In an effort to become the perfect SaberManager™, I'm taking over the 2005 Cubs and leading them through a simulated season in Out of the Park. I'll do all of the things that I've told managers they should be doing, while paying attention both to how feasible strategies are in isolation and within the context of a baseball team playing a baseball season. For more details and a full explanation of why and how I've chosen to do this—and with 2005 Cubs of all teams—click here.
Game 25 (May 1) – at Astros; Win 4-3 (12); Record: 14-11
And suddenly things are looking up. The Cubs went to Minute Maid Field in Houston and swept a three-game series, taking over second place in the NL Central. And I’ve kinda got the hang of this now. The tandem starter thing is a little more stable, although I realize that what started out as a plan to have Kerry Wood, Mark Prior, Rich Hill, Ricky Nolasco, Sergio Mitre, and Glendon Rusch all be one amorphous blob that sometimes were tandem starters, sometimes longer relievers, has turned into Wood and Hill forming a starting tandem, Prior and Nolasco forming a starting tandem, and Rusch and Mitre turning into ye olde fashioned relievers.
It means that I’m just running a six-man bullpen, with two of the relievers able to go multiple innings. And until today, that was stable. Then, Wood got hurt (surprise!) while striking out the first hitter of the game. Starter injuries happen sometimes in real life as well, and since Hill was going to pitch anyway, bringing him into the game wasn’t a big deal. Hill made it through 3 2/3 innings and was relieved by Mitre, who also pitched 3 2/3. When Mike Remlinger got the last out of the eighth inning, I turned the ball over to Michael Wuertz, filling in as closer for Ryan Dempster, who had thrown a 1 2/3 the day before.
Wuertz had a two-run lead to work with, and maybe there’s a sub-routine in OOTP for a non-closer trying to nail down a save and having difficulty with it. There’s a real life sub-routine that does this as well. The Astros tied the game in the ninth. With a thin bullpen, I had to send Wuertz out for the 10th, followed by an inning of a tired LaTroy Hawkins in the 11th. When the Cubs took the lead on an RBI double by Neifi Perez (surprise!), it was up to Rusch to be the hero and record the save. Rusch came through.
After the game, I got some bad, though not horrible news. Wood had a back injury that would keep him out for about a week. My general manager decided not to invoke the disabled list for Wood, leaving me a bit short-staffed. I hadn’t really gamed out this eventuality. In my head, I moved Hill into the “starter” role and partnered him with Mitre, but that left me with a five-man bullpen, which became an issue when we got to …
Game 27 (May 4) – at Brewers; Win 6-3 (14); Record: 15-12
This was another tandem day, and while Prior pitched four innings in his allotted 50 pitches, Nolasco pitched the fifth and sixth in 23 pitches. Normally, I would have gone with Nolasco for another inning or two to save the bullpen a little work, but his lineup spot came up in the top of the seventh, with the Cubs down one, one out, and runners on second and third. A sacrifice fly ties the game. A base hit unties it.
Could I count on Nolasco to make contact? I decided that no, I could not. Jason Dubois pinch-hit and delivered the sacrifice fly. I got what I needed, but the game stayed stubbornly tied until the 14th inning. Hawkins pitched two innings, Dempster pitched 2 2/3, and Rusch went 3 1/3, including the final three outs (again) that normally would have earned a save if he had not already been the winning pitcher. Bullpen blowout games are no fun.
The next day (again against the Brewers), new backup catcher Gary Bennett pitched. He threw a scoreless ninth and struck out Russell Branyan. I’ve always had a soft spot in my heart for major leaguers named Russell, but man, Russell Branyan can strike out against anyone.
Game 34 (May 11) – vs. Mets; Win 9-7 (11); Record: 18-16
I never thought that something like this would ever happen to me, but here’s what happened. I swear it’s true. We’d had a rough series over the weekend, losing two of three to the Phillies, who just came in and kept hammering homers. On Monday, our game against the Mets got rained out, necessitating a doubleheader on Tuesday. Game 1 went well with the Hill/Mitre combination notching me seven innings (although, honestly, it was because I pushed both of them a little. Hill threw 61 pitches, despite my usual limit of 50. I needed the length.) Wuertz and Hawkins finished off the eighth and ninth of an 8-2 win, but both had to throw more than 25 pitches to do it.
