Thursday, January 16, 2014

The top 10 postseason starting pitchers ever

Having gone through the best hitters in playoff history, the natural course of action is to switch over to the side of the game that tends to dominate October – the pitchers. So here are the 10 best starting pitchers in playoff history, in reverse order.



10. Red Ruffing
1932 WS
+41
1936 WS
-35
1937 WS
+90
1938 WS
+200
1939 WS
+188
1941 WS
+164
1942 WS
+87
Total
+736

Ruffing made ten starts in 7 World Series. He completed eight of them (with one of those an 8-inning complete game loss), and pitched 8.2 innings in another. He threw no shutouts, but had three one-run efforts, ending in scores of 2-1, 3-1, and 7-1. He also had a 3-2 complete game victory in which only one of the runs was earned.

There is a shortage of legendary pitcher’s duels in his catalog; he faced Carl Hubbell once, but Hubbell beat him easily. The next best opponent on the list is probably either Mort Cooper or Paul Derringer; the 2-1 win over Derringer in 1939’s Game 1 is Ruffing’s single best effort.

Ruffing didn’t have any especially overpowering Series outings, didn’t pitch in any legendary games or outpitch any all-timers. He just pitched well, frequently, for good teams. His Yankees went 6-1 in the Series in which he pitched, and Ruffing went 7-2 in his 10 starts with a more-than-respectable-even-if-it-wasn't-the-1930s 2.63 ERA (albeit with a good number of unearned runs allowed on top of it).

A pitcher like Ruffing is a bit tricky to write up over his postseason career. Playoff hitters often make their names with big individual moments. Pitchers, on the other hand, do it with entire games, and usually quite a few of them, and that can make it tougher to pick one thing to focus on.

9. Johnny Podres
1953 WS
-62
1955 WS
+650
1959 WS
+53
1963 WS
+114
Total
+755

On the other hand, sometimes picking a focal point for a pitcher is really, really easy. Johnny Podres is almost entirely known as the man who threw a shutout in Game 7 of the 1955 World Series, leading the Dodgers to their first-ever title. His 2-0 victory in that game earned him .623 WPA, which makes up over 80% of his career postseason contribution. Most of the rest came from an excellent start in Game 2 of the ’63 Series, which the Dodgers would also win over the Yanks. Given the lopsided Series history between the Dodgers and Yankees, Brooklyn fans could be excused for listing Podres much, much higher.

8. Sandy Koufax
1959 WS
+65
1963 WS
+119
1965 WS
+691
1966 WS
-57
Total
+818

The pitchers ahead of Koufax are so tightly bunched together that he would rank third if he’d retired after the ’65 season. That would have given him what would surely be the best last appearance in baseball history – a 3-hit, 10-strikeout shutout in Game 7 of the World Series on two days’ rest.

Koufax was great in the playoffs even before he was great in the regular season – in 1959, two years before his breakout, he had an excellent start in Game 5 of the Series, pitching 7 innings and allowing one run only to absorb a tough-luck loss to the immortal Bob Shaw (who, to be fair, was also better than Koufax during the regular season in '59). In ’63, he threw two complete games at the Yankees, striking out 15 in the first, then slowing down to K only 8 (with no walks) in a 2-1 win that finished off the sweep; both victories were against Whitey Ford.

The ’65 Series started with a tough loss to Minnesota's Jim Kaat in Game 2, but Koufax recovered to throw shutouts in Games 5 and 7 (the latter, as mentioned, coming on short rest even by 1960s standards). The last start of his career (postseason or otherwise) was Game 2 of the ’66 Series, in which the three unearned runs Koufax allowed helped Jim Palmer’s Orioles emerge victorious.

The final numbers: 57 innings, 61 strikeouts, 11 walks, 10 runs (6 earned), 2 homers allowed, and an 0.95 ERA. Favorable context notwithstanding, that’s extremely impressive.

7. Bob Gibson
1964 WS
+295
1967 WS
+468
1968 WS
+58
Total
+821

Bob Gibson, postseason stud. Who knew?

