Skip to content
CricketTaken

Explainer

The baseball shift ban explained: what the rule really says

The shift ban is not a ban on shifting. The exact positioning requirement, the two moments it is tested, the inning lock, and what a violation now costs.

By CricketTaken EditorialPublished Explainer20 min read

How this is written and checkedReport an error

A left-hander turns on a fastball and drives it on a low line into right field. Before 2023, there was a fair chance that ball was caught standing still by a man wearing a shortstop's number who had jogged out onto the outfield grass behind first base before the pitch, on the instruction of a spray chart. Since 2023 he is not allowed to be there, and the same ball is a single.

That is what the shift ban did. Almost everything else said about it in baseball coverage, starting with the name, is wrong.

The rule does not ban shifting. Clubs still position every fielder according to where they expect the ball to go, on every pitch, and they do it with more data than they had in 2015. What the rule bans is one narrow arrangement of bodies: more than two infielders on one side of second base, and any of those infielders standing on the outfield grass, at the instant the pitcher lets go of the ball. That is the whole prohibition. It is enforced by the umpire's eyes, it fits in two sentences, and the interesting parts of it are the ones nobody reads.

What the shift ban actually prohibits

The restriction lives in Rule 5.02(c) of the Official Baseball Rules, under the heading "Infielder Positioning", and it opens by granting the freedom it is about to qualify. Any fielder other than the pitcher and the catcher may station himself anywhere in fair territory, except as described below.

Then come three conditions, and they are not the same condition stated three ways.

The first says that when the pitcher is on the rubber and begins the natural movement associated with delivering the ball, the defence must have at least four players besides the pitcher and catcher with both feet completely in front of the outer boundary of the infield dirt.

The second says that when the pitcher releases the ball, the defence must have at least four players besides the battery with both feet completely in front of that same boundary, and at least two of them must be positioned with both feet entirely on each side of second base.

The third says that from the release of the first pitch of a half inning, the two infielders on each side of second may not switch sides or move to a position other than their side of the infield for the rest of that half inning.

Read them in order and the architecture is obvious. The first clause fixes the depth. The second clause fixes the depth again and adds the left-right requirement. The third clause stops the whole thing being defeated by rotation between batters.

The quantities the rule fixes
  • 4Fielders required in front of the dirt boundary
  • 2Infielders required on each side of second base
  • 2Moments in the pitch at which the alignment is tested
  • 95Radius in feet of the infield dirt from the rubber

Counts taken from Rule 5.02(c) and Rule 2.01 of the Official Baseball Rules. None of them varies by club, park or situation.

The rule is tested twice, and the second test is the real one

Most explanations of the shift restriction mention only the moment of release. The rulebook checks twice, and the reason is worth understanding, because it tells you what the drafters expected clubs to try.

Test one happens when the pitcher begins the natural movement of his delivery. At that instant, four fielders must be on the dirt. Note what is not required yet: nothing about sides of second base.

Test two happens at release. Four fielders on the dirt again, and now two of them entirely on each side of second.

Why not simply test once, at release? Because a single test at release would allow a second baseman to stand in shallow right field until the pitcher started his motion and then sprint back to a legal position, arriving a fraction before the ball leaves the hand and departing again the moment it does. The depth requirement at the start of the delivery closes that off. A fielder who is on the outfield grass when the pitcher's leg comes up has already broken the rule, whatever he does afterwards.

There is also a catch-all. The Comment attached to the rule tells umpires that its purpose is to prevent the defence having more than two infielders on either side of second base in an effort to anticipate where the batter will hit the ball before the pitch, and instructs them to assess the penalty against any fielder who, in their judgement, tries to circumvent that purpose. That sentence is doing a lot of work. It converts a geometric rule into a rule about intent, and it means a defence that satisfies the letter of both tests can still be penalised for the obvious dodge. Umpires are given the same power over creative positioning that they have over creative pitching motions in the balk rule, and for the same reason: a purely mechanical test invites a purely mechanical workaround.

