Did Joe Maddon Get Too Cute in Game 3?

It should be said that it’s players who win and lose baseball games. Jake Arrieta, Ty Blach, Kris Bryant, Madison Bumgarner, Aroldis Chapman, and Conor Gillaspie: all had a greater impact than Chicago Cubs manager Joe Maddon on the outcome of last night’s thrilling 13-inning, 6-5 San Francisco victory. Because players are often forced to make decisions quickly, however, it’s difficult to question the wisdom of those decisions. Telling Hunter Pence not to swing at a 102-mph fastball several inches out of the zone, for example, ignores the reality the situation. There’s no time for deliberation. The same isn’t necessarily true for managers.

Managers, too, make difficult decisions — decisions which, given the wealth of information they possess relative to the general public, are also frequently difficult to critique responsibly. But their thought process is easier to decipher and their decisions are less physical — excepting those that come from the gut. In Game 3 of the Division Series against the Giants, a crazy game that might have been over many times, Maddon made a number of decisions. Most, especially the successful and inconsequential, went unnoticed. What follows is a review of the less successful, less inconsequential decisions.

Using Aroldis Chapman in the Eighth

While this choice is likely to get the most attention, given that Chapman came in and blew the save, using Chapman in the eighth is pretty close to a no-brainer. When Chapman entered the game in the eighth inning and no outs, the Cubs were leading by one run. With runners on first and second and nobody out, however, the Giants actually possessed the higher win expectancy (51%). The leverage index was a massive 5.13. For comparison’s sake, when David Ross came to the plate in the top of the 13th with runners on first and second and one out, the leverage index was a very high, but less light, 4.62. At no point in last year’s 12-inning, World Series-concluding game did the leverage index ever reach 5.0. And actually, after Chapman struck out Hunter Pence, the leverage index even increased to 5.44 for Conor Gillaspie’s plate appearance. Using your lights-out pitcher in this situation is certainly the right move.

So it makes sense, turning to Chapman in the eighth. But what might not have made sense is delaying the move to Chapman until there was trouble. It’s true, Chapman only appeared in seven games this season with inherited runners — none since mid-August — and half of the 12 runners he inherited ended up scoring (League average is 30%.) Most of Chapman’s experience entering a game with runners on base has occurred in cushier situations where the previous pitcher has created a save situation for Chapman. After the game, Maddon cited Buster Posey’s success against Chapman as one reason not to use him earlier — that and generally not wanting to force Chapman to pitch six outs for the save. Maddon could have been more aggressive, although holding Chapman for the ninth would have been a much bigger blunder.

The decision not to use Chapman for six outs caused a chain reaction of substitutions for the Cubs that weren’t necessarily helpful. Substitutions like…

Double-Switching in Jason Heyward in the Bottom of the Seventh

Jason Heyward has been a disaster at the plate this season, posting an awful 72 wRC+ for the season. He’s also one of the best defensive players in baseball and still managed 1.6 WAR due to that fielding prowess. While expectations were certainly higher for Heyward’s bat given the 117 wRC+ he’d recorded in his career prior to this year, Maddon and the Cubs traded offense for defense all year, but this game was an exception: the presence of Jorge Soler in the Cubs’ lineup meant the exemption of Heyward from it. Maddon acknowledged his reluctance to take Heyward out of the lineup.

Defensively, of course, that’s something I don’t like to do – to not have your best defensive team on the field. But if we were able to grab a lead, there are things we can do in the latter part of the game.

Offensively, Soler drew a walk; on defense, he exhibited no problems with the routine plays that came his way. After the Cubs got the lead, Maddon took the opportunity to strengthen his defense, and so began his moves up and down the lineup.

Generally, a double-switch is used when (a) the manager wants a reliever to pitch more than one inning but (b) the pitcher’s spot in the order is approaching. Rather than have the pitcher bat, the manager substitutes him for a position player, bring in a second position player to occupy the pitcher’s spot in the order. In this case, the pitcher’s spot was due up fourth, likely to hit, and Soler’s spot was due up eighth. Maddon didn’t want Pedro Strop to pitch more than one inning, as Strop retired both of the batters he faced before being removed, but it seems likely he did want Travis Wood — who got the final out of the seventh and pitched to one batter in the eighth — to pitch in multiple innings. Using Wood in multiple innings and pushing back the pitcher’s spot in the lineup is consistent with the typical logic for using a double-switch.

