Ten Things I Didn’t Know Last Week

The All-Star Break made people think.

It was an interesting All-Star break, wasn’t it? Staging the event in San Francisco, where Barry Bonds is about to eclipse Hank Aaron’s career home run mark made for some interesting coverage. There were a lot of people taking time to pause and ask what it all means, or doesn’t mean. A lot of reporters were trying to find out what the commissioner thinks it all means, too. Spoiler alert: he’s not saying.

According to a a New York Times/CBS poll, 84% of fans feel that there should be an asterisk next to Bonds’ record if it is definitely proven that he took steroids. I’m in that other 16%. In fact, I’m in agreement with Mike Piazza on this issue: Let’s allow history to give us some perspective before rushing to judgment.

Not too likely, I know.

Barry Bonds needed sabermetrics, not steroids.

According to Jeff Pearlman, Bonds decided to start taking steroids after the 1998 season because he was frustrated that Mark McGwire and Sammy Sosa were getting all the media glory. In a word, he felt unappreciated.

You may remember that 1998 was the year McGwire hit 70 home runs and Sosa hit 66. Bonds hit “only” 37 homers and batted .303/.438/.609. Sosa and McGwire finished first and second in the MVP voting; Bonds finished 8th.

But while I was cruising around Jeff Sagarin’s wonderful listing of Win Probability Added results from 1957 to 2006, I found out that Bonds added more wins (6.9 WPA) than any other batter in the majors that year other than McGwire. Bonds even led all NL players in WPA for four straight years, from 1990 through 1993. No other player in this 50-year chronicle of WPA ever accomplished the same thing. Mickey Mantle, Sandy Koufax and Frank Thomas all led their leagues three consecutive times, but no one else ever did it four straight seasons.

So you see, if the mainstream media had just adopted sabermetric baseball stats twenty years ago, Bonds would have received lots of attention for his great years. He would have been plenty appreciated. Problem solved.

The best season of the past 50.

Admittedly, I lost a lot of free time in those WPA stats. There’s enough info in there to generate a ton of “Ten Things I Didn’t Know” columns. How about this one: The greatest single year of anyone other than Barry Bonds in the past 50 years was Dwight Gooden’s 1985.

Not surprisingly, Bonds’ 2001, 2002 and 2004 were the three greatest years of any player in the past 50; Gooden’s was next with more than 10 WPA. If you were alive and a Mets fan in 1985 (I was both), you remember Gooden’s year vividly. He was phenomenal, always a threat to strike out anyone who dared to approach the plate. Mets announcers and writers fell over themselves trying to come up with better and better superlatives for the guy. It was embarrassing.

Now, WPA isn’t the best metric to judge pitchers because it doesn’t include the impact of fielding. But Gooden was a strikeout pitcher who didn’t rely on his fielders as much as other pitchers. He posted a 1.53 ERA in 276.1 innings and he turned it on when it counted: Batters hit only .143 with runners in scoring position against him. I don’t have a hard time believing it was the greatest non-Bonds year in the past 50.

Following Gooden in great WPA years are Willie McCovey in 1969, McGwire in 1998, Todd Helton in 2000, Mantle in 1957 (an impressive ranking because they only played 154 games back then), Albert Pujols last year and David Ortiz in 2005.

The worst season of the past 50.

So guess who has posted the worst WPA season of the past fifty? None other than the infamous Neifi Perez, who contributed six wins less than average in 2002. As I said, WPA doesn’t include the impact of fielding, so Perez’s 2002 almost certainly wasn’t the overall worst when you include fielding prowess.

For that honor, I would nominate George Wright, an outfielder for the Rangers in the early 1980s. In 1985, Wright batted .190/.241/.242 with four stolen bases in 109 games; truly pathetic figures. His batting WPA was five wins below average, the worst seasonal total of any player not named Neifi.

As I’m sure you know, Neifi Perez was recently suspended 25 games for testing positive for stimulants (amphetamines) for the second time. So Neifi’s awful 2002 may have been chemically enhanced? Should it be asterisked?

How good was Roy Face’s 18-1 season?

