It's hard to find a place to begin after Saturday's 2-1 inning loss to the Giants. In yesterday's column, I started with a lot of positive thoughts. I emphasized that even last night's game wasn't really a must win. It was important for them to win, but it wasn't a must win. My immediate thought after the game was that the way that the Nats lost Saturday made it feel like it's going to be awfully difficult to come back from. For that reason, I decided to let the loss marinate for a bit before dissecting the grisly details.
Before we get started, let's use our opponent's own narrative against them to see if we can spark a little bit of hope....
Two years ago, these same Giants that the Nats are facing now were the third seed in the National League Playoffs. Because it was the first year with the two wildcard system, the division series was set up differently than it is now. The lower seeded team started with two home games and played the final three games on the road. The Giants started at home, losing the first two games 5-2 and 9-0 to the Cincinnati Reds. They went on to win the next three games in Cincinnati to take the series.
The Nationals have a very tough road ahead. They'll play Monday in a game that they have to win to stay alive. If they do that, they'll face the same situation on Tuesday. If they can win both of those games, they'll play in a mutual elimination game on Thursday. The Nats had twelve winning streaks of three or more this season. The Giants had ten losing streaks of three or more games this season. It's going to be incredibly difficult; It's not impossible.
It's cliche, but the Nats have to treat each of their remaining games in this series as if it's the last day of the year. If they don't, it will be. Win Monday against the Giants best starter and the doubt will start to creep in for them.
With that out of the way, let's move on to the gory details of the loss. I'll get started with the decision to pull Jordan Zimmermann in the ninth inning even though that's pretty far down the list of reasons the Nats lost this game. I generally feel like the article that I linked to is a pile of rubbish. It feels like Mr. Boswell decided to write a story where one managerial decision preempts six hours worth of offensive futility by creating fictitious gremlins, grumpkins, and baseball gods. Matt Williams made a baseball decision that he thought was the best thing he could do to help the team win the game.
I can't speak for the 44,035 that attended the game last night (or all of the other people disappointed with the result), but I can say that my top priority while cheering on the Nationals last night was that they win the game! If Jordan Zimmermann had completed his Herculean effort with the shutout, it would have been great. However, that was secondary to a Nats win.
When an offense performs (?) as the Nats did last night, it seems awfully silly to point the finger elsewhere. Still... Let's start with the decision.
Matt Williams decision to remove Jordan Zimmermann for Drew Storen with 2 outs and a runner on first in the ninth
Is it the decision that I would have made? As I said in the game thread, probably not. Was it a defensible choice given all of the factors involved? Absolutely. Let's break the decision down from a few different angles.
In the playoffs, I don't think that pitch counts are quite as important as they are in the regular season. During the regular season, you're focused on winning the ballgame, but you also have to be focused on your pitcher's upcoming starts. When you have a pitcher who is rolling in the playoffs, you stick with him a little longer than you ordinarily might. There is no tomorrow. You don't think about that next start. If he's the best player to get the job done, he stays in.
This doesn't mean you completely throw the pitch count out the window, though. It's a valuable tool in that Williams had managed 32 games that Jordan Zimmermann started this season and gotten a feel for when and how he has started to show signs of fatigue.
- Zimmermann was at exactly 100 pitches. That's a workload that he has dealt with before. He averaged 91.4 pitches per start this season.
- Based on his 32 regular season starts, Zimmermann was nearing his limit, but he wasn't quite there yet. He had 11 starts where he threw 100 or more pitches, maxing out at 114 pitches in a complete game shutout of the Padres in June.
- He threw more than 105 pitches in just three outings. While he had thrown 100 or more pitches in 34% of his starts, it's not like has a history of going too far beyond that limit.
He'd shown signs of fatigue throughout the season at this point, so even before we look at any evidence of what was happening on the field, it's reasonable to expect that he was near the end of his rope. When that's the case, you watch him more closely and have a quicker hook. That's not to say that Zimmermann was out of gas last night, but there's a lot of data that Williams had accumulated throughout a full season of managing his starts that tells us he might have been tiring.
Fourth time through the order
Zimmermann has allowed a .249 batting average to opponents in his career. As you would expect with most pitchers, that batting average rises throughout the game as the opponents have already faced him a few times and he's started to fatigue. Let's take a look at these career splits:
|Time through order||PA||BA||OBP||SLG||tOPS+|
|1st PA in G, as SP||1304||.226||.273||.355||86|
|2nd PA in G, as SP||1291||.256||.293||.389||102|
|3rd PA in G, as SP||982||.262||.313||.407||114|
|4th+ PA in G, as SP||81||.338||.363||.468||146|
Is the fourth time through the order a small sample size? Of course. Pitchers generally don't get to face a lineup a fourth time unless they're pitching extremely well. I certainly think that we can see how this works, though. As batters have faced Zimmermann multiple times in a game, their production has gone up across the board in the later plate appearances.
