Checklist for Aging MLB Superstar Pitchers to Know It's Time to Retire

We established not too long ago that, though baseball players can enjoy longer careers than most athletes, there comes a time when even the best of 'em have to know when to hang up their spikes.

At the time, we (meaning I) proposed a very simple checklist that superstar hitters can use to know when it's time to retire. In a nutshell, it's time for them to call it quits when they're old and beat up, they can't hit, hit the ball hard, run the bases or field their position.

Basically, when they're old and can no longer do the simplest of things. Told you it was very simple.

Now it's time to discuss pitchers, and their retirement checklist is really no more complicated than the checklist for hitters, really, nor is it longer.

If you'll follow me this way...

 

The Primary Test Subjects

Coming up with a retirement checklist for superstar hitters required a sample size of worthy players to focus on, and the same goes for pitchers. And like with the hitters, we need to make sure that we're talking about players that have had access to modern medical procedures and training methods.

Using WAR as a guideline, I went back to 1985 to draw up a list of the top 15 superstar hitters who are no longer active. Doing the same with pitchers on Baseball-Reference.com, the list of the top 15 no-longer-active starters looks like this:

1. Roger Clemens: 137.3
2. Greg Maddux: 104.8
3. Randy Johnson: 104.1
4. Pedro Martinez: 85.9
5. Mike Mussina: 82.8
6. Curt Schilling 80.8
7. Tom Glavine: 74.0
8. Kevin Brown: 68.7
9. John Smoltz: 66.7
10. David Cone: 61.8
11. Chuck Finley: 58.5
12. Bret Saberhagen: 57.8
13. Kevin Appier: 55.0
14. David Wells: 53.6
15. Kenny Rogers 51.2

The "superstar" label applies less and less the further you move down the list, but this is still a pretty good sample size. Starting pitchers of all types are represented here, from righties to lefties to strikeout pitchers to control pitchers and so on.

I looked at each of these 15 guys individually and determined precisely when they went from their prime years into their twilight years. No real exact science went into the process. I just looked for when pitchers first emerged as stars, and for when they started to see downturns in both their workloads and performance.

When that happens, they need to look in the mirror and consider the following questions.

 

Old and Beat Up?

When I did my checklist on hitters, this was the last point I brought up. But since it seemed to inspire a few "obvious point is obvious" remarks, I'll go ahead and quickly get it out of the way in this discussion.

The hitters I looked at all had old age in common, and the same is true of the pitchers we're looking at. Nine of our 15 pitched into their 40s, and the youngest of them—Kevin Appier in 2004—still made it to his age-36 season.

The conventional wisdom is that the peak age for ballplayers is around 27, but J.C. Bradbury found in a piece for Baseball Prospectus that the peak age is more like 29. As such, our sample size consists of pitchers who lasted way longer in the big leagues than your average hurler. All 15 of our guys got to be ancient by baseball standards.

Our sample size also consists of guys who accumulated quite a few injuries during their careers. With a little help from Baseball Prospectus' injury database, here's a look at how many surgeries our 15 pitchers had and how many trips they took to the disabled list.

Player	Surgeries	DL Trips
Roger Clemens	1	6
Greg Maddux	0	1
Randy Johnson	4	10
Pedro Martinez	1	10
Mike Mussina	0	7
Curt Schilling	6	11
Tom Glavine	1	4
Kevin Brown	2	14
John Smoltz	4	13
David Cone	3	4
Chuck Finley	1	6
Bret Saberhagen	5	13
Kevin Appier	2	7
David Wells	4	8
Kenny Rogers	2	3

The total count here: 15 players, 36 surgeries and 117 trips to the disabled list. For a quick comparison, the 15 hitters I looked at had combined for 33 surgeries and 96 trips to the disabled list.

