Today’s slate of interleague games was particularly unappealing as only one matchup featured two teams with winning records – the 2008 World Series rematch between Tampa Bay and Philadelphia. Unfortunately, the pitching matchup of Antonio Bastardo (5.21 ERA) versus Andy Sonnanstine (6.60 ERA) left a lot to be desired. The lone National League contest had two winning teams, the first place St. Louis Cardinals and the second place New York Mets, and featured a face-off between two Cy Young winners on the mound: the resurgent Chris Carpenter and struggling Mets’ ace Johan Santana. While the mainstream buzz before and after was about the Cy Young matchup, the real story of the game was the showcase of Tony Larussa’s incessant managing style which gave mixed results.
The Pitching Matchup
In Santana’s first ten games of the season, he recorded eight quality starts, had a 7-2 record with a 1.77 ERA. During these 10 games, he averaged 11.7 K/9 and he had given up only five home runs in 66.0 innings pitched, with three of those coming in his last 3 starts in May. Conversely, during his first four starts of June he had a 1-3 record with a 7.33 ERA, never struck out more than three in a game, and had given up 7 home runs in 23.1 IP.
After two years of pitching minimally in 2007 and 2008 due to injuries, Chris Carpenter has been enjoying a solid 2009, despite a lost month due to a rib-cage injury. In his seven starts since his return on May 20, he had a 4-1 record with a 1.85 ERA, with six straight quality starts entering today. Some of his great performance has been due to his unbelievably low BABIP of .191.
The Game
St. Louis got on the board in the first inning with an unearned run thanks to catcher interference by Omar Santos with two outs. Ryan Ludwick‘s swing was late on Santana’s 91-mph four-seamer and caught Santos’ glove as he was reaching for the pitch. The error resulted in men on first and second for Yadier Molina, who lined an ineffective changeup to center to score Skip Schumaker, who had singled earlier in the inning.
Santana continued to struggle in the second and third, as his pitch count climbed to fifty-seven after three. At one point, he started seven straight hitters off with a ball. Santana was able to get out of a two-out, bases-loaded situation in the second, by jamming Albert Pujols just enough to induce a deep, but harmless, fly out to left. On the other hand, Carpenter cruised through his first three frames, striking out five and allowing no hits. The only base runner for the Mets was Alex Cora who started the game being hit by a pitch that actually went behind him.
The turning point of the game was the fourth inning. Santana started the inning by walking Carpenter on five pitches. Larussa decided to play for a single run, ordering Joe Thurston to sacrifice Carpenter over to second. After a ground ball to the right side that moved Carpenter to third, Schumaker ended the threat by looking at a 93-mph fastball down the middle for strike three.
After tiring himself on the base paths in the top of the inning, Carpenter proceeded to give up a lead-off single to Luis Castillo on a ground ball through the 5-6 hole. David Wright followed with a bouncer that deflected off of Carpenter’s glove into no man’s land between second and first. Ryan Church hit a scorcher that Pujols gloved, forcing Wright out at second. Fernando Tatis then blooped a single to right field to tie the game. After Fernando Martinez bounced a grounder that forced Carpenter off the mound, Nick Evans smacked a double into the right field corner on a 1-0 cutter that didn’t have much break. Both Church and Tatis scored giving the Mets and Santana a 3-1 lead. The four hits of the inning were the only ones that Carpenter allowed through seven innings, as he set down the Mets 1-2-3 five times.
Now pitching with the lead, Santana had his first 1-2-3 inning in the fifth, allowed a solitary run in the sixth after a Rick Ankiel lead-off double, and gave up a lone single to Albert Pujols in the seventh. With 110 laborious pitches, Santana scattered seven hits and three walks, only striking out three. He did manage to limit the damage to just two runs, one of them unearned.
In the eighth, Larussa went to work. In the top of the inning, he replaced Rick Ankiel with Tyler Greene to face new reliever, lefty Pedro Feliciano. After Greene’s bloop single to center, Larussa again decided to play for one run by having Khalil Greene lay down a sacrifice bunt. The small-ball threat was extinguished again as pinch-hitter Jason LaRue softly lined out to first and Joe Thurston hit a come-backer to the pitcher.
In the bottom of the eighth, Larussa kept the bullpen phone buzzing as he used four pitchers (Dennys Reyes, Kyle McClellan, Trever Miller, and Jason Motte) to overcome a single and two walks and keep the Mets from adding an insurance run. There were only two times in 2008 when four pitchers were used in a half-inning and the opposing team didn’t score. As you might have guessed, one of those times involved Tony Larussa’s Cardinals (the other being Cecil Cooper‘s Astros). Additionally, Larussa was the only manager last year to use four pitchers in an inning that consisted of five or fewer batters; a run, however, was yielded in that situation.
Francisco Rodriguez picked up his 20th save by striking out pinch-hitter Chris Duncan and Schumaker. There was a small threat as Pujols and Ludwick each drew two-out walks, but the game ended on Yadier Molina’s line out to left.