One pitch, and the three points at which the alignment is judged
  1. First pitch of the half inning is releasedWhoever is standing on the third base side of second is locked to that side, and whoever is on the first base side is locked to that side, for the rest of the half inning.
  2. The pitcher begins his natural delivery motionFour fielders besides the pitcher and catcher must have both feet completely in front of the outer boundary of the infield dirt. Sides of second base are not yet checked.
  3. The pitcher releases the ballThe same four must still be in front of the dirt boundary, and two of them must now be entirely on each side of second base. This is the test that catches a fielder drifting late.
  4. The ball is in playEvery fielder may move anywhere. The rule constrains where a defence stands before the pitch, not where it runs afterwards.
  5. A substitution is madeThe incoming fielder may take any side. From the first pitch after he arrives, he is locked in on the same terms until the half inning ends or another substitution is made.

The sequence set out in Rule 5.02(c)(i) to (iii). A defence can be legal at one checkpoint and illegal at the next, which is the entire point of testing more than once.

The clause everybody skips: you are locked in for the inning

The third condition is the one that almost never gets explained, and it changes how a defence is built rather than merely how it stands.

Once the first pitch of a half inning has been released, the two infielders on the third base side stay on the third base side, and the two on the first base side stay there. They may not swap between batters. A club that wants its shortstop on the right side of the diamond for the left-handed hitter due up fourth cannot simply move him across when that hitter walks to the plate, because the alignment set at the start of the inning binds for the whole inning.

The single exception is substitution. Any infielder may switch sides, or move to any other position, at the time of a substitution for one of the defensive players. The carve-out excludes a straight pitching change that brings in a pitcher for someone not already in the game, so a club cannot manufacture a re-alignment simply by going to the bullpen. And the replacement inherits the constraint: any player who legally replaces an infielder during an inning is himself locked to his side from the first pitch after he arrives, until the half inning ends or another substitution happens.

This is more restrictive than it sounds. It means the shift decision is not made batter by batter but inning by inning, on the basis of the whole half inning's likely batting order rather than the man currently in the box. A defence that guesses wrong about the second and third hitters of an inning has to live with the guess. Managers who want flexibility have to buy it with a substitution they might not otherwise make, which is a real cost in a game where a bench is short and a roster spot is contested.

Where the line actually is, in feet

The rule refers repeatedly to the outer boundary of the infield dirt. That phrase would be useless if the dirt were a matter of taste, and until relatively recently it very nearly was. The rulebook now pins it down.

Rule 2.01 requires that the grass line of the outer boundary of the infield dirt be a 95-foot radius from the centre of the pitcher's plate. A club is in compliance if the average distance of that line is more than 94 feet and less than 96, and no individual measurement falls outside the same band. The same rule also fixes the strip of dirt running along the first base line at no less than 18 inches and no more than 24 inches wide.

Why this matters: the shift restriction is a rule about a physical line painted onto the ground by the home club. Before the boundary was standardised, an infield with a generous arc gave its defence more legal room to stand deep than an infield with a tight one, and a rule expressed in terms of the dirt would have meant a different thing in every park. The 95-foot radius makes the positioning rule portable. It is also one of the few places where the rulebook accepts a tolerance rather than a number, because groundskeeping is not surveying and a two-foot band is what an honest measurement can actually deliver.

The practical consequence is that an infielder's deepest legal starting point is a shade under 95 feet from the middle of the rubber, and the arc runs the same distance behind second base as it does behind the shortstop hole. Infielders play right on that edge constantly. They are not shifted; they are deep, which is a different thing and remains completely legal.

The four-man outfield died as a side effect

Seven fielders are not the pitcher or the catcher. The rule demands at least four of them in front of the dirt at two separate moments. Four from seven leaves three, and three is the largest number of players who can legally be in the outfield when the ball is released.

Nobody had to write a sentence banning the four-man outfield. It fell out of the arithmetic.

How the rule divides the seven fielders who are not the battery
57%43%
  • Minimum required in front of the dirt boundary4
  • Maximum available for the outfield3

A structural consequence of Rule 5.02(c), not a strategic choice. Four is a floor rather than a target, so a club may bring a fifth in and play three outfielders anyway.

