The Dodgers Try an Opening Gambit
Tonight, there’s only one game in town, as the Giants face the Dodgers in a winner-take-all, NLDS Game 5 slugfest in San Francisco. It’s been billed as a matchup between two borderline Cy Young candidates: Logan Webb, who humbled the Los Angeles lineup in the first game of the series, and Julio Urías, who started Game 2 for the Dodgers after a superlative 2021 season. Only, nope:
Corey Knebel will start Game 5 tonight.
— Los Angeles Dodgers (@Dodgers) October 14, 2021
This isn’t going to be a lengthy discussion of whether openers make sense. Teams clearly like the tactic as a way to fill innings, but almost never in front of a pitcher as good as Urías. I’m interested in what the Giants will do to counter it, and how that counter will determine Urías’s matchups.
When he took the mound last Saturday, the Giants set up like so:
| Order | Player | Position | Bats |
|---|---|---|---|
| 1 | Darin Ruf | LF | R |
| 2 | Kris Bryant | CF | R |
| 3 | Austin Slater | RF | R |
| 4 | Buster Posey | C | R |
| 5 | Wilmer Flores | 1B | R |
| 6 | Brandon Crawford | SS | L |
| 7 | Evan Longoria | 3B | R |
| 8 | Donovan Solano | 2B | R |
That’s a lot of righty bats, and San Francisco has played matchups all year. Three of those batters were out of the game when Urías left, replaced by lefties who could target the Los Angeles bullpen. Putting Darin Ruf and Austin Slater — in particular — at the top of the lineup meant that the Dodgers would be stuck with a tough decision: let Urías face tough righties a third time, or make a substitution and allow Gabe Kapler to counter-substitute with good lefty hitters.
As it happens, the Dodgers brought in Joe Kelly to start the bottom of the sixth inning, facing Kris Bryant to start off — Urías had retired Ruf as the last out of the fifth. LaMonte Wade Jr. entered in relief of Slater, and the substitution train was off to the races.
With Corey Knebel pitching the first inning, the Giants need to reconsider. For one thing, Brandon Crawford should move to the top of the lineup. After that, however, it’s a bind. You could, in theory, give Wade or Mike Yastrzemski a start, then substitute them out when their time to face Urías comes due. That’s a high cost, though: sacrificing their defensive versatility and solid bats to get one low-leverage plate appearance against a tough reliever is tactically shoddy.
Another option: start one of them and accept the bad matchup against Urías. That would probably mean Yastrzemski over Slater; Kapler has used Ruf against breaking-ball-heavy pitchers more so than lefties, while Slater has rarely faced righties at all. I might normally suggest using Tommy La Stella in relief of Donovan Solano as well, but La Stella aggravated an Achilles injury Tuesday night, and will likely only pinch hit. A prospective lineup might look like this:
| Order | Player | Position | Bats |
|---|---|---|---|
| 1 | Brandon Crawford | SS | L |
| 2 | Kris Bryant | RF | R |
| 3 | Mike Yastrzemski | CF | L |
| 4 | Buster Posey | C | R |
| 5 | Darin Ruf | LF | R |
| 6 | Wilmer Flores | 1B | R |
| 7 | Evan Longoria | 3B | R |
| 8 | Donovan Solano | 2B | R |
By setting the lineup this way, the Giants guarantee a few favorable matchups against Knebel. Batting Bryant second is a concession to the fact that you can’t have the platoon matchup every time out. He’s an excellent hitter regardless of the pitcher, and keeping him high in the lineup increases his expected plate appearances.
Still, I love this move for the Dodgers. The Giants don’t have any great options. They could stick with their lineup from Urías’ last start, bump Crawford up, and accept a low likelihood of scoring in the first inning — Knebel against a bunch of wrong-side platoon bats is a really tough matchup. They can substitute in some lefties to make that matchup palatable, but it will cost them in future innings, either in diminished substitutions or Yastrzemski/Urías matchups.
The best part of this move for Los Angeles? They’re cutting Kapler off at the pass. Kapler has shown what he likes to do: when the Dodgers switch from lefties to righties, he’ll pour on the pinch hitters. His lefty bats are likely the better of the two platoon options in a vacuum, anyway. If Urías went five innings and then Knebel replaced him, Knebel would likely face two solid lefties and a good righty, the same three batters (in type, if not exact identity) I expect him to face tonight.
By changing the order, the Giants can’t have their cake and eat it too. If they want to play matchups against Knebel, they obviously can — but forcing that decision to happen before the Dodgers’ best lefty goes means that you can’t hit Yastrzemski against Knebel, Slater against Urías, and Yastrzemski again against the rest of the bullpen. Going right-left-right instead of left-right-right with their first three pitchers in the game means the Giants can’t tee off.
It’s a small decision. It likely won’t change the outcome of the game. But it’s an excellent move by the Dodgers, even if it means more pitching changes and generally dulls the excitement (and slows the pace) of the best game of the year.
Ben is a writer at FanGraphs. He can be found on Bluesky @benclemens.
Rational side of my brain: great strategic move to get ahead of Kapler and force the Giants to react rather than dictate in a do-or-die elimination game.
