Proud fans of a 128-year old tradition

It is currently Mon Sep 01, 2014 10:59 pm

All times are UTC - 4 hours [ DST ]




Post new topic Reply to topic  [ 6 posts ] 
Author Message
 Post subject: GOB Ratios
PostPosted: Mon Apr 20, 2009 12:43 pm 
Offline
User avatar
 Profile

Joined: Mon Apr 28, 2008 10:10 am
Posts: 1328
I've been thinking about starting pitching lately and am proposing a new statistic which I call The GOB Ratio, where GOB stands for Good-Ok-Bad. It classifies each start as either:

    Good - ERA less than or equal to 3.00 - i.e., giving your team a great chance to win.
    Ok - ERA greater that 3.00 and less than or equal to 4.50 - i.e., keeping your team in the game.
    Bad - ERA greater than 4.50 - i.e., bring on Donnie Veal for mopup duty.

A GOB Ratio is written as three numbers that together add up to 5, indicating how many starts out of 5 fall into each of the three categories. The ideal pitcher would have a 5-0-0. Worst possible score would be 0-0-5.

Going back 10 starts, the current Pirate rotation has the following GOBs:

    Maholm: 3.5-1-0.5
    Snell: 3-0.5-1.5
    Duke: 2.5-0.5-2
    Ohlendorf: 1.4-0.7-2.9 (based on 7 starts with Pirates)
    Karstens: 1-2-2 (alas, the 2 starts that make up Karstens' "G" were his first 2 as a Bucco)

I like this approach because I care less about a starting pitcher's ERA and more about how often he gives his team a chance to win. Someone who has a high ERA because they occassionally have an awful outing is better than someone with the same ERA who is mediocre every time. While Quality Starts (QS) is somewhat useful for determining this. it doesn't distinguish between the Ok and the Bad starts which I think is important.

So, my questions to the board are: Is this a useful metric for evaluating starting pitchers? How can it be improved? And, what do you think are acceptable GOB ratios for starters 1 through 5 on a good MLB team?

Cheers,
BM


Top
 
 Post subject: Re: GOB Ratios
PostPosted: Mon Apr 20, 2009 3:15 pm 
Offline
User avatar
 Profile

Joined: Sat Apr 26, 2008 6:18 pm
Posts: 5060
Location: Scotch Plains, NJ
The problem with your approach is that you do not take into account the number of innings a pitcher pitches per start. All you look at is the ERA per start, and that tells you little about how well the pitcher kept his team in the game or gave it a chance to win. Your system would give more credit to the pitcher who gives up 1 run in 3 innings compared to the pitcher who gives up 3 runs in 6 innings. Part of a pitcher's job is to eat innings, and your GOB ratio does not account for that.

Sorry, I'm sticking with quality starts.

_________________
"For every complex problem there is an answer that is clear, simple, and wrong."
~H. L. Mencken


Top
 
 Post subject: Re: GOB Ratios
PostPosted: Mon Apr 20, 2009 3:36 pm 
Offline
User avatar
 Profile

Joined: Mon Apr 28, 2008 10:10 am
Posts: 1328
Willton wrote:
The problem with your approach is that you do not take into account the number of innings a pitcher pitches per start. All you look at is the ERA per start, and that tells you little about how well the pitcher kept his team in the game or gave it a chance to win. Your system would give more credit to the pitcher who gives up 1 run in 3 innings compared to the pitcher who gives up 3 runs in 6 innings. Part of a pitcher's job is to eat innings, and your GOB ratio does not account for that.

Sorry, I'm sticking with quality starts.

I think that your example (a starter giving up 1 run in 3 innings) is an outlier that occurs infrequently enough that it would not significantly skew the ratio. In general a starter that has only given up 1 run in 3 innings will not be removed from the game unless there is some extraneous circumstance (injury, weather). Yes, I know there are exceptions but I would submit that they are relatively rare. And if a starter gives up even 2 runs, his 3 inning start would go into the 'B' category.

I'm not suggesting that this is a replacement for QS or other statistics but I do think that it captures something that other metrics do not. Quality starts are important but Ok starts are as well. If you have two starters, one with a GOB of 2-0-3 and another with 1-3-1, which would you rather have? The former has a higher number of quality starts but the latter keeps you in 4 out of 5 games that he starts. Which is better?


