Ask a room of coaches to name Mohamed Salah’s weaknesses and you will get four answers, three of which are not weaknesses at all. One coach says he is one-footed. Another says he goes missing when it matters. A third says he is a system player who would not function anywhere else. A fourth, usually the one who has actually prepared a team to face him, says nothing for a moment and then describes a specific spatial condition under which the threat drops by a measurable amount, and explains what his own side had to give up to create that condition. Only the fourth answer is analysis. The first three are a mixture of accurate observation and false inference, and separating those two things is the entire job of this article.

A weakness in an elite attacker is almost never an absence of ability. It is a constraint on the set of actions available to him under specific conditions, and the useful question is never whether the constraint exists but what it costs each side. A defender who knows the constraint gets a repeatable plan. An attacker who owns the constraint gets a repeatable action he has rehearsed ten thousand more times than his marker has defended it. Both statements are true simultaneously, and any account of a great player’s limitations that acknowledges only one of them is propaganda for whichever side the writer supports.

The constraint dividend

The frame this article uses throughout is called the constraint dividend, and it is worth stating precisely before any specific limitation is examined, because it governs how every one of them should be read.

A strong preference, whether for a foot, a starting side, a zone of the pitch, or a category of movement, produces two dividends at once. It gives the attacker a small number of actions executed at a standard almost nobody else reaches, because volume of repetition concentrated on few patterns produces a quality of execution that breadth cannot match. And it gives the defence a small number of actions to prepare for, which is worth an enormous amount, because the hardest attackers to plan against are not the ones with the single best action but the ones whose action set is wide enough that preparation cannot narrow it.

The honest question is therefore never “is this a weakness” but “which side extracts more value from the same constraint, under which conditions, and at what cost.” That question has different answers in different phases of the pitch, against different defensive shapes, and with different personnel on both sides, and this article answers it condition by condition rather than issuing a single verdict that would be useless to anyone preparing a plan.

The constraint dividend also explains why the crowd argument and the coaching argument never meet. A supporter arguing that a limitation does not matter is usually pointing at output, which is the attacker’s side of the dividend. A coach arguing that the limitation is exploitable is pointing at plan repeatability, which is the defence’s side. Neither is wrong. They are measuring different things and calling both of them the same word.

Which of Salah’s limitations is a genuine constraint and which is a claim about variance?

Three are genuine structural constraints: the foot preference and the angle geometry it fixes, the reduced effect in central congestion relative to the wide start, and the modest aerial contribution. The supply dependency is real but shared with every wide forward. The big-fixture claim is a statement about small samples, not about ability.

The foot preference is geometry, not defect

Start with the fact everyone knows and almost nobody states correctly. Salah is a left-footed attacker operating from the right, which means the ball arrives on his right side and his most dangerous final action requires him to move it across his body onto the left foot before releasing it. That is not a description of a limitation. It is a description of a shape, and shapes have consequences that run in both directions.

The consequence that matters most is that the foot preference does not remove options randomly. It removes a specific, mappable set of angles, and it concentrates the remaining ones. This is the reason the containment plan in the counter blueprint works at all, and it is worth understanding at the level of the actual geometry rather than the slogan.

Consider a right-sided attacker in possession roughly level with the corner of the penalty area, angled infield, with a full-back in front of him and a covering centre-back somewhere behind and inside. A two-footed attacker in that position has a wide fan of final actions available. He can shoot with the near foot from a tight angle, drive to the byline and cross with the outside foot, cut inside and shoot with the far foot, cut inside and slide a pass with the far foot, roll a cutback with either, and clip a cross with either. Every one of those actions has a different release point, and the defender’s body cannot cover all of the release points at once, so the defender is genuinely guessing.

Now apply the constraint. The left foot is the release foot. Immediately, the outside-foot cross from the byline becomes a low-percentage action rather than a primary threat, because delivering it requires either an awkward right-foot contact or a body rotation that costs time the defender uses to recover. The tight-angle near-post shot with the right foot becomes rare. What remains, and what becomes correspondingly sharper, is the family of actions that begin with the ball travelling right to left across the body: the cut inside to shoot with the left foot from the top corner of the box, the cut inside to slide a pass with the left, the cutback delivered from a position slightly infield of the byline rather than on it, and the reverse ball played first time on the left before the defender sets.

The defender therefore knows the family. What he does not automatically get is a way to deny the family without opening something worse, which is the fork the counter blueprint spends its length resolving. But the family is knowable, and knowability is the defence’s half of the constraint dividend.

The angles the preference removes

The removed angles are worth naming precisely, because coaches who understand which angles are gone can position a goalkeeper and a covering defender with confidence they otherwise lack.

The first removed angle is the near-post drive from the right channel. A right-footed right-sided attacker who receives facing goal at the corner of the box can strike across himself into the far corner or drive the near post with the same foot from the same body shape. With a left foot on the far side of the body, the near-post strike from that position requires either a poor contact or an extra touch to reset. The goalkeeper’s near-post reference can therefore be set with more confidence than it can against a right-footed forward in the same zone, and that confidence is what allows the covering defender to sit a half-yard further inside than he otherwise would.

The second removed angle is the whipped early cross from wide right delivered with the outside of the right foot, the action a right-footed wide player uses to hit the far post from deep before the defence sets. That delivery is largely unavailable, which means the defensive line does not have to worry about being attacked in the air from a first-time wide delivery on that side as much as it would against a right-footed crosser, and the far-post centre-back can hold a slightly narrower starting position.

The third removed angle is the byline pull-back struck with the right foot while running toward the goal line. What replaces it is the cutback taken from infield of the byline, delivered with the left foot, which arrives at a different angle and lands in a different zone. The defending team can therefore weight its box coverage toward the penalty spot and the edge rather than toward the six-yard box, because the delivery is more likely to arrive flat across the top of the six than curled into it.

None of these three removals makes him easier to play against on their own. They make the box coverage cheaper to organize, which is a different and smaller claim, and one a coach can actually use.

The angles the preference concentrates

Concentration is the other half, and it is the half that supporters point at when they say the constraint costs him nothing.

The cut inside to shoot with the left from the right half-space is not merely available. It is available at a standard produced by a career of repetition in which almost every high-value action funnels through the same body shape. The release is fast, the contact is consistent, and the disguise between shot and pass out of the identical shape is genuine, which is the part defenders underestimate. Two actions that look identical for the first two-thirds of the movement are worth more than four actions that each telegraph themselves.

The second concentrated action is the run onto the inside shoulder of the full-back, taking the ball on the left foot as it arrives, which converts a wide start into a central finish without ever needing the right foot. This is the pattern that makes the containment problem hard, because it uses the same foot and the same body shape as the cut inside while attacking a completely different space.

The third is the arrival at the far post. A left foot on the far side of the body is an advantage rather than a constraint when a ball is delivered from the opposite flank, because the natural contact is a first-time strike or a redirect across the goalkeeper rather than an adjustment. Any account of the foot preference that treats it as pure cost has skipped this.

The angle-constraint table

This is the article’s findable artifact. It maps each final action to the geometry the foot preference produces, the defensive shape that exploits it, and the compensation his sides use to restore the option. Read the fourth column before the third, because the compensation is what tells a coach whether the exploit will actually be available on the day.

Final action	Angles the preference makes available	Angles it removes	Defensive geometry that exploits the constraint	Compensation his sides use
Cut inside to shoot	Far corner from the right half-space, disguised as a pass out of an identical body shape	Near-post drive struck first time with the near foot from the same position	Covering centre-back holds the inside lane rather than the line, goalkeeper sets a confident near-post reference, near-side midfielder shades the entry pass	Underlapping runner occupies the inside lane so the covering defender cannot hold it, or a striker pins the covering centre-back centrally
Cutback from the byline area	Flat delivery from infield of the goal line onto the penalty spot and the top of the box	Right-footed pull-back struck while running toward the goal line, and the outside-foot near-post whip	Back-post defender narrows, holding midfielder screens the penalty-spot zone rather than dropping into the six-yard box	Overlapping full-back reaches the byline instead, restoring the true pull-back from a right-footed delivery point
Run onto the full-back’s inside shoulder	Reception and finish on the left with no turn required, from a central arrival	Reception on the right with the goal on the outside shoulder, which would need a settling touch	Full-back sets his stance to close the inside shoulder first, covering centre-back defends depth rather than width	Central striker drags the covering centre-back across, opening the same run without the cover behind it
Reception facing his own goal in a congested central zone	Turn to his own left, out toward the right half-space, which is the direction he came from	Turn to his own right into the left half-space with the same speed and balance	Two midfielders set on the left side of his body, forcing the turn back toward the touchline he just left	Drop into the pocket with a bounce pass and a re-run rather than a turn
Far-post arrival from an opposite-flank delivery	First-time strike or redirect across the goalkeeper with the natural foot	Powerful attacking header at the same delivery height	Far-post defender defends the ground contact and the goalkeeper takes the aerial delivery earlier	Arrive later and deeper so the delivery is cut back to the ground rather than lofted
One-on-one against a recovering defender in transition	Push infield onto the shooting foot at the moment the defender is turned	Push outside and strike early with the outside foot before the cover arrives	Recovering defender takes an inside-shoulder recovery line rather than a ball-side one	A second runner outside him, which makes the inside recovery line indefensible

The table is the whole argument in one page. Every row contains a genuine removal, a genuine concentration, a genuine exploit, and a genuine answer to the exploit. A coach who reads only the third column will build a plan that his opponent has already solved. A supporter who reads only the second and fifth columns will conclude the constraint has no cost. Both are reading half a table.

