Starting Pitchers-Best Stat

[interloper]

2 hours ago, DrungoHazewood said:

Strikeout rate compared to league average.  K/BB ratio would be good, too.

Speaking broadly, a pitcher's career is a long (or sometimes short) battle against declining strikeout rates.  As long as you're well above average you're good.  Once you slip below a certain point you're living on borrowed time.  That point used to be around 4.5 when I was a kid, in the 1980s.  Today it's probably more like six.

Why not a stat that tells you more? Erik Bedard was still rocking a 8+ SO/9 in his bad years and only dipped to a 7.6 in his final year, which is still pretty high. You miss so much information relying on strictly strikeouts.

[Beef Supreme]

2 hours ago, Luke-OH said:

Over a huge multiple year sample size ERA- does a pretty good job. It adjusts for the park and standardizes ERA to league average. Anything two years or less, you probably want to strip out defense. So you’d go with FIP- or xFIP- depending on how much control you think a pitcher has over the quality of their flyball contact.

The fewer innings of data, the less predictive FIP is of future ERA. Only two years of FIP would not be reliable.

[Beef Supreme]

1 minute ago, interloper said:

Why not a stat that tells you more? Erik Bedard was still rocking a 8+ SO/9 in his bad years and only dipped to a 7.6 in his final year, which is still pretty high. You miss so much information relying on strictly strikeouts.

Wholly agree.

Cashner, for example, had worse results for the two seasons his K/9 approached league average. His best three years he had below average K/9. And those three years were successful well above league average.

[Frobby]

56 minutes ago, OrioleDog said:

I've been won over to strikeouts minus walks (K%-BB%) as an upgrade on K/BB.  Straight K/BB is too nice to the Josh Tomlins of the world.

Not bad for measuring current pitchers, but not so good for comparing current and past pitchers since K rates have increased steadily over time.   FWIW, of the seven all-time K/BB leaders, five are currently active, including Chris Sale at no. 1.

Here’s an unofficial list of the career K - BB leaders:

R. Johnson 3378

Clemens 3092

Ryan 2919

Schilling 2405

Martinez 2394

Seaver 2379

Maddux 2372

Carlton 2303

Sutton 2231

Jenkins 2195

There could be some low BB guys ahead of some guys towards the bottom of this list.    The only other pitchers I looked at were W. Johnson, Cy Young, Mussina, G. Perry, P. Niekro and Gibson.   So, I looked at everone in the top 15 in career strikeouts plus a couple of others.

[Redskins Rick]

2 hours ago, Frobby said:

I’m not denying that strikeouts are important.    I’m just saying focusing solely on strikeouts misses too much other information.    I’d rather use a stat that includes strikeouts but also other factors.    

Davy used to preach to Mussina that while Ks was good, outs and reduced pitch count was even better.

[Beef Supreme]

There are stats -- strikeouts, walks, etc. -- there are ratios -- ERA, WHIP, HR/9 -- and there are formulas -- WAR, FIP.

Best Stat: Innings Pitched
Best Ratio: WHIP
Best Formula: ERA-

Any conversation should consider HR/9.

[atomic]

2 hours ago, DrungoHazewood said:

Strikeout rate compared to league average.  K/BB ratio would be good, too.

Speaking broadly, a pitcher's career is a long (or sometimes short) battle against declining strikeout rates.  As long as you're well above average you're good.  Once you slip below a certain point you're living on borrowed time.  That point used to be around 4.5 when I was a kid, in the 1980s.  Today it's probably more like six.

Strikeout rate seems pretty worthless to me.  Daniel Cabera was striking out guys pretty regularly in his prime. Not a guy I would want on my team. 

[Just Regular]

39 minutes ago, Frobby said:

Not bad for measuring current pitchers, but not so good for comparing current and past pitchers since K rates have increased steadily over time.   FWIW, of the seven all-time K/BB leaders, five are currently active, including Chris Sale at no. 1.

Here’s an unofficial list of the career K - BB leaders:

R. Johnson 3378

Clemens 3092

Ryan 2919

Schilling 2405

Martinez 2394

Seaver 2379

Maddux 2372

Carlton 2303

Sutton 2231

Jenkins 2195

There could be some low BB guys ahead of some guys towards the bottom of this list.    The only other pitchers I looked at were W. Johnson, Cy Young, Mussina, G. Perry, P. Niekro and Gibson.   So, I looked at everone in the top 15 in career strikeouts plus a couple of others.

https://www.fangraphs.com/leaders.aspx?pos=all&stats=pit&lg=all&qual=1000&type=1&season=2017&month=0&season1=1947&ind=0&team=0&rost=0&age=0&filter=&players=0&sort=9,d

Here are the modern era leaders among pitchers with 1000 IP.  It can be a counting or rate stat, if you convert K/PA and BB/PA into a batting average type stat.

