The NFL catch rule explained: what a catch now requires

A receiver goes to the ground at the sideline with the ball against his chest. The official signals a catch. Ninety seconds later a voice from a monitor says the ruling is reversed, the crowd makes a noise, and not one person in the building could tell you which of the three things he failed to do.

That is the honest state of the NFL catch rule. It is not, as the annual complaint has it, incomprehensible. It is a three-part test written in plain English, and the whole of it fits in a paragraph. The trouble is that the third part is a judgement about a thing that may never have happened, made from a camera angle that may not show it, and no amount of rewriting has ever changed that.

So here is the rule as it currently stands, what it replaced, why the old version was not as stupid as everyone said, and exactly where the remaining arguments live.

What the NFL catch rule actually says

The governing text is Rule 8, Section 1, Article 3, headed "Completed or Intercepted Pass". A forward pass is complete, or intercepted, in the field of play, at the sideline or in the end zone if a player who is inbounds does three things in order.

First, he secures control of the ball with his hand or hands, or arm or arms, before the ball touches the ground.

Second, he touches the ground inbounds with both feet, or with any part of his body other than his hands.

Third, after the first two have been fulfilled, he clearly performs any act common to the game. The rulebook gives four examples: extend the ball forward, take an additional step, tuck the ball away and turn upfield, or avoid or ward off an opponent. And then the clause that carries most of the weight: or he maintains control of the ball long enough to do so.

That is the entire rule. Everything else in the section is a note handling a special case.

The order is not decorative. Each element only counts once the previous one is satisfied, which is why a receiver who is still fighting the ball as his second foot lands has not begun the third element at all, and why a receiver who catches it cleanly and gets a foot down has completed nothing until the second foot follows.

• 3Elements a catch must satisfy, in order
• 2Feet down inbounds, or one other body part
• 1Feet required by the college code
• 1Down lost when the pass falls incomplete

All four are set by the professional playing rules, except the college figure, which comes from the NCAA code. Everything else in the rule is a judgement rather than a count.

The third element is where every argument now lives

Elements one and two are close to objective. Either the ball touched the ground before he had it or it did not. Either the second foot landed inbounds or it did not. Cameras answer both.

Element three is different in kind. It asks an official to decide whether a player performed an act common to the game, or, failing that, whether he held the ball long enough that he could have. That second option is a counterfactual. The receiver did not do the thing. The question is whether he had time to.

Nothing in the sport is harder to see at real speed, and nothing produces a wider spread of opinion in a room of experienced officials.

The rulebook's own list of qualifying acts is broad and physical: a step, a tuck and turn, a reach, a swat at an incoming tackler. It is a list of things a runner does, which is the point. The rule is trying to draw a line between a player who has caught a ball and a player who has become a ball-carrier, and the line is drawn where he starts behaving like one.

In practice the officiating has narrowed that list considerably. Reviews increasingly turn on whether a third foot hit the ground, because a third step is the one item on the list that a camera can settle in a frame. The criticism inside the officiating world is that this converts a rule with four listed options and a time-based fallback into a rule about counting feet, which is not what the text says. A receiver who catches the ball, gets two feet down and immediately tucks it away has satisfied the rule as written whether or not a third foot ever lands.

That gap between the written rule and the reviewed rule is the real source of the confusion people blame on the wording. The wording is fine. The application drifted towards the part of the wording that is easiest to prove.

What the 2018 rewrite deleted, and what it kept

Before March 2018, the rule carried a fourth requirement. A receiver who was going to the ground in the act of catching a pass had to maintain control of the ball throughout his contact with the ground. If it came loose when he landed, the pass was incomplete, no matter how long he had held it or what he had done with it in the air.

That provision produced the plays everyone remembers. A receiver in 2010 who caught the ball, got both feet down, turned, took the ball to the ground in the end zone and had it pop out on impact. A receiver in the divisional round of the 2014 playoffs who caught a deep ball, took steps, reached for the goal line and lost it as he landed. A tight end in December 2017 who caught a short pass at the two, reached the ball across the plane, and had a touchdown reversed when the ball moved as he hit the turf. In each case the ruling was correct under the rule then in force, which is a large part of why the anger settled on the rule rather than the officials.

