It wouldn't be fair of me to invoke T.S. Eliot by proclaiming April to be the cruelest month for Major League Baseball fans. At the least, I figure the whole "Baseball's back!" thing is worth some points.
We can say this about April, though: On the trustworthiness scale, it's somewhere in the neighborhood of a new puppy on an expensive rug. You have to be wary of accepting what happens in April at face value.
...He said in his best Captain Obvious voice.
Yeah, I know. It's really no big secret that April isn't MLB's most telling proving ground. It's Small Sample Size Month, where wacky things happen before everyone is eventually reminded that teams and players need more time to find themselves than April can give them.
Still, it can be good to occasionally remind ourselves what we're dealing with by gaining some perspective. As always, we can get a dose of that by looking to the past.
Regarding teams as a whole, it is possible for clubs to establish who they are as quickly as April. But in looking back at the last 10 years, we can easily see that it's less than a given that they will.
I went and looked at how many clubs have gone from first place in April to first place at the end of the year, clubs that have gone from last place in April to last place at the end of the year, and at how predictive April records have been in general.
Last-place teams have been the most predictable...but only to a degree.
With help from Baseball-Reference.com, here's a look at the teams that have either been tied for or alone in last place on April 30 and ended up in last place at year's end:
|From Worst in April to Worst in the End, 2004-2013|
|Year||Worst-to-Worst Opportunities||Last-Place Ties||Went Worst-to-Worst|
|2004||7||NL West||TOR, KCR, SEA, MON, ARI|
|2005||6||None||TBDR, KCR, PIT, COL|
|2006||6||None||TBDR, KCR, SEA|
|2007||8||NL Central (3-Way)||KCR, TEX|
|2008||8||NL Central, NL West||WAS, PIT, SDP|
|2009||8||AL West, NL West||OAK, WAS, ARI|
|2010||8||AL Central, AL West||BAL, KCR, SEA|
|2011||7||AL Central||MIN, SEA, HOU, SDP|
|2012||6||None||BOS, MIN, MIA|
|2013||6||None||TOR, CHW, HOU, MIA, CHC|
There have been four seasons out of 10 in which the majority of teams that had a shot to go worst-to-worst have done so. With 35 of 70 going worst-to-worst, the overall succ...er...failure rate is 50 percent.
Keep this in mind when we get to the end of the month Wednesday. There are going to be teams either in last place or tied for last place. Fans of those teams will be wondering whether to be optimistic or pessimistic. The last decade says to be neither. It's a toss-up.
Which, I suppose, is preferable to what the last decade has to say about teams that have ended April either in or tied for first place:
|From First in April to First in the End, 2004-2013|
|Year||First-to-First Opportunities||First-Place Ties||Went First-to-First|
|2004||8||NL Central, NL West||MIN, LAD|
|2005||6||None||CHW, LAA. STL|
|2006||8||AL East, AL Central||NYY, NYM, STL|
|2007||6||None||BOS, CLE, LAA, ARI|
|2008||7||AL Central||CHW, LAA, CHC|
|2009||7||AL East||STL, LAD|
|2010||7||AL West||TBR, MIN|
|2011||6||None||NYY, TEX, PHI|
|2013||6||None||BOS, DET, ATL, STL|
There have been only two seasons in which the majority of the teams that had a shot to go first-to-first have done so. And with 27 out of 67 possible teams actually succeeding in going first-to-first, the success rate in the past decade is just a shade over 40 percent.
So yeah. As much as I hate to be the bearer of bad news, don't get too excited if you find your favorite team sitting atop the standings come the end of the month. Per the last decade, being in first place at the end of April is not a guarantee of a first-place finish.
And for the record, no, you don't have to look at just the top and bottom of things to find a lack of predictability in what clubs do in April.
In looking back at all April records from the past decade, I found that only 122 of 300 teams had a final winning percentage within 50 percentage points of their April winning percentage. That's 41 percent.
None of this should be especially surprising. Things happen in April, but then more things happen after April. By way of call-ups, send-downs, trades and good, old-fashioned injuries, players come and go. Teams have their performances impacted as a result.
Then there's the general unpredictability of baseball players as a species. Not unlike their teams, they're typically not themselves in April either. They need more than a month to stabilize their performances.
A lot more than a month in the case of hitters. According to a 2011 FanGraphs article by Russell Carleton, it takes around 500 plate appearances for on-base and slugging percentages to become stable. Knowing that, we don't really have an idea what hitters are like in a given season until around August.
