Monday, May 10, 2010

The Nationals' Knack for One-Run Wins

On July 5, 2005, the Nationals beat the Mets 3-2 to bring their record in one-run games that year to an incredible 23-7. At the time, there was a lot of praise heaped on Frank Robinson's late-game strategy, the Nationals' bullpen, and the team's overall ability to "find ways to win"--a lot of the same points you're hearing now.

By the end of the season, the team's record in one-run games was 30-31.

What happened? Did the Nationals' bullpen fall apart? Not really. Chad Cordero's ERA rose from 1.13 in the first half to 3.04 in the second, but that's hardly cratering. Gary Majewski's got better, falling from 3.21 to 2.66. Luis Ayala: 3.06 to 1.15. Hector Carrasco: 2.43 to 1.80.

Presumably the Nationals didn't suddenly lose the intangible "scrappy winner" quality we heard so much about. Frank Robinson was always a terrible in-game strategist, so there's not much point in trying to break down that particular constant.

Is winning close games a repeatable skill at all? If it is, then you'd expect the best teams to have it. Last year, the eight playoff teams were the Yankees, Red Sox, Dodgers, Rockies, Phillies and Cardinals were worse in one-run games than they were overall. The Angels had an almost identical winning percentage. Only the Twins were better. Those were all teams that "knew how to win." Most of them had pretty good bullpens and managers.

Of course, we'd need a much bigger sample size to conclude that winning close games is definitely not a repeatable skill, but someone else will have to do it. But it's certainly a questionable proposition. Nationals fans should be highly skeptical that the Nationals' good fortune in nail-biters will continue.

6 comments:

Deez Nats said...

I remember reading an article in a stathead book (I don't remember which one) that studied this and basically concluded that winning one run games is basically a 50/50 proposition and a skewed record one way or the other is luck. As I recall, one of the main points was that teams weren't consistent from year to year. I remember when I read it it made sense given our 2005 season.

Unknown said...

Remeber in '05 when the Nats swept the #Cubs in Chitown w/ Carlos Baerga batting cleanup? Batista saves are like that.

John O'Connor said...

I would fully expect playoff teams to tend to have worse records in one-run games than overall. The flip side of that is that good teams have better records in non-one-run games than they do overall. You'd expect this, as a good team probably wins a disproportionate share of blowouts.

The Nats have a couple of external factors that probably increase their chances of dfoing well in one-run games. First, the top of the bullpen has been lights out. Those are the guys used when the nats have a small lead. In addition, Riggleman substitutes for defense late, putting in relatively poor hitters for good ones. This reduces the chances of building on small leads (so those one-run leads remain one-run leads instead of becoming two- or three-run leads).

On the other hand, the bottom of the bullpen (the guys who have pitched whern the Nats are behind) have been terrible, which causes one-run deficits to grow (reducing the number of times the team will lose by exactly one run).

Unknown said...

"Is winning close games a repeatable skill at all? If it is, then you'd expect the best teams to have it."

Is this necessarily true? Does a good team, expecting to win 90 games value an individual game less than a team scrapping. To get to .500?

Here's a scenario, Nats are playing Philly and are up 1R in the 8th and decide to use Clippard and Capps in a 3rd consecutive game.

Same scenario with the Phillies up, would they use their top relievers in the same situation or take a "onger" view and put their relievers health above a single win?

Philly would probably be more willing to trust their offense to come back if the game gets tied.

I'm not sure this would hold up on deaper analysis of manager's behavior, but it is theoretically possible.

Unknown said...

Sorry for the typos, on my blackberry on the inter office shuttle.

Unknown said...

Even if Steven is right (I think he is), that doesn't necessarily spell doom and gloom for the Nats. Theoretically, we are going to be getting at least a few pitchers that will make the team better. Assuming they live up to reasonable expectations (Strasburg winning every Cy Young for the next 20 years in 3/4 of this season; Storen not throwing a single ball or giving up a single hit; etc.), the Nats will not have to continue winning so many close games. Hopefully mid-season reinforcements can hand off more than a 1-run lead and we don't get the opportunity to "even out" our unbalanced close game record.

The biggest problem with this theory is that part of the reason these games are close is that we are not scoring enough runs to have a large lead even when we get stellar starting pitching performances. If we get 7 shut-out innings from a SP, but only score 1-run, we are still in that same 1-run game that might even itself out. And the reinforcements I am expecting do not swing the bat.