Can't Pitch 6 innings? How about 1?

Last night, Rich Hill came into the game for the Red Sox in the bottom of the seventh with runners on first and third and two outs. He struck out Jack Hannahan to end the inning, then worked around a leadoff walk and a wild pitch to finish the eighth with the Sox' 4-1 lead safe and sound. In and of itself, the performance was nothing special. Hill came into a semi-difficult situation and got the job done. But it’s not what he did that I want to look at, it's where he came from that is the focus of today’s post. Time for my favorite game, Player A vs. Player B (or in this case Group A vs. Group B).

Group A: 23 games, 20.1 IP, 23 Runs, 4 HR, 11 BB, 23 K, 31 Hits, combined $9 million this year and $6 million next year

Group B: 20 games, 24.1 IP, 9 Runs, 0 HR, 10 BB, 28 K, 19 Hits, combined around $1.5 million this year, arbitration eligible next year

Group A consists of Bobby Jenks and Dan Wheeler, the two big offseason bull pen signings that were supposed to give the Red Sox a shut down reliever for every inning 6 through 9. Group B consists of Rich Hill and Matt Albers, two former starters picked off the scrap heap in the offseason for less than a million each who are actually giving the Red Sox a shut down reliever for every inning 6 through 9.

Before I get much further into my analysis, I need to say that analyzing bull pen pitchers through 20 appearances is pretty pointless. For all I know, come the end of the season, Wheeler and Jenks could be out pitching Mariano Rivera while Hill and Albers could be in AAA or released. What I’d really like to look at is where relievers come from, why it is really dumb to give big long term deals to relievers unless they have “Rivera” (and possibly “Papelbon”) on the backs of their jerseys, and how to build a bullpen. It just so happens that on this date, the Red Sox have 4 perfect examples of what I am talking about.

Bobby Jenks was drafted out of high school by the Angels as a starting pitcher. After 5 years in the minors in which he never had a season with a walk rate lower than 5.3 per 9 (very bad), the White Sox claimed him off waivers, converted him into a reliever, and his career took off. In his second year in the big leagues, he assumed the closer role for the White Sox and earned consecutive All Star births.

Pride of Providence Dan Wheeler was selected by the Devil Rays in 1996 in the 34^th round as a starting pitcher. In the minors, he started 136 of his 171 appearances posting a 4.48 ERA (which would be higher if you strip his relief appearances). In the majors he has started just 9 of his 542 games and posted a 3.96 ERA, while being death to right handed batters (.219/.270/.370).

Matt Albers was a starting pitcher in the Astros minor league system and put up solid numbers. Through 107 starts over 6 years, Albers posted a 3.63 ERA with a strong strike out rate against a slightly poor walk rate. One of the centerpieces of the Miguel Tejada Baltimore-Houston trade, Albers was immediately converted to the bull pen where he struggled the last 3 years. Since joining the Sox, he has flashed the strong strike out with iffy command profile that he had as a top prospect with the Astros.

Rich Hill came up as a starter in the Cubs system and was the darling of the stats community. A lefty with a giant looping curve ball, Hill struck out 11.7 batters per nine innings in the minor leagues, but his walk rates were concerning. Sure enough, when called up to the Majors, the walks remained high while the strikeouts dropped to about 8 per 9 innings. Some decent ERAs masked a starting pitcher who couldn’t command his pitches. After failing to regain footing with the Orioles two years ago, Hill changed his arm angle and became a relief pitcher and is thriving in a short stint with the Red Sox.

I know I am not blowing anybody’s skirt up by informing them that most good relievers are failed starters. Mariano Rivera, Billy Wagner, Joe Nathan, Joakim Soria, Dennis Eckersley, Goose Gossage, etc. were all starters first that for one reason or another couldn’t handle the job. Moving to the bullpen allows pitchers to strip away extraneous pitches and focus on their two or three best (or in Rivera’s case, one). Fastball pitchers get a chance to throw their hardest for a short time knowing they don’t need to pace themselves. Junk ballers give hitters a different look from the starter that they don’t have a chance to get used to by facing the pitcher 3 or 4 times in a game. The creation of the modern bullpen has given thousands of pitchers a second chance at extending their careers.

So, knowing that solid relievers can be found in failed starters, why do teams give multi-year, multi-million dollar contracts to middle relievers on the wrong side of 30? Young, failed starters can be found almost anywhere. They can be throw-ins in trades, signed as major or minor league free agents to sub-million dollar deals or drafted in the Rule V draft. Pretty much any pitcher that has one excellent skill (keeping the ball in the park, throwing the ball hard, a lot of strikeouts, groundball master, lefty killer) can be converted into a one inning reliever. And yet, this offseason, look at the contracts that were handed out to middle relievers of at least $10 million, with their corresponding ERAs:

Joaquin Benoit, 3 years, $16.5 million, 6.62 ERA
Jesse Crain, 3 years, $13 million, 2.35 ERA

Scott Downs, 3 years, $15 million, 0.66 ERA

Brian Fuentes 2 years, $10 million, 5.06 ERA

Kevin Gregg, 2 years, $10 million, 3.93 ERA

Matt Guerrier, 3 years, $12 million, 3.28 ERA

Bobby Jenks, 2 years, $12 million, 9.35 ERA

J.J. Putz, 2 years, $10 million, 1.80 ERA

Rafael Soriano, 3 years, $35 million, 5.40 ERA

In that list you have 4 solid to great performances, one mediocre (Gregg) and  4 disasters. And these were all relievers who pitched really well either last year or the year before. If I'm spending $5 million on a player, I don't want a 50% chance of failure. 


