Play-In Game Strategy: Skip the Starter

In a couple of weeks, we’re going to be treated to a couple of one-game, winner-take-all showdowns to determine the final playoff team in each league. The addition of the wild card play-in game changes the context of the playoff races in a new and meaningful way, and also creates some interesting questions for a handful of teams. While MLB has had tie-breaking games before, they’ve never been known to be necessary until the final day of the regular season, and teams simply had to make adjustments to play an unscheduled game at the last minute.

Now, however, it’s likely that one or more teams are going to head into the final few days of the regular season essentially certain that they’re going to be involved in the play-in game. For instance, the Braves currently have a seven game lead over St. Louis in the wild card race, and even after sweeping Washington this weekend, they’re still 5.5 games back in the NL East. CoolStandings currently gives the Braves a 96.3% chance of winning one of the two wild card spots, but only a 3.6% chance of winning their division. Barring some kind of epic collapse by either the Braves or Nationals, Atlanta is going to be one of the two NL teams in the play-in game, and they can start planning for that game right now.

And, as part of that planning, they should seriously consider the idea of skipping the starting pitcher entirely.

It’s pretty well known at this point that relief pitchers perform better than starting pitchers, as the ability to air it out for 15-20 pitches leads to increased velocity and better stuff for most pitchers. Additionally, relievers more often face same-handed hitters, so they get a larger advantage from platoon splits, and they hardly ever face the same batter more than once in a game. These advantages add up in a hurry, which is why quality relief pitchers can sustain performances that even the best starters can’t come close to.

Except for in a few cases – Justin Verlander in Detroit probably the most notable one – the team’s most effective pitcher is hardly ever a member of the starting rotation. In reality, the more innings you can allocate to good relief pitchers, the less likely the opposing team is to score runs. Starting pitchers are essentially a necessity born out of the need to play many days in a row without a day off, but that restraint doesn’t exist in the play-in game.

There’s a strong possibility that the Braves will know by October 1st whether they’re going to be in the play-in game or not. That game won’t take place until October 5th. If they win, they won’t play again until October 7th. This schedule — and the days available to prepare for it — means that the Braves should choose to begin the game with their best pitchers, and that means their relievers.

Give Craig Kimbrel, Jonny Venters, Eric O’Flaherty, Luis Avilan, and Cristhian Martinez the last couple of days of the regular season off in order to make sure they’re fresh for the play-in game. Then, announce Kimbrel as the starting pitcher, and tell him he’s getting the first six outs of the game. He’s faced 5+ batters nine times this year already, so it’s not a drastic departure from a quantity standpoint. Then, for the next few innings, you can mix and match right-handers with left-handers, taking advantage of the platoon match-ups to quash any early rallies. You’d hope to get another 9-12 outs from these four, which would take you into the sixth or seventh inning.

From there, you’re dealing with a significantly larger pool of information from which to make decisions about which “starter” you want to use to pitch the last few innings. If the game is still in doubt, you probably want Kris Medlen ready to come in to finish out the final three innings, as he’s been the team’s best starting pitcher for the last few months and has experience out of the bullpen. Paul Maholm and Tommy Hanson both have significant platoon splits, so they could be ready for match-up reasons if Medlen struggles or a need arises in a left-on-left situational spot. If the Braves offense happened to turn it into a blowout, either Maholm, Hanson, or Mike Minor could simply be asked to finish out the final three innings, thus saving Medlen for a Game 1 start in the NLDS.

By flipping roles and begininning the game with the relievers, the opposition wouldn’t be able to set their line-up to maximize platoon differential, and you’d ensure that the game wouldn’t be lost before the team’s best pitchers got a chance to pitch. Additionally, you’d ensure that you’re never wasting an at-bat on a pitcher, giving your offense a boost in the process as well. And, because of the predictable nature of the schedule, the Braves should be able to rest their bullpen before and after that game takes place, meaning that they’ll all be available for the entire division series, so that this usage pattern wouldn’t have an adverse affect on the team’s chances of winning the first round of the actual playoffs.

If they simply went with Medlen or Hudson in the play-in game and treated it as a traditional contest, they’d actually be lowering their odds of winning the NLDS by removing whichever pitcher started the play-in game from the possibility of making two starts in the first round. In starting the game with relievers and only using a guy like Medlen if necessary, they not only give themselves the best chance to win the play-in game, but they keep the possibility of their two best pitchers starting games one and two of the first round alive.