And since these are pitchers, we're naturally talking about some serious surgeries. John Smoltz and David Wells both had Tommy John surgery, and Roger Clemens, Pedro Martinez, Curt Schilling and David Cone are among those who had surgeries on their shoulders. Run through the list, and you'll also see some back surgeries and a couple knee surgeries.

Beyond the surgeries, you also won't be surprised to hear (or just be reminded) that many of these guys started piling up nagging injuries later in their careers. Martinez, Johnson, Tom Glavine, Kevin Brown and David Wells, in particular, were all dealing with something or another towards the end.

The phrase "par for the course" comes to mind. We were all taught back in Baseball 101 that pitchers have a tendency to break, and not even Greg Maddux was able to escape harm. He flirted with invincibility during his career, but a back issue in 2002 put a stain on an otherwise flawless medical track record.

If it can happen to him, it can happen to anyone. And if a certain star pitcher gets to old age and has a handful of scars to show for his time in the majors, there's nothing wrong with him thinking it might be time to call it a career.

If a pitcher's body fails to convince him, his performance should do the trick.

 

Missing Fewer Bats?

How can pitchers tell when their performance has taken a turn for the worse that can't be corrected?

They can start by looking at their strikeouts, or lack thereof. Not every pitcher relies heavily on the strikeout to perform well, but what's true for all of them is that the strikeout is a very useful thing. Even if a good pitch is made, things can go wrong when a ball is put in play. More strikeouts means fewer things that can go wrong.

To this end, almost all of our 15 pitchers were living more dangerously towards the end of their careers. Here's a look at how their strikeout percentages (strikeouts / batters faced) fluctuated from their prime years to their twilight years.

Player	Prime Years	Prime K%	Twilight Years	Twilight K%	Difference
Roger Clemens	1986-2005	23.2	2006-2007	19.5	-3.7
Greg Maddux	1988-2005	17.2	2006-2008	12.8	-4.4
Randy Johnson	1990-2005	30.2	2006-2009	22.0	-8.2
Pedro Martinez	1994-2005	28.6	2006-2009	21.5	-7.1
Mike Mussina	1992-2003	19.7	2004-2008	18.4	-1.3
Curt Schilling	1992-2004	24.5	2005-2007	19.7	-4.8
Tom Glavine	1989-2006	14.5	2007-2008	11.1	-3.4
Kevin Brown	1989-2000	17.4	2001-2005	19.2	+1.8
John Smoltz	1989-2007	21.7	2008-2009	23.6	+1.9
David Cone	1988-1999	22.8	2000-2003	17.3	-5.5
Chuck Finley	1989-1998	19.2	1999-2002	20.9	+1.7
Bret Saberhagen	1985-1994	17.3	1995-2001	14.9	-2.4
Kevin Appier	1990-2002	18.7	2003-2004	10.9	-7.8
David Wells	1990-2004	15.3	2005-2007	12.6	-2.7
Kenny Rogers	1993-2006	13.5	2007-2008	11.2	-2.3

Here we have decreases in strikeout percentages among all but three pitchers: Kevin Brown, John Smoltz and Chuck Finley.

I've got nothing for Smoltz and Finley. Smoltz didn't pitch all that much in 2008 and 2009, but when he did he compiled a strikeout rate (23.6) that was about on par with the strikeout rate he had both in his career to that point and as a starter between 2005 and 2007 (21.0).

Finley, meanwhile, was a better strikeout pitcher in the latter half of his career than he was in the first half of it. He wasn't as prolific a strikeout artist in his time with the Cleveland Indians (19.9 K%) in his final season as he was with the St. Louis Cardinals (23.6 K%) following a July trade, but he was still better than he had been earlier in his career.

Brown, however, is a different story. 

I cut off Brown's prime at 2000 because he only had one 30-start season after that, but the line could have been drawn after the 2003 season instead. His performance suffered greatly as a member of the New York Yankees in 2004 and 2005, and that was thanks in part to a lesser strikeout percentage. He struck out only 14.8 percent of the batters he faced in those two seasons.

