Of all the ailments to Darren Dreifort's major league reputation, being wild has never been one. But consider the following chart, which lists the all-time Los Angeles Dodger leaders in walks per nine innings in a season, with a minimum of 50 innings pitched:

BB/9 Year Player
7.13 1960 Danny McDevitt
6.50 1962 Phil Ortega
6.35 1999 Onan Masaoka
6.19 2002 Kazuhisa Ishii
6.19 1959 Stan Williams
6.18 2003 Kazuhisa Ishii
6.07 1962 Joe Moeller
5.94 1958 Sandy Koufax
5.88 1996 Chan Ho Park
5.84 1991 Mike Hartley
Source: Lee Sinins' Sabermetric Baseball Encyclopedia

Including today's performance against the Cubs, Ishii has improved to 5.19 walks per nine innings in 2004. However, Dreifort, who never previously broke the 5 barrier, is at 5.94.

As it happens, opponents are batting only .225 against Dreifort, which is very good. But because of the walks, each hit Dreifort allows has more potential magnitude for damage.

Since the Dodgers traded Guillermo Mota to Florida, the conversation around Dreifort has centered around whether he should retain the role as setup man for Eric Gagne. This obscures the question of whether Gagne should have a setup man at all at this point.

Is a mediocre setup man better than none?

If Gagne has at least one day of rest, what is the motivation to use anyone else in a close game in the eighth inning? The two-fold answer is fear: fear that the Dodgers will need Gagne fresh on the next day, and fear that Gagne will not be effective in the second inning. (With a token fold thrown to the usual people who would reflexively criticize the breaking of convention.)

Regarding the first point, there haven't been enough opportunities to prove that Gagne is ineffective the day after being used for two innings. Even more to the point, it much more rare than you would think that Gagne is ever needed on consecutive days.

Gagne has pitched in 50 games this season - less than half of the 115 the Dodgers have played. Several of those games have been in a mop-up role.

Gagne Pitching on Consecutive Days, 2004
Two days: 7 times in 4 1/2 months
Three days: 1 time in 4 1/2 months
Four days: 2 times in 4 1/2 months
Total: 15 times in 4 1/2 months

To hold Gagne back on the chance that he will be needed in 24 hours is to bet against the odds. It happens less than once a week.

As for a dropoff in Gagne's second inning of work, that's certainly possible, depending especially on how difficult his first inning is. But the uncertainty of the dropoff is no worse than the uncertainty the Dodgers face when they bring in Dreifort. And there remains the possibility that, as happened Friday against the Cubs, the Dodgers could score enough runs in the ninth inning to eliminate the need for Gagne to pitch an extra frame.

Ironically, the Dodgers have already used the Gagne First scenario with success. In an extra-inning game August 1, Gagne pitched before Dreifort. Facing extraordinary pressure, Dreifort got the save. It was a save that came with only a few feet of a warning-track fly ball to spare, and one that was overshadowed by the three innings Gagne pitched in his longest relief outing in years. The lessons, however, were that the Dodgers were less likely to win if they held Gagne back from pitching until after Dreifort - Gagne might not have even gotten into the game before it was lost - and that Dreifort is Dreifort, whether it is the penultimate inning of the game or the final one.

Today in Chicago, down by two runs, the Dodgers loaded the bases with two out in the eighth inning. If you were Dodger manager Jim Tracy and you weren't restricted by batting order - if you had the choice to bat any Dodger in that situation, who would you choose? A mediocre hitter, on the premise that you might want your best player for the ninth inning? Me, I would choose the best possible Dodger. I would choose Adrian Beltre.

Fortunately for the Dodgers, it was Beltre's turn to bat. Regrettably, he was called out (dubiously) on a check-swing strike three. But at least the Dodgers had their best man up. In contrast, at Cincinnati on Thursday, the Dodgers had a lead in the eighth inning, their best pitcher available and rested, with only one inning pitched in the previous four days - and did not use him. Despite little doubt that he could have handled two innings, Gagne never got into the game.

If it were game 162 of the season and the Dodgers were tied for a playoff spot, would they hold back their best pitcher on the premise that they might need him for a tiebreaking 163rd game? Of course not. It doesn't make sense to leave your best weapons in Maybeland. It's too late in the season for the Dodgers to let Gagne watch and wonder in order to nurture Dreifort's confidence. Dreifort can certainly get people out - just not as often as Gagne. Should there really be such mental anguish over acknowledging a fact as obvious as the earth is round?

The thing is, I think Jim Tracy knows that there shouldn't be. I'm not really even second-guessing him. I just think he's temporarily trapped by this need to play to the general public, to massage Dreifort and to withhold from them the ammunition that would help characterize the trade of Mota as a mistake. Eventually, as he has with Ishii, with Eric Karros and with others, Tracy will probably make it a policy, instead of a rarity, to do what needs to be done. Reducing a player's role is not the same as eliminating it.

Sometimes, the best players fail. Sometimes, lesser players succeed. And the fact is, Dreifort has cost the Dodgers only two games this season. It's way too soon to have a heart attack about it, and I'm not having one. I'm just talking about playing the percentages, about being smart.

Perhaps Dreifort should be the Dodgers' alternate setup reliever. Perhaps it should be promising, albeit hittable, Yhency Brazoban. Perhaps it should be Wilson Alvarez, if the Dodgers get healthy in the starting rotation again.

But it should be clear that the Dodgers already have a first-string setup man, and his name is Eric Gagne.