Now we've reached the starting pitcher slots. The slots are roughly in inverse order of rotation spot, so today we select the 5th starter. Since the requirements of a starting pitcher are pretty much the same for all five, I'm going to discuss a couple of alternative for each slot, but there's nothing special about any of them, they're all starting pitchers. I might do left handed alternatives with any left handed starters I might select.
Obviously, starters selections need to be actual starters. Mariano Rivera would make a poor choice, I think. Or at least a hard to justify one. Other than that, I favor picking the pitchers with high values. Pitching WAR is a good start, but I'm going to favor rate stats combined with extended track record. So my starters tend to have a strong career with a decent length and a high peak. The pitcher's bats actually do matter, but only half as much as they would in a no DH season, since I'm splitting the season into 81 games with DH, 81 games without. Pitcher fielding also adds some value.
Selection Greg Maddux
Possibly no pitcher combines peak and extended performance like Greg Maddux. His peak was longer than many All Stars' careers. From 1990 to 2001, Maddux had an fWAR over 5.7 every single year. For seven of those years, his ERA (and FIP) was less than 3.00.
Maddux's peak even had it's own peak, with his normalized ERA and FIP coming in at or below 70 every year from 1992 to 1998. And that's just the peak. In his 22 qualifying seasons, Maddux was never worse than league average. Longevity, check. A brilliant, extended peak, check. How'd he do it?
Well known as a thinking man's pitcher, Maddux never excelled at getting strikeouts. No doubt Maddux was a great observer of hitters with the ability to spot and exploit weaknesses. Legned has it that he basically outthought hitters, setting them up months in advance so that they would feel comfortable in a high leverage situation later in the season and then turning the tables. While probably apocryphal, the legend couldn't hurt Maddux, as any hitter who believed it would set himself up, no input from Maddux required.
While batted ball data is only available after Maddux's peak, it's clear that did a good job keeping the ball on the ground in those years, though not a great one. Maddux did seem to thrive on location and movement, though, particularly during his peak. His lifetime batting average on balls in play is somewhat lower than average at .281, which, combined with his low home run rate (0.63 HR/9 for his career and atn amazing four homeruns in 202 innings pitched in 1994) and only moderate strike out rate (16.5% for 6.06 K/9), suggests that Maddux not only induced weaker contact on fly balls, but had fairly strong groundball tendencies during his peak years. Maddux is the sort of pitcher who will be that much better pitching in front of a strong defensive infield, so he may see more starts with Ozzie Smith behind him than the other starters.*
*For similar K% and HR/9 rates, groundball pitchers will show higher batting average on balls in play than fly ball pitchers because groundballs in play are much more likely to be hits than fly balls.
Maddux's most obvious trait as a pitcher, though, is his extreme control. He walked only 4.9% of the batters he faced across his career at only 1.8 per 9 innings. That control combined with his ability to keep the ball in the park meant that while Maddux relied on his defense to the make the play, he gave them the opportunity to do so as well.
I should note that while Maddux didn't exhibit legendary strikeout "stuff", his fastball started out around 93 mph early in his career and still ran about 86 mph at the end of his career with the Padres. And his offspeed pitches had plenty of separation (at least at the end of his career) with his Padres tenure curveball averaging 73 mph and both his changeup and slider average about 80 mph. Maddux is the paragon of the "hit your spots, change speeds, know the hitter, have movement on the ball" skillset and, as a result, is a strong candidate for greatest pitcher of all time. He definitely deserves his spot on this team.
In many ways, the fly ball version of Greg Maddux, Schilling effectively traded Maddux's lower home run rate for a higher strike out rate. Still, lots of control (1.96 BB/9) and a lot of strikeouts (8.6 K/9). He was also some what less durable than Maddux across his career and took some time to work out his early control issues. Still, a surprisingly underrated pitcher who ought to be a no doubt Hall of Famer on his regular season performances alone, I suspect what will eventually get Schilling in are his famed post season performances with the Diamonbacks and the Red Sox. An amazing pitcher overall, with a career line of 3.46/3.23/80/74 and 86.1 fWAR. His best year was maybe 2001, maybe 2002, when he posted lines of 2.98/3.11/66/68 (7.6 fWAR) and 3.23/2.40/75/54 (9.7 fWAR), respectively.
He also hurts my theory about fly ball pitchers having lower batting average balls in play, with a .293 lifetime compared to Maddux's .281. Maybe Maddux really was that good at inducing weak contact. Oh, and Schilling is a wonderful case study for how the DH effects pitching stats. Going from NL to AL in 2004, his K/9 stable around 10.5 for the last three years, dropped to 8.2ish for the next 3. Facing pitchers can really buff your strikeouts.*
*Nolan Ryan pitched 2737 innings in DH era AL. At 2 K/9, that cost him another 608 or so strikeouts.
Halladay is another pitcher with strong similarities to Maddux. High control, good but not great strikeout rate, high groundball rate (that's been on the decline) and good home run suppression. Halladay makes a good choice for the same reason Maddux does, though his true talent might be a smidgen or three less. Likely Hall of Famer, with a career 3.25/3.31/72/75 line and 71.3 fWAR in 14 years. His best year was arguably 2011: 2.35/2.20/61/56 for 8.2 fWAR.