The central question, which is spatial rather than technical

The second real limitation is that he is less effective in central congestion than he is from a wide starting position, and it is the limitation most often stated badly.

Stated badly, it becomes “he cannot play through the middle,” which is false and easy to disprove. Stated properly, it is a claim about the relationship between a starting position and the space available at the moment of reception, and that claim is defensible and useful.

Here is the mechanism. A wide start does something specific for an attacker whose principal action is a diagonal move infield: it guarantees that the space he is moving into is the space the defensive block has the most trouble covering, because a defensive shape that is compact enough to protect the centre has to leave the outside of it, and the seam between the two is exactly where the diagonal lands. The wide start also guarantees a favourable first defender. Full-backs, as a population, are less comfortable defending an isolated inward diagonal than centre-backs are, because a full-back’s default orientation is toward the touchline and the centre-back’s is toward the goal.

Now move the same player into a central starting position. Three things change at once. The reception is more likely to arrive with his back to goal, which removes the facing-goal geometry that all of his primary actions assume. The nearest defender is now a centre-back or a holding midfielder rather than a full-back, both of whom are trained specifically to defend a receiver in that zone. And the space he wants to move into is now on both sides of him rather than one, which sounds like an advantage and is not, because the constraint means only one of the two is genuinely productive.

That last point is the one most analysts miss, so it deserves the extra sentence. A two-footed central receiver genuinely threatens both directions out of a turn, which forces the defender behind him to stay square and passive. A strong-footed central receiver threatens one direction properly, so the defender behind him can pre-commit his body to the productive side and take away the turn before it starts. In a wide position, the touchline is doing the pre-commitment work for the defender anyway, so the constraint costs almost nothing. In a central position, the constraint costs a great deal, because the defender’s pre-commitment is a choice he gets to make rather than a geometry he is stuck with.

That is why the central question is spatial rather than technical. Nothing about his technique deteriorates when he moves inside. What deteriorates is the value of the constraint, because congestion removes the compensations that the wide start supplies for free.

Why does a central role reduce the value of the same actions?

Because the wide start supplies two things a central start does not: a reception facing goal, and a first defender whose default body orientation points away from the space the attacker wants. In congestion the defender chooses his orientation freely, so a known preference becomes something he can pre-commit against.

The aerial contribution

The third genuine limitation is modest aerial output, and it is the one that requires the least argument and the most care in how it is used.

He is not a target in the air. He does not win a high proportion of contested aerial duels in the box against defenders selected for that quality, and he is not the reference point a team looks for when it needs a header from a wide delivery. That is straightforwardly observable, uncontroversial, and true of most forwards of his physical profile, which is why stating it is analysis rather than criticism.

What it means for a defending side is narrower than coaches assume. It does not mean the far post is safe when he is arriving there, because the dangerous arrival is a ground contact rather than an aerial one, which the artifact table already covers. It means something more specific: a defending side facing him does not need to allocate a dedicated aerial specialist to his zone in open play, and can therefore make its centre-back pairing decisions on the basis of the other threats in the box. In practice that is a squad-selection dividend rather than an in-match one, and it is worth roughly one selection decision per fixture.

It also has a second-order consequence that is worth more than the first. Because he is not an aerial threat, a defending team can afford to defend wide deliveries from the opposite flank with a slightly higher line than it otherwise would, since the ball that hurts is the cutback rather than the loft. A higher line compresses the pitch, which is precisely what makes the cut-inside geometry harder to execute. The aerial limitation therefore pays the defending side twice, once directly and once through the line height it permits, and the second payment is larger than the first.

The supply dependency and the absorbed weakness

The fourth limitation is the one that generates the most heat online and the least clarity, so it needs the most careful handling. The claim is that he is a system player whose output depends on a specific supply pattern and a specific partner behind him, and that in a different structure the numbers would collapse.

The claim is half right, and the half that is right is more interesting than the half that is wrong.

Start with what is actually true. An inverted attacker whose primary actions all begin from a wide-right reception facing infield needs three things supplied by the structure around him. He needs the ball to reach him in that zone often, which requires a build-up pattern that consistently finds the far side of a defensive block rather than one that funnels centrally. He needs the space behind and outside him occupied by a teammate, because if nobody threatens the outside, the defending full-back can set his stance entirely to the inside and the diagonal is dead before it starts. And he needs central occupation, because a covering centre-back who is free to slide across into the inside lane removes the cut inside without any help.

Those are structural requirements, not preferences, and a side that does not provide them will get materially less out of him. That much is honest.

Here is where the claim goes wrong. Every elite wide attacker has structural requirements of exactly this kind, and the requirements are not evidence of dependency so much as a description of what wide attacking is. A wide forward with no outside threat behind him and no central occupation ahead of him is a player in a one-versus-two, and no profile in the history of the game has been productive in a sustained one-versus-two. The test of a system-player claim is not whether requirements exist. It is whether the requirements are unusually expensive relative to peers.

Apply that test properly and the picture becomes specific. The outside-threat requirement is expensive, because it demands an attacking full-back on that side, which in turn demands either a holding midfielder who covers that flank in rest defense or a centre-back comfortable defending a wide channel alone. That is a real cost paid by the rest of the team, and it shows up in the shape of the side rather than in his own performance.

The central-occupation requirement is cheap by comparison, because any centre-forward provides it and every side plays one.

The supply-pattern requirement sits in between. Reaching the far side of a compact block consistently is a difficult build-up problem, and solving it requires either a distributor capable of switching play accurately under pressure or a structure that manufactures the switch through rotation. Sides that have neither will find him on the ball less often, not less effective when he is on it, and those are different failures that produce the same scoreline.

That distinction is the absorbed-weakness principle, and it is the correct lens for this entire debate. A great player’s real limitations frequently do not appear as failures in his own performance. They appear as constraints on the squad around him, as jobs somebody else has to do, as line heights the team cannot use, as full-back profiles the manager cannot select. Looking for the weakness in his own numbers is looking in the wrong place, because the structure absorbed it before it reached him.

What the compensations cost elsewhere

The honest accounting runs like this. An attacking full-back on his side, required to supply the outside threat, is a full-back who is high when possession turns over. That produces a specific vulnerability to a counter-attack down that channel, which the defending side can target deliberately. A holding midfielder assigned to cover that channel in rest defense is a holding midfielder who is not covering the opposite half-space, which produces a second vulnerability. A centre-forward pinning the covering centre-back is a centre-forward not dropping to link, which reduces the side’s ability to build through the middle.

Every one of those costs is real, and a team preparing against him should target them, because attacking the compensation is almost always cheaper than attacking the player. The counter blueprint develops that argument into a full plan, and a coach who wants the sequenced version rather than the diagnosis should read the counter-plan built on exactly these limitations alongside this one, since the two articles are designed to be used together: this one names the seams and that one turns them into a shape, a stance, and a set of triggers.

The defensive contribution question

A recurring line holds that his defensive work is a liability that his sides have to carry, and the honest answer is that the question is badly framed rather than wrong.

What is observably true of most high-output wide forwards, and of this profile specifically, is that the defensive job they perform is a positional one rather than a pressing-volume one. The forward holds a high starting position, screens a passing lane, and stays available for the transition rather than tracking a full-back to the byline. That is a coaching instruction as often as it is a preference, and a team that gets high transition output from a wide forward has usually decided to buy that output by excusing the tracking run.

The defensible version of the criticism is therefore not that he does not work. It is that his side has to solve a specific defensive problem that a side with a tracking-back wide forward does not: the space behind him on the flank when the opposition full-back advances. That problem is solved either by the near-side midfielder shuttling wide, which costs central compactness, or by the full-back defending a two-versus-one, which costs stability, or by dropping into a five, which costs the attacking full-back’s advanced position and therefore removes the outside threat his own game requires.

