February 6, 2008
Doctoring The Numbers
Angels, Indians, and Reds
Looking to place a wager on a sure thing? Willing to place your bet now, and wait up to 30 years for the payoff? In other words, are you a degenerate gambler? If you are, do I have the wager for you!
Whatever the current odds are that Francisco Rodriguez will make the Hall of Fame, they're too low. Rodriguez, you may recall, burst onto the scene in the 2002 playoffs by winning five games, striking out 28 hitters in 19 innings, and even earning his nickname, "K-Rod," before his rookie season started. The only player in history who comes close to this kind of post-season impact without even losing his rookie eligibility is probably Gregg Jefferies in 1988. (Although we Royals fans still have warm and fuzzy feelings about Todd Worrell's ninth inning in Game Six of the 1985 World Series.)
What you may not remember is that the next year Rodriguez came pretty slowly out of the gate in 2003, posting a 5.48 ERA as of May 21st, but then he rallied to put up a 2.14 ERA the rest of the way, and his highest ERA since his rookie season is last year's 2.81. He took over the closer's role for Anaheim in 2005, and has saved 40 or more games each of the last three years.
Here's the kicker: Rodriguez was in the majors when he was just 20, and became a full-time closer at age 23. This is historically very unusual. More than any other position, the role of closer is one that teams would much rather give to a veteran player who, fair or not, has proven he can handle the stress of the ninth inning. Also, closers often come from the rank of converted starters, and sometimes those starters have to struggle in the majors for a few years (think Eric Gagné) before they make the transition.
Rodriguez had 59 saves before his 24th birthday, which is almost unprecedented, but he doesn't even hold that distinction in his own division: Huston Street earned 69 saves before he turned 24 last August. However, Rodriguez has saved more games in each of the last three seasons than Street has saved in any season. Rodriguez has earned 146 saves, and just celebrated his 26th birthday this month. No pitcher in history has come close to recording that many saves prior to turning 26:
Pitcher Saves Francisco Rodriguez 146 Gregg Olson 131 Chad Cordero 128 Ugueth Urbina 102 Rod Beck 91
Olson, Cordero, and Street are arguably the three most successful examples of the college-trained closer who makes a successful and almost instant transition to the same role in the major leagues. Rodriguez was in the majors at an age when all three of them were still on campus.
A lot can go wrong between now and the Hall of Fame, obviously. Olson blew out his arm the following year and finished with just 217 career saves, and I'm sure he considers himself lucky he's not Ugueth Urbina or Rod Beck. If Rodriguez stays healthy-or maybe he needs a year off for Tommy John surgery, but returns as good as new-he's going to amass an absolutely ridiculous number of saves in his career. No one questions his effectiveness; he has a career 2.37 ERA, and of the top 15 players in terms of saves before the age of 26, only Olson had a lower ERA (2.36). Rodriguez also has a career strikeout rate of 11.97 per nine innings, which among those same 15 players ranks behind only John Rocker's rate (12.29 per nine).
Given the pace Rodriguez has set, and the early start to his career as a closer, and even a tool as blunt as Bill James' Favorite Toy projects him to finish his career with 509 saves, with a 15.5 percent chance at 700 saves. Trevor Hoffman currently stands at 524 saves, and figures to push that total up over 600 before he's done. No active reliever is likely to challenge Hoffman's record before Rodriguez; of the 15 active relievers with more saves than Rodriguez, the youngest is 31, and that's Gagné. If Rodriguez breaks Hoffman's save record, with the kind of success he's had to this point, you have to figure he's a shoo-in for the Hall. So place your bets now.
The difference between the 2006 Indians, who finished under .500, and the 2007 Indians, who won 96 games and came within a game of the World Series, can be summed up in two words: the bullpen. The 2006 Indians actually had the better offense (870 runs to 811), and the ERA of the starting pitchers improved only 12 points between 2006 and 2007, from 4.31 to 4.19.
