More on Strategery
On Sunday, Walt Weiss showed once again that he’s willing to buck The Book. The Book, that apocryphal collection of managerial moves one violates at one’s public peril1 contains many verses, and is definitely in flux with regards to, for example, bunting, but there are few more solid entries that: “Don’t intentionally put the winning run in scoring position.”
Weiss, on Sunday, intentionally put the winning run in scoring position. With two outs, leading by one in the bottom of the 11th, he ordered an intentional pass to load the bases, putting the speedy Leody Taveras on second where a single would win the game for the Orioles. The next batter, Sam Huff, validated the strategy with a groundout to end the game.
But how unusual is this, and how often does it pay off?
I took every example in baseball history (except this year) where the visiting team led by one after the eighth inning in which there were either men on first and second or first and third, so that a walk would put the winning run in scoring position. There is one other possibility, in which a team leads by two with the bases loaded and intentionally walks someone, which has happened, famously, once. See below for details. Note that I exclude cases where the teams are tied. I did so for two reasons: first, there could be legitimate strategic reasons in a tie game to force the opposite manager into make a substitution, for example. But where the road team is leading, such issues dwindle. All you need to do is get out of the inning with no more runs scoring to win.
So let’s look at the raw data. Here’s one breakdown:
| book | case | outs | Loss | Tie | Win | Total |
|---|---|---|---|---|---|---|
| Book | 1st and 2nd,1st and 3rd | 0 | 114 (22%) | 218 (42%) | 182 (35%) | 514 |
| Book | 1st and 2nd,1st and 3rd | 1 | 993 (44%) | 691 (30%) | 584 (26%) | 2268 |
| Book | 1st and 2nd,1st and 3rd | 2 | 2594 (76%) | 436 (13%) | 387 (11%) | 3417 |
| Book | 1st and 2nd,1st and 3rd | Total | 3701 (60%) | 1345 (22%) | 1153 (19%) | 6199 |
| Book | Bases Loaded | 0 | 61 (69%) | 27 (30%) | 1 (1%) | 89 |
| Book | Bases Loaded | 1 | 232 (70%) | 72 (22%) | 26 (8%) | 330 |
| Book | Bases Loaded | 2 | 524 (86%) | 53 (9%) | 31 (5%) | 608 |
| Book | Bases Loaded | Total | 817 (80%) | 152 (15%) | 58 (6%) | 1027 |
| Book | Total | Total | 4518 (63%) | 1497 (21%) | 1211 (17%) | 7226 |
| No Book | 1st and 2nd,1st and 3rd | 0 | 0 (0%) | 1 (100%) | 0 (0%) | 1 |
| No Book | 1st and 2nd,1st and 3rd | 1 | 2 (29%) | 1 (14%) | 4 (57%) | 7 |
| No Book | 1st and 2nd,1st and 3rd | 2 | 11 (92%) | 1 (8%) | 0 (0%) | 12 |
| No Book | 1st and 2nd,1st and 3rd | Total | 13 (65%) | 3 (15%) | 4 (20%) | 20 |
| No Book | Bases Loaded | 2 | 1 (100%) | 0 (0%) | 0 (0%) | 1 |
| No Book | Bases Loaded | Total | 1 (100%) | 0 (0%) | 0 (0%) | 1 |
| No Book | Total | Total | 14 (67%) | 3 (14%) | 4 (19%) | 21 |
So let’s start with the fact that the Book is the Book. There are 7,226 times when managers pitched to the next batter, and 21 times when they issued an intentional walk to that batter, moving the winning run into scoring position. 21/7,226 is 0.3% of the time: once every 333 times. Indeed, there are so few cases of managers going against the book, that it’s really impossible to even make any confident statements about the utility of The Book.
A preliminary note: I take no notice of the statistics of the player who is walked. We’ll look at it shortly, but since the opposing manager has the opportunity to use a pinch hitter, we can presume he used the best hitter available facing the winning run on second base.