Emotional side of my brain: HOW DARE YOU ROB ME OF A MATCHUP BETWEEN CY YOUNG CONTENDERS, DAVE ROBERTS. WAY TO MESS IT UP AND GIVE ME ANOTHER FOUR HOUR GAME. AAAAaaarrghhhhh…
Now knowing that Roberts tweeted Kapler about using Knebel as an opener, I think much of the reasoning behind the move was psychological. Roberts knows Kapler from his days in the Dodgers org, and probably figured that Urias would handle this switch-up better than Kapler would.
As someone who doesn’t care who wins the game, I hope it blows up in the Dodger front office’s face. Roberts wouldn’t d o this on his own, it’s another case of micromanaging percentage which I’m sure are correct to the nth decimal place, and I do understand that winning is the goal no matter how boring it may be.
However it’s an insult to Urias who is quite capable of going through the lineup two plus times at his best, which he has to be to beat the Giants anyway, Assuming they only give Knebel one inning,will they give Urias more than three, or will they take him out and go to the bullpen regardless of the score?
The baseball gods might force it to extra innings and a spot where Knebel would normally pitch to determine the game. Numbers are nice, but baseball has a way of flipping the script; ask the Rays.
One of the great (and underrated) qualities of Dave Roberts is that, right or wrong, his decisions, however influenced by the FO, always seem to have full team buy in. I don’t know if Knebel opening is all that significant, but whether it’s adapting to shifting, lineup changes, new positions, etc., Roberts has been a master of keeping an extremely happy clubhouse, and you have to trust his sense of how his team will react and respond more than your own from a distance.
Theoretically Knebel can just do the first inning and Urias can do however many he’s good for, theoretically still 5-7, and those just happen to be innings 2-8. But the problem I have with openers, from an excitement/narrative perspective, is that they don’t seem to just be taking a great middle reliever and moving their inning from 7 to 1, they seem to always be taking the first inning away from the starter, and then still going to the bullpen for innings 6/7+.
It depends where the lineup rolls over, as most pitchers who get this treatment are ones you are trying to delay facing the top of the order a 3rd time. Urias is good enough to get those chances, and he’s in fact more likely to get them if the top 3 of the Giants lineup looks like the one Ben proposes here, where it is better platoon matchups for him.
I too, hate innovation
I think on paper it’s a smart move and creates a suboptimal decision for the Giants. But I think it’s not worth the risk. Urias has started all of his games this season and there’s a chance that upsetting that routine could have a negative effect. Granted, he’s pitched in relief in the playoffs before and has pitched after an opener in other seasons. But he’s pitched so well this season that it seems like an unnecessary gamble to look for a 0.1% advantage that an opener might give.
Urias has extensive experience in the bullpen and in this role, and they say he doesn’t care. He is pretty high on the list of pitchers I am willing to believe this about, given his consistent thriving in all sorts of roles (including similar bulk roles) as recently as, you know, 2020.
He’s always struck me as a really cool customer, even as an MLB newbie. That’s why I’m not surprised by his success relative to his mid-rotation level stuff….could see him going full Jamie Moyer in the 2040’s 😉
Both in 2021 and overall in their careers, neither Urias nor Knebel have shown major platoon splits in their numbers. The Giants should just go with the same lineup as they’d use if Urias was actually starting, and use their usual mix and match approach when the bullpen comes in for good.
In other words, don’t out-overthink the other team.
Hitter platoon splits are stronger and more predictable than pitcher platoon splits. And a number of the Giants guys have pretty substantial ones for their careers.
The matchup of a hitter with a very weak platoon matchup against the same hand and a pitcher who shows no significant platoon difference is one that favors the pitcher, because the hitter’s weakness is still in play.
You can’t look at Urias or Knebel’s career splits and assume that a player with very strong career splits like Crawford just won’t have them because the pitcher doesn’t happen to have them.
Very good point. You’d basically have to split pitchers out between strong and weak platoon splits, and then see how much of a particular batter’s platoon split comes from pitchers of each of the four groups (RHP-s, RHP-w, LHP-s, LHP-w). I’d have to guess that for most batters with normal or larger platoon splits, you’d still have a platoon split comparing RHP-w vs. LHP-w.
Supposing Roberts and the pitching staff are prepared to save Urias for the 4th/5th and run out additional righties before going to him?
Oh, the levels!
Saved Urias for the first chance to PH for the nine hole in the lineup.
Let’s say they went with your line-up with 2 lefties at the top, wouldn’t the Dodgers be smart to keep the rightly reliever parade in at least one time though the line-up or so and then have Urias pick the 4th inning on? If Kaplar manages like the opener is only for the first inning (like your lineup implies), I would penalize him for it.
I was thinking of the same thing. It’s like a weird platoon/bench version of playing Chicken. Just daring the other manager to make the first move to the platoon switch.
If innings are innings, regardless of the order in which they occur, why not consider putting three good relievers in the first three innings and giving Urias the 6 inning save/”close”?
