Saturday, May 26, 2012

Buc Bullpen: Under Pressure

Everyone knows how to set up a bullpen. Long guy, couple of bridge guys, LOOGY, couple of set up guys, closer. Each step up the ladder ratcheted the leverage level a little higher, with ERA conversely dropping as the ninth inning got nearer.

But thanks to the Bucs inept offense and solid starting pitching, suddenly everything, from the sixth inning on, had the game hanging in the balance. High leverage at-bats and tight scores trickled down into the earlier innings so that now everyone needed to perform like a back-ender if Pittsburgh was to compete.

What is leverage? It's complicated mathematically, but at essence a measure that combines the inning and game situation as it affects a team's win probability. So the later the game, the tighter the score and the more runners on, the greater the leverage level. The average game consists of about 20% high leverage situations, with medium and low leverage situations at 40% each.

So we took a look at two stats: appearances when the game score was within a run either way, and high leverage situations faced (both from Baseball Reference). Here's the result:

Tony Watson (65% +/-1 run, 35% hi leverage)
Juan Cruz (63% +/-1 run, 45% hi leverage)
Joel Hanrahan (59% +/-1 run, 37% hi leverage)
Jared Hughes (59% +/-1 run, 29% hi leverage)
Jason Grilli (53% +/-1 run, 42% hi leverage)
Chris Resop (44% +/-1 run, 27% hi leverage)
Brad Lincoln (30% +/-1 run, 21% hi leverage)

As a unit, the bullpen has appeared 56% of the time in one run games and 33% of the time in high leverage jams. That's not too surprising, as 21 of the Bucco games (47%) have been decided by a run and 60% by two runs.

Everyone but Resop and Lincoln have made more than half their outings when the game was on the line, and all of them were above the high leverage norm. And more telling, every reliever but Lincoln was called on to work in the eighth inning or later during a one-run game.

The back-of-the-envelope analysis shows us a couple of things. First, Hanny, Cruz and Grilli have worked the same number of close games, and Hanny actually has the lowest high leverage percentage of the trio. In fact, the set-up men have faced double the standard of high leverage situations and still thrived. Hanny is the undisputed closer, but by performance, you'd have a hard time picking one over the other.

It also shows how much trust Clint Hurdle has in his young arms, Hughes and Watson. Together, 62% of their outings have been in games when the score is within a run, and they have worked over league average (32%) in high leverage situations.

A lot of credit for the Bucco bullpen success has gone to the back-enders, and they deserve huge props. Hanny, Cruz and Grilli have worked in tight, game-deciding situations more often than not with dazzling results. But the reason they're so successful is because the rest of the pen is doing yeoman duty in getting the ball to them in manageable game situations.A great late inning crew isn't worth much if the team's down by three or four before you can get to them.

How long can this team effort continue? Well, at Fangraphs, the pitching table shows that Grilli is the only reliever whose xFIP matches his actual ERA; the rest are all overachieving. Ditto for the strand rate, which is somewhere around 73% leaguewide. Six of the seven Pirates are above that, with five at 87% or better. And except for Cruz and Grilli, the Batted Ball In Play averages are way low. So yah, regression is always lurking around the corner, as we saw late last year.

But for now, the bullpen, from top to bottom, has captured lightning in a jar. They're feeding off one another, and hey - who knows how far and long they can carry the club? Pressure has brought out the best in them during the opening two months of the season. Let's hope they don't regress, at least until the Pirate bats progress.

No comments: