NBA replay review rules: what can and cannot be reviewed

A referee blows the whistle, a coach twirls his index finger at the officials, and about a third of the time nothing happens at all. The play is not reviewed. It is not that the officials disagreed with him. It is that the thing he wanted looked at was never on the list.

Almost every argument about the NBA replay review rules comes back to that list. Fans assume replay is a general-purpose appeal against error, and it is not. It is a set of named triggers, each with a named set of questions the officials are permitted to answer, and a strict rule that they may answer nothing else. Get the structure right and the frustrating parts stop looking arbitrary. They start looking like what they are, which is a series of deliberate decisions about where accuracy is worth the interruption.

This is that structure, from the trigger list to the room in New Jersey that makes most of the calls.

The NBA replay review rules sort every call into three boxes

There are three categories, and confusing the second with the third is the source of nearly all the shouting.

Automatically reviewable. The officials on the floor, or the referee sitting in the Replay Center, start the review themselves. No coach asks. No timeout is spent. These are situations where the league has decided the risk of getting it wrong is high enough that a review is mandatory whenever there is doubt.

Challengeable only. Three specific things can be looked at if, and only if, a head coach spends a timeout to demand it. If he does not, the call stands regardless of how plainly wrong it is.

Not reviewable at all. Everything else. A travel, a three-second violation, an uncalled foul, a shooting foul that was actually a clean block but was whistled in the first quarter with both teams holding all their timeouts and the coach choosing to keep them. None of it can be revisited. Replay in the NBA is not a safety net under the whole game. It is a set of specific holes cut in the floor.

The rulebook keeps these separate on purpose. Rule 13 governs the automatic side and Rule 14 governs the challenge, and they are written as two different instruments because they answer two different questions. Rule 13 asks: what must never be got wrong? Rule 14 asks: what is a coach entitled to spend a resource on?

• 20Replay stations built into the room
• 94Television monitors
• 29Arenas connected directly to it
• 15Reviewable triggers monitored per game

Figures from the league's own announcement of the facility. The arena count is lower than the team count because two clubs share a building.

What the officials review without anyone asking

The automatic list in Rule 13 is longer than most people realise, and it splits into two halves: things reviewable at any point in the game, and things that only become reviewable in the closing minutes.

The always-on group is dominated by timing and scoring, which is exactly what you would expect from a system designed around the idea that some errors are unfixable afterwards.

A field goal made with no time showing on the clock is reviewable. So is a foul called at 0:00. Both matter because a period boundary is the one place where a mistake cannot be washed out by the rest of the game. If the shot was late, the points come off. If the shot was in time, the officials may also then rule on whether it was worth two or three, whether the shooter was out of bounds, whether the shot clock had already expired, and whether anything unsportsmanlike happened in the sequence. That cascade is worth noticing: a single trigger opens a defined set of secondary questions, and the officials answer them in one go rather than requiring a separate review for each.

Uncertainty about whether a made basket was a two or a three is reviewable. So is uncertainty about whether the ball left the shooter's hand before the shot clock expired. Both of these are the bread and butter of the system, and both are handled in a way described further down that surprises people who assume the referees on the floor are always the ones looking.

A suspected clock malfunction is reviewable, and this is the widest power in the rule. If the game clock started late, started early or froze, the officials may determine how much time actually elapsed and put the correct figure on the board, then rule on whether any shot or foul in that window was timely.

Flagrant foul classification is reviewable whenever the officials are unsure whether contact met the criteria, and an altercation is reviewable so that the officials can establish who was involved, how involved each person was, and what each of them should be charged with. These two are the exceptions to almost every other rule about where decisions are made, and they are the exception for a reason covered below.

Clear-path fouls are reviewable, and the review is a checklist rather than a judgment: was the ball ahead of the offensive player in the backcourt, was there a defender ahead of him, did he have control, was he deprived of a scoring opportunity, was the foul before he began his shooting motion. Off-ball fouls are reviewable. So is a suspicion that a team has six players on the floor, which sounds absurd until you remember how substitution works during a scramble.

And there is a small, humane one: if the officials are not sure which player was fouled, they may look at the video to identify him, so the right man shoots the free throws. They may not use that review to change who committed the foul.

The last two minutes are a different game

Four situations are reviewable only in the final two minutes of the fourth period and in overtime. This is the part of the rulebook that produces the most confusion, because a play that would be examined in minute 47 is waved through in minute 20.

