
| Leicester | H | 1.17 | 0.81 |
| Southampton | A | 1.09 | 0.63 |
| Leicester | A | 1.06 | 0.80 |
| Leeds | H | 0.99 | 0.60 |
| Southampton | H | 0.94 | 0.87 |
| Everton | H | 0.69 | 0.60 |
| Average | 1.47 | 1.37 | |
| Median | 1.29 | 1.00 |
With Casemiro:
| Opponent | Home/Away | Opponent xPts per Game | Opponent Points per Game |
| Liverpool | A | 2.11 | 2.21 |
| City | H | 2.07 | 2.00 |
| Chelsea | A | 1.67 | 1.44 |
| Aston Villa | A | 1.62 | 1.81 |
| Newcastle | H | 1.44 | 1.69 |
| West Ham | H | 1.31 | 0.64 |
| Tottenham | H | 1.29 | 1.33 |
| Everton | A | 1.29 | 1.13 |
| Fulham | A | 1.20 | 1.47 |
| Palace | H | 1.18 | 1.00 |
| Wolves | A | 1.10 | 1.44 |
| Palace | A | 1.06 | 1.33 |
| Forest | A | 1.05 | 1.31 |
| Bournemouth | H | 0.72 | 0.88 |
| Forest | H | 0.68 | 0.40 |
| Average | 1.32 | 1.34 | |
| Median | 1.29 | 1.33 |
When it comes to Eriksen (and everybody else) it's somewhat challenging to isolate for more than one player at a time, half of the games Casemiro has "missed" were pre-Everton away when he wasn't starting, the other half being the one's where they were both out due to injuries and suspension.
Scott (~£50k per week) will be sold this Summer and Fred (~£100k per week) is out of contract in the Summer and is 30 (so we shouldn't extend and should be looking to sell instead). Neither are on massive money? There will be plenty of room in the wage budget, once they've been moved on.
Why?
The thread title is somewhat hyperbolic, but we can and should be signing more than one player. On importance, below is our expected goal difference when each player is playing. Of our regular starters (and therefore those who have played in a good amount of easier and more difficult games), Casemiro is the one who makes the biggest impact. Look at the difference with his back-up, Mctominay, plays instead. Our only player who causes a negative expected goal difference when he plays!
Last edited:
We’re definitely not a one man team, it’s just that we have literally nobody else in the squad that can play the holding midfielder role, not competently anyway. In some games that doesn’t matter so much, but the longer we’ve been without Casemiro, the more likely it has been that this problem is exposed.
But a much greater problem for us this season has been our inability to put away chances, meaning we’ve lost games we should have at least drawn, drawn games we should have won, and a lot of our wins have been much closer than they needed to be.
But a much greater problem for us this season has been our inability to put away chances, meaning we’ve lost games we should have at least drawn, drawn games we should have won, and a lot of our wins have been much closer than they needed to be.
Just crunch them.
Yes, we can and should be spending on more than one player. But our spending will not be in the 300-400m range it would take to bring in all the players we would want to address our needs this summer.
If we’re going to bring in a truly top striker — without any question our greatest need — that’s going to take a huge bite out of that budget. The question we all have to ask ourselves is what the next greatest need is. Cover for Casemiro is a need, but it’s not our greatest or second greatest need.
If we’re going to bring in a truly top striker — without any question our greatest need — that’s going to take a huge bite out of that budget. The question we all have to ask ourselves is what the next greatest need is. Cover for Casemiro is a need, but it’s not our greatest or second greatest need.
Spot on.
OP: “I like to crunch numbers”
Well that’s all good but if you do, you’d know that a sample size of 15 games is insignificant to prove anything really and a not jump to a “one man team” conclusion based off that.
Seriously a course on statistical significance, P-Values and stuff on YouTube would make a world of difference.
Looking at Casemiro it’s clear as day that he improves the team (when not sent off) but one should always be vary of proving stuff by such a small sample size. Many factors involved, hard to isolate just one and conclude anything (statistically speaking).
Edit: the post from @PSV above underlines this point completely
Last edited:
I’m sort of in this line of thought. I would prioritise a striker but I do think we’d be daft not to pick up a few midfielders in summer as well.
I stated in the OP that the sample sizes weren’t ideal, so there’s no need to be so condescending. Given that we don't have hundreds of games of to create a large enough sample size for statistical significance, should we not discuss the numbers at all?
That said, 15 games each way isn’t nothing, we have samples of ~1500 minutes each, with a goal being scored in the Premier League every 32 minutes. Does a discussion on a message board need to have the same rigour around statistical significance as it if it’s being published in a journal?
But your title says "Why we're a one man team" - that's concluding. Can't have one without the other.
