Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52
keyboard_arrow_uptop

I’ve been told that my never-ending pleas for fans, media, and decision makers to take the results of a month of baseball with a grain of salt gets tiring, and commenters on our site have noted that with some frustration as well. It’s a fair point, but I’m certain that following this general rule makes more sense than knee-jerk overreactions to four starts or 50 at-bats this early in the season.

I mean, if you put too much weight on one month of baseball, you say things like this:

“The [Diamondbacks] will be among the best in the league for years to come.”

At this point in the 2008 season, the Diamondbacks were 19-7 and seemed ready to run away with the NL West. Their young core was raking, with Conor Jackson at .345/.434/.583 and showing the power we’d been waiting for from him. Justin Upton was hitting .323/.371/.559. The long-awaited ascension of a young team seemed to be upon us. And I bought it, in part because it fit my preconceived notions, and in part because I simply forgot that players can do just about anything in one month. Players such as Casey Kotchman:

“Look for [Casey] Kotchman to hit .320/.410/.500 this year and continue being the Angels‘ best player.”

Yeah. Kotchman, a longtime favorite of mine whose progress had been stymied by injuries, got off to a .326/.392/.565 start. I went on ESPNews and sung his praises more than once. Kotchman even went out in the first game after I wrote that and picked up three hits, including two doubles. As you already know, it didn’t last. From the time the article dropped through July 28, Kotchman batted .274/.304/.409, drawing eight unintentional walks in nearly 300 plate appearances. He was so bad that the Angels went out and acquired two months of Mark Teixeira‘s services to replace him, sending Kotchman to Atlanta. He was worse with the Braves.

So, I got fooled. Later in May, I wrote Jim Edmondscareer eulogy. He then caught on with the Cubs and played very well for them, batting .256/.369/.568, and helping them reach the postseason.

Twenty-nine days into last season, the Orioles were 15-11. The A’s were 17-10. The Marlins were 15-10. Kosuke Fukudome was at .326/.444/.483. Brian Bannister had a 2.48 ERA, three wins, and an 18/6 K/BB. On the other hand, CC Sabathia had a 7.88 ERA. Gil Meche? 7.22. Roy Oswalt had allowed seven homers and 24 runs in 36 innings. Carlos Delgado had a 645 OPS. These are the things we were talking about, and any commentary other than “it won’t last” would have been wrong.

Now, some April stories end up being quite real. Cliff Lee did; he began the season as the best pitcher in the AL and he ended up winning the AL Cy Young Award. Carlos Quentin was an MVP candidate for most of the year after his .316/.455/.620 start. Andruw Jones and his terrible start set the tone for a lost season. Ryan Dempster really did turn a corner. The thing is, while we can use hindsight to know that Lee and Quentin and Jones were real, and that Bannister and Fukudome and Kotchman weren’t, the available evidence at the time would not likely have caused you to nail those answers. I mean, do you remember the Brian Bannister hype? Or for that matter, the Fukudome hype? Maybe not, because the performance of those two players went south and took the hype with it, and we forgot.

We shouldn’t forget, because the key point is illustrated by those players. April is fickle. April sometimes lies, and sometimes tells the truth, and discerning between when she’s doing each is hard work, sometimes even impossible work. Heck, consider the current belle of the ball, Zack Greinke. Last year at this point, the righty was 3-0, 1.25, and second in the AL in ERA to Lee. He had a good year, certainly, but he didn’t rewrite any record books, and in fact, he didn’t get a single vote in the Cy balloting. Or, if you prefer, consider J.J. Hardy, off to a brutal .167/.234/.286 start this year. He opened 2008 hitting .218/.281/.287 before pulling it together and finishing at .283/.343/.478. There’s nothing to say he won’t do it again.

It is possible to learn things in April. It’s just as possible to be misled, and a look back at the big stories through about 25 games a year ago shows just how much of what we think we know, we don’t.

By the way, I wasn’t a complete moron last year. There was this nugget in May about a team I’m known to hate.

“The buy-in for winning the AL Central eight weeks ago was 93 wins. It’s lower than that now, and if the Sox weren’t likely to reach that first number, it’s much more likely that they can get to 88 or 89, and that may be all they need.”


Thanks to Marc Normandin for his research assistance with this piece.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
Pietaster07
5/05
To be fair, the Marlins going 15-10 ended up being a Marlins going 84-77. That's 69-67 afterwards, which isn't that bad for the young team.
Richie
5/05
So can we use component analysis to separate the April wheat from the April chaff? Could greater computing power bring PECOTA into that discussion? Perhaps a list of possible predictors, things that could be checked? (schedule thus far, health factors via Will, I'm sure there must be numerous other 'usual suspects' that could be rounded up and looked into)
jsheehan
5/05
I'm not sure what the approach is, but I've been thinking a lot about how to use April's information to modify projections going forward, more for teams than players.

IOW, one way to do it is to ignore April, say that Team X was projected to be a .569 team, and project them to win (.569 x RestofSchedule + April wins) games. The other thing is to say that April shows that they're really a .535 team, and do the math that way.

The thing is, how much certainty can we put on April? Schedule, injuries, natural variation are all non-skill issues that could lead teams to have better or worse records. And while you're teasing that out, they've played a few more games, and the data's all changed.

