By In Baseball

Shooting bullets

Here’s something that’s true about any baseball move a manager makes: It could work. It doesn’t matter how stupid the move looks to the masses or how mathematically illogical the move is or even how much the players involved disagree with it. You could put your .089-hitting pitcher in the leadoff spot of the lineup and he might get the game-winning hit. You could have your best hitter bunt with two outs and two strikes and a man in scoring position and the pitcher might throw the ball away. You could have your team’s worst pitcher try to close out a World Series game with a one-run lead and he might get the save.

In this way, baseball is not like other sports. In other sports, a bad move can have almost no chance at all. I remember an ACC basketball game from years and years ago where Maryland trailed North Carolina by one in the final seconds and, for reasons that remain somewhat unclear Maryland’s legendary ol’ coach Lefty Driesell had his son Chuck in the game. He was known as Chuckie Driesell then (not now, apparently — he’s now coaching the Citadel) and he almost never played. His main role seemed to be to get good grades and raise Maryland’s team GPA. But he was in the game for that final play and not only that, but the final play seemed DESIGNED for him. It may have been that the actual play broke down leaving Chuckie as the only option. But t also may have been that Lefty was going for the element of surprise — he might have figured that nobody would expect Chuckie Driesell to try the last shot.

In that he was right, Chuckie Driesell broke open and appeared to have something that looked like an uncontested layup. Of course it only looked like an uncontested layup — it wasn’t one. Chuckie shot and Michael Jordan spiked the basketball about 843 rows into the stands and North Carolina won. But what I remember most is that if Jordan had not blocked the shot, Sam Perkins was standing right behind him and was ready to do block it 984 rows into the stands. And behind Sam Perkins was another player in position to block it into another state. Basically, it was like that scene in Airplane where people line up to slap the hysterical woman. And the point is that play — if it really was a play — had exactly 0.000000001% chance of working (the nanometer chance, by the way, is that Jordan would have blocked the ball into the basket).

Major League Baseball moves, even the obviously terrible ones, never have quite that low a chance. There are always variables that could make a baseball decision work. In other words, Don Mattingly’s bizarre choice to pinch-run Dee Gordon for Adrian Gonzalez in the eighth inning of a tie playoff game against St. Louis could have worked somehow. It’s like Henry Fonda says in “12 Angry Men” — “I don’t know. It’s possible.”

‘But it’s not probable,” E.G. Marshall responds to Fonda, and, of course, that word “probable” is at the heart of what it is to be an in-game baseball manager. Your job is to try and put your players in the best position to win the game. You use a vast array of information — some of it available to everyone. some of it available to only a few, some of it only you know. You have basic statistics, you have advanced statistics, you have a vast history of baseball to lean on, you have knowledge of the players health, you have a sense of how their personal lives are going, you have those small almost ineffable feelings that build up from different conversations and brief exchanges and small things that perhaps only you noticed. Throwing all of it together, you come up with a little probability chart in your own mind and decide accordingly.

Something banging around in Don Mattingly’s mind told him that in the eighth inning of a tie game he needed Dee Gordon out there on first base instead of Adrian Gonzalez. Something told him that the next series of events might only result in a run if the faster Gordon was out there over the slower Gonzalez. Something told him that the difference of Gordon being out there instead of Gonzalez was worth more than the value lost with Gonzalez being replaced by Michael Young in the eighth inning of a tie game.

Algebraically it might look like this.

(Speed of Gordon – Speed of Gonzalez) > (Gonzalez’s value – Michael Young’s value in eighth inning of a tie game)

The probability of the above being true, in my opinion, is very low. It is not quite Chuckie Driesell low, but it’s very low. But here’s where it gets weird. One of the very, very few ways the Gordon move might make sense is if he was to steal second base. This was a risky proposition with Yadier Molina behind the plate for St. Louis, but that seemed to be what Mattingly was thinking: Steal Gordon into scoring position. But, it turns out, that is not at all what Mattingly was thinking. The Dodgers did not try to steal Gordon. The next batter, Yasiel Puig, faced three pitches. The first was a swinging strike. The second was a ball. The third was a groundout. Gordon did not run on any of them. And based on what Don Mattingly told the media afterward, it is apparent the stolen base was not even part of the strategy.

