Starting Pitchers Success Tied To K Rate?

PhillieFever · 11-01-2009, 02:17 PM

I've just finished reading a very interesting article by Bill James in his Historical Baseball Abstract that for all intents and purposes says this,"If a starting pitcher's strikeout rate is less than 4.5 per 9 innings you can pretty much write him off as somebody who's going to have real career"
He goes on to say that since WW II many young pitchers have come up and were effective with few strikeouts:
1977-Dave Rozema 15-7 218 IP 92 K's
1978-Larry Sorenson 18-12 281 IP 78 K's
1988-Allen Anderson 16-9 202 IP 83 K's (also AL ERA Leader)
1989-Jeff Ballard 18-8 215 IP 62 K's
1985-Andy Hawkins 18-8 229 IP 69 K's

All of these pitchers and all others like them disappeared quickly after one or two good seasons.He hammers his point home by giving two ways to study the issue,here's one:
Pick a year of birth,say all pitchers born in 1960.Add 26 years to that year(1986).Look up the records for those pitchers through that season.Then split them into two groups,those who've averaged 4.6 K's per 9 and those who haven't.Then look up their won loss records for the rest of their careers.In the "before" group some of the best pitchers will be pitchers with low strikeout rates.But in the "After" group,,almost every time,all of the best pitchers will be in the high strikeout group.

Here are my questions.First,do you think James has a point? and secondly,does OOTP model this behavior with any accuracy?

Spritze · 11-01-2009, 02:24 PM

Quote:

Originally Posted by PhillieFever

Here are my questions.First,do you think James has a point?

In the modern game true enough, it is sorta known as true in the same way as with hitting where a young player with old player skills will have a short career.

RonCo · 11-01-2009, 03:19 PM

A pitcher's K-rate is a major parameter in his assessment.

To some degree it's probably fair to generalize and say a young pitcher's BB-rate gets them into the league, and his K-rate keeps him there.

Doctor No · 11-01-2009, 06:47 PM

If you assume that a pitcher has little ability to influence BABIP, then the measures of future performance should hinge upon things that they can control: strikeouts, walks, and home runs.

StyxNCa · 11-03-2009, 06:04 PM

Just off the top of my head it seems Tommy John played a pretty good while while not reaching the magic 4.5/9 for his career. I came up with him right off the bat so I know that there has to be quite a few more that are out there.

The problem is that James assumes no one will do any research on what he says since he feels he is the god of stats as do most of his followers.
I know I'm not going to bother since I don't pay much attention to James and his ilk.

RchW · 11-03-2009, 09:48 PM

Quote:

Originally Posted by StyxNCa

Just off the top of my head it seems Tommy John played a pretty good while while not reaching the magic 4.5/9 for his career. I came up with him right off the bat so I know that there has to be quite a few more that are out there.

The problem is that James assumes no one will do any research on what he says since he feels he is the god of stats as do most of his followers.
I know I'm not going to bother since I don't pay much attention to James and his ilk.

Warren Spahn

Jim Perry

Mel Stottlemyre

Claude Osteen

Mike Torrez

Doyle Alexander

Bob Friend

Joe Niekro

RonCo · 11-03-2009, 11:25 PM

Tommy John was over 4.5 K/9 for the first 14 seasons of his career.

Warren Spahn was over 4.5 K/9 eight seasons, including many of those early in his career.

Jim Perry was over 4.5 K/9 10 seasons, including some over 6.

Mel Stottlemyre was over 4.5 K/9 each of his first five seasons, and didn't drop to 3+ consistently until the last couple seasons.

Cluade Osteen was over 4.5 K/9 for eight seasons early in his career, including several over 5.

Mike Torrez and Doyle Alexander are the best exceptions--Torrez was over 4.5 four seasons early in his career before fading towards 4.0 or so. Alexander is truely an outlier to the rule.

Bob Friend appears at first to be an exception, too, though he was over 4.5 K/9 six times in the meat of his career, and at 4.4 a few others.

Joe Niekro was over 4.5 K/9 12 seasons, and up in the 4s a few other times.

RchW · 11-04-2009, 12:27 AM

Quote:

Originally Posted by RonCo

Tommy John was over 4.5 K/9 for the first 14 seasons of his career.

Warren Spahn was over 4.5 K/9 eight seasons, including many of those early in his career.

Jim Perry was over 4.5 K/9 10 seasons, including some over 6.

Mel Stottlemyre was over 4.5 K/9 each of his first five seasons, and didn't drop to 3+ consistently until the last couple seasons.

Cluade Osteen was over 4.5 K/9 for eight seasons early in his career, including several over 5.

