THE STAT GUT
Let the best pitchers throw the most innings
By BRADFORD DOOLITTLE
The Kansas City Star
Well, it’s been almost a month since I railed about the way modern big-league managers run their pitching staffs, so it’s about time for another go-round.
In a nutshell, my theory is that any advantages created by the rampant specialization of pitching roles in today’s game is more than offset by the fact that so many fewer innings are allocated to the best pitchers on each team.
Thanks to the ongoing miracle that is baseball-reference.com, let’s take a look at some more numbers on this topic that have recently become available. What we’re going to do is take last season as a proxy for modern pitcher use and compare it with 1957, a nice, round half-century ago.
Last season, big-league teams won 95.4 percent of the games in which they led going into the ninth inning. The figure for 1957 was 92.6 percent.
That latter figure was actually an unusually poor performance for teams of that era. In 1956, the number was 94 percent. In 1958, it was 95.7 — even better than today’s game.
The difference isn’t all that startling. Starting pitchers in 1957 completed 29 percent of their outings, compared with 2.3 percent in 2007. The average game fifty-one years ago had 4.9 pitchers take the mound. Last season, 7.9 pitchers appeared in the typical game.
Despite all those managerial shenanigans, the average team in 2007 won only about two more games than their brethren in 1957 when leading late.
Now, for the other part of the equation. The top eight pitchers on each big-league staff in 1957 (in terms of innings pitched) threw 89.4 percent of the average team’s innings. In 2007, that figure was 73.4 percent. The other innings went to what I usually refer to as “fringe pitchers.”
During a season, the average team today will see about 230 more innings going to fringe pitchers than they would have under the 1957 model. That’s the workload of a top-flight starting pitcher.
The average ERA for fringe pitchers last season was 5.02. For all other pitchers, it was 4.23. And that’s the crux of the issue. All these extra pitching changes means too many innings go to too many bad pitchers.
The contemporary structure of pitching staffs, the one used by every team, does work if you have the appropriate personnel and a deep staff. But not every team has those things.
The 1957 New York Yankees’ staff is the kind I’d love to see someone try to emulate in today’s game. The Yankees used only 12 pitchers all season. Seven pitchers topped 100 innings, and three more exceeded 60. Nine different pitchers threw complete games. Eight different pitchers logged at least two saves. New York led the AL in team ERA. Tom Sturdivant was the only Yankee pitcher to crack 200 innings.
For most of that season, the Yankees had a 10-man staff. In today’s game, that would enable managers to have two extra bench players. They’d be able to platoon players more effectively, choose hitters because of productivity rather than versatility and ultimately score more runs.
When it comes to pitching staffs, a little flexibility could go a long way. Is there a team out there bold enough to try?
1957 vs. 2007
| Winning percentages | 1957 | 2007 |
| When leading after 6 | .834 | .861 |
| When leading after 7 | .870 | .897 |
| When leading after 8 | .926 | .954 |
| Other categories | 1957 | 2007 |
| %Complete games | 28.9% | 2.3% |
| %Wins with saves | 31.5% | 49.2% |
| Pitchers/game | 4.9 | 7.9 |
| Top 8 IP% | 89.4% | 73.4% |
| Top 8 ERA | 3.72 | 4.23 |
| Fringe ERA | 4.55 | 5.02 |
NOTE: Top 8 refers to the eight pitchers on each staff in innings pitched
Join the discussion
Share your observations and experiences about news. Lively, open debate is the goal, but please refrain from personal attacks or comments that are racist, vulgar or otherwise inappropriate. If you see an inappropriate comment, please click the "Report as violation" link to notify a KansasCity.com editor. Thanks for your feedback.