Mattingly said: “If we don’t use him there and the next guy hits the ball in the gap, and he doesn’t score and we don’t score there, we’re going to say, ‘Why didn’t you use Dee?’”

You know what, maybe this WAS Chuckie Driesell bad. That was the reason he put Gordon in the game? On the off-chance that someone would hit a ball into the gap in such a way that Gordon would score from first base but Adrian Gonzalez would not? That was worth pulling Gonzalez out of the game? We have to rework the formula:

[{(Speed of Gordon – Speed of Gonzalez} * (percentage chance that someone would hit a ball into the gap)} * (percent chance that Dodgers would not score the run in the inning)] > (Gonzalez value – Michael Young’s value in eighth inning of a tie game).

Of course, Gordon was forced out on Puig’s groundout, so the Mattingly dream lasted all of three pitches. And then, you could hear the Baseball Gods laughing. I believe I was one of only 128 million or so people to tweet at that moment, “Surely Adrian Gonzalez’s spot in lineup won’t come up in a crucial moment.”

In the 10th inning, the Dodgers’ Mark Ellis tripled with one out. That brought up Hanley Ramirez who was, of course, intentionally walked because, yep, guess who was hitting behind him. It was Adrian Gonzalez’s spot in the lineup. Adrian Gonzalez led the National League in sacrifice flies this year. But Adrian Gonzalez was on the bench and Michael Young was in the game. Young’s shallow fly ball to right field did not seem to have the distance. And it did not have the distance. Superhero Carlos Beltran threw home, the ball beat poor Mark Ellis to the plate by about three steps, and he was called out. There was some debate about whether or not catcher Yadier Molina actually tagged Ellis, but even Ellis himself wasn’t singing that tune. “I was out,” he told reporters. When asked if he was sure, he said, “Yeah, it was pretty obvious.”

So that was one consequence of the Mattingly move. The second was, if possible, more painful. In the 12th inning, Carl Crawford led off with a single. This led to the following disastrous series of events:

Event 1: Mark Ellis sacrifice bunted Crawford to second thus reducing both the Dodgers’ run expectancy AND their win probability.

Event 2: The Cardinals intentionally walked Hanley Ramirez, perhaps while laughing.

Event 3: Michael Young, hitting still for Adrian Gonzalez, grounded into an inning-ending double play.

Eventually superhero Carlos Beltran ended things with a walk-off single — “If you have around long enough, and you’re facing Carlos Beltran, sooner or later, he’s going to hurt you,” Vin Scully said — and poor Donnie Baseball was left to mutter hunting cliches about how you’ve got to shoot your bullet when you have a chance. It’s a decent analogy. Mattingly put on a blindfold and shot a bullet. It could have hit the target. Hey, it’s possible.

Print Friendly

26 Responses to Shooting bullets

  1. BobDD says:

    Where was Dusty Rhodes when he was needed?

  2. Joe says:

    Here’s Donnie Baseball’s quote regarding the play at the plate: “It looks like he’s out,” Mattingly said. “Yadier’s out in front of the plate. The ball beats him. Holds on to the ball. Obviously, I guess, if we get replay or whatever something like that, but that’s one of those that’s hard to even argue that he didn’t tag him, you know?”

    Feels like the Cardinals swiped a game they should’ve lost.

    Dark Side of the Mood

  3. cd1515 says:

    I find him underwhelming on TV but Cal Ripken pointed this out right away as it was happening… something along the lines of “what if this game goes 12 or 13 innings?”

