Late-inning intentional walks shouldn’t look like surrender, but Candlestick Park witnessed one that defied almost every baseball instinct on May 28, 1998. The Arizona Diamondbacks led the San Francisco Giants 8-6 with two outs in the bottom of the ninth. The bases were loaded, Barry Bonds was approaching the plate, and Arizona closer Gregg Olson couldn’t find the strike zone.
Then manager Buck Showalter made his move.
He ordered Olson to walk Bonds intentionally, automatically forcing in a run and cutting Arizona’s lead to 8-7. According to the San Francisco Examiner’s account the following day, the crowd of 12,066 rose and loudly jeered Olson as Brent Mayne stepped in.
Olson had already thrown 40 pitches when the bases became loaded. The mound was wet, his command had deserted him, and Showalter faced ugly alternatives in the bullpen. Mayne worked a full count, fouling off pitch after pitch before drilling a liner toward right field. Brent Brede momentarily struggled with the lights, secured the ball, and dropped to his knees.
Arizona survived, 8-7.
But the real question wasn’t whether Showalter got away with it. Had he actually made the right decision?
The Night Buck Showalter Bet Against Barry Bonds
That ninth inning offered a lesson that would outlive the box score. Olson had already walked Bonds earlier in the game, and Arizona’s bullpen looked unappealing. According to Ed Price’s account published by the Society for American Baseball Research, Russ Springer was battling the flu while Efrain Valdez had pitched the previous two nights.
Showalter knew his closer was running out of gas. He also understood that allowing Bonds to swing might produce the tying hit, or worse. So he spent one run to avoid the hitter he feared most, leaving Olson to attack Mayne with the game still on the line.
The decision didn’t exactly calm Arizona’s dugout. Catcher Kelly Stinnett later told Price that he needed three looks to understand Showalter’s signal. Even the people closest to the action struggled to process what they were seeing.
There was another detail buried in the madness: Arizona could afford to surrender one run without losing its lead. With two outs, Showalter needed Olson to retire only one more batter. He wasn’t trying to prevent every run. He was trying to prevent the runs that would cost Arizona the game.
That distinction drives the mathematics behind late-inning intentional walks. In ordinary innings, giving away a baserunner usually hurts. But with the game hanging on one final out, managers sometimes value escaping a particular matchup more than protecting every base.
Showalter made that calculation by studying the men in front of him. Nearly three decades later, analysts could test the same instinct with detailed probabilities.
In May 2025, FanGraphs senior writer Ben Clemens examined whether opponents should intentionally walk Aaron Judge, a modern slugger capable of inspiring the same fear Bonds once commanded. Rather than relying on reputation, Clemens measured what happens when a manager replaces one dangerous plate appearance with another baserunner.
The connection to Candlestick was unmistakable. Showalter had accepted a guaranteed run because he believed Mayne offered a better chance of recording the final out. Clemens’ analysis asked when that trade actually improves a team’s chance of winning.
The answer depends heavily on one detail Showalter already understood: How many outs remain?
Why Giving Away First Base Can Ruin Everything
Picture a different ninth inning. The home team trails by one, a runner stands on second, and the visiting manager watches an elite hitter approach with one out. First base sits empty, inviting an intentional walk. The pitcher looks toward the dugout, hoping for permission to avoid trouble.
But that empty base isn’t free. Walking the batter puts the potential winning run aboard, and suddenly a gapper can end the game rather than merely tie it. It’s a brutal tradeoff: sidestep the superstar, but gift the offense another runner.
That explains why managers can’t judge late-inning intentional walks using batting averages alone. They have to consider the outs, the score, the runners, and who waits on deck. A baserunner in the third inning is a minor headache; in the ninth, he’s a siren going off.
One Free Base Sends the Risk Climbing
Clemens’ 2025 FanGraphs work also included updated run expectancy figures showing how much danger a manager creates by handing a hitter first base.
Using MLB’s 2024 results, an offense with a runner on second and one out could expect to score approximately 0.69 runs before the inning ended. Put another runner on first, and that expectation climbed to 0.93 runs.
That’s an extra 0.24 expected runs created without the batter swinging.