Mike Torrez and Doyle Alexander are the best exceptions--Torrez was over 4.5 four seasons early in his career before fading towards 4.0 or so. Alexander is truely an outlier to the rule.

Bob Friend appears at first to be an exception, too, though he was over 4.5 K/9 six times in the meat of his career, and at 4.4 a few others.

Joe Niekro was over 4.5 K/9 12 seasons, and up in the 4s a few other times.

No doubt, but James mentioned careers not seasons.

His Own Bad Self · 11-04-2009, 01:46 AM

Quote:

Originally Posted by StyxNCa

Just off the top of my head it seems Tommy John played a pretty good while while not reaching the magic 4.5/9 for his career. I came up with him right off the bat so I know that there has to be quite a few more that are out there.

The problem is that James assumes no one will do any research on what he says since he feels he is the god of stats as do most of his followers.
I know I'm not going to bother since I don't pay much attention to James and his ilk.

I don't understand this. James specifically brought up TJ in the book, pointing out that he was comfortably near average for most of his career.

And your opinion of him (and his ilk) is, as far as I can tell, completely unfounded and without merit. But suit yourself.

His Own Bad Self · 11-04-2009, 02:07 AM

The average K rate per 27 outs over Spahn's career was roughly 4.5. You can be pretty sure that when James used 4.5K/9 as the lower limits to the rule, he wasn't talking about the '50s.

Apply the 4.5K/9 rule to a modern context (the 2000s), and that's about 70% of the league average (assuming a 6.5K/9 average). 70% of a 5.5 average gives a "replacement level" of 3.8. Apply it to a 4.5 average; the RL is 3.1.

This really shouldn't be that difficult to grasp.

His Own Bad Self · 11-04-2009, 02:11 AM

Quote:

Originally Posted by PhillieFever

...does OOTP model this behavior with any accuracy?

It's supposed to. I'm not certain that it does, but someone could probably study it if they wanted to. Which I do, now...

PhillieFever · 11-04-2009, 06:58 AM

Quote:

Originally Posted by His Own Bad Self

It's supposed to. I'm not certain that it does, but someone could probably study it if they wanted to. Which I do, now...

I had all intentions of doing it the day I posted this,but unfortunately work got busy so I had to hold off for a couple of days,I should have it done by the weekend.

PhillieFever · 11-04-2009, 07:01 AM

Quote:

Originally Posted by RonCo

Tommy John was over 4.5 K/9 for the first 14 seasons of his career.

Warren Spahn was over 4.5 K/9 eight seasons, including many of those early in his career.

Jim Perry was over 4.5 K/9 10 seasons, including some over 6.

Mel Stottlemyre was over 4.5 K/9 each of his first five seasons, and didn't drop to 3+ consistently until the last couple seasons.

Cluade Osteen was over 4.5 K/9 for eight seasons early in his career, including several over 5.

Mike Torrez and Doyle Alexander are the best exceptions--Torrez was over 4.5 four seasons early in his career before fading towards 4.0 or so. Alexander is truely an outlier to the rule.

Bob Friend appears at first to be an exception, too, though he was over 4.5 K/9 six times in the meat of his career, and at 4.4 a few others.

Joe Niekro was over 4.5 K/9 12 seasons, and up in the 4s a few other times.

He did mention a couple of exceptions,Lew Burdette being one of them,but like most theories there will always be a few exceptions.As a rule though,I feel that James has hit it right on the head.

RonCo · 11-04-2009, 07:37 AM

Quote:

Originally Posted by RchW

No doubt, but James mentioned careers not seasons.

I'll need to look it up again, but my understanding is that the method looks at K-rates at age 26, checks if they are > 4.5, and predicts based solely on that one number how successful the pitcher's career will be. If I'm right, there, then a majority of these named pitchers would be considered in the "> 4.5" group.

I'll have to look it up, though, to be sure.

As a rule of thumb, it's a pretty good predictor. All rules of thumb miss a few times, though.

RonCo · 11-04-2009, 08:23 AM

Here's the actual James text in question for those who care to read it:

The New Bill James Historical ... - Google Books

professor ape · 11-04-2009, 08:32 AM

This whole conversation points to the validity of the 'rule of thumb'. Bill never stated it as an absolute and is quick to point out exceptions. The replies so far have been pointing out a handfull of notable exceptions but there isn't a huge list of pitchers with low K-rates through age 26 who go on to have long-term strong careers.

Yes, I am someone of Bill's ilk but I think that many of those who look down on his ideas are missing the point. He is always seeking better ways to analyze baseball performance. He knows that there is no such thing as a perfect statistic but it is about learning and sparking questions & conversation. He has accomplished his goal again here.