The Braves aren’t the only team in this situation either. The Orioles bullpen has been dramatically better than their starters, and this should also be a consideration for the A’s, whose best chance to advance deep in the playoffs comes from Brett Anderson getting two starts in the ALDS. By shifting innings to the relievers to start the game, these teams can ensure that their best pitchers pitch the most innings in a win-or-go-home contest while also gathering more information on whether to deploy their best starter or save him for the start of the division series two days later.

Last year, the Yankees effectively used this kind of all-hands-on-deck approach in Game 5 of the ALDS against Detroit, as Ivan Nova only went two innings and no pitcher faced more than nine batters in the contest. However, the Yankees offense was never able to score more than the two runs Nova allowed in the first inning, and so the one run allowed by the non-starters in that game proved to be the deciding margin of victory. With a little more time to prepare for a winner-take-all contest, I’d like to see a team like the Braves push this strategy to it’s full potential. Relievers first, then decide which starter is going to pitch based on the score. It might sound crazy, but it really just boils down to one simple plan – use your best pitchers to win a must-win game. And, for teams like Atlanta, their best pitchers are not their starters.

Print This Post

Dave is a co-founder of and contributes to the Wall Street Journal.

93 Responses to “Play-In Game Strategy: Skip the Starter”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Adam says:

    Luis Avilan is left handed.

    Vote -1 Vote +1

  2. Crumpled Stiltskin says:

    But relievers are also not used to starting games and that leads to a different sort of pressure. I’d much rather start a guy like hudson or Hanson or medlen and tell him he’s only going four innings so he doesn’t have to pace himself. Then they are likely to have a similar spike in stuff, aren’t they?

    Vote -1 Vote +1

    • philosofool says:

      It’s hard to know. A pitcher who is asked to pitch a little differently might need time to adjust. For example, some starters rely on their fastball more in early innings and might be uncomfortable mixing in a lot of off-speed stuff in the first inning. Etc.

      Vote -1 Vote +1

    • gobears says:

      Also, part of the point is that you have your starters ready for the LDS (no, not the Mormons) to follow….

      Vote -1 Vote +1

  3. sporkless says:

    I like the idea, but a manager won’t be brave enough to try it. He *might* be brave enough to pull his starter after 4 innings or so, even if he’s doing well.

    Nice little side benefit of your proposal for the NL teams: probably zero plate appearances by your pitcher.

    Vote -1 Vote +1

  4. Toasty says:

    Yes, very yes. I’ve long wanted to see the “no starter” plan be tried out by some out-of-contention team (much respect to the Rockies for their little experiment this year, even if it didn’t go well), but here is a situation where it actually makes sense.

    If the Brewers make it in, though, they might be wise to keep the bullpen out of the picture.

    +7 Vote -1 Vote +1

  5. Cus says:

    An added bonus would be the flexibility of pinch hitting for the reliever in early game situations and ensuring the opposing team can’t pitch around the eighth hitter to exploit the pitcher spot.

    Vote -1 Vote +1

  6. Rey22 says:

    Sounds good in theory, but I doubt any manager has the balls to pull this off.

    Vote -1 Vote +1

  7. Fredi Gonzalez says:

    There is no way I’m going to do this. I’ve just recently warmed up to the idea that there are other times to use Kimbrel than just the bottom of the 9th inning of a game that we are winning by 1-3 runs.

    And you forgot about Chad Durbin. I love Durbin. He’s a bulldog. So he doesn’t have the good “stats” like FIP, xFIP, WAR, or SIERA (whatever that is) that the internet nerds tell me about, but let me tell you he knows how to pitch.

    I am considering starting Constanza in RF for Heyward though. Let me tell you, speed never slumps. And speed is what wins games like this one.

    +84 Vote -1 Vote +1

  8. Dan says:

    This is a nice idea. Too bad creative managing is heavily discouraged.

    Vote -1 Vote +1

    • rustydude says:

      What about Joe Maddon? He’s fairly creative. Might he take a stab at this if his team makes the play-in game?

      Vote -1 Vote +1

      • Dan says:

        I agree that Maddon is probably the most creative manager, but this one is probably too dramatic for even him to try. You would need a manager with rock-solid job security, like Earl Weaver or Whitey Herzog had.

