As those pundits who reflexively criticize we sabermetric types like to remind us, baseball is not a tabletop game. As Bill James pointed out on more than one occasion, you can actually learn a lot about baseball from running through a few hundred simulated games, but leave that aside for now. The key here is the one thing you cannot get out of two-dimensional player-cards: an insight into their current ability to perform, an instinct, a feeling. The card is always ready to go, but the actual player the card represents is not nearly so dependable. He might have a cold, or a mild groin strain, or a bitter divorce, or a hundred other things that aren’t visible from the press box or your living room sofa.
No baseball simulation that I know of has an option for randomized “not that into it today” diminishment of a player’s abilities, or “severely distracted,” or “slightly out of whack.” In games, they are what they are. In real life, they vary from day to day. We must concede this, as we always have conceded it. Further, we must concede that one of the people in a better position to know these things about St. Louis Cardinals players in particular is Tony LaRussa. His information is more complete than ours. That doesn’t mean he or any manager will always be correct or wise or even lucky, it just means that he’s operating on perhaps one more level than we are.
The foregoing is a preamble to a defense of LaRussa’s decision to yank Jason Motte in the top of the ninth inning of World Series Game Two, a move that was attacked as over-managing the moment that it happened. The difficult thing about being a manager is that you’re damned if you do, damned if you don’t. Back on September 23, I took LaRussa to task for being too passive with Jason Motte in a key game against the Mets. The Cardinals took a 6-2 lead into the top of the ninth at St. Louis, and LaRussa brought Motte in to finish things out:
Motte didn’t allow a hit to the Mets, but he walked leadoff man Willie Harris, saw Nick Evans reach first base on a Rafael Furcal error that aborted a potential double play, and walked Jason Pridie. The bases loaded, Motte capped a memorable afternoon by walking pinch-hitter Justin Turner to force in a run. At that point, quick-draw LaRussa, who had not been as quick as one might have expected in this series, finally brought the hook, but neither Fernando Salas nor Marc Rzepcynski could stem the tide that had now been unleashed. The flood was exacerbated by some shaky defense from the Cardinals, not only Furcal’s error but also bad positioning on the part of left fielder Shane Robinson, who had come into the game for Allen Craig and was caught playing shallow on a drive to left field by Ruben Tejada that went for a double and tied the game. … When a pitcher like Motte, with good control, starts walking the ballpark, the manager’s radar should be set off by the uncharacteristic behavior. Leo Durocher used to say, “I wasn’t nailed to the bench,” meaning that when the game was in jeopardy he was free to act. LaRussa was nailed to the bench.
Motte wasn’t nearly as bad on Thursday night. Ian Kinsler popped a ball just past Rafael Furcal into no-man’s land to lead off the inning, then Furcal—barely—stole second base. Elvis Andrus followed by turning a fat off-speed pitch into a sharp liner to right-center that moved Kinsler to third base, Andrus taking second when the throw-in either wasn’t properly cut off or was off line, depending on how you saw the play. This brought left-hander Josh Hamilton to the plate and LaRussa out of the dugout to go for 1,000-year-old situational lefty Arthur Rhodes. That’s where the first-guessing begins.
The move didn’t work out, of course. Hamilton hit a run-scoring sacrifice fly that also moved Andrus to third. LaRussa replaced Rhodes with Lance Lynn, but Michael Young hit the second sacrifice of the inning to give the Rangers a 2-1 lead. Neftali Feliz made it stand up in the bottom of the frame.
Just because a move doesn’t work out doesn’t mean it was the wrong move. Sometimes a manager pushes the right button and gets the wrong result. In this sense, the argument for a move, the soundness of the thinking behind it, is in many ways more important than the outcome of the move itself. A manager can’t control outcomes, but he can utilize his players as best he can so that the chances of a good outcome increase. As the Hall of Fame manager Miller Huggins once said, “After all, managing is not so difficult. You just figure out the things of which your players are capable and then try to get them to do those things.”
Standing on the mound in the ninth inning last night, LaRussa thought that Motte was not capable. Keep in mind, the Cardinals were in deep trouble no matter what decision LaRussa did—keeping a one-run lead when you’re facing runners on second and third and no outs is a difficult stunt no matter who is pitching. After the game, the manager gave several reasons for making the change, including his feeling that Hamilton, strong against fastballs, would have a harder time against Rhodes's geezer-southpaw stuff. No doubt he felt this way despite the groin injury that constrains Hamilton’s swing. He’s aware of it just as much, if not more so, than you are.
He’s also more aware of Motte. In that game back on September 22, he had kept his faith in Motte and paid for it with a loss that could have dropped the Cards right out of the pennant race. LaRussa isn’t going to tell the media this, but I strongly suspect that, in this case, once bitten equals twice shy. Even if that wasn’t his reason, he has been watching Motte for four seasons now, and I’m willing to give him the benefit of the doubt as to the pitcher’s capabilities, particularly when Motte hadn’t been sharp.
