[Ed. note: The following post is part of today's Blog Swap, brought to you by the United Cardinal Bloggers. Participating blogs are:
The blogs participating are:

We'll keep the author anonymous today and let you, gentle reader, try to guess who penned typed it. Feel free to post your guess in the comments. Special thanks to, well, the blogger who wrote it.]

Revisiting the 2008 bullpen might be a topic that is more appropriate for the end of October than the end of January, what with the horror of watching it implode time after time. We all know about the talking points of 30 blown saves and ranking at the bottom of the league. Even after the signings of Trevor Miller and Royce Ring and the dismissals of Randy Flores, Ron Villone, and Jason Isringhausen, the grumblers continue to ask, how good is this pen, really?

That’s not the question I’m going to answer today, however, at least not directly. What this discussion led me to was the thought that, even with all of those flammable parts last year, it seemed to me that the bullpen got better down the stretch, especially with the additions of Chris Perez and Jason Motte. Not that it was perfect by any means, but better. You didn’t automatically hide your face when Tony LaRussa headed to the bullpen, though you may have by the end of the game.

On the gut side, this makes sense. Isringhausen was removed from his closer role and eventually went on the DL. Motte and Perez made their debuts and had success, at least the first time through. Flores was used less. Save for Kyle McClellan, it’s seems that most everyone had a better second half. Does that hold up quantitatively?

Let me make it abundantly clear, I’m not even close to the class of sabermetrician that Pip is. He will probably shake his head in despair and wish he’d never agreed to this crazy blog-swap idea, but I did some rough calculations to see how the bullpen performed, by month, during the 2008 season. (Fair warning, I compiled the numbers by going through the box scores at Baseball-Reference.com. Any errors are mine and I’d be shocked if there aren’t some.)

Let’s start with the basics. Saves, blown saves, and save percentage via month:

Ah, those heady days of April, when the Cards were in first, the bullpen did its job, and all was right with the world. Things fell off sharply after that before rebounding some in the last two months of the year. September’s save percentage isn’t great, but it did finish in the top half of the year. So, from this rough measure, there’s some basis to the improving-late bullpen.

I did add up the holds as well per month, but I’m not sure how relevant they were to the discussion. The bullpen “held” the game between 25 times (in April ) and 15 times (three different months) but there didn’t seem to be much correlation between those numbers and anything else.

Next, bullpen ERA. I know that bullpen ERA is fairly deceiving, because allowing inherited runners to score doesn’t show up in the stat. However, for a back-of-the-envelope stat, especially when you encompass the whole bullpen, it’s good enough for a discussion point.

Well, at least September looked good. The bullpen tended to fluctuate in 2008 in the ERA department, as you call tell. Without a seven runs in three innings outing on 8/26, August would have been 4.05, which would have fit in with my theory nicely. But, since about every month had a terrible outing or two, it’s not exactly fair to remove that one.

Now, of course, not all bullpen outings are created equally. Giving up seven in three is just icing when the score is already 6-1, but terrible if the Cards are up 4-1. So, let’s look at the bottom line. How often was the win awarded to a reliever, and how often did the loss get assigned to the bullpen?

How in the world did we live through June as Cardinal fans? When over half of your losses are coming from the bullpen, you are either playing a lot of close games or you are making a lot of games close. In theory, I’d think you’d want your bullpen to be low in both of these categories, getting good innings out of your starters. In that regard, August and September were on the positive side.

Finally (and I hear your sigh of relief right now), I came up with a rough stat called “heartbreak points.” It’s something I hope to tinker with and keep track of on my blog in the coming year. It’s pretty basic: if the save is blown in the sixth inning or earlier, 0 points, figuring that there is still time for the offense to take the lead back. Those games don’t hurt quite as much. If it’s blown in the seventh, one point is assigned. Two if it’s blown in the eighth and three if it’s blown in the ninth or later. Like I said, it’s not anywhere close to perfect, since it doesn’t take into account home/away or the size of the lead, for starters, but it’s something.

Again, the stat is a little skewed, since due to that large losing streak that kick off September, there weren’t too many chances to blow a save (only 9 opportunities). Most of the time the game was already lost before the pen got to it. But, at least from the stress point of view, August and September were a lot easier to watch. May not have been pleasant, but it wasn’t quite as ulcer-inducing as May and June were.

So, to wrap this up, let’s take a look at all these stats as broken up, not by month, but first “half” (pre-All Star Game) and second half. Does that indicate that things were better as time went by?

By almost any measure, the second half of the season was much stronger for the bullpen. Granted, they threw 94 fewer innings in the second half, so that probably led to some of the improvement, but when you couple this with the departures and additions, you have to feel at least somewhat better about the later innings in 2009.

This entry was posted by Pip on Friday, January 30th, 2009 at 11:00 am and is filed under group blogging. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.