        Vote -1 Vote +1

      • philosofool says:

        What Dan says. If you try this and it back fires, you could lose your job. It doesn’t matter that it makes a lot of sense. It doesn’t matter if you lose 1-0 and your pitching strategy actually worked well while the offense choked. Baseball is extremely uncomfortable with suggestions that it has been doing things incorrectly for 100 years and will punish you for suggesting it.

        +5 Vote -1 Vote +1

      • chuckb says:

        The ability to use Price probably precludes this.

        Vote -1 Vote +1

  9. Dustin says:

    It is my understanding that the teams can make roster adjustments after the play-in game, as that is a separate round of the playoffs. If true, why not jettison your third through fifth starters in favor of some bench bats or an additional bullpen specialist, assuming you’ve got someone worth having for that 1 day.

    +16 Vote -1 Vote +1

    • NatsFan73 says:

      Is that right? If so that means 5-6 extra bats/relievers all told. 3 starters you’re not going to use under any circumstances, plus 2-3 of relievers that normally only get mop-up duty. Someone like Durbin, for example, has no business on the play-in roster.

      Vote -1 Vote +1

  10. Average_Casey says:

    This is a creative idea but they would need to figure out how to get the normal starting pitcher who is going to relieve ready. Starters warm up differently and prepare differently and some (I don’t know the Braves’ starters so this may not apply) start off a little shaky and get better after a couple of innings. You wouldn’t want a starter in a high leverage situation coming into a game and trying to get a feel for his pitches.

    Vote -1 Vote +1

  11. Anon21 says:

    I love the idea, and it’s quite well-suited to the Braves’ roster, featuring no aces in the rotation but one oh-my-god-how-does-he-do-that elite reliever, a couple of above-average to good setup types, and several quality matchup specialists. Unfortunately, I think there’s a vanishingly small chance that Gonzalez (or Wren) bucks conventional wisdom and implements this strategy, especially given the way Medlen has been pitching.

    Vote -1 Vote +1

    • TKDC says:

      And who could blame them? If they did this and lost, they’d both be fired.

      Vote -1 Vote +1

      • Karyn says:

        Maybe not. They have corporate ownership, not a single owner or family ownership. If they try it, and don’t win the game, Wren might be able to make the case to the suits that it was a legit strategy by focusing on the numbers. I think it’s more likely that heads would roll in a traditionally-owned tea,.

        Vote -1 Vote +1

  12. ben says:

    This is a great idea and I’d love to see it tried out. Of course, you gloss over a major point in the phrase, “the more innings you can allocate to good relief pitchers.” GOOD relief pitchers. There are a lot of teams that this strategy would be a disaster for.

    Vote -1 Vote +1

    • Jason says:

      Good point.

      Also, with every pitcher you bring into a game you increase the odds of using a pitcher that doesnt have his best stuff that day. If you use 7 pitchers to get through 9 isnt there a pretty good likelihood that at least one won’t have their command? Its a recipe for a big inning. Going with proven starter is a risk mitigation approach.

      Vote -1 Vote +1

      • ben says:

        this is exactly what i was thinking.
        also, if you’re the braves, why not just put one of your better starters in for the first time through the lineup? eat a couple innings to save the bullpen and still have them fresh for the first playoff series.

        Vote -1 Vote +1

      • Greg says:

        I don’t buy it. If this was truly the case, you would want to leave relievers in for a couple innings per game so you minimize the number.

        Vote -1 Vote +1

  13. Irrational Optimist says:

    Skip the Starter: When Stats Attack

    Absurd. It just isn’t true that the best way to continue in the post-season is to maximize the stat-glorified pitchers in any order. Starters are expected to go out there and limit runs. Relievers are expected to go out there and get outs. I know they don’t sound like a big difference, but the mentality is very different and messing with that is bad voodoo. Even if you have to walk your stud out there on short rest, that is better than having some reliever in when the crowd is still sober enough to make their cheers cogent.

    -25 Vote -1 Vote +1

    • Irrational Optimist says:

      Alright I thumbs downed myself. Still uneasy about it, but warming up to the idea. Wait…now getting excited.

      +14 Vote -1 Vote +1

    • rarumberger says:

      “Bad Voodoo”?

      You’ve (probably inadvertently) put your finger on exactly why no manager in the game has the cajones to try something like this. Superstition and preference for qualitative over quantitative analysis is rampant in baseball.

      We’ll never get this kind of thinking applied to actual games as long as the only managerial candidates are former players.

      Vote -1 Vote +1

    • Kick me in the GO NATS says:

      I cared so little I had to tell you about it! “where is the thumb sideways!”