Whether the ball hit the rim, for shot clock purposes. Whether a defender was inside the restricted area on a block or charge. Whether goaltending or basket interference occurred. Those three are late-game only.

The reasoning is stoppage economics. Each of these is a genuinely difficult call, each is common, and reviewing every instance would add several minutes of dead time to a broadcast for calls that individually rarely decide anything. In the last two minutes they usually do decide something, so the calculation flips.

The rim-contact review is the most technically interesting of the three, and the least understood. If play has already stopped, the officials time the sequence and correct the shot clock. If play is continuous, the review happens at the first stoppage, and if the shot clock is found to have been wrong it is reset and the ball is put back in play at the nearest spot. The officials may also award possession to a team that was about to gain it. There is a deadline built in: if the review does not happen within the window the rule specifies, the play cannot be reviewed at all afterwards. That is unusual in a rulebook that mostly avoids hard cut-offs, and it exists to stop a team asking for a re-examination of something that happened ninety seconds and four possessions ago.

Why the Replay Center reviews the first 46 minutes by itself

Two of those triggers are handled without the referees on the floor doing anything, and the arithmetic in the rule tells you exactly why.

During the first 46 minutes of regulation, and the first three minutes of any overtime period, the Replay Center Official reviews two-versus-three scoring and shot clock expiration on his own initiative, immediately, without stopping the game. Regulation is 48 minutes. Overtime is five. Subtract the reviewable-by-everything last two minutes from each and you get 46 and three.

So the rule is not really about the first 46 minutes. It is about everything except the endgame. For the bulk of a basketball match, the two most mechanical questions in officiating, was that shot behind the line and did it beat the shot clock, are answered by a referee watching a synchronised feed in New Jersey while the ball is still live. The score is corrected quietly. Nobody stops. In the last two minutes those same questions revert to the on-court crew, because by then a stoppage is affordable and a visible, explained correction is worth more than a silent one.

Secaucus, and what the room actually is

The NBA Replay Center opened in Secaucus, New Jersey, at the start of the 2014-15 season. The league built it with 20 replay stations and 94 television monitors, connected by a high-speed network directly to the arenas, which at the time numbered 29 across 30 clubs.

The staffing model is the part that matters. Each game gets a dedicated Replay Operator whose job is to watch for the reviewable triggers and the referees' signals, with Replay Managers overseeing the room. When the facility opened, the operators prepared angles and the crew chief still made every decision at the scorer's table, with the room supplying better pictures than a courtside monitor ever could.

That changed a year later. From the 2015-16 season the league staffed the Replay Center with current NBA referees and moved decision authority for a defined set of situations to them. The Board of Governors split the reviewable list in two: eight categories to be decided in Secaucus, nine to remain with the crew on the floor.

47%53%

• Decided by a referee in the Replay Center8
• Decided by the crew chief on the floor9

The division approved by the Board of Governors when decision authority first moved off the floor. Counts of situation categories, not of reviews. The balance has shifted further towards the Replay Center since.

Show the numbers

The 2015-16 split, when referees were first stationed in Secaucus

Item	Value
Decided by a referee in the Replay Center	8
Decided by the crew chief on the floor	9

The eight sent to New Jersey were the mechanical ones: two-point versus three-point, made baskets at the end of a period, out-of-bounds, shot clock violations on made field goals, goaltending and basket interference, non-foul clock malfunctions, 24-second resets, and counting the players on the court. The nine kept on the floor were the judgment ones: flagrant fouls, clear-path fouls, off-ball fouls, altercations, end-of-period fouls, shot clock violations that involved a foul, free throw shooter corrections, foul-related clock malfunctions, and restricted area calls.

Read that division and the design philosophy is obvious. A question with a right answer visible in the frame goes to the person with the best frames. A question requiring a judgment about intent, severity or discipline stays with the people who were in the building and felt the temperature of the game.

The current rule has pushed further in that direction. Rule 13 now provides that the Replay Center Official determines the outcome of every review triggered under the automatic list except flagrant fouls and altercations, which the on-court officials decide with the Replay Center's assistance. Two categories out of the whole list remain with the crew, and both are disciplinary rather than factual.

The mechanics on the floor follow from this. If Secaucus is deciding, the calling official goes to the scorer's table, puts on a headset and receives the ruling. If the crew is deciding, the crew chief and a second official go to the table and rule with the Replay Center's help while the third official watches the benches. In both cases two officials must tell both benches that the play is being reviewed, which is a small courtesy with a real purpose: a coach who does not know a review is under way cannot make an informed decision about his own challenge.

