Starting pitchers rarely face the opposing lineup a third time in the postseason, and Blake Snell was about to discover how unforgiving that philosophy could be. The Tampa Bay left-hander was dealing. His fastball jumped through the strike zone, his breaking pitches disappeared beneath swinging bats, and the Los Angeles Dodgers had spent five innings searching for answers. Nine strikeouts. Two hits allowed. Just 73 pitches.
Then Austin Barnes singled with one out in the sixth inning of Game 6 of the 2020 World Series. Rays manager Kevin Cash emerged from the dugout at Globe Life Field. Snell watched him approach, knowing his night was finished despite protecting a 1-0 lead. Tampa Bay wanted a fresh arm against the top of the Dodgers’ order.
Nick Anderson replaced Snell. Mookie Betts doubled, a wild pitch brought home the tying run, and Corey Seager’s fielder’s choice gave Los Angeles the lead. Before long, the Dodgers were celebrating a 3-1 victory and their first World Series championship since 1988.
Cash had followed a philosophy supported by years of baseball research. Yet the decision left one uncomfortable question hanging over the sport. When a starting pitcher is dominating, does the third-time-through-the-order penalty justify taking the ball away?
The numbers that changed October baseball
The sixth inning once belonged to the starter. A pitcher carrying a shutout into the middle of a playoff game generally expected to keep working until something went wrong. Modern baseball has reversed that thinking.
Managers now study how hitters perform against the same pitcher during successive trips through the batting order. The results help explain why starting pitchers rarely face the opposing lineup a third time in the postseason. The concept has a name: the third time through the order penalty.
According to MLB.com’s official statistical glossary, starting pitchers allowed a .243 batting average and .713 OPS during their first trip through opposing lineups from 2016 through 2025. Those numbers climbed to .255 and .747 the second time through. By the third meeting, hitters produced a .265 batting average and .780 OPS. That’s a 67-point increase in OPS between the first and third encounters.
The numbers don’t mean a starter suddenly turns into a pumpkin when the 19th hitter approaches. Still, the pattern gives managers a compelling reason to consider the bullpen before trouble starts. There’s another complication. A third trip often begins near the top of the batting order, where teams place their most dangerous hitters. Those matchups can make a starter’s late-inning numbers look worse. However, the broader decline in effectiveness remains real. Managers aren’t just watching pitch counts anymore. They’re trying to predict what happens when a familiar hitter steps into the box again.
Hitters learn something with every swing
The first at-bat against a starting pitcher is often a feeling-out process. A hitter knows the scouting report. He might have watched hours of video, studied the pitcher’s release point, and memorized where his fastball tends to finish. None of that replaces standing in the batter’s box.
A 97-mph fastball looks different when it arrives near the letters. The slider that seemed ordinary on video suddenly disappears beneath the barrel. By the second plate appearance, the hitter has gathered useful information. He knows how quickly the fastball arrives and how long the breaking ball stays in the strike zone.
The third meeting becomes a different battle. Perhaps the pitcher used two fastballs to set up a slider during the first inning. The hitter remembers that sequence and starts looking for the breaking pitch earlier. Now the pitcher must change his approach or risk becoming predictable. Researcher Mitchel Lichtman explored this problem in a November 2013 study for Baseball Prospectus. He examined whether starters struggled primarily because of fatigue or because hitters became more familiar with their pitches. His findings challenged the traditional explanation.
Pitchers entering their third trip through the order with roughly 59 pitches performed almost identically to those entering with approximately 72 pitches, based on weighted on-base average allowed. The research suggested that pitch count alone could not explain the penalty. Other analyses have emphasized fatigue as another contributing factor. Baseball still hasn’t settled the precise balance between exhaustion and familiarity.
Yet the practical problem remains. A pitcher might still throw 97 mph, but a hitter who has already seen that fastball twice may handle it differently. That’s why managers worry about the third time through the lineup even when their starter looks physically fresh.
Blake Snell became the face of baseball’s analytics debate
Snell’s removal in 2020 remains difficult to forget because almost everything about his performance suggested he deserved another inning. The Rays were fighting to extend the World Series. Randy Arozarena’s first-inning homer had given them an early lead, and Snell protected it with one of the sharpest outings of his career.
Betts, Seager, and Justin Turner had combined for six strikeouts in six plate appearances against him. Los Angeles couldn’t solve his fastball. Cash, however, saw the approaching danger. Those hitters were coming around for another look, and Tampa Bay had spent the season trusting its bullpen in precisely those situations.
The manager also believed Anderson offered a strong chance of protecting the lead. That belief wasn’t entirely unreasonable. Anderson had dominated during the regular season and helped establish Tampa Bay’s relief corps as a major strength. But October had exposed a problem.
