June 8, 2006

Schrodinger's Bat

Swing and Miss

by Dan Fox

the archives are now free. All Baseball Prospectus Premium and Fantasy articles more than a year old are now free as a thank you to the entire Internet for making our work possible. Not a subscriber? Get exclusive content like this delivered hot to your inbox every weekday. Click here for more information on Baseball Prospectus subscriptions or use the buttons to the right to subscribe and get instant access to the best baseball content on the web.

Subscribe for $4.95 per month Recurring subscription - cancel anytime. Subscribe for $39.95 for one year a 33% savings over the monthly price! Purchase a $39.95 gift subscription a 33% savings over the monthly price! Already a subscriber? Click here and use the blue login bar to log in.

"Things could be worse. Suppose your errors were counted and published every day, like those of a baseball player."
--Unknown

Every fan knows that Babe Ruth struck out over 1,300 times to go along with his 714 home runs. Pete Rose made almost 10,000 outs in his career. And last week SABR members enjoyed a lively discussion on the their listserve discussing the players who made the most outs in a season (a hint: "Omar the Outmaker" takes three of the top ten spots).

Failure is inherent in the game, a fact that George Will, in a recent interview with our own Will Carroll, credits with making baseball "especially suited for a nation in which the politics are democratic." This is so because the game, like our republic, requires both the patience to endure that failure and the compromise of "settling for the half loaf." After all, even the best hitters fail 65% of the time, and the best teams lose 60 times per season.

Those who write about the game certainly strive to hit the mark in excess of 65% of the time, but from time to time we too strike out. So this week, we'll look at three swinging strikes from your humble author to go along with a quick take on the humidity of the front range.

Strike One: Posada's Prowess

In each of the last two columns I referenced Jorge Posada's May 17th homerun to cap the Yanks 14-13 comeback. And both times I mentioned that the Yankees probability of winning before the homerun was 46% making Posada's blow worth 54% in terms of WX.

Several readers, and our own Rany Jazayerli, quickly pointed out the absurdity of that notion. Reader Bill Watson-Canning put it this way:

"I'm wondering how the Yanks WX could possibly have been that high preceding Posada's HR. Jorge makes an out more than 54% of the time (as does just about every player not Bonds or Giambi), and this was a situation where an out would have ended the game. I could see the WX for the Yankees hitting 46% after Damon got on base to start the 9th, but with 2 outs and a man on 2nd I would be surprised if the WX was greater than 20%."

Well said. In fact, the WX for the Yankees at that point was actually 14.9%, and not 46%, making Posada's homer worth 85.1%. As Bill intuits, the Yanks WX with zero outs was in fact 46%. That's what happens when you put a '0' where a '2' should be.

Strike Two: The Walk-Off

This point gives us the opportunity to address walk-off home runs. The column relating high impact performances listed the five largest-impact plays of 2005. Unfortunately, the methodology used did not account for walk-off home runs, and as a result I missed these plays that would have been within the top five, and which I'm sure many readers will remember (whether fondly or not).

• June 2, Baltimore at Boston. With two outs in the bottom of the ninth and trailing 4-3, David Ortiz hit a three-run home run off of B.J. Ryan to send the Fenway faithful home happy. The home run raised the Red Sox WX 85%.
• June 11, Anaheim at New York. In the bottom of the tenth with two outs and runners on first and third, Cliff Floyd homered off of Brendan Donnelly to make the final score 5-3. Before the shot, the Mets had a 19% chance of winning. The homer therefore was worth 81%.
• July 15, Houston at St. Louis. Down 3-2 in the bottom of the thirteenth inning, Albert Pujols hits a two-run bomb with two outs, worth 79%.
• July 9, Detroit at Tampa Bay. Jonny Gomes smacks a two-run homer with one out in the bottom of the ninth for a 5-4 Devil Rays victory. The homer jumped their WX 78%.
• May 10, May 1011, Oakland at Boston. Most remarkably, on consecutive days Kevin Millar and Jason Varitek hit walk-off homeruns off of Octavio Dotel, each with with one out in the bottom of the ninth. Both blasts were worth 78%.

Strike Three: Missing Comebacks

After the column on comebacks was published last week, I received plenty of emails from readers who attended specific games. Many of those relived the joy, frustration, and amazement that such games produce, and so I thank those who took the time to write in.

However, several readers noted a couple other games that didn't make my list and which, after a little digging, I discovered had been omitted. Without further adieu, here are three more big comebacks and one that failed but is close to my own heart.