Gibson famously started 9 World Series games and threw 81 innings. (This does not mean he had 9 complete games. He had one 10-inning complete game and was removed after eight once, meaning he only pitched 98.8% of the innings available in his postseason starts.)

After a shaky Game 2 debut (8 innings, 4 runs allowed in an 8-3 loss), Gibson went lights out in the fifth game of the 1964 Series against the Yankees. He shut them out through eight, and it would have been nine had Mickey Mantle not reached on an error to open the last regulation frame. As it was, Mantle scored ahead of Tom Tresh’s game-tying 2-run homer. The Cards struck back immediately with a 3-run shot from Tim McCarver, and Gibson finished things in the bottom of the tenth – 10 innings, 6 hits, 2 walks, 13 strikeouts, and no earned runs allowed. He then started Game 7 on 2 days’ rest; the Yankees touched him up for 5 runs, but all of them were scored against a substantial St. Louis lead (Mantle hit a 3-run homer while down 6-0, and the other two came on solo shots in the ninth, an inning the Cardinals started ahead 7-3), and Gibson was left in to complete the victory.

The 1967 Series was even better for Gibson. It started with a complete game, 10-K, 2-1 victory, continued with a 5-hit shutout in Game 4, and concluded with another 10-K win in Game 7, a 7-2 title-clinching decision. For the Series, Gibson allowed 3 runs in 27 innings, struck out 26, walked 5, and yielded 14 hits. That is pretty impressive pitching, even by 1967 standards.

’68 started out looking like an improvement even on Gibson's previously opulent work – a 17-strikeout performance in Game 1, and a 10-K, one-run win in Game 4. Game 7 began in promising fashion, but Mickey Lolich matched Gibson’s early efforts, keeping Detroit close until Curt Flood famously fell down chasing a fly ball which gave the Tigers a lead they would never relinquish. Gibson absorbed the tough-luck loss, one which drove his World Series ERA all the way up to 1.89.

Like Koufax, Gibson is diminished by his last start – but only slightly. In neither case is the final appearance sufficient to come anywhere close to undermining the enormous value the pitcher had accrued beforehand in leading their teams to multiple titles.

6. Herb Pennock
1914 WS
+7
1923 WS
+96
1926 WS
+580
1927 WS
+104
1932 WS
+54
Total
+841

This is as good a place as any to highlight the relatively arbitrary distinction I’m making between starters and relievers. Many of the hurlers listed here appeared in both roles in the postseason at various times; I am listing their CPA totals combined between both roles, and assigning them to the role in which they accrued the most playoff value. 

For instance, Pennock is best known as a starting pitcher – but he actually relieved in nearly a third of his regular season appearances. In the World Series, that total increased to half, five out of ten. He pitched well in the five relief appearances – three fruitless scoreless innings for the A’s in 1914 as the Miracle Braves finished their sweep, saves in Game 4 of the ’23 Series (entered with the bases loaded, two outs, and a five-run lead in the eighth and retired Frankie Frisch before working the ninth) and Games 3 and 4 of the ’32 Classic, and three scoreless innings in relief of Waite Hoyt to finish off the team’s loss in Game 7 of the ’26 matchup against the Cardinals.

But much like the regular season, nobody discusses Pennock’s postseason relief efforts, because his best work came in games when he started on the mound. His first-ever Series start was a complete game 4-2 win in 1923’s second contest. The second start came in Game 6 that same year, and Pennock left it trailing 4-1. But he left for a pinch hitter, who walked to load the bases with one out; the pinch runner for that pinch hitter would score the tying run later in the inning as the Yankees rallied to take the lead and give Pennock the series-clinching win.

The effort in ’23 was fine, even good – but not exceptional. ’26 was a different story; Pennock nailed down a pair of complete game victories by scores of 2-1 and 3-2 (the second of them in 10 innings), then added the effective-but-insufficient relief outing in the seventh game. He also threw a 3-hitter at the Pirates in ’27 for his last postseason start.