In terms of the starting pitching, the June trends continued. Santana labored and seems to have lost his out pitch as he only struck out three. He did, however, pick up his ninth win, tying him for the league lead with three others. Carpenter pitched well, except for that one pitch to Evans. He did continue to keep his BABIP ridiculously low, as the Mets were just 4 for 20 on balls in play. This was a game that showed that sometimes it’s about bunching hits together and not necessarily the overall number of hits.
Key Decision Analysis
The key decisions for Larussa in this game were the sacrifices in the top of the fourth and the top of the eighth. In both situations, he had a man on first with no outs. In the fourth with a one-run lead, he had his number nine hitter (who wasn’t a pitcher due to his unique pitcher-in-the-eight-hole batting order) up. In the eighth with a one-run deficit, he had his number seven hitter up with the pitcher following (who was likely going to be pinch-hit for anyway). In both situations was the sacrifice bunt the right call?
While Larussa has a reputation for over-managing, more recently, he has been typically middle of the pack in terms of using the sacrifice bunt. From 2000 to 2003, St. Louis was in the top three in the majors in sacrifice hits, but from 2004 to 2009, the Cardinals were ranked somewhere between 5th and 9th in the National League. Is Larussa becoming more frugal with giving away outs? On the flip side, Larussa does have a weird tendency to have his pinch hitters sacrifice (eight times in 2008 which doubled any other team in the majors).
To understand the possible benefit of bunting in this situation, let’s examine the change in Win Expectnacy at the FanGraphs scoreboard. According to FanGraphs, the Larussa decision in the fourth inning reduced the likelihood that the Cardinals would win by 1.7%, and in the critical eighth, it reduced the probability that the Cardinals would win by 4.2%. One caveat to remember is that the FanGraphs Win Percentage Added only considers the position of the runners, the number of outs and the inning; it does not consider how the Win Expectancy changes based on the lineup position, or the quality of pitcher on the mound (which helps define the run environment).
To back up these numbers, I used a Run Expectancy Matrix and Run Frequency Matrix that is based on the lineup position batting. In the fourth inning, Larussa didn’t really improve his chances of scoring at least one run (going from 37.7% to 37.8%), but decreased his expected runs in the inning from 0.78 to 0.71. In the eighth inning, the sacrifice bunt was more detrimental. He did slightly increase the probability of scoring at least one run from 38.3% to 39.2%, but he decreased his expected runs from 0.82 to 0.66.
The additional head-scratcher is the fact that he was playing for the tie on the road. In the eighth, even if he ties, he is still in an unenviable position. The Mets have two chances to score an additional run (a 48.4% probability that they will), while he has only one (a 28.2% probability) until extra innings are played. The significant reduction in Win Expectancy is taking this into account.
For a manager who prides himself on squeezing out small percentage improvements in righty-lefty matchups and batting the pitcher eighth, it was surprising that he ignored the percentages in these key situations.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
The downside? In the subsequent analysis segment, his references to team sac bunt totals aren't nearly so interesting as what would have been: team-wide totals for position players' sac bunts. There were more than a few names wrong; La Russa seems to give everyone trouble, but Omir Santos too?
Finally, there's the issue of using Win Expectancy. It's cool that he took the time, but also instructive as far as the meager utility of the exercise, because doing so documented how minor the potential impact any move might make. Where Tim makes the sale about their employment to me is where he has the good sense to strike a few cautionary notes about the limitations involved. Really good stuff, all in all.
4/5 of the way through, nobody got an instant thumbs up. Is this like American Idol doing Gangsta Rap week? Do the writers really need to be so versatile that they're being pushed to do a concept or niche that they aren't really built/trained/molded for?
The "Key Decisions" worked quite well for me though. Though the overall writing style and structured worked, it seemed more time should've been spent on the "Key Decisions" instead of coming in at the end like an afterthought.
The Key Decision section had less drama to it. I was already aware sacrifices are not usually an efficient way to win games statistically. Although, I don't recall ever seeing the stats broken down for a particuly sacrifice situation with precise numbers like this, I was not sure reading all this dry stuff was worth the thrill. The numerical differences are ultimately very small - not enough to cover many factors not measuered - some of which Tim kindly suggested. There are possibly psychological factors one way or another not measured as well.
On the flip side that would be an interesting analysis. Examine pitchers' performance in the innings immediately after they have been on the basepaths versus those when they haven't to see if a pattern exists that running tires them out.
However, the use of run expectancy to show that LaRussa should have known he was hurting his chances to score/win by bunting is an article saver to me. It shows that even prior to the result, we know this is a bad decision.
Nice work. 3 thumbs up this week for me, and this will clearly be one of them.
Though the instruction was that the article had to be tied to one of the games on this day, I'm not sure why almost everyone felt the need to do a "game recap"-type article. Surely it would have been acceptable to take one aspect or event in the game and address that, as Sheehan often does. For my money, even a writer as entertaining and colorful as Christina has trouble making a straight game story hold my attention (though, to her credit, she has occasionally done so).
This wasn't.
Nice work