On an Intentional Ball Thrown to Albert Pujols

Monday night and Tuesday morning, the A’s and Angels played a game for the ages, a 19-inning affair that saw the hosts rally in the ninth and the 15th before walking off in the game’s seventh hour. The game featured 18 runs and nearly 600 pitches, and in the end, the A’s improved to 15-12, while the Angels fell to something much worse than that. Generally, such games are immediately thought of as turning points, and generally, such games don’t go on to work that way. But this was a game that few will forget, regardless, simply because the duration grew to be so extreme.

FanGraphs isn’t in the business of issuing game recaps, particularly several hours after the fact. But still, some attention to the game should be paid, and I’m electing to focus on a particular intentional ball. With two out in the top of the 11th inning, Grant Balfour intentionally walked Albert Pujols. We consider the 3-and-0 pitch.

Already, these were unusual circumstances. In extras, the Angels had a runner in scoring position, and the A’s had a righty pitcher walk a righty superstar in order to face a lefty superstar. Based on his body language, Balfour wanted to come right after Pujols, but the decision was stripped from his hands, and so four balls he threw. The first was high and away, wild. The second and third were more on the target. Of the intentional balls, here is the fourth:


The pitch clipped the outer edge of the strike zone. If you even want to call it a “pitch”. Of course, it was too high to be an actual strike, and Pujols took his base, but intentional balls are supposed to be intentional balls, not balls that come dangerously close to being strikes. This pitch was a little under five feet off the ground, in the vicinity of being over the plate, and you can see that it’s below Pujols’ shoulders when he’s standing. It was 31.6 inches from the center of the strike zone.

In the past, I’ve joked about how pitchers are pretty bad at locating their pitches. In truth, they’re incredibly good, once you account for the difficulty of the task, but if you expect pitchers to hit their targets right on, far more often than not you’re going to be left underwhelmed. Pitchers miss. Pitchers miss, even when they really don’t want to miss. To demonstrate the point, below please find screenshots of pitches from the game that were further from the center of the strike zone, despite not being in counts in which a pitcher would want to get a hitter to chase. These pitches were thrown in counts in which the pitcher would’ve wanted a strike, or at least an almost-strike. They missed by more than an intentional ball.


Jerome Williams
-12th inning


Brandon Moss led off and this happened with the count 1-and-1. Williams came inside with a cutter that somehow didn’t hit Moss in the foot. It did hit Kerwin Danley in the foot. The pitch was 32.8 inches from the center of the zone.


-Jerome Williams
-12th inning


The 3-and-1 pitch, in the same plate appearance. In extras, you don’t want to walk the leadoff guy, so Williams would’ve been badly trying to throw a strike. The pitch was 34.4 inches from the center of the zone.


Dan Straily
-3rd inning


The first pitch to Peter Bourjos nearly drilled him in the head. This was a breaking ball, not a fastball, so it probably just slipped out of Straily’s hand, but that’s not an excuse. This is a breaking ball that still missed the center of the zone by more than an intentional ball. It was thrown in a count in which Straily would’ve wanted to get ahead. The pitch was 34.6 inches from the center of the zone.


Michael Roth
-7th inning


Now, this is an 0-and-1 count, so the zone can expand a little. But Roth buried a changeup in the dirt, which wasn’t at all the intent behind the pitch. Afterward, Roth looked at his cleats and at the mound, as if he stumbled. Maybe the mound was to blame? But Roth didn’t stumble on the other pitches. Roth’s mechanics, therefore, were to blame. Which is always the case when a pitcher badly misses his spot. The pitch was 41.1 inches from the center of the zone.


-Dan Straily
-2nd inning


Again, all right, I get it, 0-and-1 count. Aggressive hitter at the plate. It makes sense to try a low, away breaking ball. But this pitch bounced in the middle of the opposite batter’s box. This pitch was a tiny little catastrophe, and Straily shelved his slider for the remainder of the at bat. The at bat concluded with the hardest-hit home run of the young season. The pitch was 47.5 inches from the center of the zone.