In total, Pennock made 5 World Series starts, pitched 44 innings, and gave up 10 runs. Add in his relief appearances, and you get a playoff ERA of 1.95. That’s fractionally higher than Gibson’s, but that fraction quickly disappears when you account for the fact that Pennock was pitching in a much, much higher-scoring period, and Gibson allowed a couple of unearned runs while Pennock did not.

5. John Smoltz
1991 NLCS
+165
1991 WS
+342
1992 NLCS
+32
1992 WS
-44
1993 NLCS
+6
1995 NLDS
-19
1995 NLCS
+29
1995 WS
-76
1996 NLDS
+49
1996 NLCS
+51
1996 WS
+210
1997 NLDS
+27
1997 NLCS
-70
1998 NLDS
+36
1998 NLCS
-4
1999 NLDS
+29
1999 NLCS
-57
1999 WS
-6
2001 NLDS
+29
2001 NLCS
+13
2002 NLDS
+13
2003 NLDS
0
2004 NLDS
+72
2005 NLDS
+17
2009 NLDS
0
Total
+844

Smoltz pitched in 25 separate postseason series; that’s more than the combined total of pitchers 6-10 (23). Unsurprisingly, he is our first of what will be three entrants from the era of the expanded postseason.

Long and distinguished as his October career was, Smoltz never quite matched his extraordinary first postseason run of 1991. He threw a 6-hit shutout at the Pirates in Game 7 of the NLCS to secure the pennant, then put up 7.1 scoreless innings against Jack Morris in the famous Game 7 of the World Series. His contribution in that game is diminished very slightly by the fact that he was pulled with one out and two runners on in the eighth, but the excellent work in the preceding innings still adds up to +.310 WPA in a title-deciding game, which is a reputation-making figure on its own.

After subsisting in relative mediocrity for Atlanta’s next three Octobers (including the team's title in 1995), Smoltz posted a ’96 playoff run for the ages. It started with nine innings of one-run ball in the opening game of the NLDS, which the Braves won 2-1 in the tenth. Up next were a pair of NLCS victories against the Cardinals, with combined totals of 15 innings, 12 hits and 2 runs allowed, 3 walks, and 12 strikeouts. Smoltz then proceeded to dominate the Yankees in the first game of the Series, coming out of his team’s 12-1 win comfortably early.

The capper game in Game 5 of the Series, with the Braves and Yankees tied at 2 games each. Smoltz pitched 8 innings, allowed 4 hits and 3 walks, struck out 10, and allowed no earned runs. He did, however, allow a single unearned run in the fourth inning, and that was enough for Andy Pettitte and John Wetteland, who combined on a shutout. Smoltz was left with a consolation prize of a 4-1 record and an 0.95 ERA for those playoffs.

After respectable postseasons in 1997 and ’98 and a rough one in ’99, Smoltz missed the 2000 season due to injury and then moved to the bullpen for four years. His October work from 2001-04 was good – 18.1 innings, 4 runs (1.96 ERA), 3 saves to one blown, and a 2-0 record. But because the Braves were starting to lose earlier than they had generally done in the ‘90s, that work did not contribute enormously to his postseason scrapbook. A good start in the ’05 NLDS and an undistinguished relief appearance in the ’09 playoffs for the Cardinals composed the end of Smoltz’s October career.

The final numbers: 15-4, 2.67, 209 innings and a postseason record 199 career strikeouts. If you give someone those numbers in a single regular season (and include the four saves), it might earn him a Cy Young. Not bad for a guy whose two most famous starts may well have been a pair of 1-0 losses.