  4. Mark Daniel says:

    This decision basicall required Puig to hit a double, since a 3B or HR would score Gonzalez. Gonzalez this year scored from 1st on a double 7 times in 20 opportunities. Gordon in his career has scored from 1st on a double 5 times in 9 opportunities. So Gonzalez has a 35% chance of scoring from 1st on a 2B, while Gordon has a 55% chance.
    Also, Puig hit 21 doubles this year. He also hit 19 HR and 2 triples, thus the odds of hitting a 2B were the same as hitting a triple or HR. There was a benefit to have Gordon on 1st, but appears to be very small, especially if you consider the fact that Puig this year hit a double once every 20 PAs, or 5% of the time. If you look at the score from 1st numbers, the chance of Gordon scoring from first, based on this 5% incidence, is .05*.55 = 2.75%, while the chance of Gonzalez scoring is .05*.35 = 1.75%. So overall, Gordon being on 1st instead of Gonzalez improved their chance of scoring when Puig was at bat by 1%.

    • Not only that, but there were no out. Let’s say Puig hits a double and Gonzalez stops at 3B. That’s still second and third and no out, with Ethier, AJ Ellis, and whoever is batting 9th coming up! So the odds are still pretty small that having Gordon on 1B instead of Gonzalez would make a difference.

      The less complex (and more stupid) decision was having Mark Ellis bunt in the 11th with Crawford on 1B and nobody out. Everybody watching (except for Don Mattingly, apparently) knew that Hanley would get walked in that situation. So what Donny did was take the bat out of Ellis’ hands (hard hit single and hard hit triple already), give a nervous pitcher a free out, and then take the bat out of the hands of the team’s best hitter. Compounding the bad decision is the fact that Michael Young can’t hit anymore.

    • Dinky says:

      I cannot agree more.

    • cd1515 says:

      “So Gonzalez has a 35% chance of scoring from 1st on a 2B, while Gordon has a 55% chance.”

      that’s a bit simplistic.
      are all doubles exactly the same?
      no context given to score, game situation, ballpark, who’s pitching, who’s in the field, etc?

      • Mark Daniel says:

        cd1515, I was just using numbers from Baseball-Reference. Of course there are different kinds of doubles, and the situations change based on ballpark, etc. But I was doing a quick and dirty calculation only for Puig’s at bat. If Puig singled, then you’d have a whole different scenario, based on whether Gonzalez / Gordon got to 3rd. As it so happens, in his career, Gordon has gone 1st to 3rd on a single about 29% of the time (7 times in 24 chances), while Gonzalez this year has only done it 19% of the time (5 times in 26 chances). So there’s another advantage for Gordon. Again, this is just an estimate.
        Going a bit further, here’s what Gordon (career) and Gonzalez (this year) have done when on 2nd base and a single is hit:
        Gordon: scored 64.7% of the time (11 out of 17)
        Gonzalez: scored 11.7% of the time (2 out of 17).

        So, there again is an advantage for Gordon based on speed. The question is whether this advantage warrants removing Gonzalez from the game.

    • Rob Smith says:

      Oh common CD. Do you want someone to foresee what type do double that Puig could have hit in a runner on first, late game situation? Even if that was somehow possible, and it doubled the advantage of having Gordon running instead of Gonzalez, we’d still be at a 2% advantage and the point would remain the same. It was a terrible managerial decision.

  5. Tom Kelly had Al Newman pinch run for Chili Davis in Game 7 of the 1991 World Series. That was why Gene Larkin, who was limping on a bad knee, had to pinch hit for Al Newman in the 10th inning with the bases loaded and one out. Larkin of course got the GW hit. I don’t recall anyone ever questioning Kelly about why he ran for Davis. Not that it was a good decision, but I see it done all the time and obviously this was a more critical game and the decision was done by a manager who had a reputation as a terrific game manager.

    • _ says:

      An interesting analogy and nothing against Tom Kelly but it was a lot easier to get by on a reputation as a “terrific game manager” in those days. It would be really interesting to see how any of the revered managers of generations past would have held up against the way the game is analyzed today.

    • Dinky says:

      IIRC, Chili Davis was the DH. Gonzalez is a near gold glove level fielder. Also, the Cardinals would have been throwing all RHP the rest of the game. If you MUST run for Gonzalez, I think the really smart move would have been to move Ethier to first base, put Shoemaker in center field, and save Young to pinch hit at a time when a double play wins the game.