For a manager nursing a one-run lead, that’s uncomfortable territory. He has made the next hitter’s job easier before throwing a single pitch. A routine single might score the tying run anyway, but now an extra-base hit threatens to bring home the winner.
With two outs, though, the arithmetic becomes friendlier to the defense. The 2024 figures put run expectancy at 0.35 with a runner on second. Adding a runner at first raised it to 0.45, a smaller increase of 0.10 expected runs.
Those decimals might seem insignificant from the grandstand. Inside the dugout, they help explain why a manager might wave a dangerous hitter toward first.
One out changes the equation. A double play becomes possible in some situations, a strikeout can finish the inning, and the pitcher needs fewer successful confrontations to escape.
Still, run expectancy only tells part of the story. A manager doesn’t receive a trophy for allowing fewer expected runs. He gets judged on whether his team wins.
Aaron Judge Makes Every Percentage Point Hurt
Imagine Aaron Judge stepping into the batter’s box at Yankee Stadium. The ninth inning has arrived, New York trails by one, and a runner dances off second base with one out. Judge settles into his stance while the opposing manager studies the lineup card.
Walking him feels tempting. The Yankees slugger can turn an ordinary fastball into a baseball disappearing beyond the outfield wall. But putting him aboard also gives New York another potential game-winning baserunner.
In his May 13, 2025, FanGraphs study, Clemens tested that exact situation. He built Judge’s offensive profile from his performance dating back to 2022, using data available through May 11, 2025.
The results exposed just how narrow the decision can become.
With Judge batting, Clemens projected a 32.7% chance of a Yankees victory. Intentionally walking him would leave runners on first and second with one out. Under the model’s league-average baseline for subsequent outcomes, New York’s winning probability rose to 33.5%.
That’s a difference of just 0.8 percentage points.
The model didn’t identify one actual pitcher or lock in a particular on-deck hitter. Clemens discussed possible left-handed batters such as Ben Rice, Jazz Chisholm Jr., and Cody Bellinger, while keeping the scenario general. His calculation used Judge’s offensive production against league-average expectations rather than a specific pitcher-versus-batter matchup.
That distinction matters. A manager facing Judge with a dominant left-handed reliever might value the next matchup differently from one relying on a fading right-hander.
Even with the model favoring pitching to Judge, the gap was small enough to leave room for a baseball decision rather than a mathematical command.
Give the Pitcher Two Outs and Everything Changes
Move the same runner to second with two outs, and the balance changes.
Clemens’ model estimated that walking Judge in that situation could lower the Yankees’ winning probability by approximately two percentage points. The intentional walk no longer helps the offense in the model. It gives the defense a slight edge.
The explanation belongs to the final out.
With one out, Judge can still make an ordinary out without ending the inning. With two, a successful pitch or routine defensive play sends everyone home.
That makes late-inning intentional walks more attractive when the opposing lineup falls sharply from one hitter to the next. The manager doesn’t need to survive another three batters. He needs one matchup he believes his pitcher can win.
But managers shouldn’t fall in love with batter-pitcher matchups just yet. A star hitter’s replacement might have only six or seven career plate appearances against the reliever. Such tiny samples offer little assurance about what happens next.
Weighted on-base average (wOBA), longer-term projections, and Statcast contact-quality data provide better starting points. The challenge comes from combining those measurements without pretending they guarantee an outcome.
Load the Bases and Dare the Next Hitter
There are moments when walking a batter changes the entire shape of the defense.
Take a tie game in the bottom of the ninth. Runners occupy second and third, with one out and a dangerous hitter approaching. The manager could pitch to him and hope the infield prevents a ground ball from scoring the winning run.
Or he could put the hitter on first.
Walking the batter sets up a force play at every base, including home plate. The catcher no longer needs to tag the runner coming from third. He only needs the ball and a foot on the plate before the runner arrives.
For a ground-ball pitcher, that looks appealing. A sharp grounder to third could produce a force at home, and the catcher might even complete a double play at first.
But the manager has also loaded the bases. Another walk, a hit-by-pitch, or a wild pitch can now end the game.
The Difference Between Losing a Run and Losing the Game
The calculation looks different with a two-run lead.
