Happy Thanksgiving! Regularly Scheduled Articles Will Resume Monday, December 1
May 23, 2001
Doctoring The Numbers
Before getting into this week's column, I would like to make a pair of comments regarding last week's feature on the amazing Ichiro Suzuki:
I stated that, assuming Ichiro could continue to get hits in 95% of his games, that his chances of fashioning a 57-game hitting streak were extremely high. I did not mean that I expected Ichiro to get hits in 95% of his games. I was merely extrapolating his performance to date, nothing more.
As it turns out, I screwed up the math anyway. As reader Andy Cleary writes, "Rany, I'm pretty sure you messed up one piece of math. The flaw is that you can't start a hitting streak on 'any given day.' You can only start a hitting streak on a day after you have been held hitless."
Andy is absolutely right. The chance that Suzuki will start a 57-game hitting streak on Opening Day is (0.95)^57, or 5.4%. However, the chance that he will start a 57-game hitting streak on the second day of the season is (5.4% x 5%), because he has to go hitless on Opening Day to start a streak the following day.
This changes the percentages entirely. Instead of having a 99.7% chance of having a 57-game hitting streak at some point during the season (again, based only on his performance to date), Ichiro's chance of passing Joe DiMaggio is just 38.6%. (However, over the course of a 1,000-game career, that chance would increase to 93.6%.)
My apologies for the bad math. And on that inauspicious note, here's this week's column.
Well, he's done it.
Alex Gonzalez, the Marlins shortstop who drew just 15 walks in 136 games as a rookie in 1999, suffered through one of the worst seasons by an everyday hitter in recent memory last year. Gonzalez hit an even .200, and with just 13 walks in 109 games, finished with an execrable .229 OBP, the lowest by any player with 400 or more plate appearances since the immortal Mario Mendoza posted a .219 OBP in 1979 (while hitting two points under the Line named after him).
In the off-season, the Marlins made it clear that if Gonzalez wanted to keep his job in 2001, he would have to show some discretion at the plate. Gonzalez got the message. "I know this year I'm going to take more walks," Gonzalez said during the first week of the season. He's done just that; in fact, Gonzalez set a career-high with his 16th walk of this young season on May 19.
In fairness, five of Gonzalez's walks are intentional (he had never been walked intentionally before this season), which makes his walk spike a little less impressive, if still noteworthy.
What is the largest increase in walks by a batter from one season to the next? A lot depends on who is eligible. If you limit the list to players who qualified for the batting title in each season, the list looks like this:
Player Year 1 AB BB Year 2 AB BB Increase
The gap between #1 and the four-way tie for #2 on this list is one of, um, Ruthian proportions. While intentional-walk data is not available before the mid-1950s, it has been widely speculated that as many as half of Babe Ruth's walks in the early 1920s were intentional. That would, of course, imply that Ruth was given a free pass 85 times in 1923, the year he set the all-time record for walks.
Regardless of the actual number, it's pretty clear that Ruth was already the most feared hitter in the game in 1922, and he didn't draw 170 walks that season. Of course, he didn't come to bat as often as he did in 1923, and that skews our results some as well.
To adjust for that, we can use "matched sets," whereby the at-bat and walk totals for each player in the season they batted more often are scaled down to match their playing time from the other season. For example, Ruth had just 406 at-bats in 1922, and if we scale his 1923 at-bat total to 406, we find that he would have walked "just" 132 times, an increase of just 48. The greatest increase in walk totals among "matched sets" in history:
Player Year 1 Matched BB Year 2 Matched BB Increase
(By way of comparison, for Gonzalez to match Richie Ashburn's record in a season of 500 at-bats, he would have to walk 108 times.)
Jimmy Wynn and Reggie Jackson accomplished their feats in 1969, when the strike zone was shrunk and the mounds were lowered, leading to one of the greatest league-wide increases in walks ever: the average NL team increased their walks by 24.5%, the average AL team by 20.1%. Gonzalez, on the other hand, is taking a leap forward in a season in which the strike zone has been enlarged and walks are significantly down throughout baseball.
It's interesting--and probably not coincidental--that two of the three biggest walk jumps in history coincided with the all-time home run record and the last .400 season.
