Hard Stuff, Stinginess, and Control in the Bullpen: A Historical Perspective--Part I

The best closers today have high strikeouts per nine innings (K/9) ratios, high strikeouts-to-walk ratios, and are stingy in allowing base runners.  This Baseball Historical Insight is the first of two examining the extent to which this combination has been characteristic of ace relievers as the use of dedicated relief pitchers has evolved since the mid-1920s.

Hard Stuff, Stinginess, and Control: A Historical Perspective, Part I

Of the 30 principal closers for the 30 major league teams in 2013, 19 averaged more than a strikeout per inning. Five had a strikeout (K/9) rate of 13 or better for every 9 innings pitched--Cincinnati's Aroldis Chapman (15.8); KC's Greg Holland (13.8); Pittsburgh's Jason Grilli (13.3); Atlanta's Craig Kimbrel (13.2); and the Dodgers' Kenley Jansen (13.0).  Jansen, Holland, and Kimbrel were also exceptionally stingy in allowing base runners; they were among only seven closers whose WHIP (walks and hits per inning) was less than one. Only Kenley Jansen among those power relievers with 13 or more strikeouts per 9 innings was among the four closers with such exceptional control that their strikeouts-to-walks ratio was better than 6-to-1.  The most outstanding reliever was Boston's Koji Uehara, who did not assume the closer's role until late June and thus recorded only 21 saves, but whose K/9 rate was 12.2 (eighth-best among the game's 30 closers) compared to walking only 1.1 batter per 9 innings for a stupifying strikeout-to-walks ratio of 11.22--by far the best in the major leagues--which explains his equally jaw-dropping WHIP of only 0.56, significantly better than Jansen's 0.86 WHIP which was second among big league closers in 2013.

Certainly since the start of this century--and arguably dating back to at least Dennis Eckersley--these characteristics have defined most of the best closers in the business.  But it was not always thus.  Until the beginning of the expansion era in the 1960s, ace relievers called upon to "close" out victories typically were not pitchers with extraordinary talent, capable of dominating games, groomed specifically to the role. With a few notable exceptions in the first half of the 20th century--specifically Firpo Marberry in the first half of his career and Johnny Murphy for all of his--pitchers who became their team's bullpen ace were either breaking into the league and being tested first in relief, pitchers trying to rehabilitate their careers after being found wanting in a starting role, or veterans who no longer had the stamina or the stuff  to be front line starters.  Statistical splits between starting and relief pitchers show that relievers generally did not match up with starters in strikeouts-to-walk ratios or WHIP, nor did they in K/9 ratio until the late 1950s.  A closer examination of those pitchers who finished the most games for their team, including getting either the victory or a save, shows this same pattern holding true.  It must be remembered, however, that until the 1950s--and then, much more in the National than the American League--few teams used the same pitcher as their relief ace over many consecutive seasons.

Firpo Marberry was the first quality pitcher to be used almost exclusively in relief when he was in his prime. In his first two full big league seasons in 1924 and 1925 (at the prime ages of 25 and 26), Marberry relieved in 91 games for Washington, earning 30 saves (15 each year), and was instrumental to the Nationals winning back-to-back pennants. A hard-throwing right-hander, Marberry's 5.1 Ks per 9 innings in 1925--when all of his 55 appearances were in relief--was nearly double the 2.7 K/9 average of AL pitchers that season.  His control, however, left lots to be desired; Marberry also walked an average of 4.3 batters per 9 innings, significantly higher than the average 3.5 walks allowed by AL pitchers, and in 1924 and 1925 combined, Marberry as a reliever allowed more walks than strikeouts racked up.  Indicative of his effectiveness, however, his 1.4 WHIP as a reliever in 1924 and 1925 was better than the league average of 1.5.

Unlike Marberry, baseball's next best relief ace, the Yankees' Johnny Murphy beginning in the mid-1930s, was not a power pitcher with a K/9 ratio better than the league average.  His manager, Joe McCarthy, made Murphy's career by astutely convincing the young pitcher that he did not have the stuff to make it long-term as a starting pitcher.  Possessed with an often-wicked curve, Murphy in 375 games in relief walked more batters than he fanned, but then again AL pitchers walked more batters than they struck out most seasons during his career.  Reinforcing his effectiveness in closing out Yankee victories, Murphy's 1.38 WHIP in his years as the Yankees' relief ace was better than the league average WHIP of about 1.5.  (Murphy was the subject of my first post--"Johnny Murphy Time" on this blog:  http://brysholm.blogspot.com/2013/03/introducing-baseball-historical-insight.html.)

Joe Page was next in the Yankees' long line of shut-down relievers, although his career longevity was short-circuited by temperance and temperament problems. His "outstanding" years were limited to two: 1947, when his K/9 ratio of 7.4 in 141-1/3 innings (95 per cent out of the bullpen) was far better than Bob Feller's 5.9 that led the league among ERA-qualifying pitchers, and 1949, when his 6.6 K/9 ratio was far better than Yankee teammate Tommy Byrne's league-leading 5.9 among starting pitchers.  Page, however, also had command issues, walking an average of 4.6 batters per 9 innings in 1947, and 5-per-9 in 1949. In both years, Page gave up fewer than 7 hits per 9 innings and had a WHIP substantially below the league average.

To this point, even as strong relief pitching was increasingly recognized as an important ingredient to winning success, few teams had followed the practice of developing talented young pitchers specifically to be relief specialists, as the Nationals had with Marberry and the Yankees with Murphy.  Most teams instead cycled through a progression of "ace" relievers from season to season, with few holding that role for more than two or three seasons.  By the late 1940s, National League teams began increasingly to designate or develop pitchers as relief specialists and to use them in that role for many consecutive years.  The most prominent included the Cardinals' Al Brazle (1950-54), who was converted to a relief ace in his mid-30s after having been a starting pitcher--which was typical in precedent; the Dodgers' Clem Labine (1953-59), who began his career as a swingman but whose talents were clearly for the end-game; the Reds' Frank Smith (1950-54), who started only seven of the 271 games he pitched; and the Pirates' Roy Face (1955-68). In their best years out of the bullpen, they were typically better than the league average in strikeout ratio, strikeouts-to-walks ratio, and WHIP, but not by much. More significantly, none were overpowering pitchers who approached the K/9 ratios of baseball's top power pitchers in starting rotations during that era.

American League teams were far behind the curve when it came to seeing value in having a dedicated relief ace who they could rely on over many seasons.  In the 1950s, only two AL teams had the same relief ace for as many as five straight years--Boston with Ellis Kinder from mid-season 1950 (when he switched from a starting role to bullpen ace on a permanent basis) to 1955, and Cleveland with Ray Narleski from 1954 to 1958.  In his best season of 1955, when he had a  9-1 record with 19 saves in 59 relief appearances, Narleski's 7.6 Ks per 9 innings would have placed him within the AL's top three starting pitchers in strikeout ratio. The AL may have been lagging, but in the 1960s it was the junior circuit that featured the most prominent denizens of the bullpen, some of whom became prototypes for the evolution to the modern classic closer.  My next Baseball Historical Insight will pick up from there.