Show the numbers
How the rule divides the seven fielders who are not the battery
ItemValue
Minimum required in front of the dirt boundary4
Maximum available for the outfield3

The reverse move survives untouched. A minimum of four is not a maximum of four, so a defence may bring an outfielder onto the dirt and play five infielders whenever it likes. The two-on-each-side requirement attaches to the four who satisfy the minimum, which means the fifth man is unconstrained and can stand wherever the situation demands: on the grass in front of second base with the winning run on third, on either side of the diamond, anywhere at all. The five-man infield is the single most conspicuous legal alignment left in the game, and it exists because the rule was drafted as a floor.

Outfield positioning is not touched by the restriction at any point. A defence facing an extreme pull hitter can still slide all three outfielders towards one line until the field is effectively vacant in the opposite gap. This is worth sitting with, because it is the clearest evidence that the rule was never about stopping teams from anticipating where the ball would go. It was about one particular region of the field, patrolled by one particular set of fielders, where the anticipation had become too effective to leave alone.

What a violation actually costs

The penalty was rewritten for the 2025 season, and the new version is more interesting than the old one because it distinguishes between two situations that the original treated identically.

Situation one: the illegally positioned infielder is the first player to touch the ball after the pitch. In other words, the shift worked, and the ball the defence had no right to be standing in front of has been fielded by the man who had no right to be standing there. The batter is entitled to first base without liability to be put out, provided he goes and touches it, and every runner advances one base without liability to be put out.

Situation two: the defence was illegally positioned but the violation had nothing to do with the outcome. The pitch is called a ball and the ball is dead.

Both situations carry the same escape hatch. If the batter reaches first base on a hit, an error, a walk, a hit batter or otherwise, and all other runners advance at least one base, the play simply proceeds without reference to the violation. The offence cannot be worse off for having been cheated, and the defence is not punished twice for an infraction that cost nobody anything.

Then there is the manager's option, added in 2025 as a Penalty Comment. If any other play follows a violation, a sacrifice fly or a sacrifice bunt for example, the manager of the offence may tell the plate umpire that he declines the penalty and takes the play instead. The election has to be made immediately at the end of the play. This is why an umpire who spots a violation with a play in progress lets the play finish rather than killing it: the offence might prefer what actually happened, and it has the right to keep it.

What happens after an illegal alignment is spotted
  1. The umpire judges the alignment illegalThe play is allowed to continue. Stopping it would destroy the option the offence is entitled to exercise at the end of it.
  2. Did the illegally positioned infielder touch the ball first?If yes, the batter is awarded first base without liability to be put out and every runner advances one base. If no, the pitch is a called ball and the ball is dead.
  3. Did the batter reach first and all runners advance anyway?If so, the violation is ignored entirely and the play stands as it happened. The offence has already got at least what the penalty would have given it.
  4. Was there another play, such as a sacrifice?The offensive manager may decline the penalty and accept the play. He must say so immediately, at the end of the play, not after a conversation in the dugout.
  5. The official scorer writes it upAn error is charged to the fielder who committed the violation, and the batter's award is not recorded as a time at bat.

The routing set out in the Penalty and Penalty Comment to Rule 5.02(c), as amended for 2025. The umpire does not kill the play, because the offence may want to keep it.

A shift violation is not an at-bat, and that is deliberate

The 2025 amendments did not stop at the penalty. Rule 9.00, the scoring code, was rewritten in several places so that a batter awarded first base for a shift violation is handled exactly like a batter awarded first base for interference or obstruction. Four consequences follow, and none of them is obvious.

The fielder who committed the violation is charged with an error. Not the club, not the manager who drew up the alignment, the individual player standing in the wrong place. When an umpire awards a base because of interference, obstruction or a shift restriction violation, the scorer charges the responsible fielder with one error no matter how many bases result.

