Braves Don’t Use Their Best Pitcher

The Phillies took a close game off of the Atlanta Braves on Friday night thanks to a tenth-inning walk-off home run from Raul Ibanez. Overall, the game was well-pitched on both sides. Roy Halladay shut down the Braves but for two runs over seven innings of work while Brandon Beachy kept the Phillies at bay through six innings. As the tie game moved into the seventh and the Braves dipped into their highly-regarded bullpen, the Phillies had to feel like they let opportunities pass them by as they were able to notch eight hits off of Beachy. Scoring even a single run off of baseball’s best bullpen is a mountain of a task.

Braves manager Fredi Gonzalez went through a normal progression after he lifted his starter from the game. He went to Eric O’Flaherty, then Jonny Venters, Scott Linebrink, and George Sherrill before going to Scott Proctor. One name you may notice is remarkably absent is Craig Kimbrel, the Braves’ young right-handed closer who leads all pitchers (min. 45 innings) in SIERA at 1.88. Coming into tonight’s series opener, Kimbrel was averaging 14 strikeouts per nine innings, which is pretty good. Most importantly, Kimbrel had allowed just one home run in 45 innings.

Certainly one would not want to lose a game without having used the best available weaponry, right? Unfortunately for the Braves, their manager plays “by the book,” which states that using the best reliever (typically the closer) in a tie game on the road is bad strategy. As such, Gonzalez opted to use Proctor, who has less than half the strikeout rate and a significantly higher walk rate without any good batted ball skills. Additionally, he had allowed two home runs in 17 innings, one more home run than Kimbrel in 37 percent of the innings. Unsurprisingly, Proctor recorded only one out before he surrendered the game-winning home run to the Phillies courtesy Ibanez. All the while, Kimbrel sat in the bullpen unused.

I have been critical of Charlie Manuel for his use of the bullpen, but it is worth pointing out that the same fallacious strategies are employed by other managers across the sport. Essentially, the “by the book” doctrine of bullpen management prays that the worse relievers get by long enough to allow your offense to take a lead, at which point the closer may then be used. Perhaps a team with an above-average offense —  one with a higher probability of scoring runs — benefits from this strategy (by and large they don’t, I’m being generous), but the Braves came into the game averaging barely more than four runs per game, the ninth-best rate in the league.

Braves fans will curse Proctor for not being good enough to shut the door, but they should really be going after their manager for not putting his team in the best position to win. Likewise, Phillies fans should do the same when Manuel makes the same egregious error. Fans of every team should be upset when their team’s manager makes easily-correctable strategic blunders. As Sabermetrics (read: basic probability theory) seeps deeper and deeper into baseball culture, we should see these maneuvers fade out over time, but there still remains far too much sub-optimal strategy in the great sport of baseball.

Leave a Reply



  1. Dan Podheiser

    July 09, 2011 09:03 AM

    How’d I know you were going to write this? It’s nice when it’s not Charlie for once, eh?

  2. Jeff G.

    July 09, 2011 09:50 AM

    Someone talked about this the other day when Manuel withheld Bastardo and Baez gave up the walk-off to Stanton. Even if the Braves had used Kimbrel in the 10th, wouldn’t they have had to use Proctor in the 11th anyway? Kind of makes sense, unless Kimbrel was going to pitch multiple innings, or if matchups dictated otherwise.

  3. LTG

    July 09, 2011 10:36 AM

    Yeah, someone would have to pitch the bottom of the 11th, but the Braves might have scored a run or more in the top of the 11th. More chances to hit = more chances to score. The best managerial strategy is to give yourself as many opportunities to score as possible. Kimbrel gives you a much higher probability of having at least one more chance to score than Proctor does. I don’t see the sense in rolling the dice in hopes of getting a lead that you can hold rather than ensuring opportunities to get the lead to hold.

    That said, it would take a lot of historical statistical data collection to account for all of the variables that affect a team’s odds of winning in extra-inning games. And that’s what it would take to demonstrate that you should always use your best relievers first, even as the visiting team. And any traditionalist will call for the proof before revising the tradition. Nevertheless, my rational intuitions agree with Bill’s on this one.

  4. Fatalotti

    July 09, 2011 10:46 AM

    Jeff G, think of it like this:

    In the bottom of the 10th (or 9th for that matter), the Braves margin for error is literally zero. They can’t give up ANY runs at all. This is why you use your best pitcher available (read: Kimbrel). If Kimbrel shuts the door and the Braves go on to score in the top of the 11th inning, then, when the Braves pitch the bottom of the 11th, they have more room for error, by definition.