Pitchers miss, and they miss a lot. These have been examples of pitches that missed the center of the strike zone by more than an intentional ball, which by design should be really far away from the center of the strike zone. Before we wrap up, let us consider one more thing. This will be something of a thought experiment, and no conclusion will be reached. Let’s look at that 3-and-0 intentional ball to Pujols again:


The pitch is within Pujols’ hittable range, so to speak. Pujols could have made contact, had he swung, and the pitch was a slow fastball at 74 miles per hour. The Oakland broadcast joked that the pitch was almost a strike, and that Pujols could’ve done damage had he been ready. As the screenshot shows, it didn’t even cross Pujols’ mind to attempt a swing, because he’s already on his way to first base before Derek Norris gloves ball four. Pujols was content to just take his walk, because at least walking provides temporary relief from running. And how often does one get thrown hittable intentional balls, anyway? How much value is there in being ready, just in case?

One is instantly reminded of Miguel Cabrera hitting an intentional ball against the Orioles. Granted, that pitcher was lobbing the ball in, but the pitch arrived in a somewhat similar location as the pitch from Balfour to Pujols. Cabrera stood up and smacked a crucial line drive. This all made me wonder: what would be the expected BABIP of swings at intentional balls?

We can’t measure anything in practice, because we have so few examples. Seldom do intentional balls end up near the zone, and seldom are hitters looking to swing. There’s a reason we still remember Cabrera after all these years. It seems to me there would be three factors. For one, the pitcher isn’t trying. There’s no deception, and intentional balls tend to be more or less similar in terms of velocity and movement. The hitter, conceivably, could time the pitches and anticipate the pitches. The hitter could try to put on a good swing.

Then there’s the matter of the defense not being prepared for a ball in play. Look at the first baseman as the pitch is on its way to Cabrera:


He’s watching, but he’s not ready to play defense. Presumably, none of the defenders were prepared for a ball in play, and they were all stunned by Cabrera’s swing. By watching, defenders would be ready to react, but they’d be delayed in their reactions, opening up more of the field. More ground balls could sneak through. More fly balls could drop or find gaps. The defenders’ zones would “play smaller” because they wouldn’t get good first steps. When there’s an intentional walk taking place, just about everyone tunes out, and it’s perfectly understandable. That’s the element of surprise that Cabrera was going for.

On the other side of the coin, intentional balls almost never end up within the actual strike zone itself. They can just get close, such that a hitter could make contact if he swung and maybe moved his feet. As we know, BABIP goes down when hitters swing at pitches out of the zone, and it’s not like hitters would be taking their best swings. You’ll probably never see an intentional ball get hit for a home run, or even a ball off the wall. A hitter would be going for a line drive, or maybe a sharply-hit grounder. With reduced swing quality, it stands to reason there would be reduced BABIP, although we can only guess at the effect.

I don’t have any idea what the expected BABIP would be in these situations. I wouldn’t know where to begin. It seems to me like it’d be worth it for hitters to be prepared to swing, just in case, because a hit’s more valuable than a walk. But is a possible hit more valuable than a guaranteed walk? If most of the hits would be singles, how many would need to drop in in order for the trade-off to be worth it? I think we’d all like to see more surprise swings at lazy intentional balls, just because we’re all fans of baseball chaos, but it’s entirely possible that would reflect a suboptimal process. You’d need to be pretty damn sure of your odds of getting a hit, and I just don’t know how sure one could be.

Print This Post

Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

34 Responses to “On an Intentional Ball Thrown to Albert Pujols”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Eminor3rd says:

    Didn’t Andres Gallaraga do that once, too?

    Vote -1 Vote +1

    • Kenny says:

      He did, but I believe he only got a single through the hole and I think it didn’t score a run. (Probably because it caught the runner off guard too.)

      Vote -1 Vote +1

  2. Rick says:

    “because he’s already on his way to first base before John Jaso gloves ball four.”
    That would be Derek Norris gloving ball four.

    Vote -1 Vote +1

  3. Scott says:

    Vlad as well perhaps?

    Vote -1 Vote +1

  4. Shuman says:

    There’s this video: https://www.youtube.com/watch?v=2_b9TCojRis where an IBB in a little league game results in a HR.