  6. Which Hunt says:

    Donnie, Donnie, Donnie. *shakes head, cries softly to self, opens box of icecream*

  7. I loved Don Mattingly the Player. I hate Don Mattingly the Manager. For several reasons. The first was that he had a habit of overworking his best relievers. Throwing your aces out there with a seven run lead burns their arms for when you really need them. Paco Rodriguez was one of the best lefties in baseball for most of the year but was gassed by September and the Dodgers had to leave him off the roster for this series. A healthy Paco would have helped last night.

    The other is his love of the bunt. How many times have I seen him bunt a runner over when the team’s best hitter was on deck, and thus would be walked intentionally with first base open? Or bunt with a perfectly fine hitter at the plate when the guys behind him can’t hit a lick? Giving your best hitters a chance to hit with a runner on base is way more important than giving up an out to advance the runner. You would think he might have learned something in the deciding game against the Braves when Juan Uribe homered after failing to lay down a bunt, but no.

    At least if his moves keep backfiring, the Dodgers will have an excuse not to renew his contract.

    • Dinky says:

      The Dodgers have excuses enough. If I owned the Dodgers, I would tell Mattingly, “If you bunt a runner to second with any non-pitcher, you will be fired.” I think keeping a happy dugout is perhaps the most valuable skill any manager has, and Mattingly has it, but he definitely lost last night’s game, three times: the pinch run, the sacrifice, and not having Young try to squeeze.

  8. Dinky says:

    This comment has been removed by the author.

    • Dinky says:

      Keep in mind, earlier this season, against a left handed starting pitcher, Mattingly started Scott Van Slyke to get the platoon advantage over Carl Crawford (reasonable). In the 5th inning, 4-3 Dodgers game with both starters looking bad, bases loaded, one out, the opposing manager brought in a righty to face Van Slyke. Mattingly let him hit instead of pinch hitting Crawford in a huge leverage situation. Van Slyke grounded into a double play. Crawford hit for Van Slyke leading off the 7th, when the opposing manager COULD bring in a LOOGY. So when I say that last night was the worst managed game I can recall by Mattingly, maybe even by any Dodger manager since Alston, that means something.

      Yes, there were lots of other reasons the Dodgers lost. A healthy Ethier (or Kemp) catches Beltran’s double, giving Greinke a 2-0 gem. Gonzalez (and Puig) might not have struck out in the first inning,second and third, one out. Puig might not have had the worst batting performance of his career (0 for 6, no balls to the outfield). Or, and here’s a crazy idea, since Mattingly loves the bunt so much, why not have Young try to squeeze in Ellis from third with one out? Donnie, your inner voice was quiet at the wrong time.

      I recall a recent column by Joe that talked about knowing when not to manage the game. Mattingly’s inexperience, in not just letting his team win the game (and in a real sense they were winning in the 8th inning, already running through the Cardinals’ bullpen, Greinke still pitching very well), cost them a critical winnable game. As I went to bed last night, I was thinking, “I hope Joe writes something so I can vent about it.”

    • Rob Smith says:

      Something we haven’t discussed…. All Managers have bench coaches to help with these decisions. If the bench coach is making these suggestions, then that’s an easily remedied situation. Do not renew his contract. But if Mattingly is overriding the wisdom of the bench coach, then Mattingly is not only bad, but also thinks he know more than his hired sage and doesn’t listen to good advice…. Which may actually be worse.

  9. I didn’t like the bunt decision either, but since everybody and their brother knew the Cards would walk Hanley with first base open, that should be factored in. 1st & 2nd with one out actually has a higher run expectancy than runner on 1st with nobody out (though a lower expectancy of scoring at least one run). I thought the decision to pinch run for Gonzalez was a much worse decision for all of the reasons cited above.