In the nightcap, when Carlos Zambrano (facing off against Mets starter Victor Zambrano) only gave me five, I fortunately had Rusch to turn to. He gave me three scoreless innings, and as we headed to the ninth up 4-3, and with Dempster ready in the bullpen, I had a good feeling that we were about to sweep. But Dempster let the Mets tie the game and the Cubs couldn’t score in the bottom of the ninth. I had to send Dempster out again for the 10th, because I was running low on bodies. The Mets pushed across another run and won 5-4.
The next day, before the game, I took a look at my bullpen and found only Mike Remlinger was even available. This was going to be interesting. But hey, Prior and Nolasco combined to give me 7 2/3 innings (although again because I pushed them to 62 and 59 pitches, respectively). In the top of the eighth, with a runner on first, two outs, a 5-3 lead, and lefty Cliff Floyd due up for the Mets, Remlinger was the perfect antidote. Floyd grounded out to second, and the Cubs were into the ninth with a lead. The problem was that Remlinger was going to have to play both lefty specialist and closer in the same game.
In the ninth, it looked like I was going to escape death. Mike Cameron grounded out, and after Mike Piazza was hit by a pitch, Victor Diaz grounded out for the second out, pushing Piazza to second. Still a little too close for comfort, but that’s life in the big city. But then Miguel Cairo walked and Brian Daubach doubled in Piazza to make the game 5-4, and Jose Reyes poked a two-run seeing eye single through the left side. After a Carlos Beltran double, what had looked like a win was now a 7-5 deficit.
In the bottom of the ninth against Mets closer Braden Looper, whom the game informed me led the National League in saves to that point with 12(?!), Calvin Murray led off with a pop up to short that Reyes dropped. Two pitches later, Neifi Perez … Neifi. Perez. (I swear that is not a misprint.) hit a two-run homer to tie the game at 7-7. The electronic Wrigley faithful went crazy. So did I. What’s funny is that it wasn’t the most improbable home run of the game.
But before we get there, I had a bit of a logistical problem. I walked over to Kerry Wood, who earlier that day had been cleared medically to start pitching again after his back injury. I had mentally figured that he would just slot back into his regular starting place which was due up a couple of days later.
“You’re going in, son.”
“You see anyone else out there?”
“Sweep the leg.”
“Nevermind, I’m getting my dreams mixed up.”
Wood went into the game, and in my head, I fretted over what would happen when Wood ran out of gas. The two teams traded zeroes in the 10th and Wood set the Mets down in order in the top of the 11th. In the bottom of the 11th, Todd Walker and Derrek Lee opened the frame with singles. I just needed Walker to score and the game would be over. I got excited. At least until Jeromy Burnitz hit a tailor-made 4-6-3 double play ball. My best chance to end the game was gone, because due up next was Wood. I couldn’t pinch-hit for him because I didn’t have a pitcher to replace him.
There already exists a “Kerry Wood Game”—the one in 1998 where Wood struck out 20 batters in a nine-inning effort. Even though I was a high school senior in Cleveland at the time, I somehow listened to the last inning of that one, when one of the local radio stations realized the historic nature of what was happening and piped in the Chicago audio feed. Had my electronic Kerry Wood game happened for real, there might have been a controversy about which one was the true “Kerry Wood Game.” On the second pitch of his at-bat, around 4:30 in the afternoon, Chicago time, Kerry Wood hit a ball into the left field bleachers at Wrigley, sending 39,000 fans home happy that they were there when it happened.
Day Off (May 12)
This seemed like as good a time as any to take stock of where I was. The Cubs were 18-16, 2.5 games behind the Pirates in the NL Central and one game out of the Wild Card. (Because in mid-May, we think like that.) The team had not been a rousing success, but they’d gotten by. I thought I noticed that my Cubs were playing an inordinate number of extra-inning games, and checking, I found that I was right. They’d played 10 (with a record of 8-2). The Cardinals had played seven, the Pirates five (three of them against the Cubs), and Devil Rays (hello, 2005!) had played five. No one else in baseball had played more than four. No wonder my bullpen was gasping for breath.
Or was it my own doing? I could feel how the tandem starters experiment was consuming my thoughts, but it had effects that I didn’t anticipate. I knew that by having six “starters” it was displacing a bullpen spot. In ye olde days, when multi-inning relief appearances were more common, this probably would have been OK, but I find myself trying to staff those spots with relievers who are only trained to go one inning. Over and over, I find myself looking at my bullpen and realizing that only two or three guys are able to go. It means that a guy like Remlinger, probably better suited to being a LOOGY, is picking up full innings out of necessity. There just isn’t any room for a pitcher who has a specialty of facing one or two hitters.