Notice that the cost circles back. The defensive arrangement that protects the flank is the same arrangement that removes the compensation his attacking geometry needs. That circularity is the genuine tactical tension in the profile, it is a real cost, and it is a far more useful thing to say than “he does not defend.”

For a defending coach, the practical instruction is straightforward: attack the space behind him early and repeatedly, not because it produces goals directly but because it forces his side into the arrangement that costs him his outside threat. Two or three sustained attacks down that channel in the opening twenty minutes will often achieve more than any marking scheme.

The link-play question

The claim that his link play is a limitation is the weakest of the commonly repeated criticisms, and it survives mainly because link play is hard to see and easy to assert.

Link play in a wide forward means three separable things, and the criticism collapses when they are separated. It means receiving with his back to goal and holding the ball for a runner, which is a centre-forward’s skill and one this profile is not asked to perform often. It means the short combination in the wide third, the give-and-go with the full-back and the near-side midfielder that breaks a compact defensive block, which is a genuine and frequently used part of his game. And it means the final pass, the ball played rather than struck, which is the disguise half of the concentrated action already described.

Of those three, only the first is genuinely modest, and it is modest because the role does not require it. Judging a wide inverted forward on back-to-goal hold-up is like judging a centre-back on crossing. The observation may be accurate and it carries almost no planning value, because no defending side is ever going to force him into a hold-up situation on purpose.

There is a narrower version of the criticism that does carry planning value. Under heavy immediate pressure at the moment of reception, with a defender arriving on his blind side, the first-touch option set narrows, because the natural protective touch takes the ball across the body onto the strong side and that is the side the arriving defender can anticipate. Pressing him from the blind side at the instant of reception is therefore more productive than pressing him from the front, and that is a coachable, specific, usable instruction that the general version of the criticism completely obscures.

The big-fixture claim, answered as a variance question

Now the argument that dominates every discussion of this player and that almost nobody handles honestly: that he goes quiet in the biggest fixtures.

It has to be handled as what it is, which is a statistical claim about small samples wearing the costume of a character claim. The costume is the problem. Once the sentence becomes “he does not turn up when it matters,” it has stopped being a claim about football and become a claim about a person’s temperament, which is not observable, not defensible, and not the business of anyone writing tactical analysis.

So take the honest version. The claim reduces to an assertion that his output rate in a defined subset of fixtures is lower than his output rate overall. Set aside whether the subset is defined consistently, which it usually is not, since the definition tends to expand to include whichever matches support the conclusion. Assume it is defined honestly. What can that comparison establish?

Very little, and the reason is arithmetic rather than loyalty. A forward’s output in any given match is a low-count outcome with enormous variance. Goals are rare events. Across a full season the count is large enough that the rate carries signal. Across the twelve or fifteen matches that anyone would classify as the biggest fixtures in a season, the count is small enough that the difference between a rate that looks poor and a rate that looks excellent is two or three moments, most of which are decided by a deflection, a goalkeeper’s reach, or a linesman’s flag. A sample that small cannot distinguish between a genuine effect and ordinary noise, and any analyst who has spent time with these distributions knows that a forward’s split across small samples will look dramatic in one direction or the other purely by chance, in most seasons, for most forwards.

That is not a defence of the player. It is a statement about what the evidence can support, and it would apply identically to a forward whose big-fixture split ran the other way, in which case the same crowd would be calling it proof of a big-game temperament rather than proof of its absence.

What the pattern does legitimately show

Having said what the sample cannot establish, honesty requires saying what it can, because the argument is not entirely empty and pretending otherwise would be its own form of dishonesty.

The biggest fixtures are, definitionally, the ones against the best-organized opponents. Better-organized opponents do three things that bear directly on the constraint mapped in the artifact table. They defend with a covering centre-back who actually holds the inside lane rather than the line, which is the specific behaviour the cut-inside geometry needs to be absent. They have a near-side midfielder disciplined enough to shade the entry pass rather than chase the ball, which reduces the number of receptions in the productive zone. And they have a full-back who has been given a stance instruction and holds it for ninety minutes rather than reverting to instinct after twenty.

Against a side doing all three, the output of any inverted wide forward drops. That is not a claim about him and it is not a claim about temperament. It is the observation that a plan built to exploit a known constraint works better when it is executed properly, which is the entire premise of this series and something that should surprise nobody.

The correct sentence is therefore: his output is lower against sides that execute the containment geometry well, those sides are disproportionately the ones he meets in the biggest fixtures, and the sample within those fixtures is far too small to distinguish that structural effect from noise, let alone to support any inference about the player as a person. That sentence is longer and less satisfying than the slogan, and it is the only version that survives contact with the evidence.

For a coach, the practical translation is genuinely useful. If you execute the containment geometry properly, you should expect to reduce his output. You should not expect to eliminate it, you should not conclude from one quiet performance that the plan works, and you should not conclude from one decisive performance that it does not. Two matches tell you almost nothing either way, which is the same lesson every analyst learns and most crowds never do.

The claimed limitations that are not real

Four widely repeated criticisms do not survive examination, and naming them is as much a part of a weakness anatomy as naming the genuine ones, because a plan built on a fictional weakness fails in a specific and expensive way.

The first is that he is easy to defend because he always does the same thing. Predictability and stoppability are different properties, and confusing them is the single most common analytical error made about this profile. A pattern is only exploitable when the defensive shape can afford to deny it without opening something more dangerous, and the whole difficulty here is that denying the primary action opens the secondary one. The counter blueprint resolves that fork with a three-player geometry, and any plan that does not commit those three players will find that knowing what is coming does not help at all. The useful-predictability test belongs to that article and is worth reading in full before anyone builds a plan on the assumption that a scouting report is self-executing.

The second is that a left-footed defender on that side neutralizes the problem. The appeal is obvious and the logic does not hold. The defender’s foot preference matters at the moment of the tackle and at the moment of the clearance, and it matters very little at the moment that actually decides the duel, which is the stance and the body angle a second before the attacker moves. A left-footed defender with a poor stance loses the duel exactly as reliably as a right-footed one, and the selection decision that genuinely matters is turn speed and cover discipline rather than which foot the defender favours.

The third is that he is finished as a threat against a deep block. What a deep block actually does is compress the space behind the defensive line, which removes one of the two primary actions, the run onto the inside shoulder. It does not remove the other, and in some respects it improves it, because a deep block concedes the space in front of it and the cut inside to shoot is executed from exactly that space. A team defending deep against him is trading one threat for another rather than removing both, and it must accept a higher volume of shots from the top corner of the box as the price of removing the through ball. That trade is often correct, and a coach should make it knowingly rather than believing he has solved the problem.

The fourth is that his numbers are a product of a favourable structure and would collapse elsewhere. This is unfalsifiable in the form it is usually stated, and the falsifiable version has already been answered above: the structural requirements exist, they are not unusually expensive relative to the peer group, and the correct place to find their cost is in the squad’s shape rather than in his output. A criticism that can only be settled by a hypothetical transfer is not analysis, and a series about preparation has no use for it.

Do serious analysts agree about any of this?

Not entirely, and the disagreements are informative. Near-universal agreement covers the foot geometry and the aerial profile. Genuine disagreement covers how much the central question costs, and how much of the supply requirement is dependency rather than ordinary role definition. The big-fixture claim divides analysts from crowds rather than analysts from each other.

What each limitation costs an opponent to exploit

A weakness anatomy that stops at diagnosis is a list. The reason this series insists on the cost column is that every exploit in football is purchased, and a coach who does not know the price will buy things he cannot afford.

Exploiting the foot geometry costs a covering centre-back’s positional freedom. To hold the inside lane rather than the defensive line, that defender has to accept being pulled out of the back four’s horizontal alignment for long periods, which means the far-side centre-back is defending a wider zone alone and the offside line is harder to hold. Against a side with a genuine centre-forward attacking depth, that price can be severe, and it is the reason so many teams talk about holding the lane and then quietly stop doing it after fifteen minutes.

Exploiting the central question costs an attacking commitment. To force him into central congestion you have to make the wide zone unattractive, which in practice means keeping numbers wide and high enough that his side prefers to build through the middle. That commitment costs central compactness, which is the opposite of what most teams want when facing a strong attacking side. Some teams can pay this and most cannot.

Exploiting the aerial profile costs almost nothing, which is why it is the one limitation every side should take. Adjust the far-post starting position, defend the ground contact, and let the goalkeeper claim the loft. There is no meaningful downside.