We baseball analysts are only now beginning to truly reap the benefits of what the great folks at Retrosheet have done. The “Retrosheet years” (1957-2006) include detailed play-by-play records of virtually every game played during those years, and it’s a great deal of fun to get dirty in the details. Luckily, you don’t have to be a data wizard to do so, because Sean Forman’s Baseball Reference has done the dirty work for you. By combining Baseball Reference and Sagarin’s Win Probability Added stats, you can learn a lot.

For instance, one of the most fascinating seasons for baseball historians is Elroy Face’s 18-1 season in 1959. His won-loss percentage of .947 is the single season record, and sportswriters were so impressed that they placed Face seventh in MVP voting that year.

But Face was a reliever and, as you probably know, wins aren’t the best metric in the world for judging relievers. WPA is better and, according to Sagarin’s numbers, Face had only 1.1 WPA in 1959, 51st in the league.

What happened, you ask? According to Face’s splits on Baseball Reference (subscription required, I believe), opponents batted .343 against Face in the eighth inning and .248 in the ninth. Also, they batted .291 when there was a one-run margin, but only .143 with two outs and runners in scoring position. One last fact: He allowed over half of his inherited runners to score. In sum, he had a curious habit of allowing teams back into games, but not enough so that the Pirates couldn’t mount a comeback later in the game.

According to WPA, Face had better years in 1958 (1.5 WPA), 1960 (2.4) and 1962 (2.6).

How relief impact has changed.

We all know that relievers have been used more and more since the days of Face. In fact, they have dominated the WPA pitching leaderboards for the past few years, as shown on Fangraphs. Now that we have historic WPA data, we can ask the question: When did this happen? What has been the overall trend of relievers at the top of WPA leaderboards?

Here is a graph to answer that question. I pulled the top leaders in pitching WPA for each of the past 50 years, and estimated the percentage of them that were primarily relievers. This graph shows the percent by year, and I’ve added a couple of dotted lines to separate what I think have been three phases of reliever growth.

Phase one was the period of the accidental top reliever, such as Lindy McDaniel in 1960 and Stu Miller in 1965. These guys were fine relievers, but they were rarely used often enough to lead the majors in Pitching WPA. Around 1980, however, relievers began to become more common among the WPA leaders.

Odd data point: remember Bruce Sutter’s phenomenal 1977, when he posted a 1.34 ERA? He finished fourth in the league in pitching WPA that year, behind Rick Reuschel, Tom Seaver and the leader in pitching WPA: Rich Gossage (another reliever who had a 1.62 ERA with the Pirates and pitched more innings).

Anyway, the success of Sutter and Gossage in the late 1970s evidently led managers to use relievers more often, and they placed among the WPA leaders more consistently. On average, about 30% of pitching WPA leaders were relievers during the second phase. In the late 1990s, relievers jumped ahead again, when the current trend of using relievers for just an inning in highly leveraged situations became common practice.

Will it this pattern change again in the future? You bet. How will it change? Got me.

New baseball graphs

As an occasional stock investor, I’ve never been sold on the value of finding stock trends by graphs. That may strike you as strange, because I’m a graphing fool, but I’ve usually preferred some analysis with a large dose of common sense and market knowledge. However, there’s a new website that is taking an investor’s approach to baseball graphs. It’s called Baseball Investor, and it’s definitely worth a look.

Ballpark progress

Are you losing sleep wondering how the new ballpark in Washington, D.C. is coming along? Worry no more, my friend, because you can now take a daily peek at the blooming park. This is actually fun—try going back in time a few months to see what things looked like before. This is what it will look like when it’s finished in time for the 2008 season.

Recognition for the U.S.S. Mariner

Congratulations to our friends at U.S.S. Mariner for making an impact on their favorite team. Exasperated with Felix Hernandez’s stubborn refusal to throw anything but a fastball in the beginning of a game, they wrote an open letter to the Mariners’ pitching coach, Rafael Chaves, pleading with him to get Felix to change his approach. Chaves actually showed the post to King Felix, who figured that if even those goofy guys thought he should change his approach, then he might as well do it.

U.S.S. Mariner and Lookout Landing are two of the best team blogs anywhere. I don’t know why Mariner readership is so blessed; must be something about the salmon.

Shane Victorino is crazy or fearless. You choose.

Here’s the evidence.

References & Resources
The Roy Face analysis was inspired by the good folks on the SABR-L listserv.