In the case of the K/9 ovservations, the evidence cleary supports his theory. Even Tommy John had a K/9 rate of about 5.5 through age 26 and he was on the low end.

In short, I am not counting on John Lannan to have a successful long-term career. He could end up as one of the exceptions but, at age 24, is on the cusp for his career and had a woeful 3.9 K/9 rate in 2009.

RonCo · 11-04-2009, 08:37 AM

Noticing a small nuance here. James is using league average K-rate as his basis for most of this. The 4.5, 4.6, 4.7 K/9 number (or whatever) is probably too proscriptive. I wouldn't get too caught up in an exact number. The learning is about what kinds of pitchers project to have longer careers or shorter careers, not specifically where you draw the line.

Malleus Dei · 11-04-2009, 10:45 AM

Quote:

Originally Posted by StyxNCa

The problem is that James assumes no one will do any research on what he says since he feels he is the god of stats as do most of his followers.

That's nonsense. Bill James is all about finding and knowing the truth and always encourages further research. In fact, if you do research that proves the opposite contention to something he has proposed you're not opposing Bill James, you're *being* Bill James.

Quote:

Originally Posted by StyxNCa

I know I'm not going to bother since I don't pay much attention to James and his ilk.

And that is a huge mistake. Ignorance may be bliss, but it will lose you a lot of baseball games. As a Texan, you should know what is carved in large letters on the side of the Main Building of the University of Texas: "Ye shall know the Truth and the Truth shall make you free". That's what Bill James is all about.

Malleus Dei · 11-04-2009, 10:48 AM

Quote:

Originally Posted by professor ape

This whole conversation points to the validity of the 'rule of thumb'. Bill never stated it as an absolute and is quick to point out exceptions. The replies so far have been pointing out a handfull of notable exceptions but there isn't a huge list of pitchers with low K-rates through age 26 who go on to have long-term strong careers.

And that's the point: pitchers without a significant K-rate by age 26 tend to have much shorter careers that those who do. It's an observable, verifiable truth. End of story.

RonCo · 11-04-2009, 10:57 AM

I suggest the reason for this is that, on average, a pitcher's ability to strike out hitters tends to peak at something around 24 (flying blind here, so I may be off a year or so too old), and fall off very gently for many years thereafter. Given that, it's reasonable to think that if he's not striking out a respectable number of hitters by age 26, it's quite unlikley that pitcher will be able to stick around for another decade.

11-01-2009, 02:17 PM	#1 (permalink)
PhillieFever Hall Of Famer Join Date: Apr 2008 Location: South Philly Posts: 5,848 Thanked 289x in 252 posts	Starting Pitchers Success Tied To K Rate? I've just finished reading a very interesting article by Bill James in his Historical Baseball Abstract that for all intents and purposes says this,"If a starting pitcher's strikeout rate is less than 4.5 per 9 innings you can pretty much write him off as somebody who's going to have real career" He goes on to say that since WW II many young pitchers have come up and were effective with few strikeouts: 1977-Dave Rozema 15-7 218 IP 92 K's 1978-Larry Sorenson 18-12 281 IP 78 K's 1988-Allen Anderson 16-9 202 IP 83 K's (also AL ERA Leader) 1989-Jeff Ballard 18-8 215 IP 62 K's 1985-Andy Hawkins 18-8 229 IP 69 K's All of these pitchers and all others like them disappeared quickly after one or two good seasons.He hammers his point home by giving two ways to study the issue,here's one: Pick a year of birth,say all pitchers born in 1960.Add 26 years to that year(1986).Look up the records for those pitchers through that season.Then split them into two groups,those who've averaged 4.6 K's per 9 and those who haven't.Then look up their won loss records for the rest of their careers.In the "before" group some of the best pitchers will be pitchers with low strikeout rates.But in the "After" group,,almost every time,all of the best pitchers will be in the high strikeout group. Here are my questions.First,do you think James has a point? and secondly,does OOTP model this behavior with any accuracy?

11-03-2009, 06:04 PM	#5 (permalink)
StyxNCa Hall Of Famer Join Date: Dec 2004 Location: Victoria, Texas Posts: 2,997 Thanked 177x in 105 posts	Just off the top of my head it seems Tommy John played a pretty good while while not reaching the magic 4.5/9 for his career. I came up with him right off the bat so I know that there has to be quite a few more that are out there. The problem is that James assumes no one will do any research on what he says since he feels he is the god of stats as do most of his followers. I know I'm not going to bother since I don't pay much attention to James and his ilk. Last edited by StyxNCa; 11-03-2009 at 06:07 PM.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

11-01-2009, 03:19 PM	#3 (permalink)
RonCo Hall Of Famer Join Date: Aug 2003 Posts: 6,566 Thanked 147x in 66 posts	A pitcher's K-rate is a major parameter in his assessment. To some degree it's probably fair to generalize and say a young pitcher's BB-rate gets them into the league, and his K-rate keeps him there.