The effects have shown up in other ways that have been surprising to me. For example, the one thing that the Cubs have excelled at is striking hitters out. (As of this point, they lead MLB.) They also tend to issue a lot of walks (fifth-most in the league). Strikeouts actually end up being fairly pitch count neutral in terms of ways of recording outs. It’s the walks that get you, not because they are particularly pitch intensive, but because they don’t record an out. There’s a feature on OOTP which allows me to request that the pitcher “pitch to contact” during the at-bat. (There is a similar button in real life.) When you run a system that requires you to be very pitch count conscious, it’s a tempting button to push.
The problem again is that the Cubs don’t have a particularly stellar defense. In my head, it seems like every time I push “pitch to contact,” the result is a base hit. That’s not actually true, but the fact that it feels that way is a problem for me. I need a button that says “don’t walk him, but feel free to strike him out.” Sadly, baseball doesn’t have Game Genie codes like that.
In 2005, MLB had a batting average on contact of .324 and a league-wide OBP of .330. On the whole, the chances of either strategy working to produce an out are about equal. With a better defense, I might feel more comfortable pushing the “pitch to contact” button, because it would represent a better way to record outs and maybe shave a few pitches off the odometer. The problem is that if you pitch to contact and the batter ends up with a single, that’s functionally the same as a walk. I had never considered how if I had a better defense, it might affect my decisions, which might affect my pitch counts, which might end up changing my bullpen strategy. That’s a line I never would have drawn.
In other news, today I had a sit-down conversation with now-former leadoff hitter Nomar Garciaparra. (Walker will take over.) At the beginning of the year, I waffled between Garciaparra and Perez as my starting shortstop, eventually picking Garciaparra’s good OBP (in the real 2004, it was .365), but questionable glove over Perez’s solid glove, but anermic bat. It turned out that with the injury to Aramis Ramirez, both are playing, and it turns out that Perez’s bat is awful (a .287 OBP), but Garciaparra has actually been worse than Perez (.262). Maybe I should have done this earlier, but I realize that I was resisting the idea that Garciaparra was really this bad. Small sample size! In retrospect, it seems obvious, but the answers are always really easy to figure out once you’ve read the answer key. I kinda wonder how many games are won or lost because a manager makes the wrong call on whether a player is going through a streak or a legitimate breakout/breakdown.
There’s one other piece that I’ve noticed about my little experiment in management: running is terrifying. Surprisingly, OOTP rarely asks me if I’m interested in trying for an “extra” base (first to third on a single?), and the few times it has, I’ve chickened out. (In fairness to me, they were calls where it probably made no sense … or so say my defense mechanisms as they pop up out of nowhere.) I’ve previously decried third base coaches as being far too conservative in their decisions to send runners, and I’ve played other sims where I had to make the send/go call quite a bit. But it’s showing up in my stolen base tendencies as well. So far, my Cubs have stolen five bases all year, tied for third-fewest in the MLB. Some of that is personnel. The real 2005 Cubs placed 23rd in that category. But even when I have a decently fast runner on first and in a good stealing situation, I find myself hesitating.
It’s generally accepted that to break even on a steal attempt, a manager has to believe that his runner has a 70 percent chance of making it, and that stolen base success rates league-wide usually hover around that same 70 percent rate. Here’s the thing: Most stolen base attempts are taken by players who are pretty fast. A sample that is overly skewed toward obviously fast runners can only make the break-even rate. It means that when I see a runner on base who is obviously not a speed racer, his chances of making it to second are probably well below the break-even point, and even when there’s a demon on wheels at first, it’s likely that his chances are at least very close to the break-even point. Even if those odds are technically a little better, that terrifying prospect of having to push the button rather than just letting everything rest in its steady state keeps me from sending runners. In my head, I know that shouldn’t be the deciding factor, but it is.
Game 41 (May 21) – vs. White Sox; Loss 1-3; Record: 23-18
I hate A.J. Pierzynski. After the off day, the Cubs went on a bit of a run. They took two out of three from the
Expos Nationals, then went to PNC Park and swept a two-game mini-series from the first-place Pirates, wins which brought them to within a half-game of taking first place. After a successful road trip, it was back home to the Friendly Confines to welcome the White Sox for some interleague, intracity fun. The Cubs built an 8-0 lead in the first game and had to hang on to win 8-6 when new arrival Jaret Wright (farewell, Sergio Mitre!) decided it would be nice to allow the White Sox to pad their batting stats a little.