    Vote -1 Vote +1

  5. Al Dimond says:

    One example is famous in Cubs lore: on May 30, 1956 (per Wikipedia) light-hitting backstop Harry Chiti tripled while the pitcher tried to intentionally walk him.

    Vote -1 Vote +1

  6. Bip says:

    Without going through the effort required to calculate the percentage dynamically given a certain base-out state, Here’s an example of one such caculation.

    Let’s say it’s -23 with 2 out, which is a very common situation to see a IBB. The RE of -23 2out is .604, and the RE of 123 2out is 0.798. In terms of run expectancy, the pitcher is giving up 0.798-0.604=0.194 runs by issuing this IBB. Let’s assume that the only kind of a hit a hitter can get in this situatoin is a single. So, for reference, if a single in this situation was also worth .194 runs, then the batter would have to expect a BABIP of 1.000 in order to justifiably try to hit an intentional ball.

    I couldn’t find the run value a particular hit per base-out state, so let’s make some assumptions and find one. Assume the runner on 2nd is slow and he will only advance to 3rd on a single. So in this case, a single scores one run (runner on 3rd) and changes the base-out state to 1-3 2out. 1-3 2out has a RE of .512, so the value of a single is 1+0.512-0.604=0.908 runs. The value of an out is -0.604 runs. However, our baseline is the IBB, since the batter can take that value guaranteed.

    -0.604-0.194 = -0.798
    0.908-0.194 = 0.714

    So succeeding is effectively gaining 0.714 runs and failing is effectively losing -0.798 runs. Since we’re already using the IBB as a baseline, we only want to find the break-even point between these two values.

    (0.714)*x + (-0.798)*(1-x) = 0
    x = 0.528.

    So if the only two outcomes of swinging at a IBB are single or out, and the runner on 2nd won’t advance on a single, the hitter would have to expect a BABIP of .528 or better on the swing in order to make the risk worth it.

    Vote -1 Vote +1

    • Bip says:

      Sorry, make that “the runner on 2nd won’t score on a single”

      Vote -1 Vote +1

      • TKDC says:

        most singles with 2 outs end up with the runner on second scoring and I think the couple of tenths of a second that you’d probably have in a delay from the outfielder would make it even more likely. Also, ignoring the possibility of a better result than a single does in fact really diminish this.

        All that said, I think it is stupid to not take your walk unless maybe it is bottom 9 (or extras) and you need one run to win (in that case, you really are just looking at the BAPIP for that hitter on that pitch versus the OBP expected from the next hitter – the run expectancy, as it is an average, is useless).

        Vote -1 Vote +1

        • TKDC says:

          And now that I think about it for 30 more seconds, it would likely be a big advantage to give it a shot if you are down 1 bottom 9 or later, as you would likely not only tie the game, but win it with a hit, whereas otherwise you would either need the next guy to get a hit, or the next 2 to both get on base in order to win (at least during that inning).

          Vote -1 Vote +1

        • Bip says:

          The very point of the calculation though is that if the break-even point required a low enough BABIP, then swinging is worthwhile in a typical situation where every run counts.

          Your points about the likelihood of the runner on 2nd scoring and the possibility of extra base hits would mean the necessary BABIP is actually lower than I said, strengthening the case for swinging at any time. If you can expect to score more runs that way, then why not do it?

          Vote -1 Vote +1

    • Travis L says:

      I applaud your efforts, but think the limitations you placed to simplify calculations affect the result too much for it to tell us very much. Unless you were to interpret it as, “as long as the hitter thinks he can have a .52% chance of getting a hit, it’s always worth swinging.”

      Vote -1 Vote +1

      • Bip says:

        No, my post is supposed to illustrate the framework by which minimum required BABIP can be calculated. Simply substituting in situation-specific values for base-out state will give the desired results.

        Also, since an IBB will only be issued in certain similar base-out states to the one above, and the value of a walk in terms of runs doesn’t vary too much, then my result serves to give a rough idea of the magnitude of BABIP we’re dealing with. If my test had yielded a BABIP of .330, we could seriously discuss if this warrants further research, whereas a result of .800 means probably not.

        Vote -1 Vote +1

    • philosofool says:

      The idea here is cool, but the main limitation is that you’re assuming a team would walk the bases loaded with 2 outs to a hitter with a 0 chance to do more than single weakly.