The award does not count as a time at bat, and it does not count as a plate appearance for on-base percentage. The scoring rule instructs the scorer to ignore instances of a batter being awarded first base on interference, obstruction or a shift restriction violation when computing on-base percentage, which means the batter's rate statistics are untouched in both directions.

It does not break a hitting streak. A consecutive hitting streak survives a plate appearance that ends in a walk, a hit batsman, defensive interference, obstruction or a shift restriction violation, and the same applies to a consecutive-game streak where every plate appearance in the game ends that way. A sacrifice fly still ends the streak. A shift violation does not.

And it appears in the formula for proving a box score, alongside walks, hit batters and sacrifices, which is the least glamorous and most telling detail of all. The scoring code had to be amended in that many places because a new way of reaching first base without a hit, a walk or an error had been created, and every accounting identity in the book that adds up plate appearances needed to know about it. The people who write these rules think in double entry, and reading a scorecard properly is a habit worth having in any sport for exactly this reason: the notation records what the rulebook thinks happened, not what you thought you saw.

Why the shift got banned when it did

The shift is not new. What was new was that it worked at scale.

For most of baseball's history, an extreme alignment was a bespoke insult aimed at one hitter. In July 1946, Cleveland's player-manager Lou Boudreau drew one on a blackboard between games of a doubleheader and put nearly his whole infield on the right side against Ted Williams. It became famous under two names, the Boudreau shift and the Williams shift, and it was famous precisely because it was strange. A defence had rearranged itself around a single opponent. Even then it was not the first, and pull hitters had been shifted against in earlier decades, but Boudreau's version is the one everybody remembers because of who it was aimed at.

What did not happen next is the interesting part. For roughly six decades, the shift stayed a novelty. Not because managers lacked the imagination, but because they lacked the evidence. To shift correctly you have to know, with confidence, where a specific hitter puts a specific class of batted ball against a specific kind of pitch. Nobody knew that. What existed instead was a scout's memory and a coach's instinct, and neither is precise enough to justify vacating a third of the field.

Then batted-ball tracking arrived and the calculation changed. Every ball hit in a professional game began generating a location, a speed and an angle, and the tendencies that had previously been impressions became distributions. A club could see that a given hitter put a specific proportion of his ground balls into a defined band of the infield, and it could work out that stationing a fielder inside that band converted more of them into outs than leaving the traditional alignment intact. The work behind that decision is the same work described in how clubs actually use their data, applied to a defensive problem rather than a batting one.

Once one club proved it, everybody copied it, because there is no proprietary advantage in a positioning decision that the opposing dugout can see with its own eyes. The shift went from a curiosity aimed at a handful of extreme pull hitters to a default alignment used against a large share of left-handed batters, and it did so within a few seasons.

And at that point it became a rules problem rather than a tactical one, for a reason that has nothing to do with fairness.

The real argument for the ban, and the real argument against it

The case the league made was about the shape of the game rather than the justice of it. A shift does not make outs from nothing. It converts a specific kind of batted ball, the hard ground ball and the low line drive pulled into the infield, into an out that used to be a hit. That is the single most common well-struck ball in baseball. Suppressing it does not reduce home runs, walks or strikeouts, all of which are unaffected by where the fielders stand, so the shift pushed the game further towards outcomes that involve no fielding at all. Fewer balls in play, fewer runners on first, fewer plays where anybody has to run anywhere.

There was a second argument, less often stated plainly. The shift did not tax all hitters equally. It fell hardest on left-handed batters with pronounced pull tendencies and ordinary speed, which is a real and identifiable group of players, and it left everybody else broadly alone. A rule change that raises the value of an entire skill, infield range, and lowers the value of a scouting insight is a redistribution, and redistribution is what rules committees are for.

The argument against is stronger than its supporters admit, and it goes like this. The shift was not a trick. It was the correct answer to a question the hitter was asking. A batter who pulls almost every ground ball is giving the defence information, freely, on every swing, and the defence is under no obligation to ignore it. The traditional alignment is not a law of nature; it is a convention that emerged when nobody could measure anything. Banning the response while permitting the behaviour that provoked it is a decision to protect a hitting approach from its own consequences.

