The Missing Two Percent
With two outs in the top of the sixth inning in a tied game and runner on second, a manager elects to intentionally walk a right-handed batter with his right-handed starter in order to have that same starter face another right-handed batter. Two singles follow, putting the manager’s team down by two, leading the team to defeat. An intentional walk leading to bad things for the pitching team is hardly a novelty or surprise, but the characters involved make it a bit more interesting.
This happened yesterday. The manager was the Rays’ Joe Maddon, and the batter was the Royals’ Jeff Francoeur. Jeff Francoeur’s game (and lack thereof) has been dissected and discussed to the point of pointlessness. I have made plenty of contributions to the field, so there is no need to belabor that point. Joe Maddon has a pretty good reputation as a manager, but analyzing any manager’s abilities as a whole is difficult for a variety of reasons. This particular sequence struck me as odd, particularly given the Rays’ reputation for trying to gain every little advantage they can.
When Maddon called for the free pass, first base was open after former Ray Elliot Johnson hit a two-out double. Francoeur came to the plate. One might think that there is never a situation in which one would walk Francoeur, who (after Sunday’s game) is hitting .214/.257/.355 (57 wRC+) and, even taking regression into account, is only projected to be slightly better (both ZiPS and Steamer rest-of-season projections have him as an 83 wRC+ hitter). Francoeur had hit a home run in his previous plate appearance in the game, but it is hard to imagine that in itself would make that much of difference given his general futility.
It might have made sense if Francoeur was hitting in front of a pitcher, or if the pitching team had a lefty that they want to leave in the game for the next batter. (Francoeur, like almost all right-handed hitters, hits lefties better, although his abilities in that respect tend to be exaggerated.) However, the former Fausto Carmona is a right-handed pitcher (who is much better against righties than lefties himself), and the batter immediately after Francoeur, Alcides Escobar, hits right-handed as well.
Escobar is a pretty terrible hitter himself, but not bad enough to justify intentionally walking Francoeur to get to him. After a decent season (for a shortstop) with the stick in 2012, Escobar is back to coming to the plate with the noodle in 2013. He has not been as bad as Francoeur this year, but bad: .247/.273/265 (63 wRC+). Now, to be fair, he is projected to be a bit worse than Francoeur: ZiPS and Steamer have his rest-of-season wRC+ at 75 and 73, respectively. Without getting all the charts and running the platoon numbers as I have in the past for this sort of thing, let’s simply say that the difference is not nearly close enough, given the game situation, to make this a “guy call.” Escobar has been having been grounding into double plays at a high rate this year, but even if one thinks that is his true talent, there were two outs, so it would not have mattered.
While any number of things could have happened after the intentional walk, what actually happened was that Escobar singled in Johnson, then, after Maddon brought in Jake McGee, Alex Gordon singled in Francoeur to put the Royals up 4-2, a lead they would not relinquish. In this case, at least, bad tactics immediately begat a bad outcome.
This rang a bell for me, and sure enough, this is not the first time an intentional walk to Jeff Francoeur has ended in tears for one of the Royals’ opponents this season. On April 25 in Detroit, the game was tied 3-3 in the top of the tenth. With runners on second and third, Jim Leyland had Phil Coke walk Francoeur to load the bases, bringing George Kottaras to the plate. However, Coke then unintentionally walked Kottaras, bringing in the game-winning run. Leyland brought in Darin Downs, who got an out before giving up a fate-sealing grand slam to Alex Gordon.
As bad as Leyland’s decision to walk Francoeur turned out in hindsight, it makes more sense than Maddon’s from yesterday. Coke is a southpaw known for his big platoon split, so pitching around Francoeur to get to the lefty-hitting Kottaras (who was followed by the lefty Chris Getz in the batting order) at least had some immediate sort of logic to it, although I have not crunched all the numbers.
Maybe there is just something about Jeff Francoeur that makes managers want to pitch around him. Leyland’s decision is at least defensible on the face of it, but it is hard to put together a rationale for Maddon’s decision. Maybe Maddon was still riding that 2008 high from getting lucky when he had Grant Balfour intentionally walk in a run with Josh Hamilton at the plate. Poor managerial decisions, individually considered, are rarely a big deal from the perspective of probability. This time, Maddon was just particularly unlucky and it showed. Nonetheless, such a poor decision really stands out when coming from a team noted for its drive to take every tiny, marginal edge it can get.
Matt Klaassen reads and writes obituaries in the Greater Toronto Area. If you can't get enough of him, follow him on Twitter.
Maybe I’m not reading this article right, but I’m not seeing a real argument here as to why this was a bad decision except because it is and it didn’t work? Is 10 points of wRC+ supposed to be nothing? Is setting up forces nothing? My prior is the walk makes little difference in this situation. WPA suggests the move was a slight minus (though less than 2%), and WPA overestimates the cost if we take the given wRC+ projections as true talent. We generally think Maddon isn’t stupid, and the article makes zero attempt to consider what he was thinking?
