The Dodgers’ Other Lefty Problem
In a few hours, the Dodgers and Nationals are going to play for their seasons; one team will advance, one will go home. The Nationals are in a pretty good position, hosting the game in Washington while throwing their ace, Max Scherzer; it’s hard to ask for much more than that if you’re in an elimination game. The Dodgers used their ace in Game 4, however, so they’re going to be mixing and matching in order to try win this one.
While they haven’t officially announced anything, the team is expected to start Rich Hill. Given his health track record, however, and the fact that he’s going on three days rest, they’re probably not going to have him work too deep into the game. In fact, reading the tea leaves from Andy McCullough’s game preview, it sounds like the Dodgers are going into this game planning for something of a tandem start.
“We talked about Rich as an option, obviously,” Manager Dave Roberts said after the Dodgers’ 6-5 win on Tuesday. “But so is Julio, and how we want to strategize to win Game 5, we’re going to talk through it.”
…
The combination of Hill and Julio Urias is unorthodox, but it would reduce the chances of Washington’s top four hitters — Trea Turner, Bryce Harper, Jayson Werth and Daniel Murphy — from getting comfortable at the plate against either. Hitters tend to perform better when they have seen a pitcher more than once in a game. The Dodgers will try not to allow that to happen on Thursday.
If the Dodgers are serious about limiting Hill and Urias to one trip through the order, that would be something indeed; they’d be limiting themselves to getting likely six or seven outs from each pitcher, and would be relying on the rest of their bullpen to get the other ~13 outs, despite being worked pretty heavily in both games in Los Angeles. It would be the most aggressive kind of bullpen game we’ve been expecting so far in the playoffs, but unfortunately for the Dodgers, I’m not sure how well set up they are to actually pull it off.
Ideally, in a tandem start, you’d like to have two different kinds of pitchers, giving hitters different looks, and making it difficult for the opposing team to set a lineup that combats pitchers of the same handedness. But like Hill, Urias is also left-handed. And when you look at the Dodgers bullpen, the lack of good right-handed options today might be a legitimate problem for the bullpen-start concept.
Kenley Jansen, of course, is a good right-handed reliever, and he’s basically a lock to pitch today. But Jansen is likely to be held back for the end of the game — Dave Roberts chose to use Pedro Baez and Luis Avilan in relief of Kershaw in the seventh inning on Tuesday, despite the fact that the Nationals had the go-ahead run at the plate — and won’t be available to break up the lefties on whom the Dodgers are leaning today.
The team’s other high-leverage right-hander, Joe Blanton, threw 35 pitches on Monday and another 19 on Tuesday, so while I’m sure he’s available after the day off yesterday, you do have to wonder about how many pitches he’ll be able to throw effectively. The team’s most well rested right-handed reliever is Josh Fields, but while Fields throws hard, he’s not all that good, and you probably don’t want him on the mound in an elimination game if you don’t have to use him.
A reliever the Dodgers do want on the mound today is Grant Dayton, their excellent setup man who held opposing batters to just a .219 wOBA this year. Dayton, though, like Hill and Urias, is left-handed. If the team plans to piggyback the two starters, and then wanted to bring in their best non-closer reliever if Urias gets in trouble, they’d potentially be throwing three straight lefties at the Nationals to start the game.
And while that would be a good plan against the Dodgers, who have struggled against left-handed pitching all year, the Nationals offense is actually one of the best in baseball with southpaws on the mound. On the season, Washington’s position players put up a 111 wRC+ against lefties, tied for the fourth-best mark in baseball. The loss of Wilson Ramos makes their current lineup a little less fearsome, but even just limiting the data to players on the team’s active roster, the Nationals hitters put up a 110 wRC+ against lefties this year.
So throwing Hill, Urias, and Dayton against that lineup, hoping to get to Blanton and Jansen in the late innings, is probably not the best of plans. So let’s see if we can figure out how Roberts might mix his pitchers today to get through the Nationals lineup.
Let’s assume that Hill won’t be limited to just nine batters; as much as the times-through-the-order penalty and the detriment of short rest come in to play, it seems more likely that the Dodgers will try to get four or five innings out of Hill if they can. And since you don’t want to really bring in a lefty to replace another lefty if you can help it, Blanton should probably be the first guy up in LA’s bullpen, ready to help Hill get out of any jams that arise as he runs out of gas. It’s probably not a bad goal to see if Hill and Blanton can get LA their first 15 outs of the game, leaving just 12 for Urias and the rest of the bullpen.
With McCullough noting that Urias won’t be allowed to go more than three innings, it does seem like the plan would be to limit him to one trip through the order, so you probably don’t want to count on him for more than six outs. But if Hill and Blanton can get the team 15, and Urias gets six, well, then all of the sudden they’re in Jansen territory, and could theoretically get through the game without having to lean on the weaker relievers in the bullpen.