At the annual league meeting in March 2018 the clubs voted unanimously, 32-0, to remove the going-to-the-ground requirement. The league said plainly at the time that the plays from 2010 and 2014 would be catches under the new standard.

Two things about that change are consistently misreported.

The first is that "the process of the catch" was not abolished. The phrase is still in the book. Rule 8, Section 1, Article 4 defines an incomplete pass as one in which "the process of the catch in 8-1-3 is not completed". The process survived; only its fourth step was cut.

The second is that the ground still matters, in a narrower way. The current rule carries a note whose heading is the phrase people wanted deleted, and it explains the surviving version: if a player has satisfied the first two elements but has not performed the third, and he then contacts the ground and loses control, there is no possession if the ball hits the ground before he regains it. Read that carefully. It is not a requirement to survive the ground. It is a statement that a player who has not yet completed the third element has not yet completed the catch, and hitting the ground does not do it for him.

The practical difference is enormous, and it is the difference between "did he keep it when he landed" and "was he finished before he landed".

• Rule before 2018
• Rule from 2018

Control before the ball touches the ground11

Two feet, or a body part other than the hands11

An act common to the game, or time to perform one11

Control kept through contact with the ground10

Each row scores 1 where the code demanded it and 0 where it did not. Only the fourth requirement was removed. The first three are unchanged, and so is the order in which they must be satisfied.

Show the numbers

What the 2018 rewrite changed, counted in requirements

Item	Rule before 2018	Rule from 2018
Control before the ball touches the ground	1	1
Two feet, or a body part other than the hands	1	1
An act common to the game, or time to perform one	1	1
Control kept through contact with the ground	1	0

Why the old rule was not stupid

The going-to-the-ground provision is remembered as an act of vandalism against common sense. It was not. It was solving a real problem, and the problem did not go away when the provision did.

A player falling with a ball is not in control of what happens next. The ground is going to hit him, hard, and the ball is going to be squeezed between his body and the turf at whatever angle it happens to be at. A receiver who "catches" a pass while airborne and has it knocked out by the ground has not demonstrated the thing the sport actually cares about, which is possession of the football against everything that is trying to take it away.

The old rule's answer was to say: prove it by keeping it. That is a clean standard, easy to officiate, and it produced correct rulings that felt monstrously wrong on television, because the eye sees a man with a ball in his hands and calls that a catch.

The new rule's answer is subtler. It says a player who is genuinely finished, who had control and both feet and then enough time to be a runner, is not undone by the physics of landing. A player who is not finished still is. That is a better answer, and it costs something: the moment of completion moved from a visible event, hitting the ground, to an invisible one, having had enough time.

You cannot remove judgement from this rule. You can only choose where to put it.

The catch test, step by step

1. 1The ball arrives with the receiver near the boundaryHe is running towards the sideline with a defender in trail position. Everything that follows is judged from the moment the ball reaches his hands, not from where he began the route.
2. 2Did he secure control before the ball touched the ground?Control means the ball is held with the hands or arms and is his. Movement of the ball does not by itself mean he lost it, and a ball that is still moving when it contacts the turf means he never had it. If the answer is no, the pass is incomplete and nothing else in the sequence matters.
3. 3Did both feet, or another body part, touch inbounds?Both feet, or any part of the body other than the hands. A knee, a shin, a backside and an elbow all count. Two hands on the turf do not. If any part of the foot touches out of bounds during the normal continuous motion of a step, that foot is out of bounds.
4. 4Was it a drag, or a delay?A player is inbounds if he drags the foot, or if there is a delay between the heel and the toe touching down. This is the difference between a toe-tap that counts and one that does not, and it is the single most-reviewed detail on the sideline.
5. 5Was he carried out of bounds by an opponent?If he would have landed inbounds but a defender held him up and carried him out, possession is granted anyway. This is the one place the rule looks at what would have happened rather than what did.
6. 6Did he then perform an act common to the game?Another step, a tuck and a turn upfield, a reach with the ball, a hand thrown out to ward off a tackler. The act must come after the first two elements, not during them.
7. 7If he did not, did he have time to?The fallback standard. He does not need to perform the act, only to have kept control long enough that he could have. This is a judgement about elapsed time, and it is the least visible decision an official makes on any pass play.
8. 8The catch is complete, and the ball is liveFrom this instant it is no longer a pass. It is a football in a runner's hands, and it can be fumbled, stripped, advanced or carried into the end zone like any other.