Let's check the batter who was coming up, who (once again) is the Giants best hitter and a former MVP who batted .311 with 22 HR this season....
|Times facing a SP||PA||BA||OBP||SLG||OPS|
|vs. SP, 1st||578||.317||.369||.486||.854|
|vs. SP, 2nd||566||.331||.387||.492||.879|
|vs. SP, 3rd||490||.301||.378||.521||.898|
|vs. SP, 4th+||50||.341||.400||.386||.786|
OK... The slugging is actually down a bit in that fourth plate appearance, but his batting average and OBP are up when he faces a pitcher for the fourth time. Again, it's a small sample size because pitchers generally don't face 30 or 31 batters (he's primarily batted third or fourth in his career). When one of the best hitters in the game who also happens to be a catcher sees how a pitcher is attacking him three times, he learns from it and makes adjustments.
So Zimmermann had faced Posey three times already. His BAA is 89 points higher than his career BAA when he's facing a hitter for the fourth time. Posey's career average when facing a pitcher for the fourth time is 33 points higher than his career average. As well as Jordan Zimmermann had pitched, there was certainly plenty of data that pointed Williams towards going to the bullpen.
Let's take a look at what was going on in the game
Let's start this with another link that I saw describing the decision. Former big league pitcher C.J. Nitkowski wrote a terrific piece on the reasons that pulling Zimmermann was justifiable for FOX which analyzes some of the critical details better than I can. He points out some of the visible signs that a pitcher can show when he's starting to fatigue, such as body language, struggles with mechanics, struggles with command, and impatience with the umpire. He points out in the article that Zimmermann showed frustration with the umpire about not getting some calls off the outside corner during the Panik AB. He points out that he overthrew a fastball that missed way up. He points out that Zimmermann looked to the dugout at one point, which certainly could have been a way of Zimmermann indicating that he was out of gas.
Let's have a look at the Panik AB:
It may have felt like Zimmermann was just missing when he walked Panik, but he wasn't. Given that umpire Vic Carapazza had been calling kind of a wide strike zone on the outside corner for much of the night, ball two was close, but it certainly wasn't in the strike zone. Nothing else really was that close outside of the 2-0 fastball down the middle... That pitch was crushed foul into the second deck along the right field line.
Going to Storen
Drew Storen was outstanding all season in 2014. He finished the year with a 1.12 ERA and 0.98 WHIP. He had taken over as the Nats closer in early September and gone a perfect 10 for 10 in save opportunities, allowing just 6 hits without walking a single hitter in those outings. He's the relief ace who had earned last night's opportunity by dominating all season. He had limited opposing hitters to a .215/.262/.278 triple slash line this season, and was even better against right-handed hitters (.181/.254/.246).
Despite the fact that Storen is the team's relief ace, you could make a pretty good argument that Matt Williams should have considered having lefty Matt Thornton ready in case Posey did what he ended up doing (single) because of Pablo Sandoval's splits (.199/.234/.319 vs. LHP). I've argued against the "Capital C Closer" tag at times this season, but I'm not really going to argue this point. I don't think that a soft line drive on a well located fastball over the outside corner screams that Williams should have gone to get his relief ace in favor of a slightly better matchup that's based on a small sample size.
A lot of us got tired of hearing the phrase "tip the cap" during the 2013 season, but sometimes you have to do just that. Let's have a look at the three (only 3) pitches that Storen ended up throwing:
All three pitches were fastballs. We'll note that the first pitch (blue 1 on the right side of this chart) was a fastball almost perfectly located on the outside corner. It was a nice piece of hitting by Posey to drive it into center field for a single. The pitches to Sandoval were also right at the edge of the strike zone. The first (the foul ball on the left side of the screen) was right over the outer edge, though it was probably up a little more than Storen would have liked. The second pitch that Sandoval drove down the left field line was at the bottom of the zone, though it probably caught a little too much of the plate. It's not like Storen came in and threw pitches that were middle-middle. The Giants executed, and you have to give them credit for that.
All in all, Williams was in a tough position. He was set up for a lot of second-guessing whether he stuck with Zimmermann or went to Storen. Both the evidence that he has of past performance and what was going on in the game said that he had two good options. He picked one. It didn't work out.
Of course, all of those reasons that he made the decision are moot if the offense looked anything like the team that finished third in the NL in runs this season. In 27 innings so far this series, the Nats have scored just three runs. In 18 innings last night, they scored just once. They pushed just five baserunners as far as second base in the entire game. The team that grinded out at bats all season long has mysteriously disappeared so far in this series.
In last night's game, the Nationals went 9 for 62. Four of those hits came off the bat of Anthony Rendon, leaving the rest of the lineup 5 for 55 with one extra base hit. They're batting a combined .160/.225/.245 through the first two games of this series. I don't care how well the pitchers have thrown. It's awfully tough to win when the offense has been that anemic.