Before Game 6, Anderson had allowed runs in six consecutive postseason appearances. His recent performance offered reasons for caution, even if his regular-season numbers remained impressive. Cash chose the bullpen anyway.
Snell walked toward the dugout, visibly unhappy with the decision. Anderson entered and immediately surrendered Betts’ double. Suddenly, the Dodgers had runners at second and third with one out. A wild pitch tied the game. Seager’s grounder then brought home the go-ahead run. Snell could only watch.
MLB’s October 28, 2020, reporting documented Cash’s decision and the starter’s frustration. The manager had followed the strategy that helped Tampa Bay reach the World Series, but this time the result was devastating. The obvious temptation was to blame analytics. That explanation, however, oversimplifies the problem.
Leaving Snell on the mound wouldn’t have guaranteed three more outs. Betts might have adjusted to his fastball. Seager could have punished a breaking pitch that caught too much of the plate. Baseball never reveals what would have happened.
The stronger criticism concerns the information Cash had in front of him. Snell still possessed his best stuff, and Anderson had struggled throughout the postseason. The third-time-through-the-order penalty describes a statistical risk. It doesn’t automatically identify the best pitcher for the next three outs. That distinction became even clearer two years later.
Zack Wheeler showed the danger of trusting a fresh arm
Philadelphia needed one victory to keep its championship hopes alive in Game 6 of the 2022 World Series. Zack Wheeler answered with five scoreless innings against the Houston Astros. His previous World Series start had raised concerns about arm fatigue and declining velocity. The Phillies gave him extra rest, hoping their ace would regain his best stuff.
The difference appeared immediately. Wheeler opened the game with a 98.4-mph fastball. His command looked sharp, and Houston struggled to put together meaningful contact. By the fifth inning, Wheeler’s fastball was still sitting comfortably around 97 mph.
Then Kyle Schwarber launched a solo home run in the sixth, giving Philadelphia a 1-0 lead. The Phillies were nine outs away from forcing Game 7. But Houston’s half of the inning brought trouble. Martín Maldonado reached after a hit-by-pitch that Philadelphia believed he had deliberately invited. José Altuve followed with a fielder’s choice before Jeremy Peña singled through the middle. With one out and runners at the corners, Yordan Alvarez approached the plate.
Phillies manager Rob Thomson had a decision to make. Wheeler had thrown only 70 pitches, but Alvarez represented one of baseball’s most dangerous left-handed hitters. Thomson preferred left-handed reliever José Alvarado for the matchup. The manager walked toward the mound. Wheeler’s night was over.
One swing erased the matchup advantage
Alvarado entered with a fastball that could reach triple digits and a devastating breaking-ball combination. Philadelphia had good reasons to trust him. Thomson later explained that the Phillies had planned to use Alvarado against Alvarez during the World Series. With runners at the corners, the manager particularly wanted a strikeout.
A fly ball against Wheeler could have tied the game. Alvarado offered a different route out of trouble. The matchup lasted four pitches. Alvarez turned on a 98.9-mph fastball and launched it 450 feet to straightaway center field. The ball cleared the towering batter’s eye as Minute Maid Park erupted. A 1-0 Philadelphia lead had become a 3-1 deficit. Houston eventually won 4-1 and captured the World Series.
MLB’s November 6, 2022, reporting confirmed that Wheeler’s early exit surprised the pitcher. He had wanted to continue but understood that Thomson controlled the decision. The manager also explained that Wheeler might have stayed if Houston hadn’t put runners at the corners. That detail matters.
Thomson wasn’t simply following a fixed rule about the third time through the lineup. He considered the baserunners, the opposing hitter, and the reliever available. His preferred matchup failed spectacularly. For Philadelphia, the result showed how quickly a carefully calculated pitching change can backfire.
Alvarado threw nearly 99 mph. Alvarez still sent the ball over the center-field wall. The modern bullpen offers different arm angles, sharper breaking pitches, and fresh velocity. But none of those advantages guarantees an out. Sometimes an elite hitter beats a great pitch. And sometimes the ace standing in the dugout might have been the better choice.
Cleveland exposed the hidden cost of shorter starts
Pulling a starting pitcher early can help win one playoff game. Doing it repeatedly can create a problem that grows throughout an entire series. Somebody must record the remaining outs. The 2024 Cleveland Guardians offered a revealing example of what happens when a postseason bullpen carries an enormous workload. Manager Stephen Vogt relied heavily on a relief group featuring Cade Smith, Tim Herrin, Hunter Gaddis, and Emmanuel Clase. Those pitchers had helped Cleveland reach the American League Championship Series. Their ability to handle difficult innings became one of the club’s greatest strengths.