4. Curt Schilling
1993 NLDS
+150
1993 WS
+28
2001 NLDS
+203
2001 NLCS
+54
2001 WS
+225
2002 NLDS
+26
2004 ALDS
+16
2004 ALCS
+27
2004 WS
+74
2007 ALDS
+29
2007 ALCS
-24
2007 WS
+40
Total
+849

Schilling’s playoff career was written up in the Hall ballot post. I’ll point out here that he is the only pitcher on the list with positive CPA scores for three separate teams, and they were all significant positives (+178 for the Phillies, +508 for the Diamondbacks, and +162 for the Red Sox). The next starter who meets the first of those qualifications is actually Schilling’s 2001 teammate, Randy Johnson, who’s well down the list in the #37 spot all-time among postseason pitchers. There does not appear to be another starter who has at least +.1 CPA for three separate franchises, let alone +.15. (Cliff Lee probably has the best shot among active pitchers to join Schilling with that distinction if he ever gets traded again; he’s currently at .231 with the Phillies and .151 for the Rangers.)

3. Allie Reynolds
1947 WS
-115
1949 WS
+350
1950 WS
+240
1951 WS
+33
1952 WS
+479
1953 WS
-128
Total
+849

The chart tells the tale quite nicely. Reynolds had three mediocre-to-poor series – 1947, in which he had a solid outing in Game 2 (CG, 3 R) followed by a relative thrashing in  the more-important Game 6 (2.1 IP, 4 R); 1951, in which he was hit hard in Game 1 (6 innings, 7 walks, 5 runs) but recovered for a complete game 6-2 win in Game 4; and 1953, with a rough start in Game 1 (5.1 innings, 3 HR, 4 runs), a fairly unimpressive save in Game 5 (the last two outs of an 11-7 win, courtesy of a Jackie Robinson double play ball), and a blown save in Game 6  (on a two-run Carl Furillo homer that set the stage for Billy Martin’s walkoff single that clinched the Series for the Yanks).

More importantly, he also had three exceptional series. The first was in 1949. Game 1 of that Series was an all-time pitching duel between Reynolds and Don Newcombe; the two aces traded zeroes through the top of the ninth, before Tommy Henrich led off the bottom of the inning with a homer that gave Reynolds and the Yankees the victory. The teams split the next two games, and in Game 4, the Yankees stormed out to a 6-0 lead with Eddie Lopat on the mound. In the bottom of the sixth, however, the Dodgers scored four runs and had the tying runs on base when Reynolds was called in from the ‘pen. He proceeded to retire all ten hitters he faced, securing the victory and propelling his team to a 3-1 lead in the Series.

1950 brought nearly-identical results to the preceding year. Reynolds faced off with the great Robin Roberts in Game 2, pitching him to a 1-1 standstill through 9 innings. Joe DiMaggio led off the top of the tenth with a solo homer, and Reynolds made it stand up in the bottom of the inning, taking a 2-1 victory. Two days later, Reynolds relieved with two out in the ninth and the tying run at the plate, striking out Stan Lopata to finish off the sweep for his team.

1952 completes the trifecta of exceptional performances for Reynolds. It opened with a spotty performance in the first game of the Series (7 innings, 2 homers, 3 runs and a loss), but Reynolds turned that around in his next outing, a 4-hit, 10-K shutout in Game 4 that tied the contest at two games each. And with the Yankees trailing 3-2 in the Series, Reynolds entered Game 6 with the team ahead by a run and a runner on second in the eighth inning and worked the final four outs to preserve the victory. In Game 7, he relieved Lopat with the bases loaded and nobody out in the fourth, allowed only one run that inning (which tied the game), and worked the next two frames as well while the Yanks backed him with a pair of solo homers that earned yet another Series win for both Reynolds and the Bronx Bombers.

The overall numbers were 7-2, 2.79; that seems less impressive until you add in the four saves, which give Reynolds either a win or a save in 11 of his 15 playoff outings. Throw in a pair of classic duels against exceptional pitchers, and you can see where the #3 spot on the list comes from.

2. Art Nehf
1921 WS
+508
1922 WS
+63
1923 WS
+349
1924 WS
+153
1929 WS
-94
Total
+859

Nehf’s postseason accomplishments include the following: A lukewarm 4-4 record, more walks than strikeouts, and a substantial role in the largest blown lead in postseason history. He never won multiple games in the same series, and his teams went 2-3 in the Fall Classics in which he pitched.