Think about it: Motte came to the majors in September of 2008. He has all the stuff in the world. The incumbent closer in that time was Ryan Franklin, who wasn’t exactly Mariano Rivera even before he pitched his way out of the majors back in June. Any manager with a pulse would think about letting Motte close given those conditions, and yet Motte has exactly 12 regular-season saves. Either we have to accept that LaRussa is a complete idiot and is blind to Motte’s capabilities, or we have to at least allow for the possibility that LaRussa has reasons for believing that Motte should not be his day-in day-out closer.
If after the game you caught the postgame show on the MLB network, you saw Mitch Williams arguing that LaRussa had now irreparably shattered Motte’s confidence for the rest of the postseason. Williams has certainly been in the position to know, but even if he is correct as to the condition of Motte’s psyche, what would he have had LaRussa do if he believed that keeping Motte in the game would lead to a loss? Chalk up a World Series game to experience because it might hurt someone’s feelings? If Motte stayed in to fall apart as he did against the Mets, or even simply allowed the tying and the winning runs to score via his own inability to execute, would that not damage his confidence as well?
Perhaps in a regular-season game there is room for a manager to gamble on a pitcher’s confidence, but this is the World Series, Motte is 29 years old, and if he doesn’t know where he stands by now, he’s not going to know. LaRussa had his conviction about Motte’s chances against Hamilton at that moment, and neither you, nor I, nor Mitch Williams will ever know what would have happened had Motte stayed in the game. We can only know, and accept, that he had his reasons and they cannot be comprehensively argued with.
Prior to the game, LaRussa had predicted this kind of second-guessing: “It comes down to you make a move, and if it works, 'Hey, what a good move.' If it doesn't work, 'What was he thinking?' That's just the name of the game.” As long as there have been managers, there have been first-guessers and second-guessers, and since most managers are no smarter than the rest of us, quite often the doubters have had a good case. This isn’t one of those times; it isn’t even close to being one of those times. Save your ammo for when someone does something worth arguing about.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
Also, while I don't think that this would have been a great idea, what about intentionally walking Hamilton to let Motte face Young?
Motte is a better pitcher and Hamilton has a slight platoon split, but still slugs over .500 against lefties.
The situation was pretty much a no-win anyway, but your goals in that situation are.
1) Strikeout or pop-up
2) Preventing at least the guy at 2nd from scoring. Getting out of the situation tied would be considered a good thing given the run expectancy.
I'd rather have my fastball pumping closer up there than an ancient lefty with a strikeout rate below 6K/9 any day regardless of confidence.
This is expanded by keeping in mind even if you get Hamilton you then have to get out at least 2 right handed hitters after him. If there are two outs, this would be defensible. With none out and the heart (all righties) of the order left to contend with, La Russa put them in the worst position possible and necessitated another pitching change after Hamilton PA no matter what.
As a Cubs fan though, I approve of this move.
One interesting point is that Motte's K/9 drops and BB/9 increases (as does FIP and xFIP) in high leverage situations...maybe something TLR knows...at the same time his batting average allowed drops to .119! Strange... nibbling?
His K/9 vs. lhp on the season is slightly better than Rhodes', but I couldn't find that breakdown by degree of leverage.
He's much more likely to induce an infield pop up against a lefty (10.7%) than Rhodes is (4.2%), though in high leverage situations that drops to 6.3% (Rhodes was at 0.0%).
Furthermore, in high leverage situations, Rhodes allowed 70% fly balls in 2011, including a whopping 21.4% that left the park. Motte was under 40% with none of them leaving the yard.
For his part, Hamilton thrives in high leverage situations, with more walks, fewer strikeouts, a higher OBP, and higher slugging percent, and a 50% fly ball rate, with his standard 19% or so turning into home runs.
He did, however, hit his most infield pop ups in high leverage moments (a whopping 18.8%, as part of 4.6% on the season; a small numbers issue with only 7 in total).
The final piece is that Motte had thrown 12 pitches, 8 of them for strikes, 10 with 7 strikes in the first game.
This was nowhere near taxing for him, based on his usage during the regular season when he averaged 14 pitches per outing and appeared two days in a row 21 times (three days in a row twice).
On the season, his ERA dropped in 19 of those consecutive appearances, including nine of ten when he threw 13-20 pitches, and went up in 4 with pitch counts of 15, 26, 27, and 29.
If this *were* a Strato game (as someone else alluded to), these would be the match-ups (using John LaManna's 2011 projections, no ballpark dimensions taken into account):
Hamilton vs. Motte: 27 hits, 32.5 on base, 50 total bases (counting walks), 2.5 ballpark HR chances, and 1.6 straight HR (out of 108 total chances).