There is also an adaptation argument. Hitters could have solved the shift themselves. The area vacated by an extreme alignment was enormous, undefended and reachable with a bunt or an opposite-field ground ball by anybody who practised it. Very few did, because the modern incentive structure rewards driving the ball hard in the air rather than placing it softly on the ground, and a hitter who retools his swing to beat a shift makes himself worse at the thing that actually pays. That is a coherent explanation for why hitters did not adapt. It is not obviously a reason to change the rule.

Both arguments are about the same underlying question, which is whether a sport should let optimisation run to its conclusion or intervene when the conclusion is unpleasant to watch. Baseball has answered that question in the same direction repeatedly, and the shift restriction arrived in 2023 alongside two other interventions with identical logic: the pitch timer, which stopped pitchers taking as long as they wanted, and larger bases, which shortened the distance a runner has to cover. All three were approved by the Joint Competition Committee for the same season. All three were about the shape of the product rather than the fairness of the contest.

What defences do instead

Clubs did not stop positioning. They started positioning inside a narrower band, and the interesting work moved to the edges of what the rule still permits.

The four legal infielders can stand anywhere on the dirt that satisfies the two-and-two requirement, which is a lot of ground. A second baseman can play at the very back of the arc, almost 95 feet from the rubber and a long way towards the first base line. A shortstop can play deep in the hole or almost on the second base bag. Neither is shifted in the banned sense, and the difference between the two extremes is worth several feet of coverage against a pull hitter.

Outfielders remain unrestricted, so the outfield now absorbs positioning decisions the infield used to make. A defence that would once have moved a fourth man into shallow right field can instead move its right fielder to the line, its centre fielder into right-centre and its left fielder into left-centre, leaving the left field line effectively unguarded against a hitter who has never used it. That is an aggressive alignment, it is entirely legal, and it costs nothing under the rule.

The five-man infield is used in the situations it was always used in, with the difference that it is now the only remaining way to add a body to the dirt. And the whole apparatus of pre-pitch preparation, the positioning cards, the pitcher-specific alignments, the adjustments by count, continues exactly as before. Defensive positioning is not less analytical than it was in 2022. It is analytical within a smaller box.

The one thing clubs genuinely lost is the ability to change alignment mid-inning without a substitution. That is a scheduling problem more than a tactical one, and it has quietly raised the value of infielders who can be trusted on both sides of second base, because a defence that can leave the same four men out there through a whole inning does not have to spend a substitution to fix a bad guess.

Who the rule actually helped

The honest answer is narrower than the coverage suggested.

A hitter benefits from the restriction in proportion to how much of his batted-ball output was being intercepted by an illegally positioned fielder. That means left-handed pull hitters who hit the ball hard on the ground, and it means almost nobody else. A right-handed hitter was rarely shifted against as severely, because the throw from short right field to first base is short and the throw from deep behind third base is long, and the geometry of the diamond has always made extreme right-handed shifts more expensive. A hitter whose contact is mostly in the air was never much affected, since launch angle and exit velocity determine whether a ball lands beyond any infielder regardless of where he stands. A hitter with real speed was already beating some shifts on foot.

The restriction also helps a class of player who never bats: the rangy infielder. If a defence may not solve a coverage gap by moving a man into it, the gap has to be closed by somebody covering more ground, and the ability to cover ground becomes worth paying for again. That is a straightforward reversal of a decade in which positioning had been substituting for athleticism, and it is the change most likely to show up in how clubs value defenders rather than in any batting statistic.

What the rule cannot do is manufacture hits from balls that were never going to be hits. A soft ground ball to a fielder standing in a conventional position is an out under any rule. The restriction only returns the batted balls that were, by the old alignment, being caught in a place where nobody would traditionally have been standing. That is a real quantity and it is not an enormous one, which is why arguments about the rule's effect on scoring should be read carefully. Anyone quoting a precise figure for how much offence it restored is quoting an estimate that depends heavily on assumptions about what hitters and pitchers changed at the same time, and the interaction between contact quality and batted-ball outcomes is difficult enough to untangle in a stable rule environment, let alone one where three rules changed in the same spring.

