Analyzing Starting Pitching In The Last Decade

[Greg Ambrosius]

I thoroughly enjoyed my time last week at the Arizona Fall League Symposium and I thought the conference was one of the better ones in the 19 years of its existence. Kudos to Ron Shandler, Ray Murphy, Brett Hershey and the entire USA Today crew for putting on another good show.

There were several good seminars at the show and I'll mention a few on the boards this week. One hot topic of conversation there and on our Message Boards involves the change in the game where pitching is now dominating hitting. We know all about that and NFBC owners are leading the change by selecting Starting Pitchers earlier and earlier each year. When you look at the numbers over the last 10 years -- which is the existence of the NFBC -- you'll see why pitching is not a hot commodity. These are numbers provided by Paul Sporer of BaseballProspectus.com:

The Last Decade In Starting Pitching
Year ERA WHIP K/9
2004 4.62 1.40 6.23
2005 4.36 1.36 6.04
2006 4.69 1.41 6.21
2007 4.63 1.41 6.30
2008 4.44 1.39 6.45
2009 4.45 1.39 6.65
2010 4.15 1.34 6.77
2011 4.06 1.32 6.75
2012 4.19 1.32 7.14
2013 4.01 1.31 7.19

Strikeouts are up almost one per nine innings since 2004 as the strikeout rate has gone from 16% in 2004 to almost 19% in 2013. Amazing. Walk rates have actually gone down from 3.18 per 9 innings in 2004 to 2.83 per 9 innings last year. The way things are going, we could see ERAs for Starting Pitchers fall below 4.00 this year, so plan accordingly.

While numbers for all Starting Pitchers are improving, it's the elite pitchers who are paving the way and that's why their draft values are going up higher and higher each year. Take a look at the Top 30 Starting Pitchers to see just how dominant they are compared to their peers:

The Last Decade of Top 30 SP Averages
Year ERA WHIP K/9
2004 3.30 1.20 7.52
2005 3.18 1.19 6.73
2006 3.54 1.24 6.96
2007 3.38 1.22 7.62
2008 3.19 1.19 7.66
2009 3.10 1.18 7.92
2010 2.96 1.16 7.91
2011 2.90 1.12 7.60
2012 3.10 1.14 7.81
2013 2.92 1.11 8.41

Look at the dominance of the Top 30 starting pitchers over the last 4 to 5 years. It's pretty amazing. Teams that are getting 3 or more of these guys on their rosters are just killing the pitching categories as their ERAs are more than a point lower than other starting pitchers while striking out one batter more per nine innings. Pretty impressive stuff.

Now, where does Clayton Kershaw, Yu Darvish, Max Scherzer, Justin Verlander and Jose Fernandez go this year? This data is pretty telling on where they might go. Enjoy.

Thoughts on this data?

[Hells Satans]

Without having done any real analysis.....I think it makes taking SP early an even bigger mistake than usual. The averages have decreased much more significantly than the Top 30. Wins are semi-random and one can usually finish with a respectable K number through proper team management. Paying early for ERA/WHIP is a mistake in my view when BA/R/HR/RBI are more scarce than ever.

[Bronx Yankees]

While pitching is dominating hitting, I'm not sure that means pitchers should be drafted any earlier. The ratios of most pitchers have improved, and the averages of certain batting thresholds have declined, arguably making stud hitters that much more valuable.

Turning to Greg's tables, I'm not sure what they mean or how they should influence drafting strategy. Let me give an example of why I am uncertain, using the first and last years shown.

In 2004:
The average starter was: 4.62 ERA, 1.40 WHIP, 6.23 K/9
The average Top 30 starter was: 3.30 ERA, 1.20 WHIP, 7.52 K/9

In 2013:
The average starter was: 4.01 ERA, 1.31 WHIP, 7.19 K/9
The average Top 30 starter was: 2.92 ERA, 1.11 WHIP, 8.41 K/9

Now, if one is to examine whether the Top 30 starters are any more or less valuable today as compared to ten years ago, wouldn't you compare the average Top 30 starter to the average starter? When I do that using 2004 and 2013, this is what I get:

In 2004:
The Top 30 starter's ERA is 1.32 lower than that of the average starter
The Top 30 starter's WHIP is 0.20 lower than that of the average starter
The Top 30 starter's K/9 is 1.29 higher than that of the average starter

In 2013:
The Top 30 starter's ERA is 1.09 lower than that of the average starter
The Top 30 starter's WHIP is 0.20 lower than that of the average starter
The Top 30 starter's K/9 is 1.22 higher than that of the average starter

Thus, while I limited my example to the first and last years depicted, in my mind the numbers indicate that, if anything, the Top 30 starters in 2004 were very slightly more valuable compared to the Top 30 starters in 2013. In other words, the benefit of a Top 30 starter compared to an average starter really was about the same in 2013 as it was ten years ago.

Just my two cents. If folks think I'm looking at this incorrectly, I'd be curious to know why. Thanks.

Mike

[Hells Satans]

That's exactly what I was trying to say, but you did a better job.