    The question then becomes, do you use your best reliever when you have no room for error and your worse reliever when you have room for error, or vice versa?

    I think the answer is pretty obvious. On the road, in a tie game, in the 9th innings and later, you use your best pitcher available, and then descend downward.

    What Gonzalez did before the 9th inning was fine (using O’Flaherty then Venters). To start the 9th inning with Linebrink and then to start the 10th inning with Proctor while Kimbrel sat around doing nothing: inexcusable.

  5. thekayell

    July 09, 2011 11:07 AM

    But isn’t there a longer-term dynamic in play? Let’s say Fredi expects his team to compete all series and wants to win all 3 games. Proctor isn’t bad so far this year. Raul Ibanez isn’t good so far this year. (If you told me Raul Ibanez is going to destroy a baseball on the way to hitting a game-winning home run, I would’ve been frankly surprised.)

    So Fredi assumes (rightly, I think) that Proctor gets through the inning AND that his team can put up runs on Carpenter/Baez/whoever else comes on for the Phils in the 11th. Fredi also assumes the Braves can do this pretty quickly, judging by just how bad the Phils’ pen is once you get past Stutestardo.

    By resting Kimbrel, he keeps his best arm fresh for the next two games in the series. Alternatively, Fredi has an ace in the hole to preserve a tight one-run lead, if needed, after the top of the 11th or 12th. Kimbrel pitched the day before; sending him out 4 days in a row (assuming, as Fredi should, that his team can/will win all three games) is a lot of work for a young pitcher.

    I agree that managers are too reluctant to use pitchers in optimal scenarios, but I think Fredi actually made the “right” medium-term strategic decision on this one.

  6. Seanster

    July 09, 2011 11:29 AM

    Different topic alert. John Mayberry and Dom Brown have nearly identical hitting stats at this point in the season, and they both recently rose above the league average OPS+.

  7. Jeff G.

    July 09, 2011 12:11 PM

    OK, I get it now. Assuming the Braves would score in the top of the 11th, you would definitely use Kimbrel in the 10th to ensure it goes to the 11th and b/c its higher leverage. I was just assuming that they wouldn’t score in the top of the 11th.

    I don’t really agree with preserving Kimbrel for future games though, you want to win the game at hand, and there may not be a higher leverage situation in the final two games so use him now.

  8. Chris

    July 09, 2011 12:31 PM

    How many times did Sherrill warm up that he could only throw 2 pitches in-game…? I mean obviously Fredi thought that he couldn’t pitch the 10th but I don’t understand why he had him warm up in the 5th/6th innings (and more? idk) only to eventually bring in O’Flaherty. Seems like just as big of an error as not bringing Kimbrel into the 10th once Sherrill couldn’t go.

  9. MIke G

    July 09, 2011 01:35 PM

    The only reason Fredi didn’t use Sherrill in the 10th was because he wasn’t smart enough to double switch when Sherrill came into the game, and he then PH for him in the top of the 10th. He should have double switched or just let Sherrill bat with two outs.

  10. Jay

    July 09, 2011 04:12 PM

    Two Things:
    1. Apparently zombies can circle the bases and walk off the field.
    2. How did 2008 Brad Lidge transform his body into a 32 year old Dominican. That is the last time I have seen such a dominant combination of fastballs and sliders. It was eerily similar. He had the slider that buckles the hitter’s knees and curves into the strike zone, and the deadly slider that looks like a fastball then just dives into the ground. Incredible.

  11. Phillie697

    July 09, 2011 07:28 PM


    You are thinking like those same idiot managers. A win in the first game of a series is the same as a win in the second or third game. Last time I checked, all wins count the same in the standings. Why would you possibly want to throw a game away in the hopes that you MIGHT win another tomorrow? To put it plainly, let say you pitch Kimbrel in game 1 and win, so he’s not available in game 2 or 3, and you lose that game. What’s the difference between that and losing game 1 and winning game 2 or 3? Fredi threw his chance at a sweep away yesterday.

  12. Scott G

    July 11, 2011 08:29 AM

    To elaborate on Phillie addressing thekayell, how can you even be sure you’ll be in a position where you need to use Kimbrel in any subsequent day? Maybe the Braves rock Lee. Maybe the Phillies rock Hanson. You know that you need your best pitcher at the current time, you use him.

Next ArticleJust A Friendly Reminder