Exploiting the supply dependency costs a targeted attacking plan. Attacking the space behind the advanced full-back on his side is the highest-value exploit in the entire list, and it costs the willingness to keep a wide runner high when you do not have the ball, which reduces your own defensive numbers. A side that is trying to defend with everyone behind the ball cannot use this exploit at all, which is a hard truth for the sides that most want to contain him.

Exploiting the blind-side pressing seam costs discipline more than personnel. It requires a midfielder willing to arrive at the instant of reception rather than a moment after it, and arriving a moment after it converts the press into a foul in a dangerous area. That is a rehearsal cost rather than a squad cost, which makes it available to sides at every level, and it is the exploit most amateur teams should start with.

How the limitations changed as the role changed

A weakness anatomy of a long career has to account for the fact that the limitation set is not static, because the role moves and the compensations move with it.

Early in a senior career, a wide attacker of this profile is typically deployed as a pure width holder, hugging the touchline and receiving with the whole flank in front of him. In that version the foot constraint is at its most expensive, because the primary action requires beating a set defender from a standing start and the removed angles are exactly the ones a touchline receiver most needs. The corresponding limitation, the one a defender in that phase of a career exploits, is the difficulty of producing a threat when the first contact is made facing the touchline rather than facing infield.

As the role matured into a narrower starting position, closer to the inside channel, with the outside space delegated to an overlapping full-back, the constraint’s cost changed character entirely. The reception now arrives already angled infield, so the removed angles matter less and the concentrated ones matter more. What arrives in their place is a new limitation, which is the dependency on the overlapping partner being present, since the narrow start only works when somebody is threatening outside.

Later still, as the role incorporated more central arrival and more penalty-box occupation, the aerial limitation began to carry more weight than it previously did, because a player arriving in the box more often meets more deliveries and the proportion of them he can attack in the air becomes relevant in a way it was not when he was starting every attack thirty metres from goal.

The general lesson matters more than the specific arc, and it applies to every subject in this series. Limitations do not persist unchanged across a career. They trade. A role change that removes one constraint’s cost almost always introduces a new one, and the scouting error is to prepare against the version of the player that existed in the phase you happened to watch most closely. An opposition report built on a touchline-width version of this profile will be wrong about almost everything if the current version starts in the inside channel, and the error will not announce itself, because the report will still contain accurate sentences.

The general principle here belongs to a later synthesis article, and readers who want the cross-career version of this argument rather than the single-subject version will find it in the treatment of what almost all great players are limited by, and why the limitations rhyme, which generalizes the constraint dividend across the whole series.

What a young winger should take from this

The practical question a young player asks after reading a limitation list is whether to fix the constraint or sharpen it, and the honest answer is not the one most coaching environments give.

The reflexive answer is to develop the weaker foot, and it is not wrong so much as badly timed and badly scoped. Weaker-foot work has a high return in a specific window and a low return outside it. In the window, roughly the years when technical patterns are still being laid down rather than refined, the return is genuine, because a young player who can execute a functional contact with both feet keeps a wider action set permanently and the cost of acquiring it is low. Outside that window, the return drops steeply, because the elite standard on the strong foot is the product of a volume of repetition that cannot be replicated on the other side within a career, and hours moved from the strong foot to the weak one are usually hours moved from an action worth a great deal to an action that will remain second-tier.

The more useful framing for a developing wide player is that the weaker foot needs to reach functional rather than dangerous. Functional means the ball can be controlled, moved five yards, and struck cleanly under no pressure, which is enough to prevent the defender from pre-committing entirely. Dangerous means the finishing standard, and pursuing dangerous on the weak side in the late teens is usually a mistake.

The sharpening path is the one that produces careers at this level, and it has a specific content. Sharpen the disguise between the shot and the pass out of the identical body shape, because that is what makes one concentrated action behave like two. Sharpen the entry angle, meaning the run that arrives at the reception already facing the productive direction, because half of what looks like a first-touch advantage is actually a positioning advantage taken a second earlier. Sharpen the second action, the one available when the first is denied, because the defensive plans that work against this profile all involve denying the primary action and the difference between an elite and a good version of the profile is what happens two seconds later.

There is a second lesson available to a young player, and it is the more valuable one. Study the compensations rather than the actions. A wide forward who understands that his own threat depends on somebody threatening the space outside him will start organizing that threat himself, calling the overlap, holding the reception an extra beat until the runner commits the full-back. That is a coaching-quality understanding, it is available to a fifteen-year-old, and it is worth more than a year of weak-foot drills.

The general technique of the duel itself, the stance, the timing, and the defender’s half of this exchange, belongs to the dedicated treatment of the one-on-one duel against an inverted wide attacker, which owns that material and goes considerably deeper into it than a weakness anatomy should.

How should an amateur coach use a limitation list like this one?

Take the cheapest exploit first. Adjust the far-post starting position and the goalkeeper’s near-post reference, because those cost nothing. Then rehearse the blind-side arrival at the moment of reception. Leave the inside-lane cover and the wide targeting plan until the cheap adjustments are habitual, since both cost defensive stability you may not have.

Where a limitation becomes a plan

The point of separating diagnosis from planning is that the two require different disciplines, and collapsing them is why so many opposition meetings produce nothing.

A diagnosis is a claim about conditions. It says that under condition A, the threat drops. It carries no instruction, because a condition is not an action. The move from diagnosis to plan requires three additional decisions that the diagnosis cannot make for you: which player creates the condition, what that player stops doing in order to create it, and what the team accepts conceding while the condition holds.

Take the clearest example from this article. The diagnosis says the cut inside is denied when the covering centre-back holds the inside lane. The plan has to name that centre-back, accept that he is no longer holding the defensive line, and state explicitly that the side is now more exposed to a runner attacking depth centrally. Without those three additions the diagnosis is a piece of trivia. With them it is a plan, and the plan can be trained, reviewed, and adjusted.

This is the discipline that separates an opposition dossier that changes a result from one that gets read out and forgotten. It is also, in practical terms, an organizational problem rather than an analytical one: the conditions, the owners, the concessions, and the triggers have to live somewhere a coach can find them on a Friday night. A coach who wants to keep these breakdowns, annotate them, and turn them into an actual match-week plan can build the opposition plan free on VaultBook and keep the diagnosis, the assignments, and the concessions in one place rather than in three notebooks and a memory.

The verdict

Mohamed Salah’s genuine limitations are three, and they are all conditional.

The foot preference fixes his angle geometry, which removes a specific and mappable set of final actions and concentrates the remainder to a standard that makes the concentration worth more than the removal in most conditions. The defence’s dividend from the same constraint is plan repeatability, which is real and considerable, and which is the reason a well-drilled side can reduce his output where a poorly drilled side of similar quality cannot.

The central question is spatial. He is less effective in congestion because congestion removes the two things his wide start supplies for free, a reception facing goal and a first defender whose orientation points away from the productive direction. Nothing about the technique changes; the value of the constraint changes.

The aerial contribution is modest, which is worth about one selection decision to a defending side directly and rather more than that indirectly, because it permits a higher line, and a higher line is the condition under which the primary geometry is hardest to execute.

The supply requirement is real and is not evidence of dependency, because the correct place to look for its cost is the shape of his own team rather than his own output. An attacking full-back on that side, a holding midfielder covering that channel, and a striker pinning the covering centre-back are all jobs performed by other people so that the geometry works, and every one of them creates a seam a prepared opponent should attack.

The big-fixture claim is a variance argument that has been dressed as a character argument, and it should be returned to its original clothing. The structural half of it is true and unremarkable: better-organized opponents execute the containment geometry better, and output falls. The inferential half is unsupported and always will be, because the sample is not large enough to carry it and never becomes larger.

For a coach preparing a side, the ranked instruction is straightforward. Take the free adjustments first, meaning the far-post position and the goalkeeper’s reference. Then attack the compensation, meaning the space behind the advanced full-back on his side, because that exploit forces his own team into the arrangement that removes his outside threat. Then, and only if your personnel can carry it, commit the covering centre-back to the inside lane and accept the depth exposure that follows. The order matters more than any individual instruction, and the order runs from cheapest to most expensive, which is the order almost nobody uses and the only one that survives a poor first twenty minutes.

The wider question of how his profile sits within the family of elite wide attackers, and which of these limitations are shared across the type rather than particular to him, is handled by the cluster’s complete guide to stopping elite wingers, which owns the general version of every question this article answers in the specific.

How to verify every claim in this article yourself

Nothing above requires a data subscription or a video analysis package to check, and a reader who verifies the claims personally will hold them far more usefully than a reader who accepts them. What follows is the observation method, marker by marker, in the order a viewer should apply it.