The review is supposed to take two minutes, and usually takes far less

Rule 13 says reviews should be conducted within two minutes, with discretion to run longer for unusual circumstances such as a possible ejection, a technical problem, an unusually difficult angle, or an altercation, which may take as long as it reasonably needs.

The league has published its own numbers on how long reviews actually take, and the trajectory in the Replay Center's early years was steep.

2014-15 season42.1s

2015-16 season31.9s

The league's published season averages. The fall follows the 2015-16 change that put referees in Secaucus and moved decisions off the floor for eight categories.

Show the numbers

Average NBA replay review time, first two seasons of the Replay Center

Item	Value
2014-15 season	42.1s
2015-16 season	31.9s

A quarter of the review time disappeared in one season, and the cause was structural rather than technological. Once a referee in the room could rule directly, the crew chief no longer had to walk to the table, watch, deliberate and return for the mechanical calls. The review happened while he was still standing near the play.

That is the whole argument for centralised replay compressed into two numbers. The pictures were already good in 2014. What was slow was the process wrapped around them.

The coach's challenge covers three things and nothing else

Rule 14 is short, and the list at the heart of it is shorter still. A head coach may challenge:

• a called personal foul charged to his own team
• a called out-of-bounds violation where his own team was not awarded possession
• a called goaltending or basket interference violation

That is the complete set. Every word in it is load-bearing.

Called means there must have been a whistle. A missed foul is not challengeable, because there is no call to overturn. This is the single most common misunderstanding in the arena and the one that generates the most futile finger-twirling.

Charged to its own team means a coach cannot challenge a foul called against the opposition in the hope of getting it wiped. He can only argue that his own player did not do it. The rule for calls against the other team is that you were already given what you wanted.

Where its own team was not awarded possession is the same constraint applied to out-of-bounds. You may challenge only the version that hurt you.

The goaltending category carries a further restriction: it disappears in the last two minutes of the fourth period and in overtime, because in that window the officials trigger those reviews themselves under Rule 13. A coach is not being denied anything. He is being told the review is now automatic, so he should keep his timeout.

Notice what is missing. Travelling. Carrying. Three seconds. Defensive three seconds. Backcourt violations on their own. Whether the foul that was called should have been a flagrant, which the officials may reach through their own review but a coach cannot demand. And, above all, non-calls of any kind. The mechanics of the fouls themselves, and why some contact is a foul and some is not, are a separate subject with its own logic, and replay does not touch that logic at all. It only asks whether the thing that was whistled happened.

The challenge sequence, step by step

1. 1A whistle goes, and the call is one of the threePersonal foul on your own team, an out-of-bounds against you, or goaltending outside the last two minutes. Anything else and the process stops here.
2. 2The head coach calls a legal timeoutNot an assistant. The team must have a timeout to spend. Attempting a challenge with none left is an excessive timeout, penalised as a technical foul.
3. 3He twirls an index finger at the officials, at the same timeThe signal and the timeout are simultaneous, not sequential. He must also tell the officials which specific event he is challenging.
4. 4The clock on the request is already runningThe challenge must come before the ball is handed to the thrower-in, handed to the free throw shooter, or tossed for a jump ball. During a mandatory or opposing timeout, he has 30 seconds from its start.
5. 5The review is conductedOut-of-bounds and goaltending challenges are decided by the Replay Center Official. A challenged personal foul is decided by the crew chief on the floor.
6. 6The call is overturnedThe challenge is successful. On a first challenge the timeout is returned and the team is awarded a second challenge for later in the game.
7. 7The call standsThe challenge is unsuccessful and the timeout is gone. There is no second challenge, and the team is down a stoppage it may badly want in the final minute.

Every one of these steps is required. Miss the timing, miss the signal, or fail to name the event and there is no challenge, only a spent timeout.

Two details in that sequence are worth pulling out because they decide games.

The first is the deadline. A coach does not have until the next dead ball. He has until the ball is put back in play, which after a shooting foul can be a matter of seconds and after a sideline out-of-bounds can be almost no time at all. Coaches who challenge well have an assistant whose entire job in that moment is to say yes or no before the referee hands over the ball. A team that hesitates has not lost an argument. It has missed a window.

The second is the delay-of-game trap. If a team is called for delay of game before it calls its timeout, it forfeits the challenge. Slow-walking the decision is not a neutral act.