And yes, if you want to actually prove something except grab stats that fits one's own agenda, then statistical significance does mean something. How else would you know your stats are valid/prove anything?
In theory( and if you look at PSV's post) this thread could have a POV about why Wan Bissaka makes us a one-man team as he's got more points pr game than Casemiro. Even the much-slandered Fred, was above the average points pr average player last season (one of the best), which really shows that the stats you choose to pick, when used on other players, can be used too to back-up whatever view you want to put out there.
And by the way I'm not arguing that Casemiro isn't impactful. Everyone can see he's a great player. But your OP makes it looks like fact based by stats, which I'm disputing. If you think stats like this can prove Casemiro's impact, then the argument is valid too that Wan Bissaka makes us a one man team even more.
It is a forum for debate, no one says it should be about having nit-picked stats, that can't prove anything, so of course you should be open to opinions that goes against your stats too. It's just common sense, not anything that says it should be able to being published in a journal.
I accept that the title was clickbaity and a bit of an exaggeration. I also accept that being challenged isn't always snark and I don't think that's the way you intended it, maybe I'm just being a bit over-sensitive.
I have run a two-factor ANOVA, with fixture difficulty defined as the opponents’ expected points in their home and away matches this season to check for correlation against our expected goal difference for each match. P-value for fixture difficulty is 0.002, and it's 0.08 for Casemiro starting or not. Whilst the latter doesn't quite meet the standard for statistical significance (0.05), it's pretty damn close.
Last edited:
How's that so? I'd strongly disagree. In fact when the push comes to shove our finishing has been pretty good.
According to Understat, only Chelsea, Everton and West Ham have underperformed in goals scored worse than us relative to xG created.
This isn’t a perfect measure, (xGOT-xG would isolate opposing goalkeeper performance), though I’m not sure anyone publishes this for free.
Watching Case, Bruno and Eriksen tear Forest a new one shows how ridiculous it was that we persisted with McFred for so long.
54 ‘big chances’ missed, and only 46 goals scored - one of the worst ratios in the league. Only Everton and Chelsea are worse.
I’d agree that a player to add to the spine of the team and to help/eventually replace Casimero is a must. As fragile as we have looked at times this season it’s still a huge improvement from how porous we looked previously. A player that can sit and protect the back 4 and initiate attacks will be crucial.
I think the influence of character has been just as important so if we can find a leader with similar traits I’d be happy. When certain players are in the team confidence can be affected for both fans and the squad and I feel he is one of those. An anchor so to speak.
Clearly we have lacked goals and a constant upfront and I’m sure this will be addressed however Cas absences at times this season has shown the profile we lack in the squad. Forward planning is in motion I’m sure and I’m hoping the same perspective is being taken with Varane. As well as other did in his absence recently.
I think the influence of character has been just as important so if we can find a leader with similar traits I’d be happy. When certain players are in the team confidence can be affected for both fans and the squad and I feel he is one of those. An anchor so to speak.
Clearly we have lacked goals and a constant upfront and I’m sure this will be addressed however Cas absences at times this season has shown the profile we lack in the squad. Forward planning is in motion I’m sure and I’m hoping the same perspective is being taken with Varane. As well as other did in his absence recently.
Yeah that would be my order of priority,mind you it looks like we could well be in for Frimpong at RB. Honestly think we should put all the groundwork in to get Maignan however long it takes.
Any of those would improve our midfield situation,so are they all able to play as either a 6 or 8 if required is the big question
I understand, I just don't agree with the comment this is hurting us. In two of the last 4 games we had xG of 3+ and won 2-0, we definitely missed some chances
However, I would like to see a proof that this has any impact on the results because I checked this in the past and in tight games we've been doing pretty good. We've not been clinical in games we've won by some margin but this is irrelevant.
I'm not sure I like the last graph much. Yes, it supports the Casemiro (and McTominay) point very clearly. But if I'm reading it right, it also suggests that Rashford, Martinez, Varane and Shaw don't make much of a difference which is transparently not the case. Far too many variables at play for a simple presentation of 'GD with/without this player' to be worth much.
Sevilla is a clear example. I don't have access to all these stats but I'm willing to bet they show we had a number of strong chances we didn't take in that game, and in the end that meant we drew rather than winning. Both the Everton and Forest games, on either side, had a very similar constitution - loads of United chances, too few of them taken - but we got away with it because we got more of the rub of the green, and defended more consistently.
If you're going to use objective data, each player in the squad would need the same kind of assessment, not just Casemiro.
It may strengthen the argument in his favour, it might not, but either way, that data must be brought forth, too.
It may strengthen the argument in his favour, it might not, but either way, that data must be brought forth, too.
Well I'd you want to be precise we had xG=1,7 and scored 2 goals. I don't argue that we probably should've scored more but ending this game with a draw and blaming our finishing is a stretch if you ask me.