The column Loss is from the standpoint of the batting team. So we see that managers who follow the book win 63% of the time and lose in that inning 17% of the time. In the remaining cases (Tie) the home team ties the game and another inning ensues. (What happens then is very interesting to me, and I’ll cover it in a later piece.) That 63% conceals big differences based on the out situation: if there are no outs and the 1st and 3rd or 1st and 2nd, the home team who is down 1 loses only 22% of the time in the 9th. But the home team loses 76% of the same games with two outs. (It should be noted that this is a count of situations, not games, so a 1st and 3rd with no outs can become a 1st and 3rd with 1 outs and then a 1st and 3rd with two outs, and represent three entries in the database. In this case, the manager had three different chances to violate The Book.
What About the 21 Times The Managers Ignored The Book?
I don’t know how old The Book is, but the first time in the data when a manager transgressed The Book was at Comiskey Park on April 22th, 1917. Leading 2-1 in the bottom of the 9th, St. Louis Browns manager Fielder Jones ordered an intentional pass to Shoeless Joe Jackson, moving Eddie Collins to 2nd with Swede Risberg already on 3rd. Happy Felsch (pictured above) came to the plate and hit a walkoff single to win the game. This is exactly the nightmare scenario which created the Book rule. But here’s the funny thing: it never happened again in the other 20 times. There were teams that tied the game (3 of them) and three other teams that walked off later in the inning, but no one ever just hit a single to win the game as the next batter except Happy Felsch.
The most famous of these games was undoubtedly this game in which Buck Showalter intentionally passed Barry Bonds with a two run lead and the bases loaded. Brent Mayne then lined out to preserve the 9-8 win for the Diamondbacks. Overall, the managers who went against the book won 67% of the time, slightly better than the 63% of the Book-followers, but there isn’t nearly enough data to suggest that this superiority is anything other than random. (Weiss’ ploy raised the win percentage to 68%.) They also lost slightly more of these games 19% vs 17%, but that is even a less significant difference.
Of the 22 violators (counting Weiss) 6 were in extra innings, one of them in the 18th inning, yet another Diamondbacks-Giants game in which Bob Brenly made a really bold move having taken a 1-0 lead in the top of the 18th. After a double (by pitcher Ryan Vogelsong) and a walk to Rich Aurilia, Brenly pitched to Bonds and got him to ground into a fielder’s choic making it first and third with one out. He then walked new Hall of Famer Jeff Kent to move Bonds into scoring position, but got Armando Rios and Benito Santiago to end the game.
Finally, rare as this managerial move is, four of them (including Weiss’) have happened since the Manfred Man rule was instituted in 2020. The variability induced by the extra inning rule is so extreme that taking chances to end the game in regulation time may have altered The Book a little.
The Game
I did not watch this game. I haven’t missed many this year, but I missed this one. The Braves won 6-2 on the strength of a four run sixth highlighted by a 3-run homer from MHII. Olson jacked one in the first as well. On the pitching side, Bryce Elder once again did not have a complete game, although he pitched quite well, but inefficiently. 93 pitches in 6 innings.
A big pitching story of this game was three excellent innings of work from Elieser Hernández. Because the rules are porously written, by the time Elieser is able to read this, he will be the property of the Gwinnett Stripers. This is a stupid, stupid manipulation that disserves the players and the fans, saving, over the course of a year, a few million dollars for owners. It’s not my few million dollars, so I understand why the CBA is written the way it is. But I am free to denounce its stupidity. I’m not sure Elieser Hernández deserves a place on a major league roster, despite the fact that his games thus far have been excellent. But if he’s good enough to bring up, he’s good enough to keep on the roster for at least a week. Having three or four rotating pitchers occupying one nominal MLB pitching slot is petty. Stop it.
- As opposed to the master sabermetric text, The Book: Playing the Percentages in Baseball by Tom Tango, Mitchel Lichtman and Andrew Dolphin which is a little dated now, although the methodologies are still really important. ↩︎
Recent Comments