But this is where I’d suggest Roberts should diverge from traditional usage. If you’re going to get through five innings with Hill and Blanton, then use Urias to try and bridge the gap in the sixth and seventh, you want a dominant right-hander ready to get out of any trouble that arises. And in the Dodgers bullpen, that’s Kenley Jansen. The Dodgers’ best chance to effectively mix-and-match their pitchers today is to free Jansen from having to be around the get the last few outs, and instead, use him as the guy to end any jams into which Urias gets himself.
In a perfect world, Hill and Blanton get you five innings, while Urias and Jansen get you four, and the Dodgers advance with just four pitchers used. But it’s probably not going to work quite that nicely for LA, and if they get into a situation where they have to choose between going to Dayton to replace a fellow lefty, or using Baez again in a high-leverage situation, I’d suggest the team is best off using Jansen earlier, then allowing Dayton to pitch after him, even if that means letting the rookie get the save opportunity if one arises. Going Hill-Blanton-Urias-Jansen-Dayton would be the team’s best shot at using their five best arms today while mixing and matching lefties and righties, and those five should be able to get 27 outs in some combination.
If Roberts sticks with the traditional path of having Jansen as the last guy on the mound, though, the team is either going to have to stack lefties together or use Baez or Fields in an earlier right-on-right situation. And those are probably not the guys the Dodgers want to see pitching with their season on the line. If the Dodgers want to avoid the lefty-stacking issue against a team that has hit lefties well this season, using Jansen earlier in the game is probably their best option.
Dave is the Managing Editor of FanGraphs.
Ross Stripling is also a right handed pitcher on the roster who can go multiple innings.
Like Fields and Baez, he’s the quality of right-hander you don’t really want to pitch in an elimination game unless forced to. And if you’re using Hill/Urias early, he’s better saved for a potential extra inning contest.
I have always wondered why teams dont do something like announce a certain starter in the playoffs, have the opposing team set their platoon hitters, then sub him out after 1 batter and replace him with a different starting pitcher. While it would not be as great in this exact game, I can see many teams and matchups where it would be.
Well the benefit is that they would have to sub for and lose their platoon bats, but the detriment is that you would have to sub and lose a pitcher… So it’s still a big sacrifice.
Yes and no. Pitchers do work in between starts right? Why not announce a pitcher, have him throw one or two pitches, (or a batter if that is the rule) and go to the next pitcher. An extreme example would be doing something like starting Kershaw on 2 days rest (his bullpen day) and take him out after one batter.
And what opposing manager would believe that Kershaw is the “real” starter on his bullpen day?
It’s been done.
Back in the deadball era when real pitchers finished what they started and sliding with spikes up was the norm. Once or twice in recent decades.
It even worked once or twice.
Not a high probability play, these days. Teams aren’t as dependent on platoon advantage as they used to be.
Don’t you just hate all the fake pitchers nowadays not finishing what they started and keeping their elbow ligaments all attached and whatnot?
also, GET OFF MY LAWN
I don’t really see the benefit. How many guys is a manager really going to put in just to platoon? Maybe 1 or 2 at most, and he can always sub 1 of them back in if he feels it necessary. Plus, you would have to do this believably by announcing a rested starting pitcher, who are competitive guys who don’t just want to be used as pawns. Not to mention in an elimination game, you wouldn’t want a quality starter to be lost after one batter.
I think the payoff is not worth the hassle. If it was, someone certainly would have tried by now.
Disagree. Lots of examples of major league managers not doing things that would make sense due to tradition. Don’t have to look much farther than rp usage. The key would be going RHP first, then switching to LHP as lhb tend to have larger platoon splits… Take a look at the remaining teams. With the exception of maybe the blue Jay’s all teams would roll out a different lineup depending on the handedness of the starting pitcher. Losing ome or two bats from a playoff team would be an advantage in a game of “inches”. Think about all the games that went to extra innings already, now imagine if bench bats and double switches were not available.
“Think about all the games that went to extra innings already, now imagine if bench bats and double switches were not available.”
Well, also think about all of those games that have gone to extra innings, then imagine that one team has burned a useful, rested starting pitcher on 1 hitter with this “fake starter” idea .
And the pitcher does need to be a useful, rested starting pitcher for your idea to trick the other team. If it’s Kershaw on his bullpen day, the other team will clearly figure out that he’s not the “real” starter but is only going to pitch something like an inning. And there’s also a good chance that they’ll figure out who the real starter will be, either because it’s obvious based on which pitchers have full rest, or maybe just by using their eyes to see who else is warming up in the bullpen prior to the game.
There’s just far more downside than upside to your idea.
It’d work best for the visiting team.
They could announce a starter.
Take their at bats while warming up the other starter.
Then bring in the new starter to start the game. They’d lose the original starter for the game, but he wouldn’t throw any pitches.