So switch Brown's prime cutoff to 2003, and the above table shows that the bulk of the top-15 starting pitchers since 1985 have had to deal with declining strikeout percentages towards the end of their careers.

The trend holds true with other pitchers as well. For example, take five guys who are stationed just below our 15 on the original big board.

Player	Prime Years	Prime K%	Twilight Years	Twilight K%	Difference
Jimmy Key	1985-1994	14.0	1995-1998	15.6	+1.6
Orel Hershiser	1984-1995	16.2	1996-2000	13.0	-3.2
Mark Langston	1984-1993	20.4	1994-1999	16.7	-3.7
Brad Radke	1997-2004	15.0	2005-2006	13.2	-2.8
Javier Vazquez	2000-2009	22.2	2010-2011	19.1	-3.1

Like Smoltz and Finley, Jimmy Key was actually a better strikeout pitcher towards the end of his career than he had been earlier in his career. But the other four pitchers here couldn't join the club, making them not unlike the bulk of the pitchers in the first table.

There are myriad explanations for the declining strikeout rates displayed here, but what matters is the big picture. Eventually, guys who have made a living fooling hitters aren't going to be able to fool hitters anymore. When that day comes, it's time to consider calling it a career.

But it's not all about the balls that are missing the bats. The balls that are hitting the bats will send a message too.

 

Taking a Beating?

A pitcher isn't going to last long in the major leagues if he's not good at limiting the damage, and the best way to limit the damage is to limit hard-hit balls. It's a lot easier to overcome a single than it is to overcome a double, triple or home run. Again, Baseball 101.

Our 15 pitchers went from having no trouble at all with having hard-hit balls in their primes to having all sorts of trouble with hard-hit balls in their twilight years. Take a look at the fluctuations in their opponent slugging percentages.

Player	Prime Opp. Slugging	Twilight Opp. Slugging	Difference
Roger Clemens	.340	.357	+17
Greg Maddux	.345	.420	+75
Randy Johnson	.340	.428	+88
Pedro Martinez	.326	.434	+108
Mike Mussina	.392	.423	+31
Curt Schilling	.374	.464	+90
Tom Glavine	.372	.454	+82
Kevin Brown	.342	.371	+29
John Smoltz	.355	.460	+105
David Cone	.347	.464	+117
Chuck Finley	.382	.398	+16
Bret Saberhagen	.370	.436	+66
Kevin Appier	.371	.480	+109
David Wells	.423	.472	+49
Kenny Rogers	.415	.480	+65

Whereas there were some exceptions to the rule in the strikeouts discussion, there are none here. Each of our 15 had issues with hard-hit balls towards the end of his career.

Even the smallest increases are not what they seem. Roger Clemens may have had a .357 opponent slugging percentage in his last two seasons, but it was significantly worse than that in his final season in 2007. After posting a .323 opponent slugging percentage in 2006, he posted a .393 opponent slugging percentage in '07.

As for Chuck Finley, he posted a solid .372 opponent slugging percentage in his final year in 2002, but that was thanks to his move to the National League more than his pitching. He had a .391 opponent slugging percentage in 18 starts with Cleveland before posting a .349 opponent slugging percentage in 14 starts with St. Louis.

Like with the strikeout trend, the hard-contact trend holds true with the other five pitchers we looked at.

Player	Prime Opp. Slugging	Twilight Opp. Slugging	Difference
Jimmy Key	.385	.416	+31
Orel Hershiser	.342	.425	+83
Mark Langston	.371	.447	+76
Brad Radke	.433	.471	+38
Javier Vazquez	.413	.437	+24

Same thing. These guys went from doing a decent job at limiting hard hits to doing a not-so-decent job at limiting hard hits.

The only guy who has a strong case to be considered an exception is Javier Vazquez, who posted a .398 opponent slugging percentage in his final season in 2011 after getting crushed as a member of the Yankees in 2010 (it seems likely in retrospect that he was allergic to pinstripes).