      Vote -1 Vote +1

  14. TKDC says:

    With the one off-day in the NLDS, this is basically trading (possibly) 2 games of Hudson and one of Medlen for 2 games of Hudson and 1 of Medlen. I’d rather let the starter know he needs to air it out, get 4-5 innings, and then have my stud relievers go out there, including innings 8-9 for Kimbrel.

    Vote -1 Vote +1

    • TKDC says:

      I meant trading for 2 games of Medlen and 1 of Hudson – you get the point. I don’t see that as that great of a gain to try a new strategy for 1 game.

      Vote -1 Vote +1

  15. chiefglockandhummer says:

    braves might be good example just because our bullpen is so good, but despite jesus having pitched the last month for us in kris medlen drag, i’m not sure that our starters are distinguished enough in quality for this to matter so much for us. one hudson or medlen start in the NLDS is only so much better than a hanson or maholm start, it seems to me. might be more important for like, the rays and a’s, right? of course, just having kimbrel and o’flaherty pitch as much as possible in the play-in is definitely good strategy, and freddie might even do that.

    Vote -1 Vote +1

    • Joe Schmoe says:

      Our you crazy? You don’t think the Braves have a significantly better chance of winning if Medlen starts rather than Tommy “Junkballer” Hanson? You obviously haven’t watched Hanson labor through every single inning he has pitched over the past three months. Hanson should not have been making starts for the last 6-8 weeks. I would much rather have seen the team pitch Delgado who gives us at least as good a chance of winning right now as Hanson and would be developing with every start. He was just starting to turn the corner it seemed to me when they pulled him out of the rotation.

      Vote -1 Vote +1

  16. Schlom says:

    Aren’t relievers for the most part starters who couldn’t cut it? I understand that pitchers are more effective over shorter periods so wouldn’t you want your best pitchers out there? Most likely that would include most of the starters and a few relievers (certainly for the Braves it would be Kimbrel and probably the lefties for matchup situations.

    Vote -1 Vote +1

    • DJG says:

      I was thinking something similar when I started reading this post, but the — “From there, you’re dealing with a significantly larger pool of information from which to make decisions about which “starter” you want to use to pitch the last few innings.” — aspect of the plan seems to be why you would want to go almost exclusively with relievers in the beginning. If you just threw your best pitchers for shorter periods, you wouldn’t be able to set yourself up for the NLDS as easily.

      Vote -1 Vote +1

  17. Chris from Bothell says:

    On paper, this is creative and a credible approach. It also puts the other team off-balance slightly, as they can’t align their starting lineup to try to do the best they can against the other staff ace or any given handedness.

    The ripple effect from all of this is probably that you get the purest read on which team is better, as both teams will put out their best lineup and best pitcher (starter or reliever) to begin the game, and then march down their rosters in order of quality of pitcher / bench bat, until one team emerges. It would be interesting to see, in that regard.

    However, in real life, this would be an abomination, because all the flippin’ pitching changes would make the one-game playoff games take 5 hours or more each. You think Yankee/Red Sox games are bad? Yikes.

    +8 Vote -1 Vote +1

  18. Mike says:

    If I’m the Braves, I let Medlen start the game and go until he has to bat, then go to the pen. That way he throws 2-3 innings and would be fine to pitch game 2 of the LDS, if they make it. After that you go to Kimbrel for two innings and then match ups the rest of the way. I agree with the earlier commenter that you replace Minor and Sheets with two extra hitters, so that if the game goes to extra innings you have a couple of extra batters to pinch hit with, and still have Hudson/Maholm/Hanson to use if you run out of relievers.

    +13 Vote -1 Vote +1

    • BillWallace says:

      This plan is exactly what I came to post.

      Vote -1 Vote +1

    • Matt says:

      Mike Minor is actually pitching quite well these days: 18.1 IP, 1 ER in September. No way do I take him off the roster for this game. But I don’t see how starting Medlen can be considered a mistake given how well he’s pitched this year.

      Vote -1 Vote +1

    • Dan says:

      Sheets and his two above average pitches are actually well suited to bullpen duty. I would have no problem with sending him in and telling him to air it out.

      Vote -1 Vote +1

  19. Mr Punch says:

    “It’s pretty well known at this point that relief pitchers perform better than starting pitchers …” And there’s the problem. It sure looks like it from the rates, doesn’t it? And yet the majority of relievers, even of pretty good relievers, are failed (or developing) starters. Relievers do have all those advantages you cite, which make them “perform better” without actually being as good. It makes much more sense, as several above have suggested, to start a starter and have him go once through the lineup.