Verifying the angle geometry

Watch a full match with the ball ignored for long stretches and the attention held on his hips instead. The marker to count is the direction of the first touch out of each reception in the attacking third. The claim in this article is that the overwhelming majority of productive first touches travel right to left across the body, and the counter-claim, that he is comfortable moving the ball either way, predicts something closer to a balance. Count twenty receptions and the pattern will be obvious in either direction, which is the point of counting rather than remembering.

The second marker is the release point of every shot. Note where on the pitch the strike is taken from, using the corner of the penalty area and the penalty spot as reference points. The prediction of the angle-constraint table is that strikes cluster in a band running from the right corner of the box toward the penalty spot, with very few from the tight right channel and very few from the left half-space. A viewer who plots twenty shots on a scrap of paper will have built a personal danger map worth more than any secondhand assertion.

The third marker is the cross origin. Note whether each delivery is taken from on the goal line or from infield of it, because the constraint predicts almost all deliveries come from infield and arrive flat rather than curled. This is the single easiest marker to check and the one that most reliably surprises people who have never looked for it.

Verifying the central question

This one requires a match in which he starts or drifts centrally for a sustained period, which is less common, so it is a marker to look for across several viewings rather than within one.

The observation is not whether he does well or badly, which is too coarse to be useful. It is the behaviour of the nearest defender behind him at the moment of reception. In a wide reception, watch the full-back’s feet: they will be square or angled toward the touchline, because that is the full-back’s default. In a central reception, watch the centre-back or holding midfielder’s feet: a well-coached one will be angled, having pre-committed to the strong side, and the reception will be squeezed toward the weak side as a result. Seeing a defender pre-commit and seeing a defender stay square are visually distinct once you know to look, and the difference between them is the entire content of the central question.

Verifying the aerial profile

The marker is not headers won, which is a noisy count. It is the arrival position on deliveries from the opposite flank. Watch three or four such deliveries and note whether he attacks the flight of the ball or arrives late and low behind the aerial contest, looking for a knock-down or a ground contact. Consistent arrival behind the contest is not a failure to compete. It is a rehearsed positional choice that reflects an accurate self-assessment, and recognizing it as a choice rather than a failing is what separates observation from complaint.

Verifying the supply dependency

The marker here is a count of a different kind. Note how many of his receptions in the attacking third occur with a teammate already beyond him on the outside. The claim is that the productive receptions cluster heavily in the subset where somebody has occupied or is occupying the outside space, and that receptions without that occupation produce a much higher proportion of backward passes.

This count is the most instructive exercise in the article for a coach, because it generalizes. Run the same count on your own wide forward and you will learn more about your team’s attacking structure in one viewing than a season of general impressions will teach you.

Verifying the pressing seam

Watch for the receptions where a defender arrives from behind and slightly inside at the moment the ball arrives rather than a beat later. They are uncommon, because most pressing is late and most pressing is frontal, but when they occur the outcome is visibly different: the protective touch goes across the body into the arriving defender rather than away from him, and the ball is either lost or recycled backward. Four or five of these across a match are enough to confirm the mechanism, and confirming it personally is what will make a coach actually rehearse it rather than nod at it.

The conditions that expose each limitation, ranked by reliability

Coaches need to know not only which conditions produce an effect but which conditions they can rely on producing it, because a plan built on an unreliable condition is worse than no plan at all: it consumes preparation time and delivers false confidence.

The most reliable condition is compression of the space behind the defensive line while conceding the space in front of it. This reliably removes the run onto the inside shoulder. It is reliable because it does not depend on any individual duel being won; it is a product of line height, which is a collective behaviour a coach controls directly. The cost, already stated, is a higher volume of strikes from the top corner of the box, and a side that cannot tolerate that volume should not use this condition.

The second most reliable is the presence of a covering defender in the inside lane at the moment of the cut inside. This is reliable in the sense that it works when it happens, and unreliable in the sense that maintaining it for ninety minutes against a side that deliberately occupies the lane is genuinely hard. It is a condition a good side can hold and an average side will hold for twenty-five minutes and then lose, which is why it belongs third in the order of implementation rather than first.

The third is blind-side pressure at the instant of reception. Its reliability depends entirely on rehearsal rather than on personnel, and a side that has practised it can produce it four or five times a match, which is enough to matter and nowhere near enough to be a plan on its own.

The fourth is congestion at the point of reception in a central zone, which is the least reliable of all, because it requires his own side to choose to play him centrally, and no opposing coach controls that choice. It is a condition to be prepared for rather than a condition to be created, and preparing for it means having a holding midfielder who knows to pre-commit to the strong side rather than staying square.

The fifth, and the one worth deliberately declining to pursue, is the attempt to force him onto the right foot in the final third. It is unreliable because the geometry of a wide-right start means the touchline is already doing most of the forcing work, and the remaining forcing has to come from a defender’s stance that simultaneously opens the run in behind. Every hour spent rehearsing it is an hour that would produce more in the first three conditions.

The debate register: what each common position gets right

Every position in the argument about this player contains something true, and identifying the true part is more useful than dismissing the position, because the true part is usually the observation and the false part is usually the inference.

The position that he goes missing in the biggest fixtures gets the observation right and the inference wrong. Output does fall against the best-organized sides, for structural reasons that are entirely explicable and have nothing to do with temperament, and the sample within those fixtures cannot support any conclusion beyond the structural one.

The position that he is one-footed and therefore limited gets the description right and the consequence wrong. The foot preference is real and it fixes the geometry, and the inference that a fixed geometry equals a reduced threat ignores that the same fixity produces an execution standard breadth cannot reach. The correct consequence is that the threat becomes plannable rather than reduced, and plannable threats are still threats when the plan is not executed.

The position that he is a system player gets the requirement right and the exceptionality wrong. The structural requirements exist and they are expensive in one specific respect, the attacking full-back, and cheap in the others. Every wide forward at this level has requirements of this kind and the honest test is relative expense rather than presence.

The position that his output depends on a specific supply structure gets almost everything right and is the strongest of the critical positions, which is why this article devotes the absorbed-weakness section to it rather than dismissing it. The refinement it needs is the distinction between being on the ball less often and being less effective on it, since those are different failures with different remedies and the debate almost never separates them.

The position that he cannot play through the middle gets the direction right and the magnitude wrong. Congestion costs him more than it costs a two-footed central receiver, for reasons that are spatial and explicable. The gap between that and cannot is very wide.

The position that criticism of him is driven by rivalry rather than analysis is the one this article is least equipped to adjudicate and least interested in adjudicating, because motive is not observable and the arguments have to be answered on their merits regardless of who is making them. What can be said is that the arguments which survive scrutiny are the structural ones, and the arguments which do not are the ones about character, and that division does not track club allegiance in any way this analysis can detect.

Facing the same profile twice

Most sides meet an opponent more than once, and the second fixture is a different problem from the first in a way that limitation lists rarely acknowledge.

In the first fixture, a defending side is executing a plan against a player who has not seen the plan. Whatever condition it creates, whether the compressed line, the committed inside lane, or the blind-side arrival, arrives as new information. The attacker’s adjustments happen inside the match, which means they are improvised, and improvised adjustments are less reliable than rehearsed ones.

In the second fixture, the plan is known. It has been reviewed, and the response has been rehearsed with the full week available. The specific responses to expect are predictable enough to prepare for. Against a compressed line the response is to receive deeper and more often, accepting a lower position in exchange for a facing-goal reception with space in front, which converts the contest from a race behind into a duel in front. Against a committed inside lane the response is an underlapping runner who occupies the lane, which forces the covering defender to choose between the runner and the cut. Against the blind-side arrival the response is a change of reception angle, checking away before the ball is played so that the arriving defender’s line is broken before he starts.

The instruction for the second fixture is therefore not to repeat the plan harder. It is to change the condition you are creating while keeping the concession identical, so that the rehearsed response is aimed at a plan you are no longer running. A side that compressed the line in the first fixture should consider committing the lane in the second, and vice versa, and should do so precisely because the response to each is different and the rehearsal does not transfer.

This is the part of opposition planning that separates coaches who prepare for opponents from coaches who prepare for fixtures. The first fixture teaches you what the response looks like. The second fixture is where that knowledge is spent, and spending it well requires having written down what you did the first time, in enough detail that the concessions and the triggers are recoverable months later rather than half-remembered.

Transferring this analysis to the player you actually face

Almost nobody reading this will prepare a side to face Mohamed Salah. Almost everybody reading it will prepare a side to face a wide attacker with a strong foot preference, a habit of coming inside, and a full-back overlapping behind him, because that profile exists at every level of football from the elite down to a Sunday league. The transfer is the point, and the transfer needs to be done carefully, because the limitations do not scale evenly.