Unfortunately, this switch also replaced the spot behind Kris Bryant and Anthony Rizzo with a pinch-hitter or bat off the bench the rest of the game, thus creating a weak link in the Chicago lineup. Using Travis Wood for one batter might not be worth that tradeoff. The entire decision was based on the logic of needing a double-switch to keep Travis Wood in the game against Brandon Belt, a batter without a platoon split. Hector Rondon came in one batter later, and would have stayed in had he retired Posey.

That, in turn, led to…

Double-Switching in Albert Almora for Heyward

Heyward had just struck out to end the previous inning and the pitcher’s spot in the lineup was now due up fourth the next inning. Bringing in Chapman in the pitcher’s spot at that time meant that Chapman was going to bat in the ninth if Dexter Fowler, Kris Bryant, or Anthony Rizzo got on base, a pretty likely scenario. With Willson Contreras already in the game as a defensive replacement, and Javier Baez and Addison Russell also positives on defense, that meant that defensive replacement Jason Heyward, who made the last out of the eighth, was replaced for another defensive replacement, Albert Almora.

Albert Almora, a very good defender who has played center field most of his career, was inserted in the game in right field. In the ninth, he made a great game-saving play on Buster Posey’s liner, one which illustrated his natural fielding talent. But Almora lacks lot of defensive experience in right field, specifically — and in San Francisco’s ballpark, as well, which features a number of challenging features for a relative inexperienced player.

While Almora didn’t appear to make any mistakes in his pursuit of Conor Gillaspie’s triple — a batted ball that becomes a hit 74% of the time based on launch angle and exit velocity, per Baseball Savant — it’s also possible, given how close he was to making the catch, that the best defensive right-fielder in baseball (i.e. Heyward) might have had the opportunity to field it.

Double-Switching in Chris Coghlan for Ben Zobrist

At this point, Joe Maddon knew that (a) he needed to get past this inning and (b) he needed a pitcher in case he got past this inning. Enter Mike Montgomery. With lefties Denard Span and Brandon Belt due up, he was a good matchup for the former and — as a pitcher with some success as a starter — decently prepared to negate the platoon advantage of the latter, as well. In order to ensure that Montgomery’s spot didn’t come up in the 10th inning (he was due up fourth), Zobrist (who made the last out of the previous half-inning) or Almora (the second to last out) needed to come out for Montgomery. Ultimately, Montgomery pitched multiple innings and pitched well before allowing the winning run in his fifth inning, so the move seemed to pay off. That said, Montgomery also batted for himself in the 12th in what would have been Zobrist’s spot — with Carl Edwards Jr. still available in the pen.

*****

These cases above certainly don’t represent all the decisions Joe Maddon was forced to make on Monday night — and some of them seemed to pay off. Hindsight being what it is, the summary critique is as follows: if you want Soler out for Heyward, put Heyward in Soler’s spot in the lineup and use a pinch-hitter for the pitcher’s spot when it comes up. With an extra bench player (Tommy La Stella never played), simply pinch hit for the pitcher, because getting one extra batter for Travis Wood in the eighth inning in exchange for using the pitcher in the cleanup spot isn’t worth it — especially when Chapman could have gone two innings. Then, were he inclined to make a double-switch later, Maddon would still have the option of putting in Almora for Zobrist or Heyward to potentially keep Zobrist’s bat in the game.

Maddon’s decision to use Chapman in the eighth was undoubtedly a good one. His decision to try and avoid using Chapman in the eighth when he knew it might be a possibility and the switches that resulted from that attempted avoidance — that’s all open to some criticism. Admittedly, Brandon Belt’s hit off Travis Wood was hardly a scorcher, but Maddon’s delay in turning to Chapman made the start to that inning possible. Even if he didn’t want to use Chapman, trying to get a clean inning from Rondon or Edwards or even Montgomery might have made more sense without requiring the sort of lineup machinations that ended up weakening his bench and limiting his options down the line. Maddon wasn’t even close to the top difference-maker in the game, and the number of decisions made can hardly be covered in a single post or even a series of posts. It’s his decision to avoid Chapman in the eighth — not the decision to use him — that deserves more scrutiny.