A Modified Pitching Rotation Proposal: The Frankenstarter




With the Blue Jays yet to decide on their fifth starter, today could go a long way to helping management choose.


NOTE: This article has been heavily updated thanks to the news that J.A. Happ will start the season on the DL and that Dustin McGowan will be the fifth starter.



In today’s Grapefruit League action in Bradenton, J.A. Happ is scheduled to start with Esmil Rogers expected to follow him. Happ and Rogers represent two of the four remaining possible fifth starter candidates with Dustin McGowan and Todd Redmond the other two.


Happ, 31, has been injured (back) a bit this spring and has only made three starts with his most recent being his “best,” going two and two-thirds innings, giving up three runs on three hits and walking four. This outing actually lowered the lefty’s ERA for the spring to 20.25. Happ is trying to reclaim a starting role after a decent start to 2013 was interrupted by a knee injury that he suffered after getting hit in the head by a line drive on May 7. He was actually better after coming back on August 7 (allowing a .724 OPS against in his final 11 starts as opposed to a .753 OPS in his first seven) and, if he pitches like that, would be a solid fifth starter.


Rogers, 28, has been better than Happ this spring (although how could he be worse?), throwing 13 2/3 innings with a 5.27 ERA and 18 strikeouts with just five walks. While the Blue Jays probably like Rogers as a long man in the bullpen, the fact that he has better command than Happ and offers a 2.18 K/BB ratio from 2013 (and a 3.60 K/BB ratio this spring) give him the edge. Rogers has been relieving more often in spring training which means that he’s been facing poorer quality of opposition than Happ has thus far which makes his lackluster numbers (5.27 ERA, 1.54 WHIP) a little more concerning.


UPDATE: Both Happ and Redmond (and everyone else who pitched) got smoked by the PIttsburgh Pirates offense. It has been announced that Happ is going to start the season on the DL and Dustin McGowan will be the fifth starter.


As far as the other two competitors go, Redmond had a good start the last time out against the Tigers but tends to get hit harder on his second and third time through the batting order (if he can make it to a third time) while McGowan had three very good innings in his last appearance and will likely get one more game in Florida before being considered to start the home opener in Toronto.


The first four spots in the Blue Jays’ rotation are pretty much set with R.A. Dickey, Mark Buehrle, Brandon Morrow and Drew Hutchison getting ready to go. I’m of the opinion that the Jays, especially if they’re bent on carrying eight relievers, should take a step towards what some other writers (here’s another) have proposed and modify their pitching rotation and use two or three of Happ, Rogers, McGowan and Redmond in what I’m going to call a “Frankenstarter” role.


In 2013, Happ put up some relatively stable numbers facing batters multiple times through the order. He had a .734 OPS against the first time through, a .714 OPS the second time through and a .775 OPS the third time. Interestingly, his walk rate went up and strikeout rate down the third time through the order. Redmond was far better through the order the first time through, surrendering just a .581 OPS the first time through the order and a .961 OPS the second time and .813 the third time. McGowan, working as a reliever in 2013, only faced batters once and had a .609 OPS against while Rogers had an .819 OPS the first time, a .792 OPS the second and .813 the third as a starter. His OPS against the first time through an order as a reliever was very similar at .817.


What this means is that pitchers could be used optimally by only allowing batters to see them once. For someone like Todd Redmond, this seems like the best strategy while there are more marginal differences in how hard batters hit Rogers and Happ in subsequent plate appearances.


When it comes to platoon splits, in 2013 Happ had negative platoon splits with a .708 OPS against righties but .802 against lefties while Rogers and McGowan both had regular platoon splits, pitching better against righties. Redmond, a righty, had OPSs against about 70 points better when facing left-handed hitters. Despite the 2013 negative splits, Happ has been better against left-handed hitters over his career (with a .703 OPS against lefties and a .767 OPS against righties) and the Jays could use a system that follows to start every fifth day.


Redmond starts. He’s by far the best pitcher the first time through the order and, with him on the mound to begin the game, will likely face a lineup with more left-handed hitters. The goal would be to get three innings out of Redmond and then, when he’s gone through the order once (or until he faces some adversity), McGowan would come in. You try to get two to three innings out of McGowan and then Rogers and Happ are available for lefty or righty matchups until you get to the seventh inning after which you can use the regular relievers.


UPDATE: I’ve decided that it’s probably best to have McGowan as the starter. By being able to have him start the games, he’s going to be able to have a scheduled warm-up plan and won’t have to get up in the bullpen and rush to get ready if needed.


You could conceivably have all four pitchers in the game but one of Rogers or Happ could be held back for the next day and any of them could be used in relief a few days later (on a normal bullpen session day). Rogers is available to be a reliever whenever he’s needed but as manager, I would try to make sure that he didn’t pitch the day before a Redmond start so that he’s fresh and able to go two or three innings if necessary.


