The third time through brought Mookie Betts back toward the batter’s box and Kevin Cash out of the dugout. Blake Snell saw his manager coming and turned away, swearing. Tampa Bay led the Dodgers 1–0 in Game 6 of the 2020 World Series. Its season depended on protecting that run. Snell had thrown 73 pitches. Nine Dodgers had struck out, and only two had collected hits. Austin Barnes supplied the second, a single with one out in the sixth. Cash took the ball anyway.
Jesse Rogers’s report that night captured the collision between the scoreboard and the manager’s judgment. Betts, Corey Seager and Justin Turner had each struck out twice against Snell. Now Cash wanted another pitcher to face them.
From the mound, the argument looked brutal. A starter had beaten the hitters waiting to see him again, without approaching a conventional pitch limit. The dugout saw another kind of danger: those hitters had already watched his best weapons twice. But removing Snell answered only half the question. Who, exactly, offered Tampa Bay something better?
The bullpen arm carried its own warning signs
Nick Anderson entered with Barnes at first and Betts waiting. A double moved Barnes to third. Anderson then bounced a pitch past Mike Zunino, allowing the tying run to score. Seager followed with a grounder to first, and Betts beat the throw home. Los Angeles had turned a one-run deficit into a one-run lead. The Dodgers eventually won 3–1. That sequence made Cash’s decision unforgettable, but the problems with Anderson had begun before the bullpen gate opened.
In his October 2020 Sports Illustrated account, Tom Verducci reported that Anderson had exceeded one inning only once during the regular season. Cash asked him to do it seven times in nine postseason appearances over 26 days. Before World Series Game 3, the manager had acknowledged the burden on Anderson’s fastball. “Partly that’s on me,” Cash said. The results supplied another warning. Verducci reported that opponents had hit .097 against Anderson’s fastball during the regular season, then .270 entering Game 6. He had also allowed runs in six consecutive playoff outings.
Afterward, Anderson described his fatigue to MLB: “Not a lot of gas. Velo was down.” That history deserved weight alongside Snell’s exposure to the lineup. The third time through might weaken a starter’s case for staying, but it cannot strengthen a tired reliever’s pitches.
Anderson still had reasons to believe in himself. A bad stretch does not erase a season of excellent work. Cash, though, had to choose between the pitchers available that night, with their actual workloads and actual stuff. Pulling Snell for the September version of Anderson would have made a cleaner argument. October had already changed that pitcher.
Why Cash feared another meeting with Betts
The top of the order approaching again gives a manager something the inning number cannot: a warning about who comes next. Betts would not walk into the box carrying only his two strikeouts. He would carry everything he had learned from them. MLB’s third-time-through statistics list starters’ OPS allowed for 2016–2025 at .713, .747 and .780 on their first, second and third lineup circuits. Those numbers explain why the third time through deserves attention. They do not tell a manager which pitcher should throw the next fastball.
A hypothetical starter might reach the sixth inning at 60 pitches after quick outs. Another might arrive at 95 after foul balls, walks and long counts. Treating those outings as identical would throw away information the game has already provided. The same caution applies to the pitcher’s talent. A diminished ace might still offer more than a fully rested middle reliever. Against a particular hitter, a starter’s changeup might remain the strongest available weapon even after two previous encounters.
Cash had a defensible concern about repeated exposure. The harder question concerned how much that risk had grown, and whether Anderson could reduce it.
The drop-off isn’t a cliff
A 2023 study by Ryan Brill, Sameer Deshpande and Abraham Wyner, available through the Wharton School, complicates the familiar explanation. The researchers examined starting-pitcher plate appearances from 2012 through 2019, accounting for hitter and pitcher quality, handedness and home-field advantage. Their analysis found little evidence of a sharp performance break when the batting order started over. Offensive expectations generally rose gradually as the game progressed.
The distinction matters when a manager studies the lineup card. Reaching the third time through need not mark the exact instant when a pitcher loses his edge. Trouble might begin earlier, or a dominant starter might remain the best choice beyond that point. The sensible response is to prepare a replacement and keep watching. Counting completed lineup circuits gives the dugout a reason to look harder. It does not relieve the manager of making the comparison.
What the catcher’s target can reveal
Consider a slider intended for the lower outside corner that hangs over the middle instead. One mistake might produce a harmless foul ball. Another might reach the seats before the bullpen phone stops ringing. Managers should compare current velocity, movement and location with the starter’s early innings and normal range. Pitch count belongs beside those observations. It should not drown them out.