And yet, the record easily backs up his placement as the second-best playoff starter ever.

Let’s go through the negatives first, such as they are. (I’m using “negatives” as an equivalent term to “losses” here; Nehf only earned actual negative WPA scores in 2 of his 12 postseason appearances.) Nehf absorbed complete game defeats in his first two October starts, Games 2 and 5 of the 1921 World Series; he allowed 3 runs in each of them while his Giant teammates scored one run between the two games. Given the high-scoring nature of the 1920s, 3-run complete games would generally get the job done, and they would have here as well but for a pair of outstanding performances from Waite Hoyt (who would make it close to the end of this list if it was a top 20).

Nehf was also credited with the defeat in the decisive Game 6 of the ’23 Series, allowing 5 runs in 7.1 innings as the Yankees beat the Giants 6-4 and took their first title. You’d think his WPA would be a large negative in that game, but it turns out not to have been negative at all – it’s +.065, a difference explained by the vagaries of scorekeeping. The Giants led 4-1 entering the top of the eighth.  Nehf recorded the first out of the inning, but then allowed two singles and two walks, forcing in a run with the bases still loaded. Still, with a 4-2 lead, the Giants could hold out significant hope of escape; their win expectancy at the time is listed at 68%. Nehf was relieved by Rosy Ryan, who walked in another run, then allowed the game-breaking Bob Meusel 3-run single (which sounds weird – it was helped by an error). Nehf gets the blame for most of the runs in traditional pitching statistics, even though they predominantly scored after he left; WPA splits the blame more evenly between starter and reliever.

Nehf’s final playoff defeat was also of the hard luck variety – a 2-1 loss to Tom Zachary in Game 6 of the ’24 Series, which immediately preceded one of the greatest baseball games ever played. He also pitched in that one, entering in relief immediately after the eighth-inning grounder that hopped over Fred Lindstrom’s head to tie the game, retiring Sam Rice with two runners on to preserve the tie, and leaving after allowing a pair of runners to reach in the ninth (one of them on an error). The Giants would get out of that jam as well, sending the game into extra innings and into legend.

The worst postseason outing of Nehf’s career is one in which he wasn’t assessed a loss. It came in the fourth game of the 1929 World Series, in which his Cubs trailed 2 games to 1, but had taken an 8-0 lead into the bottom of the seventh. Starter Charlie Root promptly spent that inning getting hammered all over Shibe Park by the formidable Philadelphia lineup, allowing them to pull within 8-4 before departing with runners on the corners and one out. Nehf entered to face Mule Haas, and coaxed a fly ball to center – which Hack Wilson lost in the sun, resulting in an inside-the-park homer that put the A’s within a run. A walk later, Nehf was pulled; four more batters (and two more pitchers) later, Philadelphia led 10-8 and was on its way to the title.

Those are Nehf’s bad postseason moments – and most of them either weren’t very bad at all, or weren’t entirely his fault. Here are the good moments.

In Game 1 of the 1922 Series, Nehf pitched 7 innings and allowed 2 runs (1 earned). The Giants rallied for a 3-2 win after he departed. Four games later, he went the distance in a 5-3 win to finish off the Series.

In Game 1 of the 1924 Series, Nehf allowed 3 runs (2 earned) on 10 hits and 5 walks, and came out with a win. This is made somewhat more impressive by the fact that the game lasted 12 innings and Nehf pitched all of them. It is not necessarily made more impressive but is definitely made cooler by the fact that the opposing starter, who also went the distance and took the loss, was Walter Johnson.

In Game 3 of the 1923 Series, Nehf faced off with Sad Sam Jones. Jones’s nickname was perfectly deserved in this one, as he pitched 8 innings and allowed only a Casey Stengel solo homer, but Nehf stuck him with the hardest of hard-luck defeats, throwing a 6-hitter and emerging with the 1-0 victory.