That’s what made Showalter’s decision against Bonds possible. Arizona could surrender one run and remain ahead. In a tie game, the same bases-loaded intentional walk would have handed San Francisco an immediate victory.
The scoreboard dictates how aggressively managers can trade runs for outs.
When the tying run occupies second, walking a batter usually leaves that runner where he stands. It also puts another potential run aboard. If the defense leads by one, that newly added runner represents the possible winning run.
With the tying run on third, the threat becomes more immediate. A sacrifice fly, passed ball, or suitable groundout can erase the lead. Intentionally walking the batter may improve the force-play possibilities, but it introduces another runner capable of changing the game’s outcome.
That’s why late-inning intentional walks demand more than the familiar question of whether first base is open. The manager must decide what each runner can do, how many outs remain, and whether the next hitter offers a genuine advantage.
The Pitcher Can Make the Numbers Lie
A manager can have the perfect matchup on paper and still watch it collapse because his pitcher cannot throw a strike.
Picture the bullpen. A hard-throwing closer stands on the mound, but his fastball has lost some life. His slider keeps missing outside, and the catcher has already made several awkward reaches.
A tired setup reliever warms beyond the outfield fence.
Now Bryce Harper approaches with the tying run on second. Imagine a rookie waiting on deck who struggles against sliders. Walking Harper might seem like an easy decision.
But what happens if the closer cannot land that slider for a strike?
The next hitter can simply refuse to chase. Suddenly, a supposedly favorable matchup turns into another walk, and the manager has manufactured a bigger inning for the opposition.
One Hanging Slider Can Wreck the Entire Plan
A pitcher who throws strikes consistently can attack a weaker hitter without living in fear of another free pass. A reliever with shaky command cannot offer the same comfort.
Modern MLB bullpen management increasingly relies on pitch-level information. Baseball Savant’s Statcast tools track velocity, movement, spin, contact quality, and how batters perform against particular pitch types.
Those measurements help clubs identify why one hitter might struggle against a specific reliever.
Still, a hitter’s weaknesses don’t automatically make him an easy out. A pitcher must actually possess the weapons to exploit them.
A rookie might chase sliders beneath the strike zone, but that scouting report means little if the pitcher keeps hanging breaking balls over the plate.
Managers can’t just run from a star. They have to be sure the guy on deck offers a better chance of escaping.
That judgment becomes even more difficult in October. A manager may have an elite reliever available, but using him aggressively can affect the following game. The choice involves the immediate matchup and the cost of exhausting another arm.
Strikeouts Offer Escape but Home Runs Bring Ruin
Baseball’s modern offensive environment has made these decisions stranger.
Home runs, walks, and strikeouts dominate much of today’s strategic conversation. Analysts call them the Three True Outcomes, because the results don’t rely on ordinary defensive fielding.
For a manager considering an intentional walk, the rise in strikeouts offers one obvious benefit. A power pitcher can potentially escape a two-out jam without allowing another ball into play.
FanGraphs’ 2025 examination of run expectancy showed how much baseball has changed. From 2000 through 2004, MLB hitters struck out in 16.8% of plate appearances. Between 2021 and 2024, that rate reached 22.7%.
More strikeouts mean fewer opportunities for runners to advance on balls in play. They also make an inning-ending punchout a more realistic escape route for pitchers with strong swing-and-miss stuff.
Yet modern hitters punish mistakes with enormous force.
A hanging slider can travel 430 feet before an outfielder takes three steps. A poorly located fastball can turn a one-run lead into a walk-off celebration.
The increased value of strikeouts must therefore compete with the damage that comes from one mistake.
One Swing Can Make the Free Pass Costly
Suppose a hitter represents a serious home-run threat with the tying run on second. The manager might walk him to avoid a two-run shot. With two outs and a weaker batter next, that choice can protect the lead.
But the walk also changes what happens if the next hitter connects.
Instead of facing one runner in scoring position, the pitcher must deal with two potential runs. A double into the corner could bring both home, depending on their speed and the defensive play.
This is where the apparent simplicity of late-inning intentional walks breaks down. Managers don’t merely choose between two hitters. They change the consequences of every possible outcome that follows.
A home run, strikeout, walk, and single all carry different costs once the base situation changes.