    Vote -1 Vote +1

    • RobBob says:

      Many times, potential starters “fail” because they don’t have the endurance to last six or more innings — unless they dial down their per-pitch effort, which makes them ineffective. Guys like Kimbrel and Chapman can thrive as relievers because they can simply cut lose for a few outs and they’re done.

      Vote -1 Vote +1

  20. MSom13 says:

    Since there is a day between the play in game and the real playoffs, can’t you take this a step further and just let 6-7 or your best pitchers pitch the play-in game regardless of whether they are a starter or reliever?

    For instance, if I’m the Cardinals – I recognize that Wainwright and Lohse are two of my best pitchers this year. I also realize that Garcia is the best lefty on the pitching staff. Can I use Wainwright and Lohse in innings 1-4 and then sub in relievers like Lynn, Garcia, Rosenthal, Salas, Boggs, and Motte to close things out?

    Sure, it might jeopardize the next series a bit, but it would give you the best chance to win that one all important game at least.

    Vote -1 Vote +1

    • BillWallace says:

      Yes, I think giving any consideration at all to the Division series is a mistake. Maximize the one game. I’d start your best starter for one trip through the lineup or until they came up to bat depending on the leverage of their AB. Then start with your best relievers, probably mixing in your 2nd best starter for a 6-9 batter appearance.

      Vote -1 Vote +1

  21. Bobby Mueller says:

    The Pittsburgh Pirates (Jim Leyland) did this in Game 6 of the 1990 NLCS. From an ESPN article:

    Ted Power, Pirates, 1990 NLCS. Power was a reliever who hadn’t started all season when Jim Leyland tagged him to start Game 6 against the Reds. The reason? The Reds had a platoon-heavy lineup so Leyland’s plan was to get Lou Piniella to start the left-handed hitters and then bring in Zane Smith. Power gave up one run in 2.1 inning and Smith one run over four, so the plan worked, but the Pirates lost 2-1.

    Power was 35 years old and had pitched in 40 games that year, all in relief. He even had 7 saves.

    +10 Vote -1 Vote +1

    • The Ronin says:

      Leyland must have grabbed someone else’s hand rolled “cigarette” that day because he is the last person I would expect to try anything out of the ordinary.

      Vote -1 Vote +1

  22. Bridgette Kerkove says:

    Does anybody know what the roster rules are for the one-game playoff? I see 3 possibilities:

    1) Treat it like an extension of the regular season, so 40-man rosters.

    2) A separate 25-man roster just for this game, that you can change before the ALDS/NLDS. So in this scenario, you could theoretically remove some of your starting pitchers off the roster for extra bench bats, and then put those SP’s back on the roster for the ALDS/NLDS.

    3) The same 25-man roster that you must also use for the NLDS/ALDS. This way you can’t get sneaky about dropping SP’s off your roster.

    I have no idea what the roster rules are for this game. Is it one of these three? Is it something else?

    Vote -1 Vote +1

    • TKDC says:

      I’d be very surprised if it weren’t #2

      Vote -1 Vote +1

    • Ron says:

      Tried researching this for a bit, it is not easy to find the answer. The only copy of MLB rules (the ones that govern the operation of baseball not playing rules) I found is pre 2012. Those rules basically said that the rosters had to be set before each series. So by that logic I agree with TKDC number 2 seems to be what it would be. It would not be a September roster unlike the tiebreaker games. I have not found a copy of the Major League Rules since the addition of the wild card game.
      While you could add a starter and wear him out and replace for the rest of the postseason it doesn’t sound like a great strategy because he probably not be the best pitcher on the team. The point of using bullpen guys is to have the best matchups to win the game.

      Vote -1 Vote +1

  23. henry says:

    i like it more for the orioles. the thing for the braves is, medlen has been unbelievable. if his dominance continues, they might want to stick with him.

    Vote -1 Vote +1

  24. ttnorm says:

    I understand the thinking but I am not a fan of this.

    1) Most teams do not have 3/4/5 bullpen guys better than their 1/2 starter.

    2) If you win, your bullpen is seriously degraded for the next series.

