September 23, 2003
During a game last Wednesday, Tampa Bay Devil Rays manager Lou Piniella complained that the Boston Red Sox relievers were watching television in their bullpen, while his team's bullpen had no television. After talking to the umps, the umps made the Sox turn off the television. Piniella said a couple of things, but mostly that by having a TV, relievers could better see batters and their approach, which gave them an unfair advantage.
There are important issues at stake here. What if there are better-quality sunflower seeds available in one bullpen? Could one team stock a nasty flavor of Gatorade, like "Glacier Freeze," in the opposing team's bullpen in hopes of knocking them out of their routine? Make the bench itself uncomfortable and wobbly, promoting inter-bullpen arguments about who's rocking it?
It's not, incidentally, cheating to steal signs. There's nothing in the rules that says you can't, because there's nothing in the rules about signs at all. Technically, this is all outside the rules anyway...except that I understand there's an MLB rule that prohibits electronic devices in ballparks entirely. Which if true, the Red Sox are breaking. Unless MLB granted them an exemption, which they do all the time when teams want to do things like build stadiums with dimensions forbidden by the rules, or violate the debt/equity rule if the team is owned by the Commissioner.
Piniella's blamed losing on sign-stealers before. Last year, his signs were so obvious I was picking them off from the stands. After the team was spanked in Toronto, having runners picked off stealing and so forth, Piniella said that he'd gotten together with his coaches and changed all their signs, because it seemed like they knew what signs were on. This ignored entirely the issue of whether the Blue Jays scouts might have figured out that Piniella, like a monkey delivered a banana every time Ichiro stole on 1-2 counts, pushed the same buttons continually.
Back to the bullpen thing. It's particularly interesting because it relates to season-long rumblings around the league that the Red Sox steal signs. Some teams have supposedly gone to always using more complicated signs usually reserved for runner-on-second situations when facing the Sox.
The unnamed sources are all on the side of sign-stealing. The New York Daily News quoted "one AL official" as saying: "Everybody in the league knows they've been cheating all year up there."
An AL scout is quoted in the Seattle Post-Intelligencer saying: "They've been cheating all year. We've been trying to find out how, but we knew there was something." And so on and so forth.
Pat Gillick complained to the Commissioner's office earlier this year about the televisions after the Mariners were swept in Boston. The same Gillick, who last week with Larry Stone said of the A's in a Seattle Times interview: "You have to give them credit, but the test is going to be how they maintain it. It's difficult, with that payroll, to maintain. We'll have to wait and see. Initially, they've gotten it done, but once Tejada and Chavez are eligible (for free agency), if they can't pay them, you'll probably see a decline in their won-loss record."
Yeah, those A's. Anyway, back to the bullpen thing again, the Seattle Post-Intelligencer quoted Gillick, when asked about Piniella's accusations, as saying: "I'd suggest that you look at the discrepancy in their on-base percentage and batting average at home and on the road and draw your own conclusions."
And by discrepancy, Gillick means "difference"--he's not implying that STATS, Inc. or anyone is intentionally manipulating events at Fenway in the Sox's favor. Gillick also said in that A's interview: "I'm not sure what the level of competition is at Class-A, rookie or amateur level." Here's a hint, from me the baseball writer to Gillick the general manager: it's lower than Double-A. Also, it's not hard to figure out how great the difference is. People like Clay Davenport do it all the time. I know it doesn't matter to you, because you're leaving after this year and all, and those players are years away, but you're one of 30 people in the world in charge of a major league team...
Back to the bullpen thing. Let's take Gillick up on his challenge. Or I will, anyway. You stay where you are and keep reading.
2003 Home: .317/.393/.524 Away: .264/.329/.458 Difference: .053/.064/.066
They're stealing signs! It's true! Wait, let's not be too hasty. Fenway's a slight hitter's park, it's natural to expect some difference. Still, that's a pretty big gap. Let's see what happens if we go back a year.
2002 Home: .273/.344/.433 Away: .280/.346/.454
Uh oh. Difference: .007/.002/.021 in the other direction.
2001 Home: .274/.343/.452 Away: .259/.325/.426
Difference: .014/.018/.026 towards playing better at home. Still, not nearly as large as the splits in 2003.
Gillick's right: if you look at their splits this year, there's something crazy going on compared to the last couple of years. You could say it's due entirely to the new offense, but even if they were designed to take advantage of Fenway's peculiar character, wouldn't they also hit on the road a little better?