The retained timeout is the whole economic design

The challenge costs a timeout, and getting it right refunds the timeout. That single rule does more work than anything else in Rule 14.

Consider what would happen without the refund. A coach in the second quarter, facing a foul call he is 70 per cent sure is wrong, would rarely challenge, because a timeout in the last ninety seconds is worth more than a foul in the second quarter. The challenge would become a fourth-quarter instrument only, used two or three times a season by each team, and the mechanism would fail at exactly the thing it was introduced to do, which is remove clear errors from the record.

With the refund, the calculation changes. A coach who is confident pays nothing. A coach who is guessing pays the full price. The rule prices confidence, not correctness, and it does it without any subjective assessment of how obvious the error was.

The second challenge, introduced by the Board of Governors in July 2023, extends the same logic one step. Be right, and you get another go. There is a deliberate limit on the generosity: the timeout used on that second challenge is not returned even if it succeeds. So the reward for a correct first challenge is one further opportunity, not an unlimited stream of them. A coach who is right twice ends the game one timeout down, and a coach who is right once ends it level.

The challenge itself began as an experiment. The Board of Governors approved it in July 2019, alongside the change that let the Replay Center start its own reviews of scoring and shot clock questions. It was made permanent in December 2020, from the 2020-21 season. The expansion to a second challenge followed in 2023.

Clear and conclusive visual evidence, and why obvious errors survive

The standard for overturning a call is that the video must provide clear and conclusive visual evidence that the call was incorrect. Not evidence that it was probably incorrect. Not a better-than-even reading of a blurry frame.

This is why so many reviews end with the announcement that the call on the floor stands, in situations where the arena and the broadcast crew are convinced. Two things are being asked, and only one of them is what the crowd is asking.

The crowd asks: what actually happened? The officials ask: does the footage available prove the original decision wrong? On a foul with contact hidden behind a body, on an out-of-bounds where the ball is obscured by a hand at the exact frame it changes direction, the honest answer to the second question is often no, even when a reasonable person watching in slow motion would guess at an answer to the first.

The standard is not a bug. A lower bar would mean reviews that overturn calls on the balance of probabilities, which is a system where every close play becomes a second, worse coin flip conducted by people with less information than the official who was six feet away. The high bar keeps the on-court decision as the default and requires the video to beat it.

The same standard governs the league's published grading of officiating after the fact, which is the next thing worth understanding.

The last two minute report grades the game nobody can change

The league publishes an assessment of officiating in the closing stage of close games. A match qualifies if the score was tied or within three points at any point in the last two minutes of the fourth period, and any overtime is included.

The report covers every whistle in that window and the non-calls the league considers notable, which broadly means those that directly affected a possession. Events that were only tangential, or only visible with technical aids such as magnified stopclock footage, are noted but not graded as incorrectly officiated. The evidentiary bar is the one used during the game: there must be clear and conclusive video evidence before the league will label a play as incorrectly officiated.

What the report does is put the league's own judgment of its officials on the record, in public, the day after, in a form anybody can read. What it does not do is change anything. A call graded incorrect in a report published on Tuesday morning does not alter Monday night's result, does not restore possession, and does not produce a replayed period. It is transparency without remedy, and reasonable people disagree about whether that combination is honest or merely uncomfortable.

The practical value of the report is that it is the only systematic public record of how often the difficult calls go wrong, and in which categories. Anyone who wants to argue about officiating standards with something other than memory has to start there.

The clock hierarchy nobody knows exists

Buried in Rule 13 is a piece of procedure that reveals how seriously the league takes end-of-period timing, and how many clocks there are in a modern arena.

When the Replay Center Official is ruling on scoring, timing or a foul at the end of a period, and the clocks disagree, there is a fixed order of authority. The backboard clock first. Then the LED lights around the backboard. Then the balcony facade clock. Then the ceiling scoreboard clock. Then, last, the superimposed clock on the television picture.

That ordering is a statement about latency. The clock physically closest to the shot, wired most directly to the timing system, is the most trustworthy. The television graphic, which has passed through a production chain before it reaches the frame, is the least. Broadcast viewers who have watched a buzzer-beater and been certain from the on-screen clock have been reading the source the rulebook trusts least.

There is an exception. Where the malfunction being investigated is the clock itself, an early start, a late start or a freeze, the digital clock in the Replay Center governs instead. The room keeps an independent record precisely so that a failure in the arena's own timing does not leave the officials with nothing to rule from.