But the workload kept building. Through nine postseason games, Cleveland’s starting pitchers had thrown just 29⅔ innings. The relievers had covered 49⅓ innings, according to MLB’s October 19, 2024, reporting. That imbalance meant the same bullpen arms had to keep rescuing games, often with limited recovery time. Eventually, the strain became visible.
Cade Smith’s disappearing fastball
Smith entered the 2024 postseason as one of Cleveland’s most dependable relievers. His four-seam fastball had been among baseball’s most effective pitches, and his ability to attack hitters made him a valuable weapon in the middle innings. The Guardians leaned on him immediately.
Smith appeared in all five American League Division Series games against Detroit. In the deciding Game 5 on October 12, he worked 1⅔ innings, helping Cleveland advance. Three days later, he delivered another 1⅔ scoreless innings against the Yankees in ALCS Game 2. Then came Game 3 on October 17. Smith retired Juan Soto, Aaron Judge, and Giancarlo Stanton on just 10 pitches. Every offering went for a strike. It looked effortless.
The following night, Vogt returned to Smith against the same dangerous portion of New York’s lineup. This time, nothing came easily. Smith walked Soto and allowed a single to Judge. Two batters later, Stanton launched a three-run homer that pushed the Yankees ahead 6-2. Cleveland eventually lost Game 4 by an 8-6 score.
The radar gun offered another clue. Smith’s four-seam fastball averaged 94.0 mph during that appearance. His combined regular-season and postseason average entering the game had been 96.0 mph. The pitch Stanton drove into the seats registered 94.2 mph.
According to MLB’s October 19 report, Vogt acknowledged signs of fatigue. Smith rejected that explanation, arguing that poor execution had caused his struggles. Both perspectives deserve consideration. A two-mph velocity drop doesn’t prove exhaustion caused a home run. Smith also fell behind hitters and missed locations, while Stanton delivered an excellent swing.
Still, the workload was undeniable. By Game 4, Smith had appeared in eight of Cleveland’s first nine postseason contests. Only one other Guardians pitcher, Herrin, had matched that usage in franchise playoff history. The club had asked its best relievers to work repeatedly because its starters rarely pitched deep into games. Now those relievers were struggling to maintain their usual effectiveness.
The bullpen has a breaking point
Cleveland’s experience revealed something the third-time-through-the-order penalty cannot measure on its own. Every early pitching change has consequences beyond the current inning. Removing a starter after four innings might create a favorable matchup against the opposing team’s best hitter. But the manager then needs five innings from the bullpen.
Repeat that approach across a long series, and the relief corps begins accumulating fatigue. One reliever might lose a little velocity. Another could struggle to command his slider. A third might need a night off precisely when the manager needs him most. Modern contenders stack their bullpens with hard throwers capable of recording four or five critical outs. Yet even the strongest relief staff has limits.
The Guardians didn’t lose the 2024 ALCS solely because of bullpen fatigue. The Yankees’ offense, Cleveland’s starting-pitching limitations, and individual execution all shaped the series. But the warning was clear. A strategy designed to protect against the third time through the lineup can create another problem when relievers shoulder too many innings.
That raises an uncomfortable question. What happens when a manager takes the opposite approach and trusts his starter to finish what he began? For an answer, rewind 21 years to a chilly October night at Yankee Stadium. There, a 23-year-old pitcher faced a championship-defining test, and his manager had little interest in making an early trip to the mound.
Josh Beckett proved an ace could still finish the job
On October 25, 2003, Josh Beckett took the mound for the Florida Marlins with a chance to end the World Series. Florida held a 3-2 series lead over the New York Yankees. Manager Jack McKeon needed one more victory to deliver the franchise’s second championship. He handed the baseball to his young ace on just three days’ rest.
Beckett didn’t disappoint. Inside Yankee Stadium, he attacked a lineup featuring Derek Jeter, Bernie Williams, and Jorge Posada. His fastball carried authority, while his breaking pitches kept hitters uncomfortable. By the sixth inning, the Yankees had already seen his best stuff twice. They knew what was coming, but that knowledge hadn’t helped them score. McKeon stayed in the dugout. The Marlins held a 2-0 lead, and Beckett kept throwing strikes.
He eventually completed nine shutout innings on 107 pitches, allowing five hits, walking two, and striking out nine. The final out supplied an unforgettable image. Posada tapped a soft grounder near the first-base line. Beckett fielded it himself, applied the tag, and raised both arms as his teammates rushed toward him. Florida had won the World Series.