Like with the fewer strikeouts trend, the more hard-hit balls trend falls under the "not fooling anyone anymore." The best pitchers know how to pitch contact, but it's hard for even the best pitchers to survive for very long when the contact is starting to get loud. When it happens, it's time to give retirement a look.

It's not just the little things that should bring retirement into focus. Pitchers should have to look at the big picture and be able to see when it's starting to fade.

 

No Longer Dependable?

Everyone wants a superstar starting pitcher for one reason: they're dependable. They're the rare guys who can go out there every five days and deliver a win more or less automatically.

But naturally, there comes a time when even getting out there every five days is hard enough. Due to dastardly things like old age and injuries, the majority of our 15 guys had a hard time making 30 starts a season as they got closer and closer to the ends of their careers.

As a result of him sitting out the first half (or so) of the season both years, Roger Clemens didn't make 20 starts in either of his final two seasons. Randy Johnson started 30 games just once in his last three seasons. Pedro Martinez threw his last pitch in 2009, but his last 30-start season had come in 2005. Curt Schilling started 30 games once in his last three seasons. And so on.

Reliability isn't just about workload, of course. It also has to do with how guys are performing in the starts they are able to make. To this end, all 15 of our guys had their share of trouble.

That's a reality that Game Score can help reveal. If you haven't heard of it, it's a stat created by Bill James to quantify how good a pitcher's start was by starting at 50 and then adding and subtracting points based on innings, strikeouts, hits allowed, etc.

Here's a look at how our 15 guys' Average Game Scores fluctuated from their prime seasons to their twilight seasons.

Player	Prime Avg. Game Score	Twilight Avg. Game Score	Difference
Roger Clemens	60	55	-5
Greg Maddux	58	49	-9
Randy Johnson	62	52	-10
Pedro Martinez	63	51	-12
Mike Mussina	56	51	-5
Curt Schilling	60	51	-9
Tom Glavine	54	47	-7
Kevin Brown	56	54	-2
John Smoltz	57	49	-8
David Cone	59	44	-15
Chuck Finley	54	51	-3
Bret Saberhagen	58	51	-7
Kevin Appier	55	45	-10
David Wells	52	46	-6
Kenny Rogers	49	43	-6

Once again, here we have drops across the board. And once again, even the smallest numbers are not what they seem.

Kevin Brown had an Average Game Score of 54 between 2001 and 2005, but it was a mere 47 in his last two seasons in 2004 and 2005. Not a shocker seeing as how he averaged slightly less than six innings per start and had issues with strikeouts and hard-hit balls.

Chuck Finley had an Average Game Score of 55.9 as a member of the Cardinals in 2002, but his average game score had been 49.2 in his 18 starts with the Indians. In those, he also averaged slightly less than six innings per start and had issues with hard-hit balls.

Sure enough, the trend holds true for our other five pitchers.

Player	Prime Avg. Game Score	Twilight Avg. Game Score	Difference
Jimmy Key	54	50	-4
Orel Hershiser	56	48	-8
Mark Langston	56	47	-9
Brad Radke	51	49	-2
Javier Vazquez	54	50	-4

Jimmy Key came reasonably close to matching his prime Average Game Score, but it's worth noting that he failed to top an Average Game Score of 50 in three of his last four seasons. Brad Radke also came pretty close to matching his prime Average Game Score, but his Average Game Scores tumbled in his last three seasons like so: 53.9, 51.4, 47.1.

It's Javier Vazquez who once again stands out as a slight exception to the rule, as his Average Game Score in 2011 was a solid 53.9. But even that was short of where he was when he was at his absolute best between 2001 and 2009, when his average game score was 55.