Top
 
 Post subject: Re: GOB Ratios
PostPosted: Mon Apr 20, 2009 4:56 pm 
Offline
User avatar
 Profile

Joined: Sat Apr 26, 2008 6:18 pm
Posts: 5060
Location: Scotch Plains, NJ
burghermeister wrote:
I'm not suggesting that this is a replacement for QS or other statistics but I do think that it captures something that other metrics do not. Quality starts are important but Ok starts are as well. If you have two starters, one with a GOB of 2-0-3 and another with 1-3-1, which would you rather have? The former has a higher number of quality starts but the latter keeps you in 4 out of 5 games that he starts. Which is better?

How would you know how many quality starts each player had, given the information you gave me? Of the starts pitched by the 1-3-1 guy, which ones were quality starts? We don't know because we have no idea how many innings he went in each start.

A quality start requires going 6 innings while giving up 3 runs or less. Without knowing how many innings were pitched in each start, there's no way to tell whether the 1-3-1 pitched any quality starts, as each start may have only lasted less than 6 innings. Is it better to have pitched only 5 innings while giving up 1 run, or pitch 7 innings while giving up only 3 runs? Given your qualifications as to what equals a "good" start and an "ok" start, your GOB ratio would say that the guy who pitched 7 innings (3.85 ERA) did a worse job than the guy who pitched 5 only innings (1.80 ERA). I would argue the opposite, as the guy who pitches only 5 innings turns the ball over to the bullpen much sooner than the guy who pitches 7 innings.

I understand you wanting to put "good" starts in context with "ok" starts, but doing so without accounting for the number of innings pitched by a pitcher seems wrong-headed in my opinion.

_________________
"For every complex problem there is an answer that is clear, simple, and wrong."
~H. L. Mencken


Top
 
 Post subject: Re: GOB Ratios
PostPosted: Mon Apr 20, 2009 5:14 pm 
Offline
 Profile

Joined: Sat Apr 26, 2008 4:22 pm
Posts: 2495
Wouldn't guys in the top and bottom third of your scale be consistent enough to not warrant such a scale? By this I mean Santana will always be pretty much 4-.5-.5 while say a Fogg will always be pretty much a .5-2-2.5 (I didn't run actual numbers btw). You can surmise that by looking at their ERAs. I suppose there could be some value in differentiating between the middle group, I'm just not sure how much.


Top
 
 Post subject: Re: GOB Ratios
PostPosted: Mon Apr 20, 2009 5:58 pm 
Offline
User avatar
 Profile

Joined: Mon Apr 28, 2008 10:10 am
Posts: 1328
Argentum wrote:
Wouldn't guys in the top and bottom third of your scale be consistent enough to not warrant such a scale? By this I mean Santana will always be pretty much 4-.5-.5 while say a Fogg will always be pretty much a .5-2-2.5 (I didn't run actual numbers btw). You can surmise that by looking at their ERAs. I suppose there could be some value in differentiating between the middle group, I'm just not sure how much.


Took a quick glance at Santana and I think your guess is spot on. I started to look at Fogg but didn't have the stomach for it...

I think it's more useful for evaluating the middle/bottom of the rotation guys. I'm always interested in considering where people like Karstens and Ohlendorf fit in vs. other 4/5 starters in the league. And while stats like W/L, ERA, WHIP and QS tell much of the story, it always seems like none of these measure the consistency (or lack-there-of) of level of effort. Maybe, I have the metrics for which starts go in which bucket right and maybe I don't (I'm willing to concede to Willton that I've given short shrift to the guy who goes 7 innings and gives up 3 runs), but I think the three buckets are useful and add some clarity to the discussion. Many of these 4/5 starters are not going to have many quality starts so having some way of distinguishing between the okay and bad ones seems worth doing.


Top
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 6 posts ] 

All times are UTC - 4 hours [ DST ]


Who is online

Users browsing this forum: Bing [Bot], Google [Bot], RTJR, sisyphus and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group  
Design By Poker Bandits