Best relief pitching performances in World Series history

The 2014 baseball season ended with an incredible performance by Giants ace Madison Bumgarner, who pitched a shutout in Game 5, then came back three days later to enter Game 7 in the middle of the 5th inning and pitched five shutout innings of relief to finish off the game and the series. A pair of singles were the only base runners Bumgarner allowed as he made himself an easy choice for World Series MVP. I've been busy with other things in the month since that game, but have been wondering about the other greatest relief appearances in World Series history. I couldn't recall anything like what Bumgarner did, but what had I forgotten about, and not been around to see when it happened?

The very first relief appearance remains one of the best. It came during Game 2 of the first World Series, back in 1903. (Both starters went the distance in Game 1. In fact, Pittsburgh's Deacon Phillippe pitched five complete games in that series, a record I'm confident will never be broken.) Bucky Veil came into the game to start the 2nd inning and pitched 7 innings of relief, allowing only 1 run.

In all the years since then, 7 innings remains the longest anyone has pitched out of the bullpen in a World Series game while allowing no more than one run. Rick Rhoden in Game 4 of the 1977 World Series is the only pitcher to have done it since 1947.

Four times in World Series history a reliever has gone more than 7 innings:

• Charley Hall (Game 7, 1912): 5 R (3 ER) in 8.0 IP
• Jesse Barnes (Game 6, 1921): 2 R in 8.1 IP
• Bump Hadley (Game 3, 1939): 2 R in 8.0 IP
• Don Larsen (Game 3, 1957): 2 R in 7.1 IP

So, the record for innings in a World Series relief appearance is 8.1. Barnes picked up the win in that game, while also setting a World Series relief record with 10 strikeouts, which would stand for 45 years, until Moe Drabowsky struck out 11 in 6.2 shutout relief innings in Game 1 of the 1966 WS. (more on that in a minute)

The last pitcher to throw more than Bumgarner's 5 IP of relief in a World Series game was Greg Harris, in Game 3 of the 1984 WS; he pitched 5.1 innings, same as his teammate Andy Hawkins had thrown two nights earlier in Game 2 of that series. All told, there have been 54 World Series relief appearances if 5+ innings, but Bumgarner's is the only one from the last 30 years.

Of those 54 relief appearances of 5+ innings, only 14 of them featured 0 runs allowed. Here they are, in chronological order:

• Ray Collins (G6, 1912): 7 IP, 5 H, 0 BB, 1 K
• Duster Mills (G3, 1920): 6.2 IP, 3 H, 4 BB, 2 K
• Rube Walberg (G5, 1929): 5.1 IP, 2 H, 0 BB, 6 K
• Jack Russell (G1, 1933): 5.0 IP, 4 H, 0 BB, 3 K
• Joe Page (G7, 1947): 5.0 IP, 1 H, 0 BB, 1 K
• Warren Spahn (G5, 1948): 5.2 IP, 1 H, 1 BB, 7 K
• Russ Meyer (G6, 1955): 5.2 IP, 4 H, 2 BB, 4 K
• Larry Sherry (G6, 1959): 5.2 IP, 4 H, 1 BB, 1 K
• Bill Stafford (G5, 1960): 5.0 IP, 3 H, 0 BB, 2 K
• Moe Drabowsky (G1, 1966): 6.2 IP, 1 H, 2 BB, 11 K
• Bruce Kison (G4, 1971): 6.1 IP, 1 H, 0 BB, 3 K (plus 3 HBP!)
• Andy Hawkins (G2, 1984): 5.1 IP, 1 H, 0 B, 3 K
• Greg Harris (G3, 1984): 5.1 IP, 3 H, 3 BB, 5 K
• Madison Bumgarner (G7, 2014): 5.0 IP, 2 H, 0 BB, 4 K

When you consider the length of each performance, how many runs and base runners were allowed, the number of strikeouts, and the specific game each one was a part of, I think Collins, Page, Drabowsky, and Bumgarner each have an argument for being the best:

Collins remains the record holder for longest scoreless relief appearance in a World Series game, but his team was behind 5-2 when he entered the game, and that didn't change. A lack of offense isn't his fault (aside from his 0 for 2 performance at the plate), but his team was in a position to afford the loss, so you could argue there wasn't the same pressure. He also struck out only one batter, though strikeouts happened far less frequently back then, so we might prorate it to ~2.5 by today's standards.

Drabowsky set a World Series record for strikeouts in relief (which still stands), and was only one out shy of Collins, in terms of innings. It should be noted that he entered the game with the bases loaded during the 3rd inning, and walked in a run. That's not charged to him, but it does seem like a slight ding on his performance.

Like Bumgarner, Page's performance came in Game 7. Page gets the edge for allowing only 1 hit instead of 2, but Bumgarner had 4 strikeouts, compared to only 1 for Page. In terms of the opposing lineup, I think the 1947 Dodgers were a tougher bunch than the 2014 Royals, so that's a point for Page, but he was pitching at home, while Bumgarner was on the road. Page's Yankees also turned a 3-2 game into a three-run lead, so there wasn't quite as much pressure by the last couple innings, whereas Bumgarner has only a one-run lead the entire time he was in the game, and Gregor Blanco's error put a man on third base in the 9th inning, cranking up the pressure on Bumgarner even farther.

Page was a reliever by trade, and had pitched only 1 other inning in the previous three days, compared to Bumgarner's 9. I'm not saying what each pitcher had done earlier in the series should necessarily be counted when rating their single-game performance, but it seems worth mentioning.

I acknowledge that recency bias could be playing a factor, but I feel like when you consider the context of the performance in addition to results, Bumgarner's game deserves the crown. Even if you side with one of the other performances mentioned here, it'd been decade since baseball fans witnessed anything like this in the Fall Classic.