I'm not exactly Grant. Hell, I'm not even a G on his rant by volume, but I do occasionally pepper my drivel with a bit of interesting discourse. The last really interesting thing I did around here was when I took a look at the SF pitching staff based on their RE24 average instead of their ERA.

That team was pretty good and the pitching staff could probably have won another World Series. This year, however, there is a lot of talk about the bullpen being our weakest link and the problems this could create. The problem is that there really aren't a lot of great stats for evaluating relievers.

Well, there is one, and I've brought this up before. You can calculate an RE24 average in order to understand how a pitcher's performance statistically changed the likelihood of a run scoring.

So, let's recap.

On September 25th, 2011, Nat Anacostia posted an article about RE24 that explained the stat better, in practical terms, than a lot of the big Sabermetric sites. Essentially, RE24 is a pure statistic. It looks at how many runs have scored over the last bazilion inning of major league baseball and says "Ok, great, so the average inning of play sees .48 runs scored" and then break down how each one of the 24 base-out positions stacks up in the same terms.

Put simply, it tells you that if you have runners at first and third with one out, the statistical likelihood of a run scoring (or the run expectancy) by the end of the inning is 65.2% . When Emmanuel Burriss grounds out softly to second base, the new run expectancy with two outs is just 28.8%. The difference is added or subtracted to the RE24 of both the pitcher and batter.

So RE is "Run Expectancy" and 24 refers to the 24 base-out states, first and third with one out, no one on with two outs, etc. There are 24 of them. Each one is correlated with a statistical likelihoods of a run scoring. You can check out the run expectancy matrix here.

* * *

ERA has a lot of problems and those problems get magnified when you look at a relief pitcher because they typically inherit runners. This means that a poor performance by a reliever quite often reflects badly on whoever they replaced while leaving them with an artificially low performance indicator.

What we can do with RE24 is look, agnostically, at how well a pitcher performed over the course of his pitcher/batter match-ups. We can look at how his delivery actually impacted the statistical likelihood of a run scoring.

If you want to understand how this is done, read Nat's article. It's quite thorough. All you need to know in this case is that I've updated the R/9 and ERA using the National League totals to date for 2012 to ensure that my numbers are accurate.

Before we jump into 2012, let's have a look at how the 2011 relief core pitched last season. Hint: they were freakishly dominant.

A caveat - the following tables use RE24/boLI as a way to correct for high leverage situations. If you want to see pure agnostic RE24 averages, you can calculate them yourselves. That makes more sense for a starter than for a reliever though.

What you see here is that in many cases, SF's pitchers were actually better than their ERA would indicate in terms of how they affected the run expectancy by their performance. Boy, that sure was fun, wasn't it? Hoo-ey... those were the days. Shame about that RISP though.

* * *

So, here we are at the end of 2012, roaring all hot & heavy into a playoff series against a very solid opponent. Do I dare evaluate the 2012 relief core by RE24 average? How will they stack up against 2011's bullpen? Will I still want dinner afterwards?

Hold onto your lunches, ladies and gentlemen. This ride is going to lurch.

Oh my dear merciful baseball gods, what?

You read that correctly. The only pitchers in the bullpen who are not appreciably worse than their ERA indicates are small sample size demons (Machi) or named Guillermo Mota (and his numbers aren't exactly encouraging). I'm not going to include guys like Loux and Penny who we can only pray won't make a playoff appearance.

So what does this all mean?

It means if relievers inherit runners they score.

In short, it means the reason you are so uncomfortable with close games this year is because inherited runners tend to score a whole lot more often than the bullpen's ERA totals would indicate. At least, they should do so statistically (and, as it turns out, frequently do).

We'd better hope that the offense stays hot because if the bats go cold the bullpen ain't getting us there.

We haven't accounted for splits.

When you look at second-half numbers and left/right matchups, the numbers do improve a bit. Bochy has a much better feel for his staff now than he did in July. It's still scary but it's not the Titanic that a year-long view would seem to indicate. Remember, many of these stats were born when the Giants were 7.5 games out of first.