The problem with relievers is that we do not know how they are going to perform from one year to the next, save for the select few (Rivera, Trevor Hoffman). The fact that they only pitch 50-70 innings on average allows for way too much variance in performance. If a one inning reliever allows three infield singles followed by a home run on a curveball that slipped (for a 36.00 ERA), it would take him 11 more appearances of scoreless work to get his ERA back to 3.00. With that kind of uncertainty, why commit the years and dollars to a pitcher who, on his own, contributes so little? Teams have even begun spending first round draft picks on relievers so they can fast track them to help out their big league bull pen, but as we have seen (Craig Hansen) this is not always a sound strategy.

So what constitutes a strong bullpen and how do you build it without going the expensive free agent route? For starters, I think it takes two relief aces (think Bard and Papelbon). These should be two versatile guys that can handle righties and lefties, generate strikeouts, limit walks and keep the ball in the park. This is where teams usually throw a lot of money around to get their 8^th and 9^th inning guys, but as is the case with the Sox, this is not necessary. 


A good way to find these aces is through your own farm system. Take a look at some of the pitchers you drafted in the first few rounds. Pick out the ones who have not established themselves as anything more than a number 5 starter. Then look at how hard they throw (92-94 in a starting role), their strike out rate and their walk rate. If the strike out rate is high (8 per 9 innings or more) and the walk rate is manageable (around 4 per 9 innings or less), give the kid a chance in the bull pen. A perfect example besides Bard (whose walk rate was astronomical in the minors) and Papelbon (who I still believe would have been a good number 2 or 3 starter), is Michael Bowden. Drafted in the Sandwich round (47^th overall), Bowden (who averaged 92 MPH on his fastball as a starter) opened strong in the minors with a strikeout rate around 8.5 per 9 and a walk rate in the 2s. When he was given a look as a starter in the big leagues, he failed to maintain these rates and struggled. Now a reliever in AAA, Bowden is dominating hitters and looks like he could be a solid addition to the Sox bullpen at any time this season.

After the top two, I think it is good to fill out the next 4 spots with relievers who have one dominant skill. A lot of teams, including the Red Sox, seem to be of the mind that you now need a 7^th, 8^th and 9^th inning pitcher. I think it is ok to assign an 8^th and 9^th inning guy as long as those innings are important and those 2 are your best relievers (and not just guys who have “done it before”), but I don’t see why there needs to be a 7^th inning guy. Late game bull pen management  should be a game of matchups. Just because one pitcher is your 7^th or 8^th inning guy, maybe he struggles some against lefties and there are 2 coming up that inning or maybe he is a fly ball pitcher and you have a one run lead at US Cellular Field (a park where a lot of home runs are hit). In this case you want to have a variety of skills to turn to. For instance, keep around a lefty and righty that are tough on their similarly handed batters in the event a string of those batters comes up in an  important spot. It would be great if those guys could face both hands, but if they could they would be relief aces and thus, too expensive. 

The other two spots are good to fill out with relievers who specialize in a different type of batted ball. Relievers that generate ground balls are excellent commodities to have when you have runners at first and third with one out and a one run lead and need a double play. Fly ball relievers can also be valuable if you play in a big park and/or have a really strong outfield defense, because fly balls that stay in the park are the easiest outs to make. Luckily, both these types of pitchers can be found on the scrap heap with solid scouting and a quick check of the numbers.

Finally, to complete your 7 man bullpen, if you must have one that large, it’s great to have a guy like Tim Wakefield or Alfredo Aceves who can throw mop up duty and take the occasional start. Again, these are guys that can be found on the cheap and for short term deals.

So that’s how to most effectively build a bullpen for short money. Use your farm system, use your scouts and scour the unwanted pile of junk to slide guys into specific roles. Notice nowhere in there did I say go out and sign the best available free agent middle reliever. When put in the right positions (don’t let righties who can’t get lefties out face lefties), a lot of freely available pitchers can succeed. Not only does this help the team on the field, but it allows teams to allocate their precious funds to starting pitchers, position players, the draft and international signings.

This offseason, Theo made a couple signings that seemed out of character for him due to the disaster of last year’s bullpen. So far those signings (Wheeler and Jenks) have not lived up to their billing, but a couple pitchers pulled from the wreckage have filled in admirably. Luckily the Red Sox will never be hamstrung by spending too much on a reliever, but it is still good practice to focus on the rejects when filling out your pen. With reliever performance changing so much from year to year, you want to think like Robert DeNiro in Heat, "don't let yourself get attached to any [bullpen pitcher] you are not willing to [release] in 30 seconds flat if [they] feel the heat around the corner."