Enforcement, and the thing that has not been automated yet

The shift restriction is called by human umpires looking at the field before the pitch. There is no buzzer. A crew that misses a violation has simply missed it, and a defence that ends up momentarily out of position because a fielder was slow returning from a mound visit is relying on the same human judgement that would forgive any other trivial breach.

This sits oddly beside the direction of travel everywhere else in the game. Ball and strike calls now have a technological backstop through the automated ball-strike challenge system, which lets a batter, pitcher or catcher ask a computer to re-adjudicate a pitch. Positioning has no equivalent. The tracking systems installed in every park know exactly where every fielder is standing at every moment, which means the data required to enforce the rule automatically already exists and is already being recorded. What does not exist is a mechanism for acting on it in real time, and there is no obvious appetite for one, because a positioning violation is not a judgement call in the way a borderline pitch is. It is a matter of fact that both dugouts can see, and self-policing does most of the work.

The likelier future is quieter: violations are rare because clubs do not attempt them, and the rule functions the way a speed limit functions on a road with a visible camera. Its effect is measured in the alignments nobody tries rather than the penalties anybody pays.

What to watch for, and where the rule stops

Four things are worth checking the next time a defence sets up, and all of them are visible from a television camera behind the plate.

Look at where the second baseman's feet are when the pitcher's leg comes up, not when the ball arrives. The depth test happens at the start of the motion, and a fielder drifting back onto the dirt as the pitch is delivered has already broken the first clause of the rule.

Look at whether the same four infielders are in the same relative positions as they were for the first batter of the inning. If a manager wanted a different arrangement for this hitter and did not get it, that is the inning lock doing its work, and it is the part of the rule that shapes team-building rather than merely alignment.

Look at the outfield, which is where the positioning conversation has moved. A three-man outfield stacked into one half of the field is a shift in every sense except the legal one.

And look at the dirt arc itself. It is 95 feet from the middle of the rubber, it is the same in every park to within two feet, and it is the only line on a baseball field that exists purely because of a rule about where people may stand. It was standardised so that a positioning rule could be written at all, which is a small piece of administrative housekeeping that turned out to be the load-bearing element of the whole thing.

More explanations of how the rules actually work, in this sport and others, are collected in the baseball archive.

Common questions

Are defensive shifts illegal in MLB?

Not in general. What is illegal is a specific alignment: having more than two infielders on one side of second base, or having fewer than four fielders with both feet in front of the outer boundary of the infield dirt, at the moment the pitch is released. Everything else about defensive positioning, including where the outfielders stand and how deep or shallow the infielders play, is still entirely up to the club.

What is the penalty for a shift violation in baseball?

It depends on what happens next. If the illegally positioned infielder is the first player to touch the ball after the pitch, the batter is awarded first base and every runner advances a base. In any other case the pitch is simply called a ball and the ball is dead. Either way, if the batter reaches first and all runners advance at least one base on the play anyway, the violation is ignored.

Can a team put five infielders on the dirt?

Yes. The rule sets a minimum of four fielders in front of the dirt boundary, not a maximum, so a club may bring an outfielder in and play five. The two-on-each-side requirement applies to the four who satisfy the minimum, which leaves the fifth free to stand wherever the coaching staff wants him.

Can infielders switch sides of second base during an inning?

No. From the release of the first pitch of a half inning, the two infielders on each side are locked to that side for the rest of the half inning. The only way out is a substitution, and the replacement is then locked in on the same terms from the first pitch he is on the field for.

Why did MLB ban the shift?

Because the alignment was converting a growing share of hard-hit ground balls and low line drives into outs, and doing it to a particular kind of hitter rather than to everybody equally. The league's stated aim was to put more batted balls in play as hits and to restore the value of infield range. It was approved by the Joint Competition Committee for the 2023 season alongside the pitch timer and larger bases.

Filed under Baseball·baseball · mlb · mlb rules · umpiring · defence · rule changes