• July 28, 2001 (game 1), Houston at Pittsburgh. This game was by far the most referenced, and for good reason. Heading into the bottom of the ninth the Pirates trailed the Astros 8-2. The first two batters were quickly retired by Mike Jackson, leaving the Pirates at their lowest WX of the game, .0142% (1 in 7,030), and the lowest of any team that came back to win since 1970. From a probability standpoint the Pirates' situation was almost twice as unlikely as that of the Indians in August of 2001.

  A double by Kevin Young followed by a home run by Pat Meares made it 8-4. A single by Adam Hyzdu and a walk to Tike Redman followed by a run-scoring single from Jack Wilson chased Jackson from the game. Billy Wagner was called in from the pen, and promptly hit Jason Kendall (as if that's hard to do), bringing up Brian Giles. On a 1-1 count Giles hit a grand slam to right field to clinch a stunning 9-8 victory. The WX chart for that game looks as follows.

  

• May 1, 1973, Pittsburgh at San Francisco. On this day, in front of fewer than 8,000 fans, the Giants trailed the Pirates 7-1 with two outs in the bottom of the ninth and a runner on first. At that point, their WX was at .0431% (1 in 2,323) as Willie McCovey came to bat. Two walks to load the bases followed by a Chris Arnold grand slam made the score 7-5. A double by Gary Matthews Sr. and walks to Dave Rader and Jim Howarth provided the opportunity for Bobby Bonds to hit a bases-clearing double for the 8-7 win. The WX chart follows:
  

• August 29, 1986, Detroit at California. At least one reader, Brian Smith, asked specifically about this game since it's one he attended. That night the Angels fell behind 3-0 in the first, 5-1 in the third, 8-1 in the fifth, and 11-3 in the 7th before making their comeback. Still it was 12-5 with one out in the bottom of the ninth and runners on first and second with the Angels WX at .0987% (1 in 1,013).

  Four straight hits plated four runs before a ground out and walk to Ruppert Jones allowed Dick Schofield to hit a walk-off grand slam off of Willie Hernandez for a 13-12 victory. Brian describes the home run as "one of those cheap Anaheim Stadium homers that curled inside the left field foul pole over the short fence at the 333 mark," reminiscent of another more famous but not more improbable walk-off.

  

But while all of these are excellent comebacks to be sure, one that I was reminded of by a reader and which has personal resonance occurred on May 17th, 1979 when the Phillies visited Wrigley Field. The wind was blowing out that day, and Cubs starter Dennis Lamp didn't make it out of the first as the Phils scored seven times, which included a home run by Phils starter Randy Lerch. When Dave Kingman answered with a three-run homer and Bill Buckner, Keith Moreland, and Ted Sizemore drove in runs in the bottom of the frame, you knew it was going to be a long day at the ballpark.

The Phillies scored eight more runs in the third and two in the fourth to make it 17-6. The Cubs then mounted a furious comeback with three runs in the fourth (on another Kingman homer) and seven in the fifth--the big blows were a Buckner grand slam followed by a Mike Vail two-run homer. I distinctly recall our family hearing the Vail home run on WGN radio (we lived in eastern Iowa and therefore did not receive WGN-TV), and being amused as the Cubs announcer rejoiced afterwards with the phrase "now the Cubs are only five behind!"

Despite our incredulity, the Cubs scored three more runs in the eighth after the Phils tacked one on in the 7th and tied the score at 22. Mike Schmidt, always a Cubs nemesis, finally won the game in the 10th on a solo home run (his second of the day) off of Bruce Sutter. The mayhem was over after four hours, 50 hits, and 11 home runs--three hit by Kingman alone.

The low point for the Cubs occurred with two outs in the top of the fifth, with the Phillies threatening and a 21-9 lead putting their WX at .15% (1 in 662). Although the Cubs lost, it'll always be a game I remember fondly.

One of the interesting things about this particular game from a WX perspective is the consideration of run environment. The calculations for this article were done based on the average runs scored per game at Wrigley Field in 1979. However, the weather conditions that day obviously served to greatly increase the run environment just for that day, and so a more accurate calculation would need to adjust the environment on a game-by-game basis. That opens the Pandora's box of applying custom run environments based not only on weather but on starting pitchers, all the way down to granular conditions including specific match-ups for each at bat (one reader made that very point when analyzing the Indians 2001 comeback against the Mariners in detail). Hey, I'm all for the reductionist approach, at least where baseball is concerned, but you can see how this could quickly get out of hand.