Two of them scale up in importance at lower levels. The angle geometry becomes more exploitable, not less, because the disguise between the shot and the pass out of the identical body shape is the hardest part of the profile to execute and the part that separates the elite version from the local one. A local inverted winger cuts inside and shoots, and the shot is the only genuine threat, so the covering defender can commit to the shooting angle without fearing the pass. That single fact makes the containment geometry dramatically cheaper at amateur level than at elite level, and it is the reason amateur sides who commit a covering defender to the inside lane succeed far more often than the elite version of the same plan does.

The aerial limitation also scales up, because amateur box defending is more chaotic and a defending side that can safely ignore one aerial threat gains more than it would in an organized elite box.

Two of them scale down. The supply dependency matters less, because amateur build-up rarely produces the clean far-side switch that makes the wide reception frequent in the first place, and the wide attacker at that level is more often receiving from a recovered ball than from a constructed pattern. And the central question matters less, because amateur central zones are less congested and the defenders occupying them are less likely to pre-commit their body angle correctly.

The practical translation for an amateur coach is a short instruction set that follows directly from that scaling. Commit a covering defender to the inside lane, because at your level the pass out of the shooting shape is unlikely and the cost of the commitment is therefore lower than the elite version of this analysis suggests. Set the far-post starting position a yard narrower. Do not attempt to force him onto his weaker foot in the final third, because it will not work and it will cost you the run in behind. And spend one session rehearsing a midfielder arriving on the blind side at the moment of reception, because it is the cheapest exploit available and it is the one that transfers most completely from the elite game to yours.

The team-building side of the constraint dividend

There is a version of this analysis that matters to a coach building a side rather than defending against one, and it deserves a section because the constraint dividend cuts both ways in recruitment as well as in preparation.

A strongly constrained attacker is a known quantity. The threat set is narrow, the execution standard within it is high, and the structural requirements are identifiable in advance. A side signing that profile knows exactly which jobs the rest of the team must perform: someone occupies the outside, someone pins the covering centre-back, someone covers the flank in rest defense. The cost is legible before the player arrives, which is worth a great deal when the alternative is discovering it over eighteen months.

An unconstrained attacker, one genuinely two-footed and genuinely comfortable across the width of the front line, is the opposite trade. The threat set is wide, which makes him harder to plan against, and the structural requirements are looser, which makes him easier to fit. What he gives up is the execution ceiling on any single action, because the repetition is spread rather than concentrated.

Neither is better in the abstract, and a coach choosing between them should be choosing on the basis of what the rest of his side can supply. A team that already has an attacking full-back on the relevant side, a centre-forward willing to occupy centre-backs, and a holding midfielder who covers a flank in rest defense should take the constrained profile every time, because the requirements are already paid for and the execution ceiling is the marginal gain. A team without those pieces should take the unconstrained one, because it will spend two seasons buying the compensations and getting less from the player than his ceiling suggests in the meantime.

That is the constraint dividend applied to squad building, and it explains something that otherwise looks like inconsistency: why the same profile can look like an elite forward in one structure and a difficult fit in another, without anything about the player having changed at all. The player did not change. The compensations did.

Phase by phase: where the constraint costs most and least

The limitations are not evenly distributed across the pitch, and a coach who understands where they concentrate can allocate preparation time proportionately rather than spreading it evenly, which is how most opposition sessions waste half their value.

In the first phase, when his side is building from the back, the constraint costs almost nothing because he is barely involved. A wide forward holding a high, wide starting position during build-up is performing a positional job rather than a technical one, and positional jobs do not care which foot performs them. The only relevant marker in this phase is how early and how wide he sets, because the width he holds at the moment of the switch determines whether the reception arrives with a defender attached or with a yard of separation. A pressing side that wants to reduce his involvement attacks the switch rather than the player, which is a first-phase problem and belongs to the pressing scheme rather than to any individual assignment.

In the middle third, the constraint begins to cost something and the cost is a specific one. Receptions in the middle third arrive with more space and more time, which sounds like it favours the attacker and in this case partially does not. More time means the option set widens, and a wider option set advantages a two-footed carrier more than a constrained one, because the constrained carrier’s second and third options are lower quality regardless of how much time he has. The defensive implication is that pressure in the middle third can be applied later and from further away than a coach’s instinct suggests, since the additional time does not produce proportionally more danger.

In the final third and the entry to the box, the constraint costs most in one direction and pays most in the other, which is why almost all of the disagreement about this player concerns this zone. This is where the removed angles are removed and where the concentrated ones are concentrated, and this is where every marking, stance, and cover decision described in the artifact table applies.

Inside the box, the constraint costs least of all. Box actions are about arrival, anticipation, and contact quality at short range, and a strong-footed forward arriving into the six-yard area with a ball rolling across him is at no disadvantage whatsoever if it arrives on the correct side, which is the far-post case already noted. The defensive implication is that box marking against him should be organized around denying the arrival rather than around any foot consideration, and defenders who try to show him onto one side inside the box are solving a problem that does not exist there.

In transition, the constraint’s cost depends entirely on the direction of the recovery run behind him. A recovering defender taking a ball-side line concedes the inside push, which is the productive one. A recovering defender taking an inside-shoulder line concedes the outside push, which is far less productive but which opens the byline if the defender is beaten. The transition instruction is therefore the inside-shoulder recovery line, always, and the acceptance of the outside beat as the price. Sides who default to ball-side recovery lines are giving away the one thing the constraint hands them for free.

On defensive set pieces, meaning when his own side is defending, his positioning matters only in relation to the outlet. A high forward left up the pitch during a defensive corner is a transition threat rather than a defensive liability, and preparing for that outlet is a different problem entirely from anything in this article.

The goalkeeper’s half of the constraint

Goalkeepers get less attention in weakness anatomies than they should, given that the last action in most of the sequences described here is a strike toward a goalkeeper, and the constraint is more useful to a goalkeeper than it is to any outfield defender.

The reason is that a goalkeeper’s positioning decision is a probability calculation over release points, and a narrower distribution of release points is worth more to him than it is to anyone else on the pitch. A defender still has to defend an action; the goalkeeper only has to be in the right place, and the right place is a function of where the ball will be struck from and where it can be struck to.

Three specific adjustments follow, and they are the highest-return goalkeeping instructions available against this profile.

The first concerns the near-post reference. Against a right-footed attacker cutting in from the right, the near post is genuinely live, because the strike can be dragged back across the goalkeeper into it. Against a left-footed attacker in the same position, the near post is a much lower-probability target, because reaching it requires either the weaker foot or a body rotation the covering defender will punish. The goalkeeper can therefore set a half-step further toward the far post than his default, which is exactly the half-step that matters on a strike into the far corner. This is one of the few adjustments in football that costs nothing and gains something measurable.

The second concerns the cutback. Because the delivery arrives from infield of the byline rather than on it, and arrives flat rather than curled, the goalkeeper’s decision point is different. A curled delivery from the goal line pulls a goalkeeper toward his line and creates the six-yard-box contest. A flat delivery from infield does not, and the goalkeeper’s better option is to hold his position and let the defenders attack the ball, staying available for the strike that follows a knock-down. Goalkeepers who commit forward to these deliveries are reacting to the wrong delivery type.

The third concerns the one-on-one when the run in behind succeeds. The reception is on the left foot with the goal ahead, which means the finishing action is most likely to be a strike across the goalkeeper into the far corner or a lifted finish over him if he commits early. The instruction is patience and a slightly delayed set, because the constraint makes the disguised alternative, the roll under the body with the outside of the foot, less likely than it would be against a two-footed forward. A goalkeeper who knows the distribution can afford to wait a fraction longer, and a fraction longer is the entire difference in that duel.

What the analysis does not cover

A weakness anatomy has boundaries, and stating them is part of the integrity of the exercise.

This article has not addressed anything outside publicly observable professional football, and it will not. It carries no assessment of the person, no assessment of mentality, no commentary on any matter beyond the pitch, and no claim about anything that is not visible to anyone watching a match with attention. That is not a limitation of the format so much as the entire reason the format is worth anything: the moment a limitation list starts drawing conclusions about a person, it has stopped being useful to a coach and started being useful to an argument, and those are different products.

It has also not tried to weigh the limitations against the strengths and produce a verdict on the player’s overall standing, because that is not a question a coach preparing a plan needs answered and it is not a question this format can answer well. The relevant statement is simply that all of the limitations described here belong to a forward whose concentrated actions are executed at a level very few players in the history of the position have reached, and that the entire purpose of studying the limitations is that no other route to a plan exists. Nobody prepares against a great player by hoping he plays badly.