Raw K% tracks respectably with traditional batting average at the top of the scale - Chris Sale's 36.2 K% last year (or .362 strikeout average, if you will) just outdid Jose Altuve's .346 last year, but there are a lot more .300 hitters than 30.0 K% pitchers among qualifiers.  Relievers are on a whole other scale.

[Beef Supreme]

Quote

One single stat only to judge starting pitchers...

As for pure stats, Innings Pitched has a tremendously high correlation to success. And I know I am in for it worse than the attendant who cleans the monkey house at the zoo, but the other stat which has a tremendously high correlation to success for starting pitchers is...
Wins.

Yeah, everybody hates them. They scream bloody murder should Wins even be mentioned in a post. It is blasphemy! Stone that poster who dare speaketh of them. "Go back to the 50's and read your baseball cards, you effin idiot!" And so on...

But if you look at the top career Win leaders in the modern era, they are a great reflection of the best starting pitchers of all time.

N.B.: I realize that things have changed and continue to change regarding 5-man and even 6-man rotations as well as pitch counts, etc. that result in lower win totals for starters, making the validity of the stat as a reflection of success less than it had been in previous decades.

[Beef Supreme]

Some may like Quality Starts, but they have been impacted by the changes than Wins have. Maybe Quality Starts versus league average would be a good barometer. But that would be a formula rather than a stat.

[DrungoHazewood]

2 hours ago, interloper said:

Why not a stat that tells you more? Erik Bedard was still rocking a 8+ SO/9 in his bad years and only dipped to a 7.6 in his final year, which is still pretty high. You miss so much information relying on strictly strikeouts.

Of course you look for other things, but when I hear about a team acquiring a pitcher the very first thing I look at is where their strikeout rate is going.  When I look at a roster of minor league pitchers I pretty much ignore everyone who isn't striking out 7 or 8 per nine.  It's kind of the foundation. 

[Can_of_corn]

1 hour ago, cimota said:

Strikeout rate seems pretty worthless to me.  Daniel Cabera was striking out guys pretty regularly in his prime. Not a guy I would want on my team. 

Daniel Cabera had a prime?

Where was I?

[atomic]

13 minutes ago, Can_of_corn said:

Daniel Cabera had a prime?

Where was I?

2005 to 2008.  He was a league leader in several categories in his prime seasons:

2006 led league in BB and Wild Pitches

2007 led league in Loses, Earned Runs and BB

2008 led league in hit by pitches and wild pitches

 

 

 

[interloper]

1 hour ago, DrungoHazewood said:

Of course you look for other things, but when I hear about a team acquiring a pitcher the very first thing I look at is where their strikeout rate is going.  When I look at a roster of minor league pitchers I pretty much ignore everyone who isn't striking out 7 or 8 per nine.  It's kind of the foundation. 

Gotcha. Yeah, that's definitely fair.

[Beef Supreme]

1 hour ago, DrungoHazewood said:

Of course you look for other things, but when I hear about a team acquiring a pitcher the very first thing I look at is where their strikeout rate is going.  When I look at a roster of minor league pitchers I pretty much ignore everyone who isn't striking out 7 or 8 per nine.  It's kind of the foundation.

Cashner is an interesting example to consider. In 2013, Cashner's first (almost) full year as a starter, his K/9 dropped -- as would be expected, compared to his rate as a reliever -- to 6.6 but he posted an ERA of 3.09 and an ERA+ of 111. The next year his K/9 was 6.8 with an ERA of 2.55 and an ERA+ of 132. Cashner pumped up the K rate in 2015, his third season as a starter, to a respectable 8.0 -- good for 26th amongst all ML starters. But he saw his ERA rise to 4.34 and his ERA+ drop to 86. Similarly, his 2016 numbers were 7.6 K/9, 5.25 ERA and a 76 ERA+.

His K/9 plummeted in 2017 to 4.6, ranking 44th of 45 qualifying starters. Yet he had the best ERA+ in his career as a starter at 138 with a 3.40 ERA.

I am as dubious of Cashner's future success as many on OH are. But maybe there are some pitchers whose results get better despite their K/9 going down. Here's hoping Cashner is one of those, even though 4.6 seems going too far down to me. Greater success as a starter at the same time as a falling strikeout rate may seem counter-intuitive, but it may be more common than this one example.