Would you walk in the tying run?by Mike Fast
July 16, 2008
On April 26, 2003, Aron Ralston was hiking alone through Utah's Blue John Canyon. Seven hours into his hike, while Ralston attempted a difficult traverse, a boulder fell, crushing his right forearm and pinning his hand underneath the large rock, trapping him. After five days without rescue and multiple failed attempts to move or chip away the boulder, Ralston finally amputated his right arm in order to free himself. He rappelled down a cliff and hiked out of the canyon before meeting other hikers who helped him find a search and rescue team.
What would you have done if you were in Ralston's place? Would you have had the courage and strength to cut off your own arm in order to save your life? Not to hopefully trivialize his story too much, but his dilemma is the one that came to my mind when I read this question posed by White Knight on the Baseball Fever message board:
Here's the situation. It's a pretty important game, and your team is up by one run (let's just say it's 5-4) and it's an away game. There are two outs, and it's the 9th inning. However, the bases are loaded, and Barry Bonds is up, in all his 2001-2004 glory. Up after him is a mediocre batter, and the SF Giants have no great hitters on the bench. Do you walk Barry Bonds, and hope for the best next inning? I think I do.I can't imagine any manager or pitcher ever wanting to purposefully allow the lead to slip away. But maybe it's the right thing to do, the better part of valor, to sacrifice your arm (getting Bonds out and winning the game right there), so to speak, in order to survive and live another day (hoping to take the game in extra innings). When I read that post, my gut, my heart, my head, and any other part of me screamed, "Heck, no!" You don't give up the lead like that! Are you a man or a mouse? You're a man. You go after Bonds and hope for the best.
But not only is it the macho thing to pitch to Bonds, it's also the smart thing, as Sky Kalkman ably demonstrated in his response. If you walk him...
Assuming a below-average hitter next with a .300 OBP, you've got a 30 percent chance of losing the game immediately and a 50 percent chance of winning in later innings. The other 70 percent of the time, you've got a 50 percent chance of winning in later innings.This yields a 35 percent chance of winning the game if you walk Bonds. On the other hand, if you pitch to Bonds...
In 2004, when Bonds wasn't intentionally walked, he did this...Which all adds up to about a 58 percent chance of winning the game if you pitch to him.
If Bonds makes an out, you win 100 percent of the time. With a walk, it's the same as the scenario above. With any hit, let's assume two runs score and you lose—that happens 27 percent of the time.
Clearly, the odds dramatically favor going after the batter while clinging to the one-run lead, even if that batter is the almost-superhuman Barry Bonds of 2004. But maybe, just maybe, there's some scenario where it would make sense? If Barry Bonds were playing against a bunch of Little Leaguers, and a 90-year-old grandma was painstakingly making her way into the on-deck circle with her walker and a Louisville Slugger, maybe then?
Well, probably. Assuming grandma's good, old-fashioned ash bat isn't quicker than you think, and the umpire doesn't squeeze the zone out of sympathy for grandma's advanced age, that only ups your chances of winning by walking Bonds to about 50 percent. Even in this scenario, Bonds needs to up his batting average against our theoretical Little League pitcher to .500 in order to even the odds.
In any real major league game, the on-deck batter isn't going to be a .000 hitter, and if he is a very poor hitter, the manager will surely pinch hit for him. Moreover, even in the clutch, the best Hall-of-Famer doesn't hit .500. My favorite clutch hitter, the guy any Royals fan would want at the plate with the game on the line, George Brett, only went 2-for-6 in his career with runners in scoring position and two out in the ninth inning with the Royals trailing by one run. You're not going to find a remotely realistic major league combination where it makes sense to sacrifice the tying run in hopes of winning the game in extra innings.
In fact, when you look back through the Retrosheet era extending back to 1956, no manager has ever intentionally walked in the tying run with two out and the bases loaded in the bottom of the ninth. That's unsurprising; not only does it not make much sense according the numbers, but no manager would want to face the media or the fans after ordering such a move. But maybe some managers or pitchers decide that they need to pitch around certain tough batters in that situation, hoping to get them to bite on a bad pitch but being willing to let them walk and force in a run rather than give in and give them a pitch over the plate. We can check for this.
From 1956 through 2008 at the All-Star break, pitchers have faced 483 batters with the bases loaded, two out, and clinging to a one-run lead in the bottom of the ninth inning. They walked 32 of these batters and hit another four batters with a pitch. First of all, it's interesting to see how those situations turned out. When they put the batter on base to force in the tying run, 13 of those 36 times, they lost the game on the next batter. Five of those losses were as a result of yet another walk or hit batsman, seven losses were on base hits, and one came on a passed ball (to Phil Mankowski, who was pinch hitting for rookie Alan Trammell. That's probably one of the few times the Tigers ever won a game by pinch hitting for Alan Trammell.)
That still leaves 23 times that the team that walked home the tying run was able to escape the ninth inning jam. In those 23 games, the team that gave up the tying walk won 12 of them in extra innings, lost 10, and...well, tied one. I didn't think that baseball had ties, but as best I can tell, the result from the 12+ innings played by the New York Yankees and Kansas City A's on June 15, 1960, did not count for anything in the standings.