It's Not So Much the Heat, it's the Humidity

But let's move on to something a little more relevant to this season. Here in the shadow of the Rocky Mountains the talk has turned to humidity (at least before yesterday's 16-9 Rockies victory over the Pirates). A spate of articles at the end of May both locally and in the national media have recounted the fact that Coors Field seems to be playing as more of a neutral park this year, and Kent Mercker even chimed in by saying, "I like pitching here. This park has become one of the best pitchers' parks."

Most of the stories done on this topic at some point mention that it looks like the humidor the Rockies use to store baseballs and installed after the 2001 season is finally "kicking in." Because the Rockies have said that they've always stored the balls at 70 degrees Fahrenheit and 50 percent humidity, and continue to do so, one might find it hard to believe that only now would the humidor's effects be felt.

To see whether this season is truly an outlier, let's start by considering the following graph which shows the ratio of home versus road scoring and home runs per out made for the Rockies and their opponents from 1995-2005.

What the graph indicates is that for three consecutive years prior to this season, home runs have declined at Coors relative to road games, while run-scoring had held basically steady since 2001 (with some year-to-year variance). However, it is clear that the 1995-2000 period was different from the 2001-2005 period, and so one could certainly argue that the humidor has played a role in that despite the fact that offensive levels were up in 2002 relative to 2001. However, it should be noted that with the opening of Minute Maid Park in 2002, Great American Ballpark in 2003, and Citizen's Bank Ballpark in 2004, the advantage that Coors Field had has shrunk relative to the league. More hitter's parks raise the league's overall level, making Coors less extraordinary.

But where does 2006 stack up? Through games of June 4th, run-scoring stood at .99 and home runs at 1.10. In other words, through last weekend Coors was playing basically league-neutral in terms of run scoring and about 10% higher in terms of home runs. Of course, this includes just 26 home and 30 road games. Because of the small sample size, one might imagine that what we're seeing is the effect of early-season weather in Denver, which can be very cold and windy and therefore depress run scoring.

To see whether early season weather may be a factor, we can restrict our view to games played through May 31st of each season, as shown in the graph below.

You'll notice that there is obviously more variability when looking at so few games, but what shows up is that 2006 has indeed been by far the lowest through the early part of any season. Only 2002 approaches this year, when run scoring was at 1.12 and home runs at 1.28, still 12% and 14% higher respectively.

So it would appear that something has indeed changed. Although we don't at this point have definitive answers, we might speculate that there are least three factors that may be playing a role.

• Bad Offense, Good Pitching? Probably the second most-discussed topic on the front range has been the Rockies' pitching. It's clear that both the Rox starters and especially relievers are better this year than in years past. It's also clear that the offense is among the worst, ahead of only the Cubs in run scoring despite taking advantage of some great technology. It stands to reason that because baseball is an inherently non-linear game, a better pitching staff will be more consistent both on the road and at home, and therefore likely depress the difference between home and road run scoring. In the same way, a poor offense may inherently mean smaller differences in run scoring even when playing in a hitter's park.
• Soggier Baseballs? Interestingly, struggling Rockies shortstop Clint Barmes mentioned in an interview on MLB.com Radio last week that the Rockies are leaving the balls in the humidor longer. This points to the fact that the team has experimented with the humidor, possibly including the temperature and humidity levels, in order to find the right mix. In an article by the Denver Posts' Mike Klis last May, Klis noted that under the mix at that time the balls put into play weighed an average of 5.12 ounces, still below the maximum weight of 5.25 ounces, and therefore a bit more lively. Perhaps tweaking the settings and exposing the balls for a longer time period has brought them closer to the maximum weight and size.
• Longer Grass? Rockies TV announcer George Frazier mentioned on the June 5th broadcast that the team is also keeping the grass a bit longer, thereby making it more difficult for grounders to scoot through the infield. While that wouldn't have any effect on home runs, it may indeed mean fewer hits and/or fewer extra base hits, and therefore lead to fewer runs. As anecdotal evidence, I noticed twice in last weekend's series sweep by the Marlins that rookie shortstop Hanley Ramirez was able to stretch what would normally be singles into doubles as the balls slowed considerably after landing in the outfield. Perhaps Rockies outfielders will need to adjust accordingly.

In the end it may be too early to say anything definitive, but it certainly appears as if playing at altitude is not what it used to be.