But the bullpen, which went 16-27 with a 4.73 ERA in 2006, dropped its ERA by nearly a full run to 3.75, and nearly inverted its record with a 26-17 mark. Limiting your opponents to 50 fewer runs does not normally result in an 18-game improvement in your record, but the timing of those runs was particularly critical. The Tribe was just 18-26 in one-run games in 2006, but 29-24 last season. Not coincidentally, a team that had underperformed its Pythagorean record by 11 games in 2006 actually outperformed its Pythagorean record by five games last year. As an aside, this is why Pythagorean records are so important: the Indians won just 78 games in 2006, but on paper they were nothing like a 78-win team, and they proved that last season.
The improvement in the bullpen certainly didn't come in the ninth inning, where Joe Borowski once again proved that a closer is just a reliever who arrives fashionably late. Borowski had a 5.07 ERA but still managed 45 saves; of all the relievers who have ever made 40 saves in a season, he had the highest ERA in history, by a margin of more than three-quarters of a run. He blew eight saves, but the Indians came back to win four of them anyway.
It wasn't at the end of the game where the Indians saw their turnaround-the improvement came in middle relief. Among the notables who created that turnaround, Rafael Perez emerged from obscurity to give Cleveland a dominant left-handed reliever. Even more importantly, Rafael Betancourt had himself quite a year, arguably the greatest season by a middle reliever of all time. Let's start by defining a middle reliever. Make it as simple as possible: any pitcher who made five or fewer starts, earned fewer than 15 saves, and pitched at least 55 innings is a middle reliever. Betancourt is one of just 20 middle relievers in history to put up a single-season ERA under 1.50, but that alone does not give him claim for the best season ever. His 1.47 ERA wasn't even the best of 2007; Carlos Marmol had a 1.43 ERA.
What made Betancourt so unique was his incredible ability to keep men off base. In 79 1/3 innings, he surrendered just 51 hits and six unintentional walks. He also did not put a single batter on base with a hit-by-pitch, and in face he has just one HBP in his career. Counting hits, unintentional walks, and hit batsmen, Betancourt surrendered fewer baserunners per nine innings than any middle reliever (min: 55 IP) in history. It's not even especially close:
Year Pitcher IP H UIBB HBP BR BR/9 2007 Rafael Betancourt 79.1 51 6 0 57 6.47 1991 Jeff Gray 61.2 39 6 1 46 6.71 1964 Dick Hall 87.2 58 11 0 69 7.08 1966 Hoyt Wilhelm 81.1 50 15 1 66 7.30 2002 Arthur Rhodes 69.2 45 12 0 57 7.36
If we lower our threshold to 40 innings, then Cla Meredith's 2006 season sneaks to the top of the list; in 50 2/3 innings he allowed just 6.22 baserunners per nine, and with a 1.07 ERA to boot. But aside from the huge difference in innings, Meredith's performance was the result of incredible luck in the hit department; he struck out just 37 batters, and owed most of his success to an absurd .199 BABIP. (In contrast, last season his BABIP was .344, and he allowed 94 hits in 80 innings. Maybe Voros was right.)
Betancourt, on the other hand, struck out over a man an inning, and there's nothing about his season that screams "fluke." The Indians just signed him to a two-year, $5.4 million contract with a club option for 2010 for another $5.4 million. At those prices, he's not just the best middle reliever in the game, he's also one of its biggest bargains.
Jeff Keppinger gets no respect. Scouts like him for being a gamer that plays above his tools, and analysts admire his ability to make contact and hit for high averages, but general managers treat him like lint in their pockets, as something to be discarded as soon as it's noticed.
The Pirates drafted him in the fourth round in 2001, and after an unimpressive debut season in 2002 when he hit .276, he's done nothing but contend for batting titles ever since. He batted .325 for Lynchburg in 2003, finishing third in the Carolina League. He moved up to Double-A the following year and was hitting .337 when the Pirates threw him into that bizarre exchange of Kris Benson for Ty Wigginton and Jose Bautista trade, the kind of trade that only the Pirates could make-if they wanted Bautista that bad, they could have protected him in the Rule 5 draft the year before. Keppinger hit .348 in the minors for the next month before the Mets called him up, and he hit a perfectly respectable .284/.317/.379.