Opposing teams would probably take offense to this.
Because teams reasonably don’t want to burn a player – a rested starting pitcher – who might be useful for much more than one batter later in the game.
What if the “real” starter comes in and ends up needing to be pulled by the third inning? Sure would be nice to have that other opposite-handed starting pitcher available for multiple innings, but the team has burned him pitching to one hitter in the first inning.
What if the game ends up a long extra innings game? Again, it would be nice to have that starting pitcher on full rest available after the team has gone through its best bullpen arms.
In practice, the situation that you’re talking about would arise from a pair of starters where one of them is the #5 guy bumped from the playoff rotation: two fully rested starters, opposite-handed, close enough in quality that either is a plausible starter for this game, but neither of them needed to start a game after this one. That is a very reasonable situation to set up a planned piggyback with the idea that each of them pitches no more than 3 or 4 innings. That makes more much sense than having one of them pitch to a single hitter to try to force the opposing team to burn 1 or 2 bench players.
Sorry, but you really haven’t thought through this idea, including that teams don’t normally run out of bench players in 9 inning playoff games, even in the NL.
I’d expect teams should also plan which pitchers go when depending on who’s at the plate. So maybe instead of looking at outs, the better way would be to look at lineups. Washington’s top 7 has been the same in the games against lefties. Essentially, you probably want right handed pitching for the Rendon-Zimmerman-Espinosa trio when possible, so to me that would leave it as:
Hill: 13 batters
RHP: 6 batters (Rendon through Turner)
LHP: 3 batters
and essentially alternate that when possible, although you can stretch guys sometimes
So doing things that way, Hill/Blanton likely only get through the 4th or 5th inning (the first 19 batters). Then Urias would get Harper-Werth-Murphy. If he’s pitching well, or you’re up, maybe you try to leave him another time through the lineup and go to the matchup game after that.
Either way, I assume you basically don’t want any RHP other than Jansen to face either Harper or Murphy, and I’d expect that they have enough arms to make sure that’s the case.
If I’m Dave Roberts, I want to avoid one matchup as much as possible: a LHP pitching to Werth. Werth is smashing the ball this series and has extreme platoon splits. Murphy is excellent against RHP but still very good against LHP. Same for Harper. Same for Turner.
One slight complication: do the Dodgers have a strong preference for bringing Urias into the game at the start of an inning?
He has been used a bit as a reliever – I see 3 relief appearances with the Dodgers this year and 4 more at AAA – but has far more experience as a starter. I don’t know if any of his relief appearances have been entering the game in the middle of an inning or not.
Solid piece as always Dave but I don’t expect Dave Roberts to break from tradition now with regards to bullpen usage. It would take an act of God for us to see Jansen come in before the 8th inning.
Well, well, well. I stand corrected!
Your strategy would work well because Joe Blanton is their most effective pitcher off the bullpen, other than Jensen. I don’t see Dave Roberts as a non traditional strategist as a manager and would be surprised to see him use Blanton earlier than seventh inning. I’m still trying to figure out Dodgers’ refusal to walk Murphy. I wouldn’t let him beat me, especially with runners on base.
Murphy came up with the bases loaded in Game 4…still give a free pass?
I understand the handed-ness argument for the tandem starters, but flipping Kenley and Dayton doesn’t sound like it nets any advantage since they probably won’t be facing the same batters. You’d be changing the look for some batters when Jansen comes in, but Dayton would still face batters whose last PA was against a lefty. I’d think it would be more advantageous to just have Kenley pitch to the best bats, whenever that happens late, rather than worry about the negative effects of stacking lefties at that point.
Josh Fields not all that good??? Are you kidding Dave? The guy *only* strikes out more than a batter per inning and throws 99, what more can he do? He’s been elite for a few years now and I have no idea how the Astros let him get away. Meanwhile you’re praising Will Smith in another article and dumping on a better reliever here…smdh.
Maybe Dave has other reasoning he’d like to share, but Fields has been awful against left-handed hitters. The Nationals will alternate R-L hitters 1 through 5. Espinosa is a switch hitter who will probably hit 7th. So the situations where Fields makes sense against the Nats’ lineup are very limited.
Fields’ career splits aren’t that awful: .316 wOBA allowed to LHB (3.56 FIP) vs. a .289 wOBA (2.58 FIP) allowed to RHB. Though I suppose that could be skewed for a reliever if he’s purposefully kept away from really good LHB’s but is used to pitch to really good RHB’s.
No mention of Hill having reverse splits this year? His K-BB% to RHB is nearly 9 points higher than to LHB. If Hill is pitching well, removing your second best pitcher by some margin to have a platoon advantage (even though they would still be worse than Hill vs.RHB) by putting in slightly above average reliever would be a grave mistake. In my opinion, your suggested pre-determined match-ups by TTO would be over managing.