Finally, it has not produced the plan. It has produced the diagnosis, the conditions, the costs, and the verification method, and the move from those to a shape, a stance, a set of triggers, and a fallback belongs to a different article by design, because collapsing diagnosis and plan into one document produces a document that does neither job well.

The closing position

The most useful sentence to take from this article is not any individual limitation. It is that the same fact about a footballer can be a strength and a weakness at the same time, depending entirely on who is holding the clipboard and what they are trying to do with it.

A strong foot preference gives an attacker a repeatable, high-quality action, and gives the defence a repeatable, high-quality plan. Both dividends are real. Both are paid out of the same constraint. The question that decides a match is not which of them exists but which side is better organized to collect, and that question is settled by preparation rather than by argument.

That is the constraint dividend, and it applies to every subject this series examines. It is also why the criticism most commonly aimed at this player, that everyone knows what he is going to do, is simultaneously the most accurate observation in the discourse and the least useful conclusion drawn from it. Everyone does know. Knowing has never been the hard part.

The limitation set changes with the defensive system

One reason coaches talk past each other about this profile is that they are describing accurate experiences of different systems. A limitation that is available to a back three is not available to a back four, and a condition a mid-block can hold is one a high press cannot. The list is stable; the accessibility of each item is not.

Against a back four with a wide midfielder ahead of the full-back

This is the most common shape and the one in which the full set of conditions is theoretically available and practically expensive. The wide midfielder can shade the entry pass, the full-back can set his stance, and the covering centre-back can hold the inside lane, which means every condition described in this article can be created.

The problem is that all three jobs have to be performed at once, and each one weakens the shape elsewhere. The wide midfielder shading the entry pass is not pressing the ball, which concedes time to the passer. The full-back committed to the inside shoulder is not tight enough to prevent the reception. The centre-back holding the lane is not holding the line. A back four with a wide midfielder can therefore create every condition and cannot afford all of them simultaneously for ninety minutes, which is why plans in this shape should pick two and rehearse the sequencing between them rather than attempting the full set.

Against a back four without a wide midfielder

Here the entry pass cannot be shaded, which means the reception will happen and the plan starts a phase later. What remains available is the stance and the cover, and the correct allocation is to abandon any ambition of preventing the reception and commit fully to the geometry afterward.

The specific consequence is that the covering centre-back’s inside-lane job becomes the whole plan rather than one third of it, and the far-side centre-back has to accept that he is defending a wide zone alone. Sides in this shape often try to solve the problem by dropping the near-side central midfielder into the wide area, which produces a back five in everything but name and costs the midfield its central presence. That trade is sometimes correct and it should be made deliberately, with the concession named, rather than emerging by accident in the thirtieth minute.

Against a back three or a back five

This is the shape in which the limitations are most accessible, and it is worth stating why, because it is not the reason most coaches give.

The usual explanation is that a back five simply has more bodies, which is true and not the mechanism. The mechanism is that a wing-back and a wide centre-back can perform the stance job and the lane job simultaneously without either of them abandoning a line, because the third centre-back is holding the line behind them. The structural tension that makes the plan expensive in a back four does not exist here. The wide centre-back can step into the inside lane and the remaining two hold the line, which is the single cleanest resolution of the geometry available in football.

The cost is elsewhere and it is significant. A back five with wing-backs pinned deep concedes the whole middle third to the opposition, which means the pressure on the ball at the moment of the entry pass is minimal, the switch is available all match, and the side has almost no attacking outlet on the flank. Teams pay this price all the time and should know they are paying it.

Against a high press

The high press is the shape in which the limitation set is least relevant, and this is worth understanding because it is counterintuitive and because it points to the most effective plan available to strong sides.

A high press does not attempt to defend the geometry at all. It attempts to prevent the situation in which the geometry matters, by winning the ball before the far-side switch is available and by keeping his side’s build-up compressed toward one flank. If the ball never reaches the productive zone with a facing-goal reception, none of the angle constraints signify.

The trade is stark and familiar. A press that fails once produces exactly the transition condition the constrained forward is most dangerous in: space behind a high line, a recovering defender, and a facing-goal reception. That is the single highest-value situation available to him, and a side that presses and loses the trigger has handed it over.

The verdict on the press against this profile is therefore conditional and should be stated as such. Press if your press is genuinely coordinated, because prevention beats containment. Do not press as a compromise, because a partial press produces the transition condition without producing the prevention, and a partial press against an attacker of this profile is the worst plan available. Sides who are uncertain about their pressing coordination should adopt a mid-block, which produces neither the best outcome nor the worst, and which is the correct choice under uncertainty.

Against a man-oriented scheme

A dedicated man-marker following him across the pitch is the plan most often suggested by supporters and least often used by coaches, and the reason is instructive.

Man-marking works against a receiver whose danger comes from the reception itself. It works poorly against a receiver whose danger comes from the geometry after the reception, because the marker is behind him at the moment the geometry activates and behind is the least useful place to be. The man-marker also removes himself from the defensive shape, which means the compensations the rest of the plan depends on are unavailable.

There is a narrow version that works, which is worth naming because it is genuinely useful: a man-oriented assignment applied only inside a defined zone, with a handover at the boundary. The near-side midfielder takes him wherever he goes within the wide right channel and hands him to a central midfielder if he drifts inside. That is a zonal scheme with man-oriented behaviour rather than man-marking, it preserves the shape, and it is what most well-organized sides actually do while supporters describe it as man-marking.

Two limitations that belong to the role rather than the player

A final distinction prevents a common category error, and it is the error that produces the least useful opposition reports.

Some things that look like limitations in an individual are properties of the job he has been given, and preparing against them as if they were personal produces plans aimed at the wrong target.

The first is the absence of sustained defensive tracking. This is a role property. A side that wants a wide forward available high for the transition has bought that availability by excusing the tracking run, and the same instruction applied to any forward would produce the same visible behaviour. The plan that follows should therefore attack the arrangement rather than the individual, which means targeting the space behind him with a runner and forcing his team into the deeper shape. Preparing to punish a player for not tracking is preparing to punish a coaching decision, and coaching decisions do not get embarrassed.

The second is the concentration of touches in one third of the pitch. A wide forward asked to hold a high, wide starting position during build-up will register very few actions in the first phase, and a viewer who counts touches by zone will conclude he is uninvolved in build-up. That conclusion is accurate and meaningless, since involvement in build-up is not the job. The tactical consequence is real, though, and worth naming: because he is not part of the first phase, pressing his side’s build-up does not reduce his threat directly. It reduces it only through the switch. A press that funnels play toward his flank rather than away from it will produce more receptions in his productive zone, not fewer, which is a mistake pressing sides make regularly because the press is designed to trap the ball rather than to control where the ball ends up when the trap fails.

That second point deserves its own emphasis, because it inverts an instinct. Pressing sides usually decide which flank to overload on the basis of where the ball is and which side of the pitch their trigger fires on. Against an attacker whose threat depends on receiving with the whole width of the pitch already behind him, the more important question is where the ball will be if the trap breaks. A trap that breaks toward his flank hands him the reception he wants with the defensive shape already displaced toward the opposite side, which is the single worst arrangement a defending team can produce. Traps should therefore be set on his side rather than away from it, so that a broken trap sends the ball toward the flank where the threat is lower, and this is a decision made at the whiteboard rather than in the match.

Both of these are role properties rather than personal limitations, and a coach who understands the difference will build a plan that survives a substitution. A plan built against a player is obsolete the moment he is replaced. A plan built against the arrangement continues to work against whoever is standing in the same position.

Frequently Asked Questions

Q: What is Mohamed Salah’s biggest weakness?

The most exploitable limitation is the fixed angle geometry produced by his foot preference, because it removes a specific and mappable set of final actions and lets a defending side organize its box coverage, its covering defender, and its goalkeeper’s near-post reference with unusual confidence. Calling it the biggest weakness is accurate only in the sense that it is the most plannable one, since the same constraint concentrates his remaining actions to an execution standard that breadth cannot match. A defending coach gets plan repeatability from it; he gets execution quality from it. The genuinely costly limitations behind it are the reduced effect in central congestion and the modest aerial contribution, both of which are conditional and neither of which a side can create at will.

Q: Is Salah predictable when he cuts inside?

Yes, and predictability is not the same property as stoppability, which is where most versions of this argument fail. The family of actions is knowable: the ball travels right to left across the body, the release comes from the right half-space, and the shot and the pass emerge from an identical body shape. What knowing that does not give you is a way to deny the action without opening the run onto the full-back’s inside shoulder, which is the more dangerous of the two. Denying both requires a third player, usually the covering centre-back committed to the inside lane, and committing him costs the defensive line its horizontal alignment. Knowing what is coming is worth nothing unless the shape can afford the cover, which most shapes cannot hold for ninety minutes.

