Every year, teams confuse having a handful of veteran names on the roster with having depth, or “roster flexibility.” After signing a marquee player, they’ll surround him with a handful of overrated, overpriced scrubs and cancel out many of the positives that the star contributes.

Shrewd general managers, however, aren’t afraid of a little roster creativity. They know how to supplement their stars by identifying and scooping up other teams’ undervalued castoffs – cheap, freely available talent acquired via trades, the waiver wire, the Rule V draft, or using non-roster invites and minor-league deals in the spring.

The 2002 Angels, for example, taught us that you can spend very little and still have dominant middle relievers. Minor-league veterans Ben Weber and Brendan Donnelly seemingly came out of nowhere to help Anaheim make its first playoff appearance in 16 years and, ultimately, the first World Series championship in franchise history.

Both relievers had almost completely lost the “potential major leaguer” label and earned the league minimum that year. If dominant middle relief can be had for the league minimum, there’s simply no reason to pay more than that for a less-than-average pitcher like Pittsburgh’s Jose Mesa, who will earn a pointless $2 million in 2005.

Weber and Donnelly had one very important thing in common: Most general managers fixated on the things they did badly instead of looking at their performances to see what they’d done well. Consequently, they were labeled damaged goods, which drove their prices down. The Angels smartly moved in, signed them for practically nothing, and got inexpensive production.

The teams that best capitalize on these market inefficiencies give their clubs the same kind of edge you’d expect from an extra $10 million in budget room. Here’s a look at five major leaguers who appear poised to provide their new teams with crucial production at rock-bottom prices after being discarded by less savvy organizations.

To begin, the Dodgers have a couple of relievers in the Weber/Donnelly mold with proven minor-league success who will be earning the league minimum in 2005.

The first is D.J. Houlton, who was a 55th-round draft choice of the Seattle Mariners in 1997. He did not sign and, like fellow 55th-rounder Aaron Heilman, re-entered the draft in 2001. Selected by Houston in the 11th round this time, Houlton moved through the Astros organization quickly before being snapped up by the Dodgers in the Rule V draft.

Houlton must therefore remain on the 25-man roster all season, but his minor-league track record suggests that the Dodgers are hardly living dangerously by giving him 80 innings of middle relief. In fact, his past performance indicates that he should be about a league-average pitcher, with a decent chance to be slightly better.

In 547 minor-league innings, Houlton gave up less than one homer per nine innings, walking 2.17 batters and striking out 8.6 batters per nine. At $316,000 a year, Houlton looks like he’ll be a fantastic bargain.

The Dodgers will also be looking for decent production from Steve Schmoll, a submariner out of the University of Maryland. Though not a castoff from another team like others on this list, Schmoll signed with L.A. as an undrafted free agent in 2003, about as humble a beginning as a U.S.-trained player can have.

After the success of the A’s previously unheralded submariner Chad Bradford, many believe Schmoll has a similar future in store. Like Bradford, Schmoll is stingy with the long ball; in 2004, he didn’t surrender a single home run in 84.2 innings of work. Also included on his resume is a dandy 9.62 K/9 ratio, plus ERAs below 2.00 at each of his most recent stops.

Baseball Prospectus uses a proprietary series of algorithms to predict player performance based on three-year averages called Pecota (Player Empirical Comparison and Optimization Test Algorithm). Using this projection system, Schmoll’s minor-league success will translate to the major-league level, predicting a better-than-league-average performance in H/9 (8.4), K/9 (6) and HR/9 (0.8).

Schmoll is a good bet to perform somewhere between Bradford and longtime righty-killer Steve Reed. If used properly, he’s a great, cheap asset and gives the Dodgers a second bullpen arm who’s a good bet to outperform quite a few more expensive pitchers around the league.

The performance analysis community has been waiting for years for Mets catcher Ramon Castro to break out. Since 1999, the promising young slugger was inexplicably stuck behind no-hit backstops like Paul Bako and Mike Redmond on the Marlins’ catcher depth chart. Castro has never had more than 138 at-bats in season, but with Mike Piazza starting behind the plate in Flushing, Castro could see more playing time than in past years.

He’s unlikely to become a superstar, and Shea Stadium will probably depress his power numbers, but Castro’s power is still very real. The .135 AVG/.231 OBA/.260 SLG line Castro put up in Florida in 2004 in 96 at-bats is skewing his 2005 Pecota forecast southward, largely the result of an annoying toe injury (plus some off-season legal trouble that may also have contributed); he’s a much better bet to exceed his .218/.292/.416 baseline forecast.

The poster boy for undervalued and seemingly flawed talent, the Royals’ Calvin Pickering is a 6-foot-5, 295-pound slugger who once had to have two separate pairs of pants sewn together for him during a brief stint with the Reds. Now 28, Pickering is quickly running out of time to establish himself as a major leaguer.

Not terribly gifted with the glove at first, general managers have been passing on him for years because of his size and defensive shortcomings. In response, all he’s done is hit, including a .246/.338/.500 line in 122 at-bats in Kansas City last season.

The Royals, who aren’t going to compete in 2005 anyway, need to be patient with Pickering. He walks a ton and strikes out even more; if they can look beyond that, as the Red Sox did with bargain-basement second baseman Mark Bellhorn (and as the Reds have with Adam Dunn), they’ll have a huge addition to their offense for just over $300,000.

Like Castro, Rangers outfielder Gary Matthews was hurt by an off year. The .204/.250/.327 line he put up in 2003 in Baltimore (in 162 ABs) was totally out of line with his career numbers; Matthews finished 2003 in San Diego, where he put up a .271/.346/.379 line. In limited duty last season in Texas, he performed admirably: .275/.350/.461 in 280 at-bats. While his 11 home runs were definitely ball-park-assisted, he still gets on base enough to be a real asset.

The switch-hitting Matthews, 30, also offers the bonus of being able to play all three outfield spots. As a starting center fielder, one could do a lot worse, especially for the league minimum contract.

Like most of his comrades on this list, Matthews is unlikely to blossom into a superstar, but he should be a lot better than what five or six AL teams fielded in 2004. If Laynce Nix ever starts hitting enough to reclaim his starting job (he began the 2005 season in the minors), Matthews would make a fine fourth outfielder, as well.