The elements are resolved in this order, and each one only begins once the previous is satisfied. Any of them answered against the receiver ends the sequence with an incomplete pass.

Feet, toes and the parts of the body that count

The second element is written more carefully than most people realise. Both feet, or any part of his body other than his hands. A knee, an elbow, a hip, a shoulder, the seat of the trousers. All of them end the question in one touch. The hands are excluded because a receiver's hands are already occupied holding the ball, and letting a hand establish possession would make almost every sideline dive a catch.

Then there is the boundary itself. If any part of the foot hits out of bounds during the normal continuous motion of taking a step, heel to toe or toe to heel, the foot is out of bounds. A player is inbounds if he drags the foot, or if there is a delay between the heel and the toe touching down.

That distinction between a continuous step and a delayed one is where sideline catches are actually decided, and it is far more precise than the "toe-tap" shorthand suggests. A receiver who lands toe first and rolls onto a heel that touches the white is out, because the step was one continuous motion. A receiver who taps a toe, pauses and then puts the heel down has two separate events, and only the first one has to be inbounds.

The other sideline provision runs the opposite way. If a player who would have caught the ball inbounds is held up and carried out of bounds by an opponent, the catch is granted, and he does not have to keep hold of the ball when he lands outside. This is the only surviving piece of the old force-out idea, and it exists because a defender should not be able to convert a completion into an incompletion by picking a man up.

Bobbles, ties and the ball that never stops moving

Three notes handle the situations that generate most of the shouting.

Movement of the ball does not automatically result in loss of control. The ball can shift, roll, or be juggled, and that alone is not an incompletion. What matters is whether control was ever secured before the ball touched the ground.

A pass caught simultaneously by two eligible opponents, with both retaining it, belongs to the passers, which is to say the offence. But it is not simultaneous if one player gains control first and the other subsequently gains joint control. That second sentence does almost all the work. Genuine simultaneous possession is rare; what looks like it is usually a receiver who had it first and a defender who got his hands in afterwards, and the ball goes to the offence either way.

And a completed catch is a live ball. Once the three elements are satisfied, a ball that comes loose before the player is down by contact is a fumble, not an incompletion, and anybody can pick it up. In the receiver's own end zone it is a fumble; in the opponent's end zone it is a touchdown or a touchback depending on what happens next. The catch rule does not only decide whether a play counts. It decides which kind of play it was.

The end zone version, which is where it hurts most

Every element of the rule applies identically in the end zone. The rulebook says so in as many words: the possession test applies in the field of play, at the sideline and in the end zone. There is no special goal-line catch rule, which surprises a great many people who assume there must be.

What changes is the consequence. In the field of play a disputed catch costs a team a down and some yards. In the end zone it costs seven points, and it usually happens in the last two minutes, which is why the plays that made the old rule famous were almost all end zone plays.

Two wrinkles are worth having straight.

A receiver who catches a ball in the end zone and satisfies all three elements has scored, and it does not matter what happens afterwards. The ball is dead the instant the third element completes, because a touchdown ends the play. There is no window in which he can lose it. This is exactly the point the old rule kept colliding with: under the going-to-the-ground provision, a receiver in the end zone had not yet scored while he was still falling, so the ground could take the touchdown away. Under the current rule, if he was finished before he landed, the play was over before he landed.