Hamilton vs. Rhodes: 15 hits, 24.5 on base, 52.5 total bases (worse), 8 ballpark HR chances (much worse), and 6.7 straight home runs (also much worse).
Young vs. Motte (assuming he stayed in and got a "cold out" against Hamilton or semi- or intentionally walked him): 19 hits, 22.5 on base, 30 total bases, 0.0 ballpark HRs, 0.95 straight home runs.
Beltre vs. Motte (assuming he got Hamilton and walked Young): 13 hits, 14.5 on base, 23.5 total bases, 4 ballpark HR chances, and 1.85 straight home runs.
Young vs. Lynn (almost as good as Motte): 21.5 hits, 25.5 on base, 35 total bases, 0.5 bp HR, 1.0 straight HR.
Wasn't really trying to say anything about predictive power, but this is part of the debate. Are there meaningful metrics that should inform decision making, or should it be manager hunches?
LaRussa made the move based on whatever he did, we can debate whether it was the right move or not, but don't have full access, and the Rangers ended up winning that game, whether TLR made the right moves or not.
Historically, Beltre grounds into more than his share of DPs. A quick glance at BR says he's hit into 57 DPs over the past 3 seasons, and he missed a decent amount of time in two of those three seasons. He's also not particularly patient, so less likely to hurt you with a walk with the sacks full.
All of this said, I have no issue with TLR's decisions. I agree that he knows his players better than any of us, and my analysis is more from a replay standpoint where it is simply about playing percentages.
I don't always trust LaRussa's opinion on players. He too often favors experience over ability beyond any evidence in performance, and Rhodes (and previously Ryan Franklin and Miguel Batista) will be trotted out to honor their years of experience despite their current lack of performance.
I can understand people wanting to give Pujols a pass on a lot of things, but I'm guessing if the 1B was Carlos Pena then he'd be under a bus right about now.
Personally I would have left Motte in to try and strike Hamilton out. Unless Motte was injured I think that would have been the Cardinals' best bet. Then if he struck Hamilton out I would have walked Beltre and hoped for the double play. A lot of ifs to be sure and no guarantee of success, but more likely than Rhodes and Lynn being able to retire the side without the ball being put in play.
LaRussa has had a long and successful career and will deservedly go into the Hall of Fame once he retires. He is not the first nor will he be the last HOF manager to have a decision back fire on him in a crucial game situation.
The situation would have been Andrus at first, Kinsler at third, Motte on the hill against Hamilton.
Steal situation, certainly, but with Yadi behind the plate? And with Kinsler at third, do the Cards' even throw for him? Do the Rangers try to "antler" a steal of home (or is it the other one)?
How's Motte at holding guys? John LaManna projects him as a +3 hold, Rhodes a -1 (on his inter-league card, -5 with Texas).
So TLR may have made the move to Rhodes v. Hamilton even if Andrus were still at first, to try to foil a steal.
Would have been fun to see.
I was texting with a Cardinals-supporting friend last night, and after the game, I said that the headline will be "Motte blows lead" rather than "Cardinals offense fails to support fine overall pitching effort." Some defensive plays didn't get made in the ninth, to be sure, but let's not lose sight of the lack of offense by the Cardinals either.
It's kind of a nice touch, actually.
Actually, wouldn't the best move, if you want to avoid burning all of your bench players, be to put in a starting pitcher who runs decently? Not all starting pitchers have great instincts for running the bases, of course.
Whatever, I'm done thinking about it, serenity now.
I can't agree with removing Motte for Rhodes. Motte is more likely to record a K against Hamilton, putting me at 1 out with runners on 2nd/3rd and completely removing the possibility that two deep sac flies beat me. It's not that Motte is overwhelmingly likely to strike someone out, but it's enough that I'll take it over Rhodes and consider it a reasonable risk/reward rate.
TLR pulled his best bullpen pitcher for one of his worst because the handedness matched up and he got beat.
***
Am I missing something? If Lohse runs for Molina isn't Molina's bat and glove "burnt"? There aren't courtesy runners in MLB last I checked.
Plus, in all reality, if Pujols makes the cut off, we're playing extra innings!
However, I always looked at bringing in a new pitcher as a bit of a crapshoot- you never know if that NEXT guy you bring in is not going to have it that day (like hitting on a 13 in blackjack, you may get an 8, but you may get a 10). As a result, I would go with fewer relief pitchers for longer outings each instead of flipping through my entire bullpen for every little platoon advantage. Thoughts?