It's a really interesting question, and I don't quite know how to answer it systematically. It starts to seem more..."squishy"...than quantifiable. I can tell you why teams have a better record than expected through one month, and I can tell you why they're supposed to regress (or the converse); it's harder to know, really know, when Cliff Lee is happening.

I really wish I were as arrogant as advertised. It'd be easier.
blakesley
5/05
George Santayana, from The Life of Reason, Vol.1: "Those who cannot remember the past are condemned to repeat it."
dfiala
5/06
You are the rudder of sanity. It's a thankless task -- everyone wants instant "analysis" and to discern the big surprises before everyone else -- even if it is only from a handful of events. Your relatively long-term outlook and measured take on the game is a welcome antidote to the aimless hyperbole readily available elsewhere.

Stay the course and continue to write about the perils of "knee-jerking" in an entertaining manner. That's one of the things I've come to enjoy about April baseball.
illgamesh
5/06
The rudder of sanity?
eighteen
5/06
Yes, as opposed to its mizzenmast.
tinseltown
5/06
There is an implicit acknowledgement of the small sample size problem in traditional lore when, as you noted in an earlier article, a win in April is not treated as seriously as a win in August. (Though of course it would be better to look at factors such as strength of schedule when deciding if a team's April record is indicative of a sustainable level of play rather than just dismissing it entirely.)

One key problem with repeating yourself ad nauseum is that after a while, I tend to skip over articles that seem to cover the same old ground, and can easily miss any new aspects contained within. I'd rather you work more on fleshing out new points of discussion, rather than go over what I've read many times before (in the same month, to make it worse).
jsheehan
5/06
From the writer's standpoint, the inspiration to revisit this was the examples from 2008. I think it's constructive to look back at what we were experiencing at the same point in other seasons, to show that these "great stories" often end up being just random spikes.

It was in doing the research that I discovered my own work from last year, which I thought would be a second hook. Come on, who doesn't like a good round of mocking the writer?

In any case, I do think it's a point worth making repeatedly, largely because as I look around, I don't see the tide turning. Coverage of baseball is so attuned to these small spans of time, to its detriment. If you read everything I write...well, thanks, first of all...I can see how it'd be frustrating. On the other hand, I'm also trying to influence the wider discussion, and as always, I'm writing for an audience that includes both people who read r.s.b in the 1990s and people whose parents met in the 1990s.

OK...off to Camden Yards. (More in Unfiltered later.)
beitvash
5/06
I love your approach for this article, Joe, and I love that you used your own quotes.
illgamesh
5/06
I don't know that it's your "never ending pleas" that people object to so much as the meta-journalism that's making up more and more of your output. Most of what you have to say about the media ("they get hysterical over steroids," "they see patterns where there are none") is so typical of media-criticism about all media (as opposed to just baseball media) as to come off as kind of facile. Everyone knows that mainstream media in this and all other countries simplifies the story to the point of parody and appeals to the lowest common denominator. We don't need you (who, I am the first to admit, is a top-notch baseball writer) doing your best Fire Joe Morgan without the wit impression. It just seems like you're banging your head against a wall that your target (and somewhat captive audience) climbed a long time ago.
irablum
5/06
So, who are the biggest surprises that you don't expect to be maintained? As a Ranger fanatic, there are some surprises that have made the AL West look a bit upside down. Will they be sustained?

1) Millwood managing an ERA of 2.78 and averaging over 7 innings per start with a BABIP of .220

2) Andruw Jones hitting .393 all year. Ok, that won't be sustained.

3) Frank Francisco not allowing an earned run. That probably won't be sustained either, though stranger things have happened. Anyone know the longest scoreless streak by a reliever? Francisco is up to 26.1 innings without an earned run since August 22 of last year. He hasn't allowed an earned run or unearned run (though he has allowed 2 inherited runs to score). Eck's best during 1989 was 22 innings.

TGisriel
5/06
We all have a tendency to accept as true the trends that confirm our own hypthoses, or those we would like to believe, while rejecting as small samples the trends we don't like or don't want to believe.

For example, as an Orioles fan, I am certain that Adam Jones' April is the beginning of the emergence of a star player, while Pie's April is confirmation of the suspicion we had that he will never hit in the big leagues.

Then again there are the trends you can count on because they are a continuation of what has come before. Markakis really is that solid, and the O's starting pitching really is that bad.
Oleoay
5/06
Sample size is always a bit quirky. Players can have a fluke month, or a fluke season. How much of a season is determined by skill as opposed to luck? Even if you took a player who is well below replacement level, they can still have that one hot month where they look like the best player in the league. Conversely, a star player can have a bad month or a bad season and it is uncertain how much of that is due to a decrease in skill or to bad luck. We can make some guesses or hypotheses, based on the data, on how much "might" be attributable to luck, but we can't ever really seem to "know".
ddrezner
5/06
To what extent do Bayesian approaches help you out here?
mglick0718
5/06
Had to laugh as I found this article immediately after reading Joe's: http://sports.yahoo.com/mlb/news;_ylt=AvILgoOWF4XLSFcKwD9Y2loRvLYF?slug=jp-ibanez050609&prov=yhoo&type=lgns