    Vote -1 Vote +1

    • philosofool says:

      That’s why he suggests that the Braves rest their pen a lot for the last few games of the season. With 40 man rosters, giving Kimbrel et al. three days off is easy. Then each guy pitches throws about 15-30 pitches in the play in game, and maybe pitches the next day, which is normal.

      Vote -1 Vote +1

    • Gomez says:

      Degraded? You sure? They get a day off before the LDS and no reliever will likely throw more than a couple innings.

      Vote -1 Vote +1

  25. baycommuter says:

    I don’t like this for the A’s… Doolittle, Cook and Balfour for 2 innings each are not significantly less likely to give up runs than Anderson for 6 iinnings with his 0.91 WHIP. It might make some sense for a team without a true ace (Baltimore).
    Griffin has been as unhittable as Anderson. If the final Wednesday is meaningless, it would make sense for the A’s to skip his turn and then use him in the first game of the ALDS if they get there.

    Vote -1 Vote +1

  26. Harry says:

    Tony Larussa won a world series last year by taking his starter out after 3-4 innings and making it a bullpen game. Of course that could have been that the starters(outside of Carpenter) stunk.

    Vote -1 Vote +1

  27. Bryan Curley says:

    I think the idea itself is great, but unveiling it when the stakes are highest isn’t smart. We’ve long heard pitchers lament about not having a defined role and frequently see pitchers who struggle when their role is even slightly adjusted, so the risk of this happening in a winner-take-all game is just too high.

    Vote -1 Vote +1

    • Ron says:

      But isn’t doing this when the stakes are high part of the point?
      The current structure of a mlb pitching staff is to get the team through the season. The season is a marathon but one game is a sprint. A starting pitcher is the endurance runner but in a game game series you don’t need that. You need a combination of sprinters. One game is exactly the situation to do this, either in a one game series or a 7th game.

      Vote -1 Vote +1

    • vivalajeter says:

      This is my thinking as well (although I don’t necessarily think the idea itself is great).

      They’ve played the entire season in defined roles, and they could wind up with 7-8 more wins than the team they’re playing. They’re at home, and they’re the favorites. So in an effort to slightly increase your odds in the next round, you’re willing to completely change everybody’s schedule and workload for a single game?

      Plenty of relievers say they like consistent work, and don’t feel as sharp if they have too much rest. Yet we’re looking to give them plenty of rest in the final week, only to ask them to pitch more than a full inning in a do-or-die game? And you’re asking your closer to pitch in the 1st inning, and having your starters pitch in later innings with the game on the line?

      If this is a video game, it might make sense. In reality, it’s very risky. Players like defined roles, and there’s a chance a great pitcher wouldn’t be as great if you change his role for a day. The risk isn’t offset by any benefit, in my opinion, especially since there’s no evidence of there even being a benefit.

      Vote -1 Vote +1

  28. ToMcN says:

    Why wouldn’t the road team in a play in game bat a position player second (in the order) as the pitcher? Said team could then just put in a reliever after the top of the first and proceed with the bullpen game.

    +9 Vote -1 Vote +1

    • Ron says:

      And have a left and righty warm up to hide who they would start. Sneaky.

      Vote -1 Vote +1

    • saskatunes says:

      interesting idea… I like it. Why not do it all the time as a road team?

      Vote -1 Vote +1

    • Marcus Andrews says:

      Well I like the idea, theoretically the second at bat of the game has a much lower effect on the win probability than later at bats. So yes, you’re skipping your pitcher first time through, but it’s bringing him up more often later in the game in a likely more meaningful situation.

      And deciding which hitter to use would be difficult. You don’t want to use your best bench bat, but you also don’t want a guy like Jose Costanza batting in the first inning.. Well maybe Fredi would, but he shouldn’t want that.

      Vote -1 Vote +1

      • Brendan says:

        I think this first part is wrong if I understand correctly. It’s essentially batting your pitcher 11th rather than 9th. Your pitcher won’t get any extra at bats or be more likely to come up in a crucial situation at the end of the game, and if he did you could just pinch hit. I think you’re second point is relevant though…you want your best hitters batting more, and putting a bench bat in the 2 hole over a good starting hitter is not necessarily a good move.

        Vote -1 Vote +1

  29. Brooks says:

    When the Braves end up essentially certain to be in the play-in game, why not try this for one of the last regular season games (maybe 4 games from the end) so that the first time Kimbrel starts a game isn’t in the must-win game. If everyone feels good about the way it works, you use it for the play-in game. If not, you have time to figure out if it’s the tactic or just an off day from the relievers.

