It’s been an up-and-down season for the Washington Nationals. Depending on the week, they have appeared to be either a mediocre squad or a playoff lock. Despite a measly 1 1/2 games separating them from the wild-card lead, the Nats still face a fundamental problem: There isn’t enough talent on the roster to spark the type of run needed to make the playoffs.

On the morning of May 29, the Nationals’ record stood at 24-25, putting them 3 1/2 games out of first in the tight NL East. A .490 winning percentage might be disappointing to many squads, but in 2004 the team formerly known as the Montreal Expos lost 95 games. Baseball Prospectus’s projection system, PECOTA, pegged the team for a 76-86 record while predicting the Nats would score fewer runs than any other team in baseball – partly due to their pitcher-friendly home park, but largely due to their anemic offense. Compared to expectations, and amidst the love fest that was baseball’s first few months back in Washington, D.C., the Nationals’ 24-25 start was far from disappointing.

Then fantasy took over. From June 1 through July 4, the Nationals went on a huge 25-6 run, winning eight series, four by sweep. The team’s incredible one-run record drove much of that streak. The Nats went 11-1 in one-run contests during their June-July run. By the time fireworks lit up the July 4 sky, the Nationals’ record in one-run games stood at 22-6 for the season. The strong performance of the Nationals’ bullpen, headed by lights-out closer Chad Cordero, played a big role in the team’s success, as did the strength of Washington’s starting pitching.

On July 4, Baseball Prospectus’s Clay Davenport ran a series of simulations on the rest of the season. He determined that the Nationals, then holding a 5-game lead in the division and a nine-game advantage in the wild-card race, had a 66.4% chance of making the playoffs.

But the team’s lofty record in one-run games raised a red flag. Over the long haul, a team’s record in one-run games compared to all other contests tends to even out. If the Nationals’ near historic pace of one-run wins came to an end – as the odds strongly suggested it would – the team’s overall record could plummet in the second half.

That’s exactly what happened. Since their season peak, the Nats have crashed to Earth, going 14-28 and falling five games back of the first-place Braves. A reversal in the Nats’ record in one-run games has played a role. But perhaps the most telling stat for the Nationals has been their total runs scored and runs allowed. Even when the team was surging ahead in first place, those numbers were running about even, suggesting Washington was merely an average team.

Though the Nats remain above .500 and in the race, they’ve now allowed more runs than they’ve scored, as their anemic offense has done them in. Baseball Prospectus’s most recent Playoff Odds simulations give Washington just an 8.8% chance of seeing October.

So do the Nationals have another winning streak in them? Let’s take a look at their hitting, starting pitching, and bullpen to determine whether there is room for the team to improve beyond their current .524 winning percentage.

On the hitting side of the ledger, the Nationals have been even worse than expected. The team is averaging 3.93 runs per game, dead last in the majors. The spacious dimensions of RFK Stadium are influencing these figures, but a look at run scoring while teams are on the road shows the Nationals to be the thirdworst team in baseball.

A scan of the roster makes it hard to see how the Nationals’ offense could improve. Most of the team’s best hitters are already performing above their heads. First baseman Nick Johnson’s .303 AVG/.420 OBA/.500 SLG batting line is a huge bump from his career averages and makes him the Nationals’ most productive hitter by a long shot, but as someone who hasn’t played in more than 100 games in a season since 2002, he’s vulnerable to another extended stint on the disabled-list. In addition to Johnson, key hitters like Jose Guillen and Brian Schneider are also hitting well above their career averages.

New third baseman Vinny Castilla has played worse than was hoped, but the expectations placed on him were unfairly high. Castilla’s current line of .246/.317/.393 fits his recent history, especially when you consider that his one true offensive strength – his ability to hit for power – is thwarted by his home park’s cavernous dimensions.

Center fielder Brad Wilkerson’s .405 slugging average this year is 73 points below the cumulative slugging rate from his three previous seasons, which is partly due to RFK Stadium. An irritated nerve in his forearm has also sapped his power.

As for the rotation, it’s also unclear where the Nationals could reasonably expect enough improvement to make a serious run. As with most teams, the Nats have auditioned and jettisoned several subpar back-of-the-rotation arms, but Claudio Vargas, Zach Day, and Sun-Woo Kim all looked horrible in their auditions.

Still, current starters Livan Hernandez, Esteban Loaiza, Tony Armas Jr., and John Patterson – along with the traded Tomo Ohka – have taken 101 of the team’s 124 starts, all pitching as well or better than projected. Patterson’s amazing 147.3 innings of 2.32 ERA ball have played a huge role in keeping the Nats in the race all year. Meanwhile, Armas is enjoying his best season since 2001 and Loaiza is having one of the best years of his career at age 33. It’s hard to see where the team can improve, considering its dearth of front-line pitching prospects.

If the hitters and starters are playing as well as can be expected, how much more can we hope for from the bullpen? Baseball Prospectus’s Reliever Wins Added looks at how well a reliever performed and multiplies it by how important the situation was to the game. According to this new statistic, the Nationals bullpen is the best in baseball. Chad Cordero has accumulated more Reliever Wins Added by himself than 18 entire major league teams.

It’s hard to see how this bullpen could get any better. Outside of Cordero, Luis Ayala’s 2.55 ERA is the best of his career and Gary Majewski’s 3.08 is certainly respectable for a second-year player.

Even adjusting for offense-crushing RFK Stadium, the Nationals’ rotation and bullpen have greatly outperformed expectations. But if adding lackluster veterans like Guzman and Castilla to an already light-hitting crew looked like a losing bet last winter, this season’s offensive woes have only confirmed it. With 38 games remaining, it’s possible the Nats could put together another streak similar to the one they managed in June. But to make that happen – and outpace three better teams to boot – they’re going to need a lot more luck, because the talent just isn’t there.

Mr.Gorman is a writer for Baseball Prospectus. For more state-of-the-art commentary, visit their Web site at www.baseballprospectus.com.