The problem of Darren Dreifort is about as easy to solve as that famous 1,500-piece jigsaw puzzle, "Coastal Fog at Night."

I mean, what do you do? Where do you start?

A year ago Thursday, I explored the popular notion that Dreifort should try to salvage what's left of his career by adding full-time pinch-hitting to his pitching duties.

The discussion was inspired by the intent of Milwaukee's Brooks Kieschnick to do the opposite - add pitching to his hitting duties.

Interestingly, with the 2003 season behind us, we can see that Kieschnick succeeded more than he failed.

Kieschnick's pitching was below average, with an ERA of 5.26 and an ERA+ of 83. But keep in mind that Kieschnick was the last man on the staff, and his ability simply to eat up 53 innings that otherwise would have gone to the next-to-last men on the staff had some value in itself.

Further, according to Baseball Prospectus, Kieschnick had a positive VORP (Value Over Replacement Player) of 0.3, which means that while he wasn't as good as the average major leaguer, he was as good as any minor leaguer that might have filled his spot.

On top of that, Kieschnick managed to turn in a pretty stellar year as a reserve hitter, OPSing .967 and OPS+ing 145, thanks to seven home runs in 70 at-bats. His VORP on offense was 11.9, higher than, for example, Colorado's Larry Walker.

Anyway, in looking last year at Dreifort's hitting abilities, I saw that despite his Wichita State pedigree of 25 home runs in 314 at-bats, and his occasional home run with the Dodgers, Dreifort was as much of a major-league hitter as Anna Nicole Smith is an actress. He had never had an OPS over .600. His career OPS is .517.

"Even with extensive batting practice, even if he is truly healthy, it stretches credibility to imagine that Dreifort could raise his hitting skills to the point of being more than an emergency batsman," I concluded back then. "There are many better hitting pitchers out there today.

"Too bad. It looks like he's going to have to earn that salary on the mound."

And that, we all know, is a tough proposition.

When Dreifort does pitch, he has moments of brilliance before finishing with an average ERA. The oxymoronic combination of Dreifort's flashes of talent and degenerating physique puts the Dodgers in a very tough spot.

Dodger manager Jim Tracy cannot look at Dreifort to be a starting pitcher, nor even as a thrice-a-week reliever in the Paul Shuey mold. Instead, what Tracy has in Dreifort is a pitcher who probably needs two days off one week, five days off the next week, and perhaps two weeks on the disabled list the next month, before restarting his cycle.

To make use of Dreifort's talent when he is active, the Dodgers might have to carry 12 pitchers on their roster. And for that effort to be worthwhile, either Dreifort or that 12th pitcher, who would be someone like Steve Colyer, has to provide more value than the reserve hitter he would replace.

And so, we now turn to examine the candidates for the final bench spot on the Dodgers. Guys like Joe Thurston, Olmedo Saenz and Jose Hernandez.

Earlier this month, I identified Saenz, for example, as a dark-horse candidate to make the team and even contribute. He had an above-average OPS+ of 113 in 178 plate appearances with Oakland in 2002.

But Saenz can't pitch.

And this is really the conundrum for the current Dodgers. Who is more valuable to have on the roster - Dreifort or Saenz? Colyer or Saenz?

Let's look again at Brooks Kieschnick. He produced four Win Shares last season: two as a hitter, two as a pitcher.

Dreifort, in 60 innings with an ERA of 4.03 last year, produced three win shares as a pitcher, none as a hitter. If he can average three innings a week, taking into account at least one stay on the disabled list, Dreifort might get 80 innings this season and four Win Shares.

If he can hit at all - which, again, is questionable, but maybe he can do better than Daryle Ward - Dreifort might generate a fifth Win Share.

Colyer, who enters the picture as the heir apparent to Tom Martin as the lefty specialist, had two win shares last year. Martin had four, which might be a reasonable expectation for Colyer in 2004.

Can Saenz, or Thurston, or whoever, as the last man on the roster, generate more win shares than the modest totals of the 12th Dodger pitcher?

Not necessarily.

Five Win Shares is tough to come by off the bench. Mike Kinkade only got two in 2003. Ron Coomer got zero.

Perhaps this is will all become moot for the Dodgers. Perhaps Dreifort will be on the disabled list so much that he will essentially remove himself from the equation. In general, major-league rosters are in flux.

But while Dreifort is healthy, he may have worthwhile value as a 12th pitcher. And if he can hit - hit at all - that only adds to it.

The Dodgers should have this in mind when they schedule Dreifort's duties in 2004, in games and in practice. He doesn't even have to hit as well as the worst position player on the 25-man roster to be worth splitting into two.