With this system, you’d pretty much be piggy-backing Redmond and McGowan (as well as providing them with designated relievers to mop up) which works well for two very important reasons. The first is that Redmond was excellent the first time through the order in 2013, something that he showed again in his strong start against the Detroit Tigers in his latest Grapefruit League start. If he’s good his first time through the order as a starter, let’s not waste that and let him start and throw three or four innings. This method also lets the Jays control how many innings McGowan throws while making sure that they’re getting a solid return on their investment. If he throws three innings per game, once every five days, he’ll still get up to about 125 innings on the season which is a big jump from what he’s been able to do. If you think that he’s getting tired, you can dial that back to one or two innings at a time and if he looks strong, you can flip flop him and Redmond or relegate Redmond to the bullpen (or cut him loose). This way, you’re not relying on McGowan to shoulder the burden of throwing five or six innings every start and you’re maximizing the results he gets by only letting batters see him once or twice.


UPDATE: There was a lot of chatter on Twitter yesterday as to whether the McGowan would be able to get his pitch counts up to the requisite 90-pitch threshold to start the season. The Frankenstarter solution allows him to gradually get to the point where he can give the team five or even six innings without feeling any pressure to rush himself to get there.


Happ and Rogers have both shown to be okay in a starting role but both have been maddeningly inconsistent. This allows them to be used in controlled and limited roles with the possibility of riding one or the other if he gets hot.


As it stands now, the Blue Jays pitching staff is likely going to consist of the first four starters (Dickey, Buehrle, Morrow and Hutchison) and five key relievers (Janssen, Santos, Delabar, Cecil and Loup). The other four pitchers that the Blue Jays seem hell bent on keeping (because they’re out of options) could be put to better use by using a “Frankenstarter” system that takes the best parts of each of them and puts them in a role in which they will (hopefully) succeed.


The fact that the Blue Jays have four regular starters means that they can implement a Frankenstarter system for just one of their five regular start days. It also doesn’t necessarily tie up all four pitchers to just pitch the one day; all four should be available for limited relief situations the rest of the time. Instead of trying to get 200 innings out of one guy and have the bullpen pick up the rest of the slack, you could conceivably get 400 innings out of four pitchers (125 each for Redmond and McGowan with 75 each for Rogers and Happ), keeping them all fresh and active throughout the season.


Since a team will need about 1450 innings (give or take) to complete the season, getting 400 from four of your pitchers would get you a long way. With two pitchers almost locks to throw 200 each (Dickey and Buehrle) that gets you to 800 innings, requiring just about 650 innings combined from the remaining seven pitchers on the Opening Day roster.


The ball’s in your court Blue Jays, take a risk and move into the future of pitching, the Frankenstarter.


Don’t forget about The 2014 Toronto Blue Jays Minor League Handbook, available March 31 as an ebook at Smashwords.com, Amazon, Apple iBooks, Kobo and other fine retailers for $7.99. Pre-order information coming soon! You can also preview the book at our Smashwords.com page!


If you’re in Toronto, you can also join us for the Book Launch of the 2014 Toronto Blue Jays Minor League Handbook! Find all the details at our Facebook event page! You can also “like” us on Facebook!

2 thoughts on “A Modified Pitching Rotation Proposal: The Frankenstarter

  1. I like the Frankenstarter in principle, but it always gets tough figuring out how to get to 1450 innings overall. Going with the extra pitcher helps but still…I don’t think you can ask either half of the Frankenstarter to pitch ~4 innings every five days and then log random bullpen time as well with the way that starters prepare. And then to get to 125 innings they’re both going to have to average 4 innings a game, which eliminates the benefit of once through the lineup and seems unlikely — very hard to count on 250 innings out of your two Frankenstarters alone when the 1/5 of the season they’re pitching in only has 291.6 innings. And if the idea is to get to the 7th inning and use the five relievers left, then their workload is going to be high compared to a typical season. I’m sure it could be done but with the ebb and flow of a season it’s no small task to manage a bullpen with two guys in it that you kind of need to keep in reserve for the 5th start while also finding regular innings.

    1. Thanks for the comment halejon. I like your logic but in my innings count, I was including the fact that the two Frankenstarter guys would be available to relieve on their regular bullpen days for an inning or so each. In effect, they become long-men once through the rotation (on the day of the Frankenstart) and shorter-term relievers another day through the rotation. Rogers would be kept out of the games the day before a Frankenstart so that he’s available to mop up but can be used in normal one and two-inning stints otherwise. That’s how I estimate that each of the Frankenstarters can get to 125 innings: I’m not strictly appearances to one game a week. With McGowan, you might want to be a bit more careful but I’d say that you could use Redmond and Rogers at your discretion.

Comments are closed.