    +11 Vote -1 Vote +1

  30. snoop LION says:

    people can laugh all they want at the “voodoo” stuff, but some pitchers just don’t react well to adjusting their roles. but in this case with the Braves, I think it could work well with having a guy like Medlen whose had extensive experience starting and relieving in the bigs (even just this year).

    Vote -1 Vote +1

  31. Jon says:

    Interesting idea, but you could possibly burn three of your starting pitchers and your most effective bullpen guys? Not sure this makes sense, but its worth advancing.

    Vote -1 Vote +1

  32. Brendan says:

    1) I think this would have to be something discussed in advance. If I were the manager thinking about doing this, I would talk to the GM to make sure he’s on board with it as well as the pitching staff. Theoretically, the inning shouldn’t make a difference (as a college pitcher, I’ve never thought any differently about it or struggled to change), but for some reason, some pitchers have trouble adjusting to a different inning, so I would make sure they are okay with it and don’t have a problem with it. If Kimbrel says he can’t start and can only throw the 8th or 9th, I wouldn’t mess with his head.

    2) While some relievers are failed starters, that does not necessarily mean that they are worse in a relieving role than a current starter. Some pitchers are moved to the bullpen because they don’t have the stamina or a third pitch, but they can be extremely successful as a one or two inning reliever. For example, Kimbrel is a better option for one inning than probably any of the Braves starters. In addition, a starting pitcher might have trouble if he goes and airs it out for one inning rather than pacing himself like he is used to and has done and practiced all year. So in that respect, many relievers used in one inning stints is better than a starter.

    Vote -1 Vote +1

  33. Matt says:

    Not sure if this point has been brought up, but there will be 25 man rosters in these “playoff” games. Not 40ish.

    Vote -1 Vote +1

  34. schlomsd says:

    I don’t think this is feasible. There are too many unknowns involved for you to get enough guys warmed up in the bullpen. Let’s say the Braves do this and start with Kimbrel. He walks the first two batters so you are forced to warm up a potential reliever, maybe two since there is a left-handed batter with a big platoon split coming up. Now if he walks another batter you might pull him which presents another whole set of problems. But maybe he gets out of it with a triple play on the next pitch and then starts mowing down batters on the first or second pitch. You’ve had guys warming up in the bullpen this whole time so you have to sit down those guys to give them rest (as they can’t warm up for 3 innings) and get two different relievers up. And so on.

    This seems like something that would work in Pursue the Pennant but not in real life.

    Vote -1 Vote +1

  35. Tom says:

    So Dave is proposing using the best pitcher in one of the lower leverage spots (1st inning, noone on base)?

    Isn’t Kimbrel the type of pitcher you would leverage with men on base, where a strikeout has even more value?

    If you plan to get 6 outs from him I’d rather have him end an inning where there is a man(men) on base, pitch a (hopefully) clean inning and then get 1 or two outs in the next inning where he can hand off a low leverage situation to a lesser reliever (1 or 2 out, noone on)

    Even if you were to go with an all bullpen approach, it doesn’t seem wise to start your best reliever.

    +13 Vote -1 Vote +1

  36. jpg says:

    The biggest issue I see is that you’re relying on 6 or 7 different relievers to all “have it” that day. If you throw a starter you could always just give him a short leash or just pull him when his spot in the order rolls around. If that starter has a hot hand you can ride him out.

    What happens if Venters is getting rocked and can’t get out of the third inning. And let’s say he’s penciled in as the third and fourth inning guy. Now you need O’Flaherty to put out Venter’s fire, which becomes an issue because he was supposed to get you through the fifth. No problem…of course until Fredi realizes that Venters was due up in the fourth…

    On paper it looks like a solid plan but it’s predicated on a bunch of guys all pitching well. I myself would probably go with my starter and situational guys to get me through 5 , have Venters and Kimbrel get the last 12 outs and hope for the best.

    Vote -1 Vote +1

  37. BillyF says:

    Having Kimbrel start or pitch anytime before the 6th won’t work. Closers are known to need the adrenaline hyped, which means he’ll perform as his career only when the game’s on the line: high leverage, 8th or 9th and close game.

    -6 Vote -1 Vote +1

  38. Ryan says:

    The Yankees did it because Nova got hurt. Not cause they chose to. Nova was pitching well enough to go at least five innings that game. Maybe this writer should know about the game rather than just looking at the box score and seeing Nova only went two innings.