I agree that if things go well and the Giants only stack lefties on top of the order it would be quite reasonable for the Dodgers to use another good right handed reliever to follow Knebel and push the issue. This is especially true if they can scratch a run or two early and put the Giants in a position where they need to try to equalize.
One of the advantages of this strategy in an elimination game is that if you can get an early lead, even a small one, you put enormous pressure on the opponent to not concede plate appearances. I do not think Logan Webb is going to throw 15 shutout innings against an offense as good as the Dodgers in this series, and this kind of move is -really- valuable if you can get ahead.
Forcing Kapler to consider a pinch hitter literally any time they get a baserunner is huge pressure with long-term rewards (creating more favorable matchups later on). This also exists with the pitcher himself, something we saw in Game 7 in the 2018 NLCS when the Dodgers got up early against the Brewers, and they were forced to use Josh Hader to maintain the deficit only, and consequently also forced to never concede a PA by having a pitcher like Hader who was cruising hit for himself. The ideal situation for the Dodgers with this is to get a run or two in the first or second, and immediately put the Giants in a position where they have to treat every PA as offensively critical, especially if anyone reaches base. That situation is lose-lose for the Giants, forcing them to either concede scoring opportunities with sub-par matchups, or to substitute early and open themselves up to bad matchups later in the game.
Knebel in the proposed batting order could theoretically go through the lineup once, no? Or bring in Treinen? Leverage doesn’t matter in a do or die game. If their best move is Urias for 5, Knebel, Kelly, Treinen and Jansen for one each it doesn’t matter the inning as much as the matchup. It’s actually a bit strange to use Knebel as opener instead of Jansen who should have a less pronounced platoon split. Nobody else should pitch barring injuries or extras.
I’m not convinced either that the plan is for Urias to appear in inning two.
It will depend on the Giants lineup and how guys look, for sure. Urias will pitch 5-6 innings of this game, but I dont think the Dodgers would have any problem with that being 3-8 or even 4-9.
And as I mentioned above, if they can score at all, they can really put the Giants in awkward positions regarding substitution. When a game is scoreless, or you are ahead, you can make long-term optimal choices. In an elimination game, if you’re behind, you cannot do that for very long, and every opportunity with a baserunner is critical.
Ballsy move, anyway you look at it on a really big stage!Wonder if there has ever been anything comparable in a winner take all post season game.
Dodgers are probably just wanting Urias to leadoff the third:).
ah yes, Julio Urias cosplaying as the lead-off Shohei Ohtani pinch hitter/pitcher!
He means the Dodgers probably want Urias to start pitching in the bottom of the 3rd.
Maybe I’m old and just not following the logic, but I thought FG crowd would love this kind of move. Something Boone/Cashman were never accused of doing.
Urias has A LOT of experience coming in the middle of the game during the playoffs to boot.
Actually, the Yankees pulled a similar move last year. After routing Tampa Bay in the first game of the 2020 ALDS they started Deivi Garcia in the second game, replaced him after one inning with JA Happ and then watched Happ get bombed in a 7-5 loss after which Happ complained that he was mentally put off by not starting in the game. That doesn’t mean it won’t work for the Dodgers with two much better pitchers involved
And the fact that Urias proved he is more than capable coming out of the pen.
credit to Boone for giving JA Happ a plausible excuse for what likely would have been a shelling regardless (as a part-time NYY fan, Happ starts were always uncomfortable!)
Thanks for that – there I really was showing my true colors as a bitter, bitter Yankee fan. I knew there is a counterexample somewhere but apparently blocked out that Happ(less) episode in my mind that I couldn’t recall one.
I’m a dodgers fan and I like this move strategically, but man mlb really needs to quicken the pace of these games, way too many commercials…
What about the fact that Urias has reverse platoon splits? Last time they faced Urias they had a very righty dominant lineup and he had a very solid game. Urias’ FIP this year vs RHBs is pretty elite. So Giants should just play the lineup they would against Urias, which would also be an advantage against the RHP Knebel.
Reverse splits simply aren’t predictive until you have many, many seasons of data. A single year’s reverse splits should not govern any decisionmaking, and does not represent anything sustainable.
While I agree that the Giants’ best option was to just stick with a Urias-based lineup, it’s more because of their dependence on leveraging platoon splits in their lineup and not having to waste those tradeoffs early in the game (like they did with MYaz). Knebel’s platoon splits are probably best looked at as neutral, and not an overriding factor.
What about the benefit of having more professional hitters instead of pitchers hitting? Dodgers likely pinch hit the pitcher spot the first time through and could potentially hit for the pitcher every time with a platoon advantage.
I think that was the end game all along.
NL games basically have one pitcher AB per team anyway these days.
Since La Stella was probably only available to pinch hit anyway and is much better against righties, the move caused the Giants to bat him leadoff in likely a lower-leverage situation than he otherwise would have hit.
It also got Yazstremsski into the starting lineup when he didn’t start again Urias in game 2. The Giants can end up in a spot where Urias sees him 3 times instead of seeing Ruf or Wade.
Wasn’t the idea also around Knebel being a reverse platoon guy then going to another RHP in the middle?