According to MLB’s historical account of the game, Beckett became the first pitcher since Jack Morris in 1991 to record a shutout in a World Series-clinching game. He also earned World Series MVP honors. No parade of specialists. No sixth-inning hook with the opposing lineup approaching again. Just a dominant starting pitcher finishing the biggest game of his career.
Why Beckett’s performance still matters
The obvious difference between 2003 and modern baseball involves how teams manage risk. McKeon allowed Beckett to keep pitching because he trusted what his ace was delivering. The young right-hander maintained his command, limited damage, and gave New York few opportunities to build a rally.
That doesn’t mean the statistical penalty disappeared. The Yankees saw Beckett’s pitches multiple times, and every additional appearance offered another chance to adjust. But the Marlins starter kept winning those confrontations.
His performance demonstrated that the third time through the lineup doesn’t automatically make a pitcher ineffective. Some starters possess the command, pitch variety, and physical durability to survive those dangerous matchups. Beckett combined those qualities with exceptional execution. Of course, asking every starter to repeat his achievement would be foolish. His outing remains memorable partly because so few pitchers can dominate for nine innings under championship pressure. Still, it challenges the notion that an early pitching change always represents the safer decision. Sometimes the best bullpen move is leaving the bullpen alone.
The next evolution of October pitching decisions
The next chapter of postseason pitching strategy won’t necessarily involve removing starters even earlier. Baseball already understands the third-time-through-the-order penalty. The harder challenge involves recognizing when that league-wide tendency applies to the pitcher standing on the mound.
FanGraphs writer Jay Jaffe documented the shift in an October 22, 2025, analysis of postseason workloads. Through that year’s League Championship Series, starters averaged 4.35 innings per playoff start, compared with 5.19 during the regular season. The previous postseason showed an even wider gap, with starters averaging 4.25 innings against a regular-season mark of 5.22.
Those numbers help explain why starting pitchers rarely face the opposing lineup a third time in the postseason. Teams increasingly prefer fresh relievers to familiar starters, particularly when the top of the batting order approaches. But the decision still demands judgment.
Modern pitch-tracking systems measure velocity, movement, spin, and release characteristics. Coaches can tell whether a starter’s fastball has lost its usual life or whether his breaking pitches have stopped finishing sharply. A manager watching his ace pump 97-mph fastballs with late movement has evidence that the pitcher remains physically effective.
That evidence, however, doesn’t reveal everything. A hitter might have recognized the starter’s pitch sequences. Another could have timed his delivery after two earlier plate appearances. Meanwhile, the bullpen carries its own uncertainty. A reliever who dominated the previous night might struggle with command. Another could feel sharp despite throwing on consecutive days. Managers must balance all those possibilities while considering how many games remain in the series.
Sometimes removing a starter protects a narrow lead. Other nights, the same decision exposes an overworked bullpen or wastes an outstanding performance. Snell, Wheeler, Cleveland’s relievers, and Beckett illustrate different sides of that calculation. None supplies a permanent answer.
The numbers will continue shaping October baseball because they reveal genuine risks. Ignoring the third-time penalty would mean ignoring valuable information about how hitters and pitchers perform. Yet the best managers must understand when a particular game demands something different.
Before long, another postseason ace will reach the sixth inning with a low pitch count and a one-run lead. The opposing leadoff hitter will step into the box for his third appearance. Somewhere behind the outfield wall, a reliever will begin warming up.
The starter will glance toward the dugout. His manager will study the matchup, consider the bullpen, and weigh the consequences of one mistake. The decision might decide a championship. Baseball’s third-time penalty will still favor caution, but October has never guaranteed that the cautious choice will work.
When your ace has spent five innings making the opposing lineup look helpless, is the bigger gamble leaving him on the mound or taking the baseball away?
READ MORE: What early-inning swing rates tell us about October plate discipline
FAQs
Why do starting pitchers rarely face a lineup three times in the playoffs?
Hitters often perform better after seeing the same pitcher repeatedly. Managers frequently bring in fresh relievers to reduce that risk.
What is the third-time-through-the-order penalty in MLB?
It’s the tendency for hitters to improve against a starting pitcher with each trip through the lineup, particularly during their third meeting.
Why did Kevin Cash remove Blake Snell in the 2020 World Series?
Cash wanted a fresh reliever against the Dodgers’ best hitters. Snell had thrown only 73 pitches, making the decision especially controversial.
Why did the Phillies pull Zack Wheeler in Game 6 of the 2022 World Series?
Rob Thomson preferred left-handed reliever José Alvarado against Yordan Alvarez. The decision backfired when Alvarez hit a three-run homer.
Who threw a complete-game shutout to win the 2003 World Series?
Josh Beckett pitched all nine innings for the Florida Marlins, allowing five hits and striking out nine against the Yankees.