That earned him a repeat trip to the minors in 2005. Once again, he hit .337, and once again, he drew more walks than strikeouts. Unfortunately, he broke his tibia in a baseline collision in mid-June, and missed the rest of the year. Still, he had proven pretty definitively that he could hit minor league pitching.
The Mets disagreed, sending him back to Triple-A in 2006. (Hey, Chris Woodward had to play.) Keppinger hit an even .300 for Norfolk, and in mid-July the Mets traded him to Kansas City for Ruben Gotay in a classic "my unwanted second baseman for yours" swap. Keppinger hit .354 for Omaha, and in September gave the Royals a .267/.323/.400 performance. The Royals were so impressed with what they'd gotten that last January they designated him for assignment, eventually dealing him to Cincinnati for an A-ball pitcher. The Reds, naturally, returned him to Triple-A; apparently they wanted to see if he could hit .400 or something. He didn't, but he did hit .368/.424/.469. Grudgingly, the Reds called him up in July.
In his first ten days on the roster Keppinger saw action just three times, all as a pinch-hitter. Then Alex Gonzalez missed time to be home with his ailing infant son, and Keppinger started at second base, then third base, then shortstop. And he hit all over the field. He hit .320/.382/.500 in July, and by the time Gonzalez returned in early August, Pete Mackanin had to keep in the lineup somewhere. Two weeks later Gonzalez left the team again, and Keppinger played mostly shortstop the rest of the way. This in itself is amazing enough; Keppinger was a second baseman from the day he signed his pro contract, and in six minor league seasons he'd played all of eight games at shortstop. He played in 47 games at shortstop for the Reds last year, and played at least passably well.
While learning a new position at the major league level, he hit as well as he had in the minors, if not better. He followed his hot July with a torrid August, hitting .393/.460/.596. Despite cooling off in September, Keppinger finished the season at .332/.400/.477. Just as impressively, he continued to show remarkable plate coverage, coaxing twice as many walks (24) as strikeouts (12).
Let's think on how impressive that last feat is. Prior to 1956, major league pitchers as a whole had never managed to strike out more than one batter every other inning (i.e. 4.5 strikeouts per nine innings) in any season. Since 1956, they have never failed to do so. Prior to 1956, it was not unusual for a batter, even a middle infielder, to walk twice as often as he struck out. In 1925, Joe Sewell walked sixteen times as often has he struck out, which is what can happen when you whiff just four times in an entire season.
Since 1956, no middle infielder with a minimum of 40 games at second or short had ever batted .330 in a season with a walk-to-strikeout ratio of 2-to-1. Until Keppinger:
Year Player AB Avg BB K BB/K 2007 Jeff Keppinger 241 .332 24 12 2.00 1956 Harvey Kuenn 591 .332 55 34 1.62 1975 Rod Carew 535 .359 64 40 1.60 1974 Rod Carew 599 .364 74 49 1.51 1996 Chuck Knoblauch 578 .341 98 74 1.32
If we lower the batting threshold to .320, Keppinger's walk-to-strikeout ratio drops to third, behind another Reds middle infielder of some repute-Joe Morgan, in his twin MVP years of 1975 and 1976.
Typically, Keppinger still has an uphill battle for playing time on his hands this year. Alex Gonzalez is still under contract for the next two years, and is the much better fielder. Keppinger could move back to his natural position, except Brandon Phillips-one of only two second basemen (the other is Alfonso Soriano) to put together a 30/30 season-has that job locked up. Keppinger is currently slotted for super-utility duty, which is a shame.
The circumstances that brought Keppinger to Cincinnati are eerily similar to what happened ten years prior, when the Royals gave the Reds another second baseman they didn't know what to do with in a lopsided trade. While Chris Stynes didn't have Keppinger's plate discipline, he did hit .348 in his first season with the Reds, and .334 three seasons later in close to full-time duty, while playing all over the field. Stynes didn't hit much in the seasons in between, and his bat died soon after leaving Cincinnati. There's reason to think that Keppinger's ability to make contact and stronger minor league resumé will lead him to a better fate.