  10. hasnain raza says:

    Lols Gag is the the Best Lol Network Ever, where you can every thing is lol and Funny, Troll Images, Prank Peoples, Funny Peoples, funny planet, funny facts, funny cartoons, funny movies pics, iphone funny, funny jokes, Prank Images, Fail Pictures, Epic Pictures, Lols and Gags, Lol Pictures, Funny Pictures, Lol is the Laugh out of Laugh where you can Fun Unlimited and Laughing Unlimited.

  11. Unknown says:

    I think Joe is being overly hard on Mattingly here and not fair in his presentation of the data. Joe comes up with a very technical equation for the left hand side of the equation and compare it to the very vague ” (Gonzalez value – Michael Young’s value in eighth inning of a tie game)” on the right hand side. Why not give the right hand side of the equation equal attention? Joe’s analysis doesn’t mention that it was the 8th inning and Gonzalez had just hit (9 hitters to go before he comes up again).

    The right hand side of the equation should be refined to be [Chance that Gonzalez will get up again *(Gonzalez value as a hitter-Michael Young’s Value as a hitter]. You can use any number of offensive statistics to further refine this. For simplicity, let’s use batting average (any kind of hit would have won the game in the 10th). Gonzalez batting average is .293 compared to .279 for Michael Young. That is 14 points or 1.4% difference.

    So now the righthand side of the equation is (.014*Chance that Gonzalez spot will come up again). That is a very low number on the right side.

    So while putting the faster Gordon in to run in the 8th may have only give the team a very small edge–statistically it also only had a very small downside. I’m not saying it was the right decision, but I just don’t think it was so clearly the absolute, in-arguably, wrong decision as Joe presents it to be. I think the article unfairly makes Mattingly look like he made a bone-head call.

    Hindsite is 20/20 and this article is taking the known facts that hindsite gives us and making it seem like it was the obvious outcome.

  12. Fosberry says:

    Unknown – Tom Tango would seem to agree with your point:

    His take is that it was likely a good move, in large part because Gonzalez wasn’t likely to bat again, but also because the gap between Gonzalez and Young isn’t *THAT* big (Tango estimated it as .04 runs/PA), whereas he thought having Gordon instead of Gonzalez in that spot was worth 0.10 runs.

    Don’t forget that having Gordon forced at second right away, while certainly possible, was not too likely. So the value of Gordon isn’t just scoring from first on a possible Puig double, or to try to steal against Molina with Puig at bat. Suppose instead Puig and Uribe struck out, leaving Gordon on first with 2 out. Now there’s really a strong case to run, even against Molina, and that possible steal would indeed greatly increase your chances of scoring a run compared to Gonzalez. Or maybe Gordon’s speed results in an infield hit where Gonzalez would have been easily forced at second. Gordon is much more likely to go from first to third on a single.

    There are plenty of possible ways Gordon’s speed could have let the Dodgers score a run in that inning where they would not have with Gonzalez on base. That instead Gordon was immediately forced out makes it easier for Poz to ignore them, but we don’t know about the force out when making the move.

    • Rob22 says:

      Where I think this argument lacks in the “threat” of Gonzalez at the plate and his defense. But, sticking to the “threat”, even if Mattingly chooses to bunt later on, with Gonzalez still in the lineup it’s unlikely that they walk Hanley Ramirez to face Gonzalez with runners on first and second. That keeps the bat in HanRams hands & you get two shots to knock the run in. The entire strategy changes with Michael Young, because instead of a pick your poison decision, the bat is automatically taken out of HanRams hands & you’re counting on Michael Young. Add into the equation that HanRam’s been an extra basehit machine during the playoffs and the decision looks even worse.

  13. dave says:

    of course, this is an injured Hanley Ramirez because the cardinals were smart enough to drill the dodgers best hitter in the first inning of the series. surprised they didn’t drill kershaw or greinke as well. cheaters.

  14. Mike says:

    Loved Lefty! Chuck was two years ahead of me at Springbrook High School in MD. Thanks for the memories, Joe!
    I might be remembering incorrectly but wasn’t there also a ball hit to first that Young couldn’t make a play on that Gonzalez might have gotten to?

Leave a Reply

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