Eh. Compared to putting a man on base, 10 points of wRC+ is pretty much nothing.
And Francoeur has been the worse of the two, season-to-date (57 vs. 63 wRC+).
Not really disagreeing with the conclusion here, but I would imagine a batter’s BA is lower when runners are on first and second then when a runner is only on second because of the two forces. May have played into his reasoning a bit.
Some day, after Jeff Francoeur moves on to the great right field in the sky, we’re going to find in his personal effects a treasure trove of compromising photos of nearly everyone in any position of power in the hierarchy of Major League baseball. It’s going to be so massive and so solidly incriminating, we’re all going to wonder how we missed what was happening all this time.
On the contrary, he just has nude photos of Bud Selig, but no one wants to see those published.
It’s the Turkey underwear. Teams want the gobble gobble mojo he brings with it.
The real problem with this move, it seems to me, is that it puts Alex Gordon on deck (or in the hole in the earlier instance) with men on base. Since Gordon can actually hit, there’s significant risk in this. Why do it? I wonder if it’s because Francoeur is bad in an unpredictable way (the way Vlad Guerrero was good) – since he might swing at anything, he can’t be “pitched to.”
Exactly; I’d rather make 2 guys with sub-.275 OBPs get on (less than 7.5% chance if you’re doing the math) before I have to face Gordon than 1.
It seems very logical to me, that in the American League, where the pitcher does not hit, that you would ALWAYS want to pitch to both of these players. Does anyone disagree?
Barring a situation like the Detroit game, anyway, where there were men on second and third in extra innings with less than two out.
I’ve been doing research for six weeks now on an advanced metric called QFGA (Quality fangraphs articles), and it has yielded surprising results. Average QFGA for Monday 3.7, Tuesday 3.3, Wednesday 3.2, Thursday 2.1, Friday 1.7. Look at that clear downward trend. The problem is that demand for QFGA actually goes the other way with higher demand when we are checking out of our jobs. Today, this is one of many QFGA, helping Monday become even more dominant. I’ll update again in the future, but let’s reduce the volatility (that was another QFCA from today).
You may disagree with Maddon’s decision, but it’s no mystery. Francoeur came into that plate appearance with 2 home runs off Hernandez in 13 career at-bats (plus a double, a couple of singles, and even a pair of walks). Earlier in the same game, he’d lined out to center before hitting that second homer.
This may also be why Ned Yost started him in the first place.
Yeah this seems to be the clear logical reason. Maddon isn’t looking at Francouer’s numbers overall but specifically versus Hernandez. He saw Francouer had been hitting the ball really well off Hernandez and didn’t want to go to the pen just yet. Therefore he figured the percentages were in his favor versus Escobar to get the final out rather than watch Francouer square up another pitch an hit it hard somewhere. Dont have the numbers for the Escobar matchup but I bet that they heavily favor Hernandez. It didn’t work, but the line of thinking isn’t that unreasonable or hard to see…he just didn’t like the batter pitcher matchup at that specific moment.
Along the same lines, Francoeur has hit ground ball pitchers a lot better than fly ball pitchers — 116 tOPS in his career and 205 tOPS in 2013. On the other hand Escobar seems to be equally terrible against everyone. This may or may not have sample size issues and needs to be regressed and all that, but it should be a better reason than the specific batter vs pitcher matchup.
Either Jon or you is probably correct on Maddon’s reasoning.
I tend to think that Maddon must be a great manager because of the results he gets from one of the lowest payroll teams year after year, but he does make a lot of decisions that completely mystify me.
I like to believe he had a better reason than the batter-pitcher matchup, but maybe it was just a brain fart.
This is sarcasm, right?
There would have been an argument – not necessarily a good argument, but an argument – if the Rays had actually been up a run, despite the old “don’t put the go-ahead run on base” adage. The idea would have been that Francoeur may be more likely to hit a HR than Escobar is to get an extra-base hit. Francoeur’s ISO against righties is .148 for his career and .142 this season. Escobar’s ISO against righties this year is .058 this year and .098 for his career. So, that would be the argument: it would be more likely for the Royals to take the lead with a Francoeur HR than with a Francoeur walk and anything Escobar could do. If you’re framing the argument, I guess that would be it…
Had it been a one-run game and not a tie game. All this did was shift the lineup from bad-hitter/bad-hitter to bad-hitter/good-hitter, with an extra man on base. I don’t get it.
I see a real double standard here. If this were, say, Don Mattingly making the same decision, he would be crucified in this comment section. Instead, because it’s Joe Maddon, actual FanGraphs readers are making arguments based on the batter-hitter matchup.