      Vote -1 Vote +1

      • Bip says:

        That assumption was based on Jeff’s statement that a an intentional ball probably won’t be hit with much power, since it’s so far out of the strikezone.

        Vote -1 Vote +1

  7. Rick says:

    The win probability graph of this game is epic. The A’s were down to a 1.2% chance of winning at one point.

    Vote -1 Vote +1

  8. Impossibles says:

    No reference to Mr. Baseball or Tom Selleck’s moustache, no care.

    Vote -1 Vote +1

  9. UncleCharlieVT says:

    Manny definitely did this, and I think it was against the A’s.

    Vote -1 Vote +1

  10. J6takish says:

    Watched that Cabrera clip again, Hanley scored from 2nd on a single to center. Did he know Cabrera was going to swing? Hard to see him making it home if he was caught off guard

    Vote -1 Vote +1

    • Soledad says:

      You can see Hanley extending his lead as the pitch is thrown, but definitely not aggressively, rather he was walking it out a bit. However base runners are undoubtedly thinking about the possibility of a wild pitch in this situation as it is very common for these pitches to be very poorly thrown. Here is an example of the Mariners beating the Marlins back in 2011 on a wild pitch .

      Vote -1 Vote +1

  11. asa lasky says:

    I seem to remember when I was a wee lad, my favorite player Willie McCovey slapped a double while being “walked”. Maybe it was only a dream.

    Vote -1 Vote +1

  12. dannyrainge says:

    I am salivating over the potential Miggy IBB hit gif. Someone make it true.

    Vote -1 Vote +1

  13. Jason H says:

    “That’s the element of surprise that Cabrera was going for.”

    …actually, Cabrera is just batshit crazy (but in a good way in this case).

    Vote -1 Vote +1

  14. Hurtlockertwo says:

    I think Johnny Bench got an intentional strike in the WS when the other team faked an intentional walk.

    Vote -1 Vote +1

    • Robin says:

      Game 3 of the 1972 WS. With a 3-2 count the Oakland A’s set up for an intentional walk. Tenace jumped back behind the plate and Rollie Fingers struck out Bench. There was quite a bit of play-acting by either the manager or the pitching coach who went out to the mound, gesturing toward the open bag and the guy on deck. Great stuff.

      Vote -1 Vote +1

  15. badenjr says:

    “I think we’d all like to see more surprise swings at lazy intentional balls, just because we’re all fans of baseball chaos, but it’s entirely possible that would reflect a suboptimal process.”

    I think this is a case where context is everything. A pitcher doesn’t throw intentional balls to a batter for no reason. These are almost always going to be situations where a hit (or even a “productive out”) produces one or more runs – often at a point in a game where one run is critical. If playing for one run is sufficient, I’d imagine that having a (generally good) batter swing unexpectedly at a lobbed, hittable pitch with no movement to speak of would be very likely to produce the desired outcome for the offense (even if coupled with an out), and I’d be very surprised if it, on average, did turn out to be “sub-optimal” from the perspective of bringing in that key run.

    Vote -1 Vote +1

  16. Doug Lampert says:

    My father has a theory that batters should ALWAYS swing on 3-0 of an intentional walk. Force the pitcher to throw a deliberately bad ball one more time, force the catcher to actually catch one more pitch that’s no where near the strike zone.

    What are they going to do? Decide to strike you out now that it’s 3-1 rather than the 0-0 it was when they first decided to give you a free pass? You’re still well ahead in the count.

    Vote -1 Vote +1

    • Snag says:

      Only problem with that, the pitcher might get pissed and decide to put the batter on base via HBP.

      Vote -1 Vote +1

      • Doug Lampert says:

        And since it will obviously be deliberate if he does it in that at bat he’ll be fined and out of the game (at the minimum) after trying it.

        Mind, the batter might be in trouble the NEXT time he faced that pitcher.

        Vote -1 Vote +1

  17. Phil Livingston says:

    I may be way off the mark here, but wouldn’t the definite walk be better than a potential base hit? I guess the situation would dictate this a little but still… A guaranteed base or possible base hit/home run?

    Vote -1 Vote +1