Q: How is Salah with his right foot?

Functional rather than dangerous, which is the correct standard to apply and the one that most descriptions get wrong in both directions. He can control, move, and strike the ball with the right foot competently enough that a defender cannot pre-commit his entire body to the strong side without risk, and that functional threshold is what stops the constraint becoming crippling. What the right foot does not provide is the finishing standard, the disguised delivery, or the first-time strike from the right channel, which is why the angle-constraint table lists the near-post drive and the outside-foot early cross as removed rather than reduced. Functional on the weak side and elite on the strong side is the normal shape of an elite inverted forward, not an anomaly worth criticizing.

Q: Does Salah struggle against a deep block?

A deep block removes one of his two primary threats and slightly improves the other, which is a trade rather than a solution. Compressing the space behind the defensive line takes away the run onto the full-back’s inside shoulder, because there is no space behind for the run to attack. It simultaneously concedes the zone in front of the block, and the cut inside to shoot is executed from exactly that zone. A side defending deep against him should expect a materially higher volume of strikes from the top corner of the box and should decide in advance whether it can tolerate that volume. Teams that believe a deep block has solved the problem are usually measuring the absence of the through ball and ignoring the shot count.

Q: Is Salah’s defensive work a weakness for his side?

The honest version is that his defensive job is positional rather than tracking-based, which is a coaching decision as much as a preference, and it creates one specific problem: the space behind him on the flank when the opposition full-back advances. Solving it costs something in every available direction. A near-side midfielder shuttling wide costs central compactness. The full-back defending a two-versus-one costs stability. Dropping into a five costs the advanced full-back position that his own attacking geometry depends on. That circularity is the genuine tension in the profile, and it is far more useful to a defending coach than the general complaint, because it points at a repeatable exploit: attack that space early and repeatedly to force his side into the arrangement that removes his outside threat.

Q: Does Salah rely on service from the right?

He requires a supply pattern that reaches the far side of a defensive block consistently, which is a genuine structural requirement rather than a preference, and sides without a distributor capable of switching accurately under pressure will find him on the ball less often. The distinction that matters is between being on the ball less often and being less effective when he is on it, because those are different failures with different remedies and the debate almost never separates them. A poor supply pattern reduces volume rather than quality. The requirement is real, it is shared with every wide forward at this level, and the correct place to look for its cost is the shape of his own team rather than his output.

Q: Is Salah as effective through the middle?

Less effective, for reasons that are spatial rather than technical, and the gap between less effective and ineffective is very wide. A wide start supplies two things a central start does not: a reception facing goal, and a first defender whose default body orientation points away from the space he wants. In a central zone the reception more often arrives with his back to goal, the nearest defender is a centre-back or holding midfielder trained for exactly that duel, and that defender is free to pre-commit his body angle to the strong side rather than having the touchline make the decision for him. Nothing about the technique deteriorates centrally. What deteriorates is the value of the constraint, because congestion removes the compensations the wide start provides at no cost.

Q: How is Salah in the air?

Modest, and observably so, which makes stating it analysis rather than criticism. He is not a target for wide deliveries and does not win a high share of contested aerial duels against defenders selected for that quality. The useful marker is not headers won but arrival position: watch three or four deliveries from the opposite flank and note whether he attacks the flight or arrives late and low behind the aerial contest looking for a ground contact. Consistent arrival behind the contest is a rehearsed choice reflecting an accurate self-assessment. For a defending side this is worth roughly one selection decision directly, and rather more indirectly, because it permits a higher defensive line, and a higher line is the condition under which his primary geometry is hardest to execute.

Q: Why do critics say Salah goes quiet in the biggest fixtures?

Because output does fall against the best-organized opponents, and the observation gets converted into an inference the evidence cannot carry. The structural half is straightforward: the biggest fixtures are against sides whose covering centre-back genuinely holds the inside lane, whose near-side midfielder shades the entry pass rather than chasing the ball, and whose full-back holds a stance instruction for ninety minutes. Any inverted forward’s output drops against sides executing all three. The inferential half fails on arithmetic. Goals are rare events, the subset of matches anyone classifies as the biggest is a dozen or so, and across a sample that size the difference between a poor split and an excellent one is two or three deflections. That sample cannot separate a structural effect from noise, and it certainly cannot support a claim about temperament.

Q: Does Salah need a particular system to be effective?

He needs three structural conditions: a build-up that reaches the far side of a block, a teammate threatening the space outside and behind him, and central occupation that prevents the covering centre-back sliding freely into the inside lane. Those are requirements rather than dependencies, and every elite wide forward has requirements of this kind, because a wide attacker in a permanent one-versus-two has never been productive in any era. The honest test is relative expense, not presence. Two of the three are cheap, since any centre-forward supplies the central occupation and most structures can find a switch. The third, an attacking full-back on that side, is genuinely expensive, because it forces a rest-defense arrangement that creates a specific counter-attacking vulnerability.

Q: Has Salah changed his game over time?

The role moved and the limitation set moved with it, which is the general pattern for long careers and the reason opposition reports go stale without announcing it. A touchline-width version of an inverted forward pays the highest price for the foot constraint, because beating a set defender from a standing start is exactly where the removed angles hurt. A narrower starting position with the outside space delegated to an overlapping full-back reduces that cost sharply and introduces a new one, the dependency on that overlap being present. A version arriving in the penalty area more often meets more deliveries, which raises the weight of the aerial limitation. Limitations trade rather than persist, and scouting the version you happened to watch most is the error that follows.

Q: Is Salah’s link-up work a limitation?

Mostly not, and the criticism survives because link play is three separable things that get discussed as one. Back-to-goal hold-up for a runner is modest, and judging a wide inverted forward on it is like judging a centre-back on crossing: accurate and useless, because no defending side will engineer that situation deliberately. The short combination in the wide third is a genuine and frequently used strength. The final pass is the disguise half of his concentrated action and is one of the harder parts of the profile to defend. There is one narrow version that carries planning value: under blind-side pressure at the instant of reception, the protective touch travels across the body onto the strong side, which is the side the arriving defender can anticipate.

Q: What can young wingers learn from Salah’s limitations?

That the weaker foot needs to reach functional rather than dangerous, and that hours beyond functional are usually better spent sharpening the constraint than fixing it. Weaker-foot work returns well in the window where technical patterns are still being laid down and returns poorly afterwards, because the strong-foot standard is the product of a repetition volume that cannot be replicated within a career. The sharpening path has specific content: the disguise between shot and pass out of an identical body shape, the run that arrives at the reception already facing the productive direction, and the second action available when the first is denied. The most valuable lesson is not technical at all, though. Study the compensations. A winger who understands that his threat depends on someone occupying the space outside him will start organizing that himself.

Q: Is Salah overrated by some supporters?

The question is not one this analysis can settle, and the reason is worth stating rather than dodging. Overrated is a claim about a gap between reputation and value, and reputation is not observable in the way a body shape or a release point is. What can be examined is whether the specific arguments hold, and the division is consistent: the structural arguments survive scrutiny and the character arguments do not, and that division does not track club allegiance in any way this analysis can detect. A supporter pointing at output is measuring the attacker’s half of the constraint dividend. A coach pointing at plan repeatability is measuring the defence’s half. Both are right about what they are measuring and both call it the same word, which is why the argument never resolves.

Q: Do analysts agree on what Salah’s weaknesses are?

Not entirely, and the disagreements are more informative than the agreements. Near-universal consensus covers the angle geometry produced by the foot preference and the modest aerial profile, both of which are observable in a single match by anyone counting first-touch directions and arrival positions. Genuine disagreement covers how much the central question actually costs, since the sample of central deployments is small and the conditions vary, and covers how much of the supply requirement counts as dependency rather than ordinary role definition. The big-fixture claim divides analysts from crowds rather than analysts from each other, because the arithmetic of small samples is not a matter of opinion among people who work with distributions regularly.

Q: Does Salah’s threat drop when he is doubled early?

It drops in one specific version and rises in another, which is why doubling is a decision rather than a solution. Doubling early and from the front, with the second defender arriving in his eyeline, produces very little, because he can release the ball backward before contact and the two defenders are now committed to a zone the ball has left. Doubling from the blind side at the instant of reception is materially different, because the natural protective touch travels across the body into the arriving defender, and that is the exploit worth rehearsing. Both versions cost the same thing: a defender removed from the shape somewhere else, which the overlapping full-back and the far-post arrival will attack. The blind-side version is worth the cost and the frontal version is not.