If runs are more scarce, elite hitters become more valuable not less valuable (and vice-versa for pitchers). This is particularly true when you consider what is available in FAAB. Run production is virtually impossible to replace after the draft in a 15 team league.

[Gekko]

I might not take a SP until the 10th round this year

[DOUGHBOYS]

When players were on steroids like Bonds, McGwire, etc. an emphasis to get more hitting wasn't needed. Quite the opposite.
The hitting was readily available. Pitching was more scarce.
Now that it has swayed the other way, I won't put an emphasis on pitching. I agree with all the posts after the original. If anything, I'll lean towards getting more hitting.
It is not like pitching is top heavy. It is prevalent throughout the stats, meaning that pitchers can be had in later rounds.
So why should pitching be selected earlier?

[fwicker]

I must be missing something.... if hitting points have no effect on pitching points and v.v. (e.g. HRs and Wins aren't counted in the same category; nor, e.g. saves and steals in the same category) how does the emergence of pitching at the expense of hitting have any bearing at all? So what if a 2004 .270 hitter is now a 2014 .255 hitter; if he's the 83rd best hitter in either year it does't really make a difference, it's relative to the pool for that year. Same reasoning for improved pitching, it's still relative to the pool. The SGPs should remain constant across the spectrum of performances even though that spectrum or range is shifting up for pitchers and down for hitters. I suppose this could be incorrect if the standard deviations of performances are radically changing but I don't think that's true. Again... I may be missing something here and would love to hear compelling arguments.

This does beg the question of why there aren't an equal (or proportional --- 14:9) number of pitchers taken during each round. I think that's because pitchers' roles are less certain than hitters. But I'd like to hear comments on this too.

[Bronx Yankees]

This does beg the question of why there aren't an equal (or proportional --- 14:9) number of pitchers are taken during each round. I think that's because pitchers' roles are less certain than hitters. But I'd like to hear comments on this too.

In my opinion, the reason why earlier rounds are more hitter-dominant than the starting lineup ratio (14:9) are at least fourfold. First, a top-tier hitter theoretically can contribute to all five hitting categories, while a top-tier starting pitcher only contributes to four pitching categories (I.e., excluding saves). Second, I think many believe that pitchers are more injury prone, or prone to lengthy injuries, and, therefore, they may be "riskier" than hitters in the early rounds. Third, I think many believe there is more year-to-year variability in terms of top pitchers than hitters, making it safer or more appealing to "wait" on pitching beyond the early rounds. Fourth, similarly, there is so much "good" pitching out there, many folks like to use earlier picks on more scarce positions.

All that being said, I just completed a 50-round draft by email and took two starters with my first four picks, something I typically do not do, but that's how the draft presented itself to me when my turn came up. Also, I'm not sure I agree with your conclusion that fewer pitchers are taken than the 14:9 ratio would suggest. That statement certainly strikes me as true in the early rounds but am not sure it is correct with respect to entire drafts. For instance, in that 50-rounder, I wound up taking 26 hitters and 24 pitchers. In leagues with a 30-round draft, after the 14:9 starting lineup, typically four out of my seven bench spots are pitchers. Not sure if that is the best strategy, but it makes my team more pitching dominant (17 hitters:13 pitchers) than the 14:9 ratio.

Mike

[fwicker]

In my opinion, the reason why earlier rounds are more hitter-dominant than the starting lineup ratio (14:9) are at least fourfold. First, a top-tier hitter theoretically can contribute to all five hitting categories, while a top-tier starting pitcher only contributes to four pitching categories (I.e., excluding saves). Second, I think many believe that pitchers are more injury prone, or prone to lengthy injuries, and, therefore, they may be "riskier" than hitters in the early rounds. Third, I think many believe there is more year-to-year variability in terms of top pitchers than hitters, making it safer or more appealing to "wait" on pitching beyond the early rounds. Fourth, similarly, there is so much "good" pitching out there, many folks like to use earlier picks on more scarce positions.

All that being said, I just completed a 50-round draft by email and took two starters with my first four picks, something I typically do not do, but that's how the draft presented itself to me when my turn came up. Also, I'm not sure I agree with your conclusion that fewer pitchers are taken than the 14:9 ratio would suggest. That statement certainly strikes me as true in the early rounds but am not sure it is correct with respect to entire drafts. For instance, in that 50-rounder, I wound up taking 26 hitters and 24 pitchers. In leagues with a 30-round draft, after the 14:9 starting lineup, typically four out of my seven bench spots are pitchers. Not sure if that is the best strategy, but it makes my team more pitching dominant (17 hitters:13 pitchers) than the 14:9 ratio.

Mike

Thanks Mike, I agree with your reasonings on pitchers value vs hitters in the early rounds. More impactful injuries (Harvey, e.g.), the need to fill hitter positions, and variability (Cliff Lee wins in 2012), are all good reasons. On the 14:9 ratio I should have clarified I meant in the first 20 rounds or so.... in a DC50 I think that 20+ pitchers are needed. I just finished the first 23 rounds of an NFBC DC and the breakdown was 234 hitters and 91 pitchers, or 28% ... vs a 14:9 ratio which would be 39%.