The Nats patient approach at the plate has been a staple all season. They finished third in the NL and 7th in MLB with 517 walks. It goes beyond just drawing walks, though. Working deep counts and (ideally) getting ahead in the count means that the Nats have generally ended their plate appearances on better pitches to hit. In Saturday's game, the Nats drew four walks. The first of these walks occurred in the twelfth inning after they had already struck out twelve times.
- Wilson Ramos saw 25 pitches in his 7 PA. This wouldn't necessarily seem low if four of those didn't end with strikeouts. Someone mentioned in the game thread last night that it looks like Ramos' stance is even more open than it was throughout most of the regular season and that he's having even more trouble getting to pitches on (or off) the outer half. The Giants have done a wonderful job of attacking that weakness over the first two games.
- Perhaps the Nats most patient hitter (actually, Werth led the team in P/PA) is Adam LaRoche. LaRoche saw just 22 pitches in 7 PA last night and failed to reach base once.
- Denard Span has fallen back into his early season habit of trying to pull everything. Through the first two games of this series, Span has had twelve PA. He's walked once. He's struck out once. He is 0 for 10 when putting the ball in play. Nine of those ten balls in play were to the right side (ironically, the one where he didn't hit the ball to the right side was after Nate Schierholtz's leadoff double in Game 1). Seven of them have been on the ground. He has hit two line drives, both of which ended up being bad BABIP luck as they were struck right at Brandon Belt and Hunter Pence.
Cabrera ejection/Vic Carapazza
As is the case with one managerial decision, we can't place too much of the blame on an umpire. Since it's impossible to expect for an umpire to perfectly call the rule book strike zone, there's only one thing that you can really ask for... consistency. Let's start by looking at Carapazza's called strikes last night against LHH....
We'll note the pretty big outside corner to both lefties and righties that I mentioned above. The Giants weren't the only team taking advantage of that. We'll also note the 4 called strikes (out of 82) that were above the zone. Why?
As we can see, two of those four strikes that he called above the zone were in Asdrubal Cabrera's final at bat. The highest strike that he called all night is the one over the inner half that was a delayed strike two call. Cabrera does need to reel it in there, though. It's a playoff game. The game was in extra innings. They simply couldn't afford to lose him and shorten the bench. I don't care how big a game is. When you slam your bat and helmet down on the plate after an umpire rings you up, you're getting the gate. Carapazza made one really bad call (and it was delayed) in the at bat. The third strike call was borderline. They were correct to eject Cabrera for overreacting.
Carapazza inserted himself into the game numerous other times as extra innings progressed. He called a third strike on a borderline 3-2 pitch to Anthony Rendon in the 11th inning. He did the same on a 1-2 pitch to Hunter Pence in the 14th inning. He put a lot of hitters behind in the count when they shouldn't have been by calling strikes on pitches that Pitchf/x data had as out of the zone. Per Baseball Savant, 20 of the 30 pitches that he called strikes that weren't in the zone went against the Nationals.
The Belt Home Run
I'll bring up Carapazza again. Once again, he was not the reason that the Nats lost the game. However, that 2-2 pitch to Brandon Belt?
It's pitch number 7, which we can see on the left-hand side at the top of the zone. The pitch was lower than either of the pitches that he rung Cabrera up on. The pitch was just a hair off the outside corner, but he had called a big outside corner all night, including a pitch earlier in the at bat (1) that was even farther off of the plate. The next pitch was a 3-2 fastball over the inner half, and Belt deposited it in the second deck to win the game. The 2-2 pitch wasn't a rule book strike. It was in a location that Carapazza had called a strike on earlier in the game. Consistency...
Matt Williams decision to have Anthony Rendon steal a base with 2 outs and Adam LaRoche at the plate in the 8th
At the time, the Nats were still leading 1-0. The Giants had gone to Jean Machi to face Jayson Werth and I was absolutely certain that Javier Lopez was coming in to face Adam LaRoche. I've talked about how avoiding that matchup was going to be one of the keys to this series for the Nationals. LaRoche didn't just struggle against LHP this season (.204/.284/.336). He's also got some history with Lopez, the Giants only true LOOGY. LaRoche is 0 for 10 with 9 strikeouts and 2 walks against Lopez in his career.
Anyway, Giants manager Bruce Bochy didn't make the expected move when LaRoche came to the plate. He stuck with Jean Machi. Rendon stole second base on the first pitch. Then Bochy went out and called for Lopez with an 0-1 count. Three pitches later, LaRoche struck out. There's no guarantee that Bochy doesn't pull that odd maneuver even if Rendon hadn't stolen second, but it did force his hand a bit when a single could suddenly push the lead to 2-0. Since the first pitch was in the zone (swinging strike), it's not like Bochy had called for Machi to be overly careful with one pitch just to give Lopez another few pitches to warm up.