Here's one way to test that theory: look at the splits of regular players who've been with the team. Did their home stats go up abruptly this year? Let's look at this chart:
2003 home 2003 away Who AVG OBP SLG AVG OBP SLG Nomah .369 .411 .659 .245 .286 .396 Manny .330 .450 .579 .317 .402 .586 Nixon .319 .409 .537 .294 .387 .611 Damon .292 .367 .430 .255 .240 .386 Varitek .301 .383 .550 .249 .327 .489 2002 home 2002 away Who AVG OBP SLG AVG OBP SLG Nomah .328 .362 .522 .293 .343 .533 Manny .336 .441 .659 .360 .458 .635 Nixon .258 .343 .456 .254 .334 .482 Damon .284 .366 .426 .287 .348 .458 Varitek .263 .335 .406 .268 .330 .380
If anyone's stealing signs, it's Nomar Garciaparra. Nomar's hitting hugely better at home: 124 pts of AVG, 125 of OBP, 263 points of SLG at home compared to on the road, and that's much better than his splits last year.
Meanwhile, if the bullpen's stealing signs, they're not consistently getting back to home plate in time for the rest of these guys. Look at their year-to-year improvement in splits using quick and dirty OPS:
Nomah 0.380 Manny 0.034 Nixon -0.035 Damon 0.185 Varitek 0.086
Damon's 185 points of OPS is nothing to scratch at. It's nearly the difference between Barry Bonds and Albert Pujols, for instance, without adjusting for parks (which tips the scales further in favor of Bonds). Or Alex Rodriguez and the next-best shortstop in the major leagues.
If the Red Sox are cheating, I'm not seeing it. Why would Nixon hit worse? Why wouldn't Manny, a dude who can crush pitching anyway, not absolutely destroy the league? Looking at different hitters, a 100-point fluctuation in total OPS season-to-season is nothing that raises eyebrows. That fluctuations within the much smaller samples of splits would be much greater is entirely reasonable. That they're all up is suspicious, though.
Is there then a case for the Red Sox as a bunch of cheaters? Not really. The lineup makeover means it's harder to pin this down, but from what we can look at, only Nomar is doing as well as you'd expect if they were consistently stealing signs.
Here's the problem, though--the Red Sox are a smart bunch. If they were to steal signs at home, and it was a huge advantage, perhaps they've thought this through, and they're only pushing the advantage when it's critical: close and late, runners on.
So welcome, dear readers, to Small Sample Size Theater.
2003 home empty 2003 home RISP Who AVG OBP SLG AVG OBP SLG Nomah .403 .440 .685 .340 .403 .619 Manny .360 .433 .680 .282 .488 .494 Nixon .346 .431 .626 .303 .415 .447 Damon .296 .364 .441 .263 .344 .329 Varitek .272 .330 .515 .338 .449 .588
Standard disclaimers apply. Actual consumption of these statistics may cause dizziness, nausea, and loss of indie and/or street cred.
In four of the five cases, these batters become significantly less dangerous with men in scoring position:
OPS drop Nomah .103 Manny .131 Nixon .195 Damon .132 Varitek -.192
Weird thing about that, the all-AL average for bases empty is .261/.323/.423, and with runners in scoring position it goes up to .272/.353/.427 (what, no pitches to drive in those plate appearances? Don't they tell us that's when the batters see more meaty fastballs?).
Where the average hitter in the AL (Adam Kennedy, say) hits a little better and draws a few more walks with runners in scoring position, the Boston hitters at home aren't hitting the ball as often, as hard, though they're drawing a couple more walks.
If the Red Sox were stealing signs, we could expect to see top-to-bottom improvement in their hitting, and if they were being shifty about it and using it selectively, the only obvious conclusion available is that this signaling is distracting the batters, causing them to squint off past the pitcher and take more pitches than normal.
Unless the Red Sox are protecting their methods like the Allies protected their breaking of Enigma--maybe they're using it selectively, say against divisional and Wild Card rivals, where the sample sizes are going to be so small they're effectively useless for evidence, and they're not using their advantage the rest of the time in order to better conceal its existence in a much greater sample of normal, unaffected outcomes.
In this vein, it makes sense that as the Allies concealed their advantage with decoys and dummy intelligence operations the Germans could believe as explanations for unlikely convoy sinkings, the Red Sox realized that since their stealing had gotten around the league, they would offer their enemies an easy explanation--the bullpen television--and allow that to be shut down, while retaining their other better sign-stealing mechanism(s).
Which is to say that if the Red Sox are cheating, they're being pretty devious about it, and it's not the kind of thing that easy stat perusal is going to reveal, or turning off the offending television is going to solve.