Quote:
|
Originally Posted by SkyDog
2055: Closer had 37 saves, 3.03 ERA. A total of 6 relievers had ERA's of 3.73 or less.
2053: rookie closer, 26 saves, much weaker bullpen overall
We may have the answer there.
|
Perhaps this is as much an indictment of current MLB reliever usage as the game's AI, but why would a reasonable manager remove a pitcher with a high endurance performing as Sevier did in 2055 (0.95 WHIP, 1.67 ERA, 7 HRA in 253 IP) for those relievers? I wonder if there is some sort of "winning by three runs or less in 8th inning on" instruction that automatically inserts a setup man and then the closer in the 9th. If so, it replicates real-life to an extent, but pitchers
this good would still get more than 1-3 complete games in a season. Pedro had seven in 29 starts in 2000 with a good bullpen and an endurance rating that I'd place certainly below 90.
If Sevier was on my team, I would throw a 'slow hook' on him, he'd have 15-20 complete games per season, and my team would have more wins as a result. He didn't even win half his starts in 2055 (for a team that won 86 games). For his career, 28% of his starts resulted in no-decisions. Compare that to some of MLB's current stars:
Pedro - 19.6%
Maddux - 20.7%
Clemens - 23.3%
R. Johnson - 22%
Schilling - 20.3%
Mussina - 20%
All of them have garnered decisions in signficantly more of their starts than Sevier, and others I've noticed, and that's in an era where relievers are used with greater frequency than ever.
My concern, and it's really not a major one, from looking in detail at Sevier and glancing at some of the other star pitchers from the 100-year history of this test league, is that the AI pulls the best starting pitchers too frequently.