I don’t really know if anyone much cares about this topic, but it is one that I have felt an important one for many years.  Why is it that some guys always seem to be “tough luck” losers in tight games and some guys seem to be the “lucky” beneficiaries of scoring largesse?  I am certain that luck actually does have some role in the phenomenon, much in the same way Mike Shannon always says Bob Gibson was the luckiest pitcher he ever saw.  After all, Gibson always had the good fortune to pitch on the days the other team didn’t score many runs.  Seriously, there is some luck in who the pitcher pitches against, how hot an opposing team is at a given point in time, how hot his team might be and so on.

Nevertheless, I have always held the completely unsubstantiated belief that pitchers who held the opponent down early in the game were more likely to get the opportunity to pitch with a big lead.  In general, a pitcher with a lead can be a more confident and aggressive pitcher who should generally perform more effectively.  Conversely, a lineup that is always playing catch up has to battle a more aggressive and confident opposing pitcher. 

So, I decided to do another one of my not particularly scientific, sort of small sample size, analyses in the hope of learning a little something and stimulating some discussion.  All I did was look at the 2007 starting rotation, specifically Wainer, Looper, Wells, and Reyes to determine in which part of the game they surrendered the most runs.  Then I looked to see if I could find any correlation with run support.  I thought about including WonderBrad as well, but less than half his appearances were starts and I couldn’t find a way (or at least one not too time consuming) to separate out his data. 

I looked at each inning, but the primary determinant I used was how many runs were allowed in the first three innings versus how many runs were allowed in innings four through six.  I must admit that I am using total runs allowed, as the fine statistics at baseball-reference.com only give total runs allowed in the inning-by-inning splits.  What I came up with was pleasing to me as it fit my pre-conceived notions nicely.  It remains to be seen whether that means much, but I will leave it up to the community to argue that point.

Here is the data:

Pitcher              Run Support                 Runs 1-3          Runs 4-6

Wainer                         4.92                 3.66                 4.17

Looper                         4.79                 4.60                 4.73

Wells                            3.55                 5.01                 7.09

Reyes                           3.32                 7.79                 4.60

Let me explains the runs numbers above.  I took the total number of runs allowed in each inning, divided by the number of times that particular pitcher pitched in that inning, and then multiplied the quotient by 9.  The intent is to give it the look and feel of an ERA for familiarity’s sake, but it counts all runs allowed and doesn’t take into account partial innings as that data was not available to me. 

It looks like a pretty clear conclusion for this dataset was that there was a strong inverse correlation between runs allowed in innings 1-3 and run support.  I can’t really explain the big delta in run support between Looper and Wells as their numbers are pretty similar.  In fact if you look at Wells numbers through the fourth inning, they are nearly identical to Looper’s.  It might have something to do with Wells runs allowed number in the fifth of 11.88 (the single worst inning of all starters).  Like I said this post is designed to promote discussion, so please feel free to chime in if you have an informed opinion.

Interestingly enough, Reyes got a big fat zero for run support in six of his 20 starts (30%) and Wells got a goose egg in four of 26 (16.67%).  Wainer got at least one run to work with in every one of his 32 starts and Looper was only shutout in one of 30. 

A final tidbit from looking at this years numbers:  the starters are getting over half a run better support (4.85 v 4.26) and have much, much better numbers over the first three innings.  The 2007 squad had a runs allowed average of 5.01 over the first three innings (or 1.67 per game) while the 2008 squad has allowed an average of 3.50 runs over the first three inning (or 1.17 per game).  The difference is particularly pronounced in the first two innings with the 2007 staff averaging 1.14 runs per game and the 2008 staff at an impeccable .56.