11-01-2009, 06:47 PM	#4 (permalink)
Doctor No Minors (Single A) Join Date: Jul 2007 Location: Denver, Colorado Posts: 76 Thanked 3x in 3 posts	If you assume that a pitcher has little ability to influence BABIP, then the measures of future performance should hinge upon things that they can control: strikeouts, walks, and home runs.

11-03-2009, 11:25 PM	#7 (permalink)
RonCo Hall Of Famer Join Date: Aug 2003 Posts: 6,566 Thanked 147x in 66 posts	Tommy John was over 4.5 K/9 for the first 14 seasons of his career. Warren Spahn was over 4.5 K/9 eight seasons, including many of those early in his career. Jim Perry was over 4.5 K/9 10 seasons, including some over 6. Mel Stottlemyre was over 4.5 K/9 each of his first five seasons, and didn't drop to 3+ consistently until the last couple seasons. Cluade Osteen was over 4.5 K/9 for eight seasons early in his career, including several over 5. Mike Torrez and Doyle Alexander are the best exceptions--Torrez was over 4.5 four seasons early in his career before fading towards 4.0 or so. Alexander is truely an outlier to the rule. Bob Friend appears at first to be an exception, too, though he was over 4.5 K/9 six times in the meat of his career, and at 4.4 a few others. Joe Niekro was over 4.5 K/9 12 seasons, and up in the 4s a few other times.

11-04-2009, 02:07 AM	#10 (permalink)
His Own Bad Self All Star Starter Join Date: Jan 2006 Location: Phoenix, AZ Posts: 1,209 Blog Entries: 12 Thanked 14x in 9 posts	The average K rate per 27 outs over Spahn's career was roughly 4.5. You can be pretty sure that when James used 4.5K/9 as the lower limits to the rule, he wasn't talking about the '50s. Apply the 4.5K/9 rule to a modern context (the 2000s), and that's about 70% of the league average (assuming a 6.5K/9 average). 70% of a 5.5 average gives a "replacement level" of 3.8. Apply it to a 4.5 average; the RL is 3.1. This really shouldn't be that difficult to grasp.

11-04-2009, 08:23 AM	#15 (permalink)
RonCo Hall Of Famer Join Date: Aug 2003 Posts: 6,566 Thanked 147x in 66 posts	Here's the actual James text in question for those who care to read it: The New Bill James Historical ... - Google Books

11-04-2009, 08:32 AM	#16 (permalink)
professor ape Major Leagues Join Date: Jan 2005 Posts: 435 Thanked 89x in 31 posts	This whole conversation points to the validity of the 'rule of thumb'. Bill never stated it as an absolute and is quick to point out exceptions. The replies so far have been pointing out a handfull of notable exceptions but there isn't a huge list of pitchers with low K-rates through age 26 who go on to have long-term strong careers. Yes, I am someone of Bill's ilk but I think that many of those who look down on his ideas are missing the point. He is always seeking better ways to analyze baseball performance. He knows that there is no such thing as a perfect statistic but it is about learning and sparking questions & conversation. He has accomplished his goal again here. In the case of the K/9 ovservations, the evidence cleary supports his theory. Even Tommy John had a K/9 rate of about 5.5 through age 26 and he was on the low end. In short, I am not counting on John Lannan to have a successful long-term career. He could end up as one of the exceptions but, at age 24, is on the cusp for his career and had a woeful 3.9 K/9 rate in 2009.

11-04-2009, 08:37 AM	#17 (permalink)
RonCo Hall Of Famer Join Date: Aug 2003 Posts: 6,566 Thanked 147x in 66 posts	Noticing a small nuance here. James is using league average K-rate as his basis for most of this. The 4.5, 4.6, 4.7 K/9 number (or whatever) is probably too proscriptive. I wouldn't get too caught up in an exact number. The learning is about what kinds of pitchers project to have longer careers or shorter careers, not specifically where you draw the line.

11-04-2009, 10:57 AM	#20 (permalink)
RonCo Hall Of Famer Join Date: Aug 2003 Posts: 6,566 Thanked 147x in 66 posts	I suggest the reason for this is that, on average, a pitcher's ability to strike out hitters tends to peak at something around 24 (flying blind here, so I may be off a year or so too old), and fall off very gently for many years thereafter. Given that, it's reasonable to think that if he's not striking out a respectable number of hitters by age 26, it's quite unlikley that pitcher will be able to stick around for another decade.