If hitters stop chasing the breaking ball, the catcher may need another way to finish an at-bat. Repeated misses above the target can also force a starter into counts where hitters wait for a fastball. The third time through becomes dangerous when the pitcher loses his out pitch and has to challenge hitters with something worse. Waiting for the scoreboard to confirm the problem can cost the lead.
A pitching change commits the dugout to more than one matchup
The on-deck hitter watches the reliever’s warmup pitches too. A manager might love the matchup at the plate and dislike the one standing a few feet behind it. MLB’s three-batter minimum makes that concern concrete. Relievers generally must face three hitters or finish the half-inning, with exceptions for injury or illness.
The bullpen choice therefore needs to survive more than its first comparison. Can the new pitcher handle the next left-handed hitter? Does he have a second pitch when the fastball misses? Who follows if the inning continues? Postseason bullpen management also reaches into tomorrow’s game, but elimination changes the priorities. Saving an arm offers no benefit if the team never plays again.
The third time through forces the manager to balance those costs before the next hitter digs in. Calling for relief does not remove risk. It changes which pitcher carries it.
Milwaukee gave Snell another chance to finish the argument
Five years later, Snell stood on the other side of the Dodgers’ decision. In the 2025 NLCS opener against the Milwaukee Brewers, he pitched eight scoreless innings, allowing one hit, walking nobody and striking out 10. The Associated Press’s account of the opener reported that Dave Roberts removed him after 103 pitches with Los Angeles leading 2–0. This time, Snell had already worked through the lineup deep into the night.
According to MLB’s game report, Snell’s changeup generated 14 of his 22 swings and misses. He used the pitch 37% of the time, up from 23.6% during the season. That gives the comparison something firmer than a claim that Snell simply wanted it more. Milwaukee kept missing a pitch he continued to use. Roberts let him keep attacking.
The ninth still brought trouble. Roki Sasaki allowed a run, and Blake Treinen eventually struck out Brice Turang with the bases loaded to preserve a 2–1 victory, according to AP’s game account. Snell had delivered eight innings of relief from the bullpen’s problems. Even then, the pitching change nearly surrendered his work.
The next hitter still gets a vote
Snell’s Milwaukee performance does not prove that Cash should have left him in five years earlier. Different nights demand separate judgments. It does show what an automatic hook can cost when a starter still owns a pitch the opposition cannot handle. The third time through should begin a sharper conversation in the dugout. Coaches need to know which pitches still work, which hitters threaten them and which relievers can offer an improvement. Those answers should develop before a single puts the manager on the top step.
A catcher can see a breaking ball stop finishing. The pitching coach can track whether the fastball has lost movement. A reliever can admit that his arm feels heavier than his season numbers suggest. That last conversation matters. Anderson’s workload exposed the danger of examining the starter closely while assuming the replacement remained intact. Every arm in the decision deserves the same scrutiny.
For the pitcher, none of that makes surrendering the baseball painless. Nine strikeouts feel different from a forecast. He has watched hitters miss, heard the ball hit the mitt and earned the right to believe the next pitch will work.
The manager must decide whether that belief still matches the best available odds. Sometimes he should take the ball before the crowd understands why. Other nights, he should leave his best weapon alone.
Then the hitter settles into the box. How much has he learned, and does the pitcher still have an answer?
READ MORE: How cold weather sharpens breaking balls and tests pitchers in October
FAQs
What does third time through the order mean?
It means a starting pitcher faces opposing hitters for the third time in the same game.
Why did Kevin Cash pull Blake Snell in the 2020 World Series?
Cash wanted to avoid letting Mookie Betts and Corey Seager face Snell a third time, despite Snell’s dominant performance.
Why was Nick Anderson a risky replacement?
Anderson had allowed runs in six straight playoff appearances. His heavier postseason workload also raised concerns about fatigue and his fastball.
Should managers always pull a starter after two trips through the order?
No. Managers must weigh the starter’s remaining effectiveness against the available relievers, upcoming hitters and game situation.
How did Blake Snell pitch in the 2025 NLCS opener?
Snell threw eight scoreless innings against Milwaukee, allowing one hit and striking out 10. The Dodgers held on to win 2–1.