And a homely fallback sits at the end of the section: if the replay system is not working or is unavailable, the officials use the nearest courtside broadcast production monitor. The best system in professional sport, with a designated backup that is somebody's television.

The flopping wrinkle

The Board of Governors introduced an in-game flopping penalty in July 2023, and its interaction with replay is a small masterpiece of rule-drafting.

A flop is a physical act that reasonably appears to be intended to cause the officials to call a foul. Calling one is a non-unsportsmanlike technical foul, worth one free throw to the opposition, which may be taken by any player on the floor at the time. Nobody is ejected for it. Officials need not stop live play to call it and may wait for the next neutral stoppage, so a flop does not cost the flopped-on team its scoring opportunity. Both a foul and a flop can be called on the same play, which is the honest outcome when a player is fouled and then embellishes.

The replay connection is the interesting part. A flop is not itself challengeable. A coach cannot demand a review to have an opponent penalised for flopping. What officials may do is call a flop out of a review they were already conducting for another reason, and the reviewable-matters lists in Rule 13 name flopping violations again and again as a secondary question attached to other triggers.

So flopping is reviewable in practice but never on its own account. It rides along on reviews that were happening anyway. That is a deliberate design: the league wanted the deterrent without adding a new category of stoppage, and this is how you get one without the other.

What the NBA replay review rules still cannot fix

The honest limits are worth stating plainly, because they explain most of what fans experience as unfairness.

Non-calls are almost entirely outside the system. With narrow late-game exceptions, if there was no whistle, there is nothing to review. A missed foul in the last thirty seconds is discussed in the two minute report and nowhere else.

Judgment about contact is not transferable to video. Officials review whether the events forming a clear-path foul occurred, whether a defender's feet were in the restricted area, whether a shot beat the clock. They do not review whether a defender's contact was enough to warrant a whistle, and the rulebook is careful never to give them that power.

Travelling and the ball-handling violations are untouchable. The most argued-about calls in modern basketball are outside both Rule 13 and Rule 14 entirely.

Reviews are scope-limited by their trigger. This is the constraint least understood outside officiating circles. Each trigger carries its own list of questions the officials may answer, and they may answer no others. Officials looking at a buzzer-beater who happen to notice an uncalled foul earlier in the possession cannot act on it. The review is a window onto specific facts, not a general re-examination of the play.

Correcting the shot clock does not undo what followed. Where a clock error is found in continuous play, the rule resets the clock and restarts from the nearest spot. It cannot restore a possession that was played out under the wrong number.

Set against those limits, the system does a narrow job well. Scores are right. Buzzer-beaters are right. The out-of-bounds calls that decide possession in the last minute are right far more often than they were. The line between two and three is now essentially never wrong, which is not something anyone could say of the sport twenty years ago. Other sports have drawn the boundary elsewhere: cricket puts the decision itself in the hands of a review with a player-initiated appeal and a deliberate margin of error built into the ball-tracking, and tennis has gone further still by replacing the line judge with the machine. Basketball chose to review facts and leave judgment alone, and every complaint about NBA replay is really a complaint about that choice.

What to watch for during a review

Four things tell you what is happening while the officials are at the table, and none of them requires knowing the rulebook by heart.

Who walked over. If a single official is at the scorer's table with a headset and is not looking at the monitor for long, the ruling is coming from Secaucus and the category is mechanical. If the crew chief and a second official are both there and the third is minding the benches, it is a flagrant foul or an altercation and the decision is being made in the building.

Whether a coach twirled. No twirl means Rule 13 and no timeout at risk. A twirl means Rule 14, a timeout already spent, and a bench that will be one stoppage poorer if the call stands.

How much time is on the clock. Inside two minutes of the fourth period, the reviewable list is at its widest and goaltending has moved from the coach's hands to the officials'. Outside it, the list is much shorter than the crowd assumes.

What the officials announce afterwards. The wording is not decoration. A call that is overturned met the clear and conclusive standard. A call that stands did not, which is a different statement from a call that was confirmed, and the difference is precisely the gap between the video proving something and the video failing to disprove it.

The rest is on the list, published, and rather more logical than it looks from a seat in the upper bowl. More on how this sport is governed, refereed and argued about is in the basketball archive, and if you want to see the raw material the officials and everyone else are now working from, the optical tracking systems that record every position on the floor are the other half of the same story.