The other wrinkle runs the other way, and it is brutal. If a receiver completes the catch in the field of play, becomes a runner, and then fumbles the ball forward through the opponent's end zone and out of bounds, the ball goes to the defence as a touchback. The offence loses possession entirely, at the twenty, for a play that gained sixty yards. Nothing about that is a catch-rule question, but it is only reachable because the catch was complete. Complete the catch and you inherit every rule that applies to a runner, including the harshest one in the book.

A reach for the goal line is also the clearest available example of an act common to the game. The rulebook lists extending the ball forward first among its examples, which is not an accident. A receiver who catches it at the two, gets both feet down and stretches the ball at the pylon has performed the third element in the most visible way the sport offers, and whatever the ground does to him afterwards is irrelevant.

A completed catch does not mean the play is over

Here is an interaction the rule's critics almost never mention, and it decides more plays than the third element does.

In the professional code a player is down only when he is contacted by an opponent. There is no rule making a player down for touching the ground on his own. So a receiver who completes a catch, falls over untouched, and gets back up may stand and run, and the play is live the whole time.

Combine that with the current catch rule and something quietly significant follows. Under the old wording, a receiver going to the ground was in a defined state, the process of the catch, that ended when he stopped moving. Under the current wording, a receiver who has satisfied all three elements while falling is simply a runner who happens to be on his way to the floor. If nobody touches him, he can get up.

Officials manage this by whistling the play dead when a receiver is on the ground and contacted, which is the overwhelming majority of the time. But it is why you occasionally see a receiver scramble to his feet with the ball while eleven defenders assume the play is over, and why coaches spend practice time on precisely that scenario.

It cuts the other way too. The most damaging officiating error in this whole area is not a wrong ruling. It is a quick whistle. Once officials blow a play dead, the down is over and replay cannot give back the yards a player would have gained, as the reversal rules above make explicit. An official who lets a doubtful catch play out and rules on it afterwards has preserved every option. An official who kills it early has destroyed one of them permanently, and no review can rebuild it.

That is the reason officiating instruction on close catches leans towards ruling a completion and letting the play continue. A completion that turns out to be an incompletion is fully correctable in replay. An incompletion that turns out to be a completion is only partly correctable, because the football that would have been run has not been run.

It is not really a catch rule at all

Here is the piece almost every explanation misses. The three-part test is not confined to receivers.

Rule 3, Section 2, Article 7 defines player possession, and it applies the same three elements to any player gaining possession of a loose ball that has been caught, intercepted or recovered. Control with the hands or arms, both feet or another body part down inbounds, then an act common to the game or the time to perform one. The rulebook states explicitly that this applies in the field of play, at the sideline and in the end zone.

So the rule people call the catch rule is really the possession rule, and it governs a defender diving on a fumble as surely as it governs a receiver at the boundary. That is why a scrambled recovery in a pile can be ruled no possession, and why an interception at the sideline is judged by exactly the same standard as a catch at the same spot. One test, three codes of behaviour, applied to everybody.

Knowing this changes how you watch the sport. A great deal of what looks like inconsistent officiating across different kinds of play turns out to be the same rule doing the same job in an unfamiliar setting.

What replay can and cannot do about a catch

Whether a pass was complete or incomplete is expressly reviewable. It is the first item listed under plays involving possession in the replay rule, which tells you how central it is to the way the professional game handles reviews and challenges.

Two provisions attached to it are the source of a recurring and entirely legitimate grievance.

If a ruling of incomplete is changed to a catch, the ball is placed at the spot where it was when the receiver's second foot or body part touched the ground with control. No advance is awarded. Everything the receiver did after that point is gone, because the officials blew the play dead and he was not permitted to do it.

If a ruling of incomplete is changed to a catch and a fumble, the ball goes at the spot of recovery to whichever team recovers it in the immediate continuing action, and if nobody clearly recovers it, back to the team that lost it at the spot where possession was lost.

Put those together and you get a situation that looks like an injustice and is in fact the only fair option available.

Ruled complete on the field, and it stands35yd

Ruled incomplete, and it stands0yd

Ruled incomplete, changed to a catch on review15yd

Ruled complete, changed to incomplete on review0yd

Constructed illustration. A receiver gets his second foot down fifteen yards past the line of scrimmage and runs a further twenty before being tackled. The four rows apply the actual rules to the same play under four different rulings. The yardages are invented to make the arithmetic legible.