Also, the poster I quoted first said that finishing is our biggest issue, while I'd say over the course of the season it's not an issue at all.
OK, some good feedback, so I thought I should try another approach.
I've ranked each of our fixtures for difficulty by looking at the opponent's expected points per game. As an example, for our home game against City, I've taken their expected points per game in all of their away matches this season. The season is not yet complete, so their will be some variability in terms of who has played who, but most teams have played 14 or 15 out of 19 home and away fixtures so far. Nevertheless, it's pretty sensible looking.
Next, I've averaged the fixture difficulty for each of the significant appearances a player has made. This has been judged to be at least 40 minutes played. Then I've bound that to FBREF's measure of xGD when a player is on the pitch. That gives us the below chart:
Fixture difficulty isn't weighted by the minutes a player played in any one match.
I've ranked each of our fixtures for difficulty by looking at the opponent's expected points per game. As an example, for our home game against City, I've taken their expected points per game in all of their away matches this season. The season is not yet complete, so their will be some variability in terms of who has played who, but most teams have played 14 or 15 out of 19 home and away fixtures so far. Nevertheless, it's pretty sensible looking.
| Fixture | Difficulty |
| City (A) | 2.40 |
| Arsenal (A) | 2.18 |
| Newcastle (A) | 2.16 |
| Liverpool (A) | 2.11 |
| City (H) | 2.07 |
| Brighton (H) | 1.85 |
| Arsenal (H) | 1.85 |
| Brentford (A) | 1.75 |
| Chelsea (A) | 1.67 |
| Aston Villa (A) | 1.62 |
| Newcastle (H) | 1.44 |
| Brentford (H) | 1.38 |
| West Ham (H) | 1.31 |
| Tottenham (H) | 1.29 |
| Everton (A) | 1.29 |
| Leeds (A) | 1.29 |
| Liverpool (H) | 1.25 |
| Fulham (A) | 1.20 |
| Palace (H) | 1.18 |
| Leicester (H) | 1.17 |
| Wolves (A) | 1.10 |
| Southampton (A) | 1.09 |
| Palace (A) | 1.06 |
| Leicester (A) | 1.06 |
| Forest (A) | 1.05 |
| Leeds (H) | 0.99 |
| Southampton (H) | 0.94 |
| Bournemouth (H) | 0.72 |
| Everton (H) | 0.69 |
| Forest (H) | 0.68 |
Next, I've averaged the fixture difficulty for each of the significant appearances a player has made. This has been judged to be at least 40 minutes played. Then I've bound that to FBREF's measure of xGD when a player is on the pitch. That gives us the below chart:
- Our expected goal difference with each player tracks almost completely with fixture difficulty, with the exceptions of Casemiro (who improves us significantly), Sabitzer, Garnacho and McTominay (who cause us to underperform). In other words, we consistently outplay bad teams. This isn't the case for the last couple of seasons - but that's another thread
- The drop-off in our performances in more difficult fixtures is severe
- I think Garnacho can be explained by the fact that we're better when Rashford plays on the left instead of him
- Casemiro has not made a significant appearance (as defined above) with either Sabitzer or McTominay all season
- The chart shows that Lindelof, Maguire and Wan-Bissaka's overperformance relative to Dalot, Varane and Martinez on expected goal difference can be explained by the fact that they've played in easier fixtures on average
Fixture difficulty isn't weighted by the minutes a player played in any one match.
In terms of priority and effort, yes. But a prime striker (under 28 y old) will cost you 80m, a backup DMF 30-50m.
If it's purely back pascal grob would have been a good cheap option but then see he has extended his contract at brighton, he has done a good job of late as the deepest midfielder for Brighton and goes under the radar
It's hard to ignore the Frimpong rumors, but we'll see if that's nothing more than lighting a rocket up the asses of AWB and Dalot.
As far Maignan, there's no denying that Maignan would be an outstanding acquisition.
Where I'm coming from is the assumption that we're not going to spend more than 200m this summer, even if the Qataris are our new owners, and if I'm right about that I just don't think there's enough in the transfer budget for a striker that's going to cost us around 100m, two midfielders who together will cost us around 100m, a RB that will cost us around 40m and then a keeper who would cost us around 80m.
That said, we will make some money on sales but I have serious doubts about hauling in a significant sum for McTominay Van De Beek and whoever else we sell.
Tonight's performance has really highlighted how wrong the thread title is. We came in tonight missing 4 of our best 5 performers this season and look how terribly we played!
Casemiro has been terrific this season, but it's clear as day just how important Martinez, Varane, Bruno and Rashford are for us too.
Casemiro has been terrific this season, but it's clear as day just how important Martinez, Varane, Bruno and Rashford are for us too.