According to the note on Retrosheet, "Rain broke out in the bottom of the 13th at 11:57 PM and after a 30-minute wait the game was called. The game reverted back to the last complete inning, the 12th. The game was rescheduled for 08/07/60 as part of a doubleheader. Stats from 13th inning don't count." On August 7, 1960, the Yankees and A's played two complete nine-inning games, seemingly ignoring the 12 innings played on June 15. So we'll count the total ledger for teams walking in the tying run with two out in the ninth as 12-23-1, or wins in somewhere between 33 and 35 percent of those games, depending on how you want to count the tie game.
What happens when the pitcher doesn't walk or hit the batter? In those 447 at-bats, the batters got 92 base hits, for a batting average of only .206. This figure may seem low until you peruse the list of pitchers involved, a Who's Who of relief aces extending from the Brooklyn Dodgers' Clem Labine in 1956 to the Los Angeles Angels' (of Anaheim) Francisco Rodriguez in 2008. Two games also ended with two runs scoring on a fielding error. Of the 92 base hits, 12 were singles that scored only one runner, either because the second runner was thrown out while attempting to score (four times) or because only one runner made for home. In addition, two batters reached on an error that scored only one run, and one batter reached after striking out on a wild pitch.
In these 15 games that were extended past the original batter on something other than a walk or a hit batsman, four times the pitching team lost the game to the next batter in the ninth, and in extra frames, the pitching team went 6-5. The total record for teams that didn't walk in the tying run with two out in the ninth is 356-91, which equates to winning 80 percent of the time.
On average, a team is more than twice as likely to win by going after the batter rather than walking him. However, what about those cases where the batter at the plate is a superior one? Several Hall of Famers have been walked in the two-out, bases-loaded, bottom-of-the-ninth situation. The Astros' Larry Dierker walked Joe Morgan on August 4, 1973 in order to face the Reds' third baseman Dan Driessen. Dierker got Driessen to line out to shortstop, but in extra innings, the Reds' bullpen couldn't escape Morgan, who singled to drive in the winning run in the 11th inning.
On April 18, 1979, the Giants' Gary Lavelle walked Dave Winfield to force in the tying run and bring the Padres' first baseman Mike Hargrove to the plate. The Retrosheet game notes inform us that Gary Lavelle was ejected with a 3-1 count to Hargrove. Either Hargrove so frustrated Lavelle by living up to his nickname, or Lavelle and the plate umpire didn't exactly see eye to eye on the size of the strike zone. Joe Coleman replaced Lavelle on the mound and issued ball four to Hargrove, ending the game.
In a game between the Indians and Twins on July 20, 1996, Eddie Murray was issued a game-tying walk by Dan Naulty, who instead faced Indians' second baseman Carlos Baerga, who lined out to shortstop. However, the Indians went on to win the game in the 11th inning on a solo home run by Alvaro Espinoza.
So, the three times that a Hall of Fame batter was walked with two out in the bottom of the ninth to force in the tying run, it didn't work out well for the visiting team. More often, however, a Hall of Fame hitter gets to swing away and end the game one way or another. Duke Snider, Ernie Banks, and Carlton Fisk all struck out in their chances. Ted Williams, Dave Winfield, and Gary Carter grounded out. Lou Brock and Brooks Robinson flied out. Luis Aparicio popped out and flied out in his two chances, Al Kaline grounded out and flied out, Robin Yount flied out and struck out, and Hank Aaron flied out twice. Rod Carew singled to win the game, as did Nellie Fox in both of his two opportunities. That's a total of three hits in 19 at bats, for a batting average of .158. Hall of Famers don't appear to be fearsome in that situation at all, and certainly not worth walking at the cost of giving up the lead.
It's worth noting, however, that being a Hall-of-Fame slugger like Murray, who has 504 career home runs to his credit but only a .287 batting average, is not necessarily as troublesome to pitcher in our situation as facing a lesser slugger with better plate coverage and a higher batting average. A single ends the game almost as often as a home run, and a loss counts just the same whether it's by one run or three. Instead of Hall of Famers, what if we look at hitters with at least a .295 career batting average?
|Eric Gagne considers offering his glasses to the umpire after ball four to Miguel Tejada loads the bases in the bottom of the ninth inning, May 4, 2008 (Icon/SMI)|
That gives us what is probably a better list of tough hitters for this particular scenario: Ted Williams, Albert Pujols, Rod Carew, Derek Jeter, Magglio Ordonez, Howie Kendrick, James Loney, Don Mattingly, Ralph Garr, Alex Rodriguez, Hank Aaron, Bill Madlock, Hal Morris, Manny Mota, Lance Berkman, Al Oliver, Mike Greenwell, Juan Pierre, Mike Sweeney, Kenny Lofton, Dante Bichette, Bobby Abreu, Minnie Minoso, Jim Rice, Garret Anderson, Julio Franco, Matt Kemp, Joe Torre, Gary Sheffield, Al Kaline, Hideki Matsui, Jeff Bagwell, Mark Loretta, Manny Sanguillen, Duke Snider, John Olerud, Bob Nieman, and Barry Larkin.