Show the numbers

Worked example: what one contested catch is worth

Item	Yards the offence ends up with
Ruled complete on the field, and it stands	35yd
Ruled incomplete, and it stands	0yd
Ruled incomplete, changed to a catch on review	15yd
Ruled complete, changed to incomplete on review	0yd

The third row is the one people argue about. The receiver ran thirty-five yards. The offence gets fifteen and a first down, because the play should never have been whistled dead and the rules will not hand a team yards it did not earn under live conditions. The alternative is worse: awarding the full distance would mean replay was inventing a result rather than correcting a ruling.

Reviews of catches also reach the officiating department in another way. The league office may advise on-field officials on objective aspects of a play where clear and obvious video evidence exists, and completed or intercepted passes are on the list of things it may address. This is a lower-friction route than a formal challenge and it is the reason a certain number of obvious errors are now fixed before anybody throws a flag.

Every code that added video review has had this argument. The professional football version of it is at least explicit about the trade, which is more than can be said for the way football's video review handles the same problem, where the equivalent decisions are made without a published rule about where the ball goes afterwards.

College is a different rule, and it shows up in scouting

The most consequential difference between the codes is short enough to state in a line: the college game requires one foot inbounds, the professional game requires two.

That sounds minor. It is not. A receiver who has spent four years learning that a single foot at the sideline finishes the play arrives in a competition where he must land a second one in a strip of grass he was never trained to think about, against defenders who know it and are working to push him towards the white. Sideline production frequently does not carry over, and it is not a change in the player's hands or route running. The rule underneath his feet changed.

The same logic runs through a good deal of what looks like adjustment from one level to the next. The contact rules for defenders change too, which is why coverage that works in one code can look completely different in the other, and why so much scouting effort goes into projecting a college performance onto a rulebook that did not govern it. Anybody working through the rest of the American football coverage will find the same pattern in almost every rule that differs between the two codes.

Where the arguments still live

The rewrite removed one category of dispute and left three.

The first is control before the ground. A ball fought for all the way to the turf is genuinely ambiguous, and different camera angles support different answers. No wording fixes this, because the fact is not visible.

The second is the second foot. Toe drags, heel taps and the exact instant a shoe crosses the white are settled by frame rate as much as by rule, and the difference between a continuous step and a delayed one can be a matter of a few hundredths of a second.

The third is the act common to the game, and it is the largest. As long as the standard includes "or maintains control long enough to do so", an official is being asked to price a hypothetical. The narrowing towards counting a third step is an understandable response to that, and it is also a departure from the rule as written.

There is a fourth, quieter consequence that rarely gets discussed. A receiver is considered a defenceless player throughout the entire process of the catch, and until he is capable of avoiding or warding off the impending contact of an opponent. That means the catch rule is also a safety rule: the moment the process ends is the moment a hit that was a fifteen-yard foul becomes a legal tackle. Any change to when a catch is complete is a change to when a receiver stops being protected, which is a large part of why the competition committee moves slowly on this particular paragraph and why it interacts with the other fouls that turn on what a defender did and when.

How to judge a catch yourself, in about four seconds

Next time a ruling is disputed, work down the list in order and stop at the first one that fails.

Did he have it before the ground touched it? If the ball is still moving when it meets the turf, stop there. It is incomplete and nothing else is relevant.

Did two feet, or a knee, hip or elbow, get down in bounds? If it was a foot at the sideline, ask whether the step was continuous or whether there was a delay between the heel and the toe. Continuous, and any part of the foot on the white kills it. Delayed, and only the first contact has to be clean.

Then ask the only genuinely hard question: after both of those, did he do anything a runner does, or did he simply have the ball, upright and settled, for a beat with nobody hitting him?

If the honest answer to that last one is "I am not sure", you have located the rule's real boundary, and so has the official. That is not a failure of the wording. It is the residue that any wording leaves behind, and the reason this particular paragraph has been rewritten before and will be again.