Pitchers are going to have plenty of other ways to know when they just can't hack it anymore in old age, but this is a fairly definitive overall measuring stick for them to use. They may be able to still get out there and gut it out on at least a semi-regular basis, but this is a stat that can show them when they're just not capable of eating innings and carving up lineups like they could in the good old days.

When it happens, they can either keep soldiering on as lesser pitchers, or they can realize that their best days are well in the past.

 

What About Superstar Closers?

Up until now, we've been focusing strictly on superstar starting pitchers. They're the guys who make the big bucks and get all the accolades, after all.

But yes, relievers are people too, and it's certainly possible for them to achieve superstar status. Not as easy, mind you, but possible. All they have to do is work their way into ninth-inning duty and then do enough to get branded as a "Proven Closer."

The big question: When should proven closers know when to call it quits?

Like with starting pitchers, old age and wear and tear are two things that should push them towards retirement. Beyond that, it's not when their save totals start to decline, as the save really isn't the best measure of the quality of a closer. Not as long as a save that features a couple hits and an earned run counts the same as a three-up, three-down save, anyway.

Win Probability Added is a better measure of a closer's dependability. Baseball-Reference.com defines it as being the "sum of the differences in win expectancies for each play the player is credited with," and it's particularly partial to late-inning relievers due to the high-stress innings they have to work. 

A quick search of post-1985 players on Baseball-Reference.com yields six retired closers who managed to save at least 350 games. As their prime and twilight average WPAs can show, none of them was as reliable towards the end of his career as he used to be.

Player	Final Year Age	Surgeries	DL Trips	Prime Years	Prime Avg. WPA	Twilight Years	Twilight Avg. WPA
Trevor Hoffman	42	3	2	1994-2007	2.4	2008-2010	-0.3
Billy Wagner	38	2	9	1997-2007	2.4	2008-2010	0.5
John Franco	44	2	7	1986-1998	1.1	1999-2005	0.1
Lee Smith	39	0	2	1983-1993	1.7	1994-1997	0
Dennis Eckersley	43	0	2	1988-1992	3.6	1993-1998	-0.6
Troy Percival	39	3	13	1996-2004	2.2	2005-2009	0.1

As you can see, these six guys all got to be pretty old, and they also suffered their share of wear and tear. You can also see that they weren't so reliable towards the end.

Trevor Hoffman came pretty close to matching his prime WPA in 2009 with a WPA of 1.9, but his WPA was under zero in 2008 and 2010. Billy Wagner had a very strong final season with the Atlanta Braves with 37 saves in 2010, but his WPA was only 1.0. Dennis Eckersley managed a WPA over zero just once in his final six seasons.

Then you have guys like John Franco and Lee Smith, both of whom were finished as primary closers by the time they got to the end of their careers. Troy Percival enjoyed a return-to-form season in 2008 with the Tampa Bay Rays, but the last year his WPA topped even the 1.0 mark was in 2004.

It's obviously a lot more important for teams to have dependable starting pitchers than it is for them to have dependable closers, but I'd say dependability is a more important selling point for closers because, well, it's really their only selling point.

Once they no longer have it, it they need to consider whether they no longer have it in general.

 

Final Thoughts

That's all there is to it. Top-shelf starting pitchers need to seriously consider retirement when they're old and beat up, and when they're missing fewer bats and surrendering more hard-hit balls. 

In addition, both top-shelf starters and top-shelf closers need to know when they're not as dependable as they used to be. When in doubt, they can always look at the numbers.

Not exactly complicated stuff, to be sure, but I warned that was going to be the case. And I'll make the same parting point here that I made in my checklist for hitters: Simplicity is necessary in this case because baseball is a simple game at heart. For pitchers, it doesn't get much simpler than throw ball, fool hitters, repeat, repeat, repeat.

When they can no longer satisfy these simplest of requirements, they've had enough.

 

Note: All stats courtesy of Baseball-Reference.com.

 

If you want to talk baseball, hit me up on Twitter. 

Read more MLB news on BleacherReport.com