It's important to examine what he does as a manager and how much of his team's success has nothing to do with him and everything to do with the extraordinary amount of hitting talent he has. There's been a lot of praise heaped on his bullpen management recently, and his "making all the right moves" down the stretch. But that neglects that he started the season with Franklin, Batista, Augenstein, Tallet, Miller in his bullpen working high-leverage situations, while Salas, Sanchez, Lynn were in AAA. His bullpen/roster management early in the year is what led to the team needing a late-September gift from Atlanta to be where they are now. He really hasn't changed. He's just rode a hot streak. I'm glad BP is a place where an alternate look, beyond the oversimplified "The Rasmus trade got the team here" view that broadcasters and the mainstream media are delivering.
The narrative about TLR over-relying on decrepit old players may be emotionally satisfying. That doesn't make it correct -- in this particular case. If you want to cast aspersions on his roster management, look to the abomination he was running out there as a middle infield for the first half of the year, not the bullpen, where he didn't really do nearly as badly as he was accused of, and would have done quite well indeed if Sanchez and Lynn hadn't got hurt.
Look - bashing the Cardinals and LaRussa is sort of pointless in my opinion. I think the national media should turn their binoculars the other direction - to the daring (and excellent) baserunning that kept the Rangers from going back to Arlington down 2 games to none. Kinsler had the most narrow of windows to steal that base against one of the better catchers at stopping the running game, and just barely did so. Elvis Andrus is one of the best baserunners and the league,a nd it showed last night. Elvis forced the most minor of mistakes by taking second base, and it won the game for Texas. give credit where it is due, and stop bashing LaRussa.
Sometimes, that's just the way baseball go.
He's a great manager, but he's no Einstein.
BTW, I'm fairly confident it should read "criticize US sabermetric types" not "criticize WE sabermetric types."
Yes, it was a blunder. And I'm a TLR fan.
Oh, and of course he should've pinch-run Theriot. You need 1 run to tie or otherwise lose, you put your best runner in there. I'm a TLR fan, and he made two blunders last night. In a game he projects to lose anyway, even making the right moves.
It's been awhile since I last re-read it, but didn't J Henry Waugh's game had something along these lines?
As for the 9th last night, assigning blame to any individual for that inning is an angels on pinheads exercise.
In fact, early in the book there’s this passage: “No, somehow, he [J. Henry Waugh] had to get him out of there! He sought for some excuse. Something Bancroft saw in the way the kid was exercising the bat as he moved toward home plate? A kind of slump or twitch in his pitching shoulder? Why not? Look close, Barney!â€
There’s a reference late in the book to Waugh finding “one pretense or another – personal problems, minor illnesses, obscenity on the field – shaken up the Knick lineup…†in hopes that they would lose their games, but that was the proprietor’s imagination, not a roll of the dice.
And when he finally introduces his friend Lou to the game, he’s benched the Knicks’ only 4-star hitter (Bran Maverly), because he had “been in a slump†(though it’s clear that was another subjective move to encourage the Knicks to lose).
Statis-Pro Baseball has the following for pitchers in the advanced rules section:
"Good and Bad 'Stuff' for Pitchers...this rule is meant to simulate days when certain pitchers are unhittable (as Len Barker was when he pitched his perfect game in May, for example) and others when they can't get a man out."
Based on a random turn of a Fast Action Card before a pitcher starts (or enters as a reliever), he has Great, Good, Normal, Bad, or Terrible stuff, which adds to or subtracts from the play number result derived from a combination of his PB ("control factor") and the batter's ratings.
Relievers can only achieve Goodness, not Greatness.
Pitchers don't vary in their propensity for one or another daily effect (could use Ron Shandler's DOM/DIS to introduce that concept), but good pitchers start with a higher PB that then gets adjusted, so maybe that's good enough.
Also, both players would know a priori what the situation was, which doesn't necessarily replicate that one manager would likely have more insider knowledge.
And for batters, every day is Groundhog Day.
Statis-Pro, Extra Innings, Gil Hodges' Pennant Fever, Strat, Calcu-Ball, and Cadaco All-Star Baseball (as well as MicroLeague and Tony LaRussa Baseball for the PC) are the ones I have, and only Statis-Pro (c. 1982) has this feature.
Don't know about APBA, Big League Manager, Pursue the Pennant/Dynasty, Sports Illustrated Superstar Baseball, Replay, Out of the Park, Ball Park Baseball, Sherco, or Negamco, but nothing I've found online about these, which I haven't played, suggests any have this innovation.
It's interesting, too, that in Waugh's game ratings for the next season were based on how a player did *in the simulation* during one season, not like in "our" games, where it's actual MLB performance that determines the "cards."
This is not Monday morning QBing. In the game in question, Rhodes' performance was actually pretty neutral -- i.e., the odds of winning were roughly the same when entered as when he left. It's more a question of expected results going forward.