The 1921 World Series was the last best-of-nine series in baseball history (unless they change the rules again), which is why it included a Game 8. In the first inning of Game 8 of the 1921 Series, the Giants scraped together a run on two walks and an error. With Hoyt starting again and at his best, it would be the only run they scored all day. And it would also be the only one they would need, as Nehf threw a Series-clinching four-hit shutout.

A pitcher with a 2.16 career postseason ERA in the high-scoring 1920s, one who outdueled Walter Johnson in a 12-inning game and had not one, but two 1-0 shutout wins, one of which clinched a title? That sounds like a respectable list of achievements for an all-time great postseason pitcher.

1. Jack Morris
1984 ALCS
+53
1984 WS
+285
1987 ALCS
-47
1991 ALCS
+7
1991 WS
+1008
1992 ALCS
-115
1992 WS
-165
Total
+1027

Like Schilling, Morris was analyzed more fully in the Hall of Fame ballot post. Also like Schilling, Morris pitched in the postseason for three different teams; the difference on that front is that Schilling was good for all of his teams in October, where Morris was bad for Toronto in the ’92 playoffs, and followed that up by being bad enough in the ’93 regular season that the Blue Jays avoided him entirely that October.

But Morris was terrific for the Tigers in 1984, and I assume you all know what he did for the Twins in 1991. Speaking of which, I’m going to take a moment to address what I’ll refer to as the Lonnie Smith argument, which goes something like, “Nobody would talk about Jack Morris if Lonnie Smith hadn’t screwed up.”

For those unfamiliar with the play under discussion, it went as follows: Game 7 of the 1991 World Series was a scoreless tie into the top of the eighth, of course, between Morris and the previously mentioned John Smoltz. Lonnie Smith led off that inning with a single, and NL MVP Terry Pendleton then crushed a 1-2 pitch into the left-center field gap. Minnesota’s rookie second baseman, Chuck Knoblauch, pretended to field a ground ball and throw to second for a force, and the decoy maneuver appeared to fool Smith at least briefly, as he slowed down enough to keep him from scoring what should have been a routine run. Instead, he stopped at third, with Pendleton at second, and the Braves didn’t score in the inning; the Twins would win the game in the bottom of the tenth.

I have two comments on this play. First, it is entirely clear that Smith should have scored. Had he done so, it seems likely (though not certain) that the Braves would have won the game and the Series, leaving Morris as the guy who pitched really well in a loss. There are worse categorizations to fall under; I suspect that CPA would still be highly fond of Morris’s production had he lost 1-0 in 9 rather than winning 1-0 in 10, even if he wouldn’t be on top of the list (and would quite possibly fall entirely out of the top 10).

Second, while Smith’s play was clearly a blunder, its dreadfulness has been exaggerated by hindsight. Think of a horrific baserunning miscue you’ve seen recently. (The instance that leaps immediately to mind for me is this one, but feel free to substitute your own.) I’d be willing to bet that a significant majority of the examples you’d think of would have something in common – namely, they resulted in one or more outs, thereby thoroughly devastating the rallies amid which they occurred.

Smith’s foulup did not do this. Yes, he should have scored on the play – but the Braves were still left with runners at second and third with nobody out, which is hardly a disastrous outcome. And yet, they did not score, despite the highly favorable situation, because Morris induced a weak grounder to first from Ron Gant, intentionally walked David Justice, and got Sid Bream to hit into a 3-2-3 double play.

It was partly Lonnie Smith’s mistake that made Morris’s Game 7 effort legendary. But it was also Morris’s legendary performance that made Smith infamous. If Morris had allowed the Braves anything else at all in the eighth inning, they would have scored and taken Smith off the hook. As it was, Morris got a break, but he still had to cut through the middle of the Atlanta order to cash it in. The fact that he did just that, preserving the shutout and giving his team time to mount the winning rally, makes him deserving of at least a substantial portion of the acclaim his performance is given.

No comments:

Post a Comment