    Vote -1 Vote +1

  39. Craig Nasty says:

    Interesting idea, definitely enjoying reading all the comments. Unfortunately, we all know how much of a purist’s game baseball is, so I wouldn’t expect to see this strategy used any time soon. It would def be exciting to watch, as long as it wasn’t MY team doing it.

    The best argument against it is relying on so many relief pitchers to all pitch well on the same day. The more pitchers you use the lower those odds get. God forbid one of them doesn’t have their stuff that day, has a rough inning in the 2nd thru 4th, and now you’re down a couple runs. Now one of your better relievers is out of the equation, you’ll most likely need to pinch-hit for the pitcher from here on out, and you’ll be relying on a starter to come in and hold the line.

    We’ve seen this used before in deciding playoff games, a top starter being brought in late in a game, the ‘all hands on deck’ thing. My latest memory of it is Lincecum coming in in game 6 of the 2010 NLCS in Philly. For as dominant as he was that postseason, he was far less effective in that little relief stint.

    The mental/pre-game preparation aspect of this strategy is what’s being glossed over the most. All starters are different. They all have different routines on their day to start. Crazy superstitious stuff like eating the same meals or sacrificing live chickens. After 162 games, and qualifying for this wild-card game, NOW you ask them to throw all that out the window and come in in the 7th inning with possibly runners on base? That’s just flirting with disaster.

    Too risky if ya ask me. I might try this in my MLB The Show playoff games though.

    Vote -1 Vote +1

  40. Curious George says:

    But if the Orioles used Chris Davis as their first pitcher they’d eventually lose his bat in the lineup, amirite?

    Vote -1 Vote +1

  41. Jason H. says:

    Great plan. Use all your best relievers and all of your pinch hitters early in the game so that the manager has zero flexibility late in games when the stakes are highest. This is a plan only the opposing team could love!

    …in actuality, baseball managers have been managing these types of games correctly as long as I have been watching baseball. You start your best starter available. You then use him until the first sign of trouble. It is understood that there is a quick hook in must win games. Then you use your bullpen similar to how Cameron wants except, you save the best guy for high leverage situations in the last two innings. …sometimes, baseball is conservative because they’ve already hit on the best strategy.

    Vote -1 Vote +1

  42. Ruki Motomiya says:

    I like this idea, but I have a few reservations about it.

    1) The aforementioned many times in comments “what if one guy is off” dealio. I think the article presents only the best case scenario and doesn’t seem to offer enough of an alternative for what happens if something goes wrong, such as if someone is just off that day or otherwise gets shelled and, say, loads the bases or whatnot. Seems a tad risky.

    2) Even if you rest your relievers, this is using a lot of relievers warming up a fair amount of pitches and going, say, an inning each, then going into the next game after one game of rest: What happens if your #1 starter sucks and needs to go early? You’ll need to do that again and your bullpen will be quite depleted from all the work over three days, so you’ll be on a thin rope.

    3) Wouldn’t you rather save your better relievers for the higher leverage situations to maximize their value?

    And, to me, #4 and most importantly…

    4) Why not do this the first game of the playoff series instead? If it fails, you don’t lose your one-game playoff and instantly go out, you still have games to make it up. Your #1 starter still gets use in the highest leverage games he can(The one-game, winner take all playoff, and he should be ready again around the elimination games), the rest of your starters are able to go as planned. Unless I am missing something, all you miss is the single off day, which I don’t think will matter too much given the strategy, and you still maximize the leverage stats? Am I wrong here or missing something?

    Vote -1 Vote +1

  43. John Thacker says:

    The Orioles have actually been doing a variety of this strategy, but based around the fact that they have a ton of guys who have not been good enough to really be starters, but can be pretty effective for 2+ innings.

    So if you look at recent games, you’ll see the Orioles do stuff like “Randy Wolf starts and throws 4 innings, then Tommy Hunter for an inning, then Brian Matusz for two innings, then Jake Arietta for an inning,” and so forth.

    Most teams don’t have three guys who started 16+ games for them this year who are now not in the rotation, though. But even those 6th to 8th starters might be more effective for 2 innings each than a lot of higher in the rotation guys, especially when the O’s don’t have a true ace.

    Vote -1 Vote +1

  44. Pip says:

    This is essentially the premise of my “10-man rotation” idea. The plan can help “starters,” too, by preventing them from facing the lineup more than a couple of times a game.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>