Fish-Cap: Thank Josh Johnson, Jerry Manuel for Marlins win

Last night’s game between the Marlins and the New York Mets had all the makings of a classic pitcher’s duel, and indeed the duel between Josh Johnson and Johan Santana was just as we expected. The two pitchers went toe to toe, equaling each others’ performances, and both left with the game still in the balance in the late innings.

Thankfully, the Marlins don’t have Jerry Manuel managing the team. Manuel made a series of odd moves that may have cost the Mets the opportunity to win the game. I figured we should talk about a few of the managerial moves of last night’s contest.

The View from Afar

Aces duel to a tie

As expected, Johnson and Santana shined in the early goings of the game. Both pitchers allowed only one run, but you have to think that Johnson got the best of the duel last nigh.

Johnson: 7 IP, 1 R, 3 H, 7 K, 0 BB, 0 HR
Santana: 7 IP, 1 R, 6 H, 5 K, 1 BB, 0 HR

Either way, both dominated their opponents’ respective lineups. Johnson threw 95 pitches, getting 62 in a generalized strike zone (65.3% Zone%), similar to Ricky Nolasco’s recent start. JJ recorded 10 whiffs on his pitches, split evenly between his fastball, change-up, and slider. The fastball and change must have been working in terms of ground balls as well, as Johnson induced 10 grounders out 18 balls in play. Essentially, you could not have drawn up a better start from JJ, it was close to flawless. Outside of the timing of the second inning (all three hits came in the second), Johnson was facing the minimum throughout.

Quick Aside: An Anti-Barden Message

Look, I have no problems with Brian Barden, the person. I’m sure he’s a nice guy. But Ted and Dave at Marlins Diehards are right; Barden had no business being the #2 hitter for the Fish last night. I estimate it may have cost us 0.18 runs perhaps against a lineup I make, maybe 0.12 runs if we just bat Gaby Sanchez second. That may not seem like a lot, of course, but it’s costing us something. Also, it got Dan Uggla’s bat out of the lineup, and I’ll bet the reason wasn’t to rest him but rather to keep his 1-for-19 performance against Santana away from the game. This move actually did impact the game. Back to the story.

Fredi makes right move, pulls Johnson, but for wrong guy

In the seventh inning, Santana walked Brett Carroll following a Sanchez double, creating an opportunity with men on first and second and two outs. Here are some relevant numbers with regards to that situation:

Marlins Win Expectancy: 57.6%
Marlins Run Expectancy: 0.47
Leverage Index: 2.86

What this means is that the PA mentioned was 2.86 times more important than the average PA, that teams were expected to win 57.6% of games in which they have reached this base/out/inning/score state, and that teams in this situation were expected to score 0.47 runs by the end of the inning on average.

The walk brought up Johnson, and Fredi Gonzalez had to make a decision on JJ’s presence in the game. I had an argument about the decision with the fake Fredi Gonzalez on Twitter, but I believe I have the information to back my claim that pinch-hitting for JJ is the right move. Consider the best case scenario for keeping JJ in the game:

ZiPS in-season projected FIP, Johnson: 3.10
ZiPS in-season projected FIP, Clay Hensley (the man who replaced JJ in the eighth inning): 4.44

That means that the projected difference between these pitchers over nine innings is 1.46 runs. In one inning, however, it would be 0.162 runs. And that does not take into account that Johnson is on his third trip through the order, and that pitchers do traditionally worse on their third try.

Now consider the pinch-hitting situation. The average hitter, facing an average pitcher, would be expected to produce 0.47 runs in that case. But Santana is not an average pitcher and Johnson is certainly not an average hitter. Neither is Wes Helms, the man who replaced Johnson, or Dan Uggla, the man who should have replaced Johnson. Santana is far better than average, but Helms is significantly worse (ZiPS projected .304 wOBA) and Uggla a good deal better (projected .356 wOBA). Both would be pinch-hitting, so both would receive similar penalties, but I can’t imagine that this matchup for Uggla would yield anything less than an expected 0.25 runs. Instead, however, sending Helms probably would yielded a lower value than sending Johnson out again, given Uncle Helms’ lack of hitting talent.

There is probably enough variance in the Uggla move that a risk-averse person would air on the side of keeping Johnson in the game, as those moves are probably fairly close. However, there’s no arguing that sending Helms was the wrong call.

Manuel helps the Marlins with puzzling moves

At least Fredi had the right idea (mostly), just the wrong guy. Manuel, on the other hand, just made some wrong moves. In the ninth inning, Luis Castillo singled and advanced on a Marlins error, putting him at second base with no one out. The average run expectancy in that situation was 1.20 runs, already over a run expected to come home. Manuel had his three best hitters, Jose Reyes, Jason Bay, and David Wright, coming up to hit against Leo Nunez, hardly considered a world beater. The odds were actually likely to score more than 1.20 runs based on the identities of the players involved. However, Manuel inexplicably attempted to sacrifice bunt Castillo to third. Now, it’s not always cut and dry in terms of determining whether a bunt is a right or wrong move, but given the fact that Reyes signaled the move and the Marlins were prepared after the first attempt, and also considering that Manuel is not well known for this advanced sort of thinking, I think I can put the game theory application of this maneuver out of the question.

Manuel apparently did not trust his three best hitters to drive in a run via a legitimate hit, and thus decided to attempt the bunt. The bunt failed for them, but that’s not what made it the wrong move. Rather, the move was probably wrong to begin with.

Manuel further put his team in the hole in the bottom of the inning. The Marlins put up a first and third situation with only one out, in part thanks to the strong baserunning of Cody Ross. Ross moved from first to third on a single by Ronny Paulino. At this point in the game, the leverage index was at 5.0, the highest of the game. This meant that the situation was five times more important than the average PA in deciding the outcome of the game. The Marlins sent in Uggla to pinch hit for the pitcher. The Mets countered by staying with Fernando Nieve, who had pitched himself into the jam. This would not be a huge problem if the team didn’t have a better reliever, but it just so happens the club has Francisco Rodriguez, whom they pay a lot of money to pitch high-pressure situations. Instead of warming up K-Rod to pitch in the most important part of the game, Manuel let him sit out the game in the hopes that the Mets would at some point build a lead for K-Rod to save. Nieve then threw the wild pitch that ended the game, thus wasting a potential appearance by the Mets’ relief ace.

In many of these situations, managers will argue that the save situation will be more important. However, with the winning runner on third base, it was highly unlikely that another situation later in the game would have come up that had a higher LI than this. Manuel wasted the opportunity by saving his best pitcher available for the save that never happened. From us Marlins fans, thank you Jerry Manuel.