This list isn't particularly relevant to Gonzalez, though, as all five of these players were pretty patient hitters before their walk spike. Gonzalez, on the other hand, was one of the most free-swinging hitters on the planet before this season, and it would be interesting to see whether other brain-dead hackers have been able to develop plate discipline overnight.
Here's the same list, limited to players who drew less than one walk per 20 at-bats in Year 1 (since 1900):
Player Year 1 Matched BB Year 2 Matched BB Increase
Two rookies appear on this list, and it just so happens that both of them went into the Hall of Fame. Both Joe DiMaggio and Al Kaline appeared headed for stardom even as rookies, though. Larry Bowa's 1979 season was a complete fluke; it was the only season in his 16-year career in which he walked even 40 times.
The player on this list that resembles Gonzalez the most, unfortunately for him, is Brian Hunter, who like Gonzalez was in his third full major-league season when he stopped swinging at everything. Hunter's patience fled as quickly as it arrived, though; he drew only 36 walks the following season as his career began its sink into oblivion.
Hunter, however, was 26 years old when he had his walk spike, while Gonzalez is just 24. Perhaps younger players who develop the skill of plate discipline are more likely to hold onto it. So let's look at one more list, which like the last one is limited to hitters who walked less than once every 20 at-bats in the year before their walk spike, but who were 24 or younger in the year of their walk spike:
Player Year 1 Matched BB Year 2 Age Matched BB Increase
Aside from the aforementioned Hall of Famers, the list includes another first-ballot inductee (Paul Molitor) and one of the greatest phenoms of the last 50 years (Cesar Cedeno). Cedeno's walk spike coincided with his emergence as a bonafide superstar, as he hit .320 with 22 home runs, a league-leading 39 doubles, 55 steals, and a Gold Glove, all as a 21-year-old playing in the Astrodome in a pitchers' era.
If you look at our Player Card for Cedeno, you can see that if his performance that season was translated into an average park of today, he would have hit .346/.413/.632 with 33 home runs and 52 doubles, again as a 21-year-old Gold Glove outfielder. Let's put it this way: Cedeno, at the conclusion of that season, was a more valuable commodity than any player active today, up to and possibly including Alex Rodriguez. Yet he finished his career with barely 2,000 hits, and he never won a Gold Glove, made an All-Star team, or hit 20 homers after he turned 26 years old. Cedeno is a sobering reminder that on the road to the Hall of Fame, even the most reliable car can veer off course if the driver isn't paying attention.
Back to our study... Dick Bartell and Ted Simmons both maintained their solid walk rates for the rest of their distinguished careers (Bartell, known as "Rowdy Richard," was perhaps the best defensive shortstop in the NL in the 1930s.) Tony Gonzalez also maintained his walk spike during his sophomore season, and was a fine outfielder for the Phillies throughout the 1960s before his recent reincarnation as the best tight end in the NFL.
Bobby Tolan's walk rate began to slide almost immediately, and he never again had a season as good as his 1970 campaign (he did miss the entire 1971 season, and in all honesty, I don't know why). Lee Tannehill and Phil Lewis were both generally awful players whose walk rate dipped as soon as pitchers in the dead-ball era figured out that they had nothing to lose by throwing strikes to either player.
What does this all mean for Alex Gonzalez? Only three of the ten players on the list did not go on to enjoy long and productive careers after their walk spike, and of the three, two played in the dead-ball era and one missed the entire season after his walk spike. For a guy whose career was on the brink last season, that's a pretty good sign.
Keep in mind, of course, that Gonzalez still has to show that he's capable of maintaining his newly-found discipline all year long. Early returns are not encouraging: after drawing 12 walks in 22 games in April, Gonzalez has walked just four times in 18 games in May. Not coincidentally, after hitting .307/.409/.453 in April, his numbers have tumbled to .167/.231/.200 this month.
But there's still time for him to rebound and try to wedge his name onto a list that contains three Hall of Famers. And let's not forget that less than a year ago, his offensive ineptitude evoked comparisons to Bill Bergen, the worst hitter ever. That's progress.
Rany Jazayerli is an author of Baseball Prospectus. Contact him by clicking here.