And those costs grow sharper as the final out approaches.
A Spreadsheet Cannot Feel a Closer Losing His Grip
Front offices have spent years refining win probability models to help managers make better decisions. These systems consider the inning, score, base-out state, and expected offensive performance.
The most sophisticated approaches can also account for specific hitters, pitchers, baserunners, and defensive alignments.
But baseball still refuses to behave like a controlled experiment.
An elite closer might have a devastating slider, only to discover that his grip feels wrong. A weaker hitter might enter the box after seeing the pitcher’s entire repertoire from the on-deck circle. A fast runner could score from second on a softly hit ball that another player would never attempt to advance on.
Such details matter because the difference between the competing decisions may amount to only one or two percentage points.
That doesn’t make the analytics useless. It makes them a guide to the risk a manager accepts.
To get this right, teams must weigh both a hitter’s raw talent and how his swing matches up against the pitcher on the mound. They must also consider whether the next batter offers enough relief to justify another baserunner.
MLB simplified the mechanics in 2017 by introducing automatic intentional walks. Managers can now signal the home-plate umpire without requiring four deliberate balls.
The change removed the danger of a wild pitch during those intentional deliveries. It didn’t eliminate the strategic risk of putting another man on base.
The manager still has to live with what happens next.
The Next Ninth Inning Will Not Wait for Perfect Math
Baseball’s best late-game decisions often look ridiculous before they work.
A manager can intentionally walk a slugger, watch the next hitter roll into a double play, and receive praise for his courage. Another skipper might make the same mathematically defensible choice and surrender a walk-off double.
By morning, the television panels will have rendered their verdicts. One man will look brilliant, while the other will face questions about why he avoided the best competition.
Neither judgment necessarily explains whether the decision improved his team’s chance of winning.
That distinction separates thoughtful baseball strategy from results-driven second-guessing.
The deeper lesson from late-inning intentional walks is that managers must calculate the cost of the decision before they know the outcome. They cannot rely on whether the next line drive finds a glove or the gap.
For the team on defense, the goal is simple: survive the inning with the lead intact. But reaching that goal can require surrendering something valuable along the way.
Showalter understood that on a cold, uncomfortable night at Candlestick Park. His pitcher was laboring, the opposing superstar was waiting, and the crowd wanted to see Bonds swing.
Instead, the manager chose Brent Mayne.
The ball nearly made him regret it. Brede’s catch ended the game, but Mayne’s sharp line drive showed just how thin the difference was between escape and disaster.
Today, managers have better information than Showalter possessed in 1998. They can study hitter projections, run expectancy, bullpen performance, and the probability of every possible outcome.
Yet the final decision still belongs to someone standing in a dugout while thousands of people demand an answer.
Before long, another closer will find himself staring at a tying runner on second. The opposing team’s most dangerous hitter will step into the box, and the manager will have seconds to choose between attacking and retreating.
Late-inning intentional walks remain baseball’s most uncomfortable calculation: Sometimes the smartest way to protect a lead is to give the other team exactly what it wants.
But when the next hitter drives a ball toward the outfield wall, will anyone remember what the percentages said?
READ MORE: Intentional walks backfire when postseason managers chase the safer matchup
FAQs
Q1. Why do MLB managers intentionally walk hitters late in games?
Managers walk dangerous hitters to face someone they believe offers a better chance of recording an out. But the extra baserunner can make things worse.
Q2. Why did Buck Showalter intentionally walk Barry Bonds in 1998?
Showalter wanted to avoid Bonds with the bases loaded and two outs. Arizona sacrificed one run, retired Brent Mayne, and escaped with an 8-7 win.
Q3. Does an intentional walk make more sense with two outs?
Often, yes. With two outs, the pitcher needs just one more out to escape. That can make avoiding a superstar worth the extra runner.
Q4. Can a team intentionally walk a batter with the bases loaded?
Yes. The walk forces in a run, but a manager might accept that cost to face a weaker hitter. Showalter famously did it against Barry Bonds.
Q5. When did MLB stop requiring four pitches for an intentional walk?
MLB introduced automatic intentional walks in 2017. Managers now signal the umpire, who awards the batter first base without any pitches.