In this game, Carlos Zambrano was dealing. Big Z had thrown seven innings of two-hit shutout ball and was still going strong in the eighth inning, making it look like the one run that the Cubs offense had scored would be enough. It didn’t seem like such a big deal when Aaron Rowand singled to lead off the inning. But when A.J. smacked one into the left center seats, my heart sank. Suddenly, a 1-0 lead was a 2-1 deficit and I may have yelled out a rather Oedipal term when it happened. The Pirates also won their game, so even if I had won, I wouldn’t have slipped into first place, but I began feeling the wave, whether or not there was one to feel.
I feel like I’ve reached a point where I have a lot of the tactical maneuvers down. There have been a few off days, so my bullpen isn’t so beleaguered. I felt like I had gotten the hang of things. And then, A.J. happened. I’m at once an emotional wreck and also very aware that there are still 120 more games to play. They don’t give playoff spots out in May 21, but it would have felt so nice to say, “I’m in first place.”
Game 47 (May 27) – vs. Rockies; Loss 3-4; Record: 26-21
This particular game wasn’t all that remarkable, aside from the fact that we lost it in the top of the ninth when the Rockies plated a run against Rusch. He has become the secret weapon that I didn’t know I needed. He has become a high-leverage long reliever. The original role that I had for Rusch was pitching multiple innings in low-leverage situations, mostly games that I was losing. His job was supposed to be to sponge up innings (along with the now-departed Mitre) so that the “real” relievers (Dempster, Wuertz, Remlinger, and Hawkins) could pitch the situations that really mattered.
The thing is that Rusch might be the best reliever in my bullpen. He has pitched 37 innings in 17 games so far, posting an ERA of 1.70. Even his faux pas in this game (yes, I probably should have had a fresh Dempster in the game in the ninth, rather than Rusch in his third inning of work, but I don’t feel too bad about that one even though I probably would have yelled at another manager for making the same call), can’t diminish the work he’s done for the Cubs this year. In a weird way though, his performance gives me pause.
Part of the idea behind having a Zambrano-Tandem-Maddux-Tandem rotation was that I didn’t need to find a fifth starter, and that Zambrano and Maddux could pitch every fifth day. Given Rusch’s performance, perhaps I should return to a Zambrano-Maddux-Wood-Prior-Rusch five-man rotation (and a seven-man bullpen, with Nolasco and Hill still able to throw multi-inning stints.) But would Rusch still be as effective in a six-inning stint as he has been in three? Has his performance been a small sample size mirage? Part of the problem with a reliever who shows the ability to throw three good innings is that it’s easy to fantasize about him throwing six good innings, and to push him to do it, even if he’s not able.
I have, however, given up on one of my official SaberStrategies™, as I have stopped batting my pitcher eighth. The idea behind the pitcher hitting eighth is that he’s usually the worst hitter on the team, and because lineups are best when they have bunches of good hitters stacked together, it’s not a good idea to have your weakest hitter stacked up with your best hitters at the top of the lineup. The math works out that having a non-pitcher in that ninth spot is better for overall run scoring, although the effect is rather small.
I’ve written before that the problem is that in the NL, hitting the pitcher eighth also forces the manager to make that choice, when the pitcher’s spot comes up in a key situation in the sixth inning, but the starter could go longer, more often. The manager must choose between pinch-hitting and removing an effective pitcher, plus burdening his bullpen with another inning or potentially leaving a key offensive situation to a guy who has a .123 OBP. There’s no correct answer to that question. You lose value either way, and the value that you lose more or less wipes out the value that you gain from the increased offense.
But the killer for me is that if I pinch-hit, I know that I need to fill another inning with the bullpen, and I’m realizing what a giant headache those words are to a manager. You’re going to have to get through three innings with the pen instead of two. Maybe I have the horses in the bullpen tonight that I feel confident will get me through three innings, but if I pitch a guy tonight, it might spoil him being available tomorrow, and eventually, you start looking at your roster and saying, “If I do that, I’ll only have one or two guys who can pitch at all tomorrow night.” I am very used to evaluating strategic decisions on the basis of how they impact today’s game. I rarely think about how they impact tomorrow.