In 46 plate appearances, there were four walks and two hit batsmen which resulted in three wins and three losses for the pitching team, including the May 4, 2008 game where Eric Gagne walked Lance Berkman to force home the tying run. (Hunter Pence won the game for the Astros with a home run in the 12th inning.)
In the other 40 at bats, these batters produced 12 hits, for a batting average of .300, and a win-loss record of 29-11 for the pitching team. Clearly, the batting average of the batter matters, but even with these excellent batters at the plate, it was still a losing proposition to put them on base and tie the game.
What if we had a tough hitter at the plate, say someone of the class of Williams, Pujols, or Carew, and the next batter was practically helpless, a pitcher or someone of similar ability. What about then? We've already covered the theory that says this is a losing proposition, but have any such scenarios occurred historically?
The closest I could find was the game between the Red Sox and Twins on May 19, 1986. With the bases loaded and two out, Jim Rice (career batting average of .298) came to the plate with the Red Sox trailing 7-6. Twins' pitcher Ron Davis walked Jim Rice to knot the score at 7-7. In the eighth inning, Red Sox manager John McNamara had pinch run for starting catcher Rich Gedman, so in the ninth inning, backup catcher Marc Sullivan was forced to hit with the game at stake. For his career, Sullivan was a .186 hitter with a .236 on-base percentage. Davis hit Sullivan with a pitch to force in the winning run, so we'll never know what would have happened had Sullivan been forced to swing the bat.
Walking Rice to face the weak-hitting Sullivan is one of only two cases where the pitcher walked in the tying run to face a hitter with a career OBP lower than the career batting average of the original hitter. The other case occurred on April 19, 2006, when Lew Ford (career average of .272) was walked to face Juan Castro (career OBP of .269). Castro grounded out to end the threat in the ninth, but Michael Cuddyer homered in the tenth to win the game for the Twins.
You may wonder, then, if the numbers lean so heavily in favor of pitching to the batter, even if the batter is a tough out, why do pitchers walk anyone? Are they trying to be too fine around the corners? Or are we simply seeing the limit of pitchers' ability to control pitches?
With the bases loaded and the pitcher protecting a one-run lead with two outs in the bottom of the ninth inning, seven percent of the 483 plate appearances resulted in a walk or a hit batsman. For comparison, let's look at the situation where the game is tied with the bases loaded and two outs in the bottom of the ninth. In that situation, a walk or a hit batsman loses the game for the pitcher's team. Still, 12 percent of those 982 plate appearances resulted in a walk or hit batsman. The pitchers may be trying to be too fine around the corners, but they don't appear to be doing it because they think they can afford a walk. They do it even when they clearly cannot afford a walk.
Let's close with a great example of how to go after a tough hitter with the game on the line.
On September 16, 2007, the Boston Red Sox and New York Yankees faced off in the rubber game of a three-game series. The Yankees were on a roll in the second half of the season, charging back from a poor 21-29 start to take the lead in the wild card chase and had closed within 5.5 games of the Sox in the Eastern division. In this game, Derek Jeter hit a three-run home run in the eighth inning to give the Yankees the lead, and Mariano Rivera was on the hill to protect the 4-2 lead in the ninth inning.
Rivera started the inning by walking Jason Varitek, but he got Eric Hinske and Coco Crisp on ground balls. However, Lugo followed with a double, scoring Varitek, and Rivera hit Jacoby Ellsbury with a pitch and walked Dustin Pedroia to load the bases and bring up none other than David Ortiz. In the on deck circle was Red Sox third baseman and cleanup hitter Mike Lowell, sporting a .326 batting average and .382 on base percentage. Yankees manager Joe Torre visited the mound. "If you get this guy out, then we win the game."
"We were thinking to go and attack him—attack him in," Rivera said. "That's my best pitch, and that's his powerhouse. We just went strength to strength. There's nowhere to put him. But to face him, you have to give him everything that you've got."
Rivera started Ortiz with a cut fastball up and over the plate, and Ortiz fouled it back for strike one. Rivera followed with a fastball, missing the strike zone just as he had with that pitch to Pedroia earlier in the inning. Rivera then challenged Ortiz with three straight cutters inside, the first of which missed high for ball two. Ortiz grounded the next cutter foul down the first base line, breaking his bat and evening the count at 2-2. Finally, Ortiz swung at the perfect cutter, up and in, just off the black, and lifted a soft fly which settled into Derek Jeter's glove out behind second base to end the game.
At the end, we've come to the conclusion that was fairly obvious from the beginning—with the bases loaded and clinging to the slimmest of leads, a pitcher should still go after that tough hitter, even if it's Barry Bonds, Ted Williams, or David Ortiz, and good things will happen more often than not.
References and Resources
Historical game data is courtesy of Retrosheet and was parsed using the Baseball-Reference Play Index. Pitch trajectory information came from the Sportvision PITCHf/x data available through MLB Gameday.
Mike Fast is a Royals fan who enjoys investigating baseball questions using data of many sorts. He is a member of Complete Game Consulting. He welcomes comments via e-mail.