I don't have the mathematical know-how to disprove a notion like that, but we can illustrate the point by asking a simple question: How many of the top April hitters every year actually finish the year as top hitters?
In the last decade, not many.
Using Weighted Runs Created Plus (wRC+) as a measuring stick—you can read more about it in my recent Sabermetrics for Dummies article, but the short version is that it's a park- and league-adjusted metric that measures a player's offensive value in runs above average—I looked at how many of the top 10 qualified hitters in the last 10 Aprils have finished the year as top-10 qualified hitters.
With an assist from FanGraphs, we behold the following:
|From April Top-10 wRC+ to Final Top-10 wRC+, 2004-2013|
|Year||Top-10 to Top-10 Opportunities||Tied for 10th||Top-10 to Top-10|
|2006||11||2||Pujols, Thome, Hafner|
|2008||10||None||Pujols, Quentin, Jones|
|2009||10||None||Gonzalez, Pujols, Youkilis|
|2010||10||None||Pujols, Cabrera, Konerko|
|2011||11||2||Votto, Bautista, Berkman, Holliday, Kemp, Cabrera, Fielder, Braun|
|2012||10||None||Posey, Encarnacion, Willingham|
|2013||10||None||Davis, Cabrera, Choo|
Outside of when everyone and Uncle Everyone went from the top 10 in April to the top 10 at the end of the year in 2011, it's been rare that the year-end wRC+ leaderboard has closely resembled the April wRC+ leaderboard.
Which makes sense, as you don't need to narrow it down to wRC+ to know that April is a time of unusual suspects putting up huge offensive numbers. This year's happenings are a good example, as Matt Joyce is leading in OBP and Charlie Blackmon is leading in average, slugging and OPS.
The unusual suspects that pop up in April tend to go away eventually. Sometimes it's a case of pitchers figuring them out. Other times, injuries do the trick. Other times still, it's simply luck running out. Alas, the baseball gods can be cruel once they decide a hitter has received enough of their charity.
Regarding pitchers, Carleton found in a 2013 Baseball Prospectus article that strikeout rates actually stabilize pretty quickly at around 70 batters faced. Everything else takes more time, and there's nothing there for when results-based stats like ERA tend to stabilize.
But one look at the past decade tells us it's certainly not April.
Courtesy of FanGraphs, we see that the April ERA leaderboards from the past decade have been about as predictive as the April wRC+ leaderboards:
|From April Top-10 ERA to Final Top-10 ERA , 2004-2013|
|Year||Top-10 to Top-10 Opportunities||Tied for 10th||Top-10 to Top-10|
|2005||10||None||Willis, Peavy, Clemens|
|2006||10||None||Webb, Arroyo, Carpenter|
|2007||11||2||Penny, Haren, Peavy|
|2008||10||None||Sheets, Lee, Peavy, Lincecum|
|2009||10||None||Jurrjens, Cain, Greinke|
|2011||10||None||Shields, Weaver, Halladay|
|2012||10||None||Cueto, Zimmermann, Lohse|
|2013||10||None||Sanchez, Kershaw, Iwakuma, Bumgarner, Harvey|
That's about a 30 percent wire-to-wire success rate, and the picture only marginally improves if we look at statistics less influenced by luck than ERA.
There's FIP, for example. Short for Fielding Independent Pitching, it's an ERA estimator that evaluates pitchers based on their strikeouts, walks, hit-by-pitch numbers and home runs, effectively cutting bad defense and luck out of the equation.
In the last 10 years, only 38 of 100 possible pitchers have gone from the April FIP leaderboard to the year-end FIP leaderboard.
Then there's SIERA. Short for Skill-Interactive ERA, it's a more complex ERA estimator that tries to make something of balls in play rather than ignore them. It's also park-adjusted, which is ideal. But even SIERA is only so capable of brightening the picture. In the last 10 years, 44 of 102 possible pitchers have gone from the April SIERA leaderboard to the year-end SIERA leaderboard.
Pitchers are derailed by the same forces that affect hitters: the league adjusting to them, injuries and, of course, the baseball gods. Though metrics like FIP and SIERA are designed to ignore the influence of any deities, they still have the power to disrupt those metrics as they see fit.
Alright, I'd say we've gone far enough. We came in with a built-in awareness of how treacherous trusting April baseball can be, and we've just illustrated the point by considering the lack of predictive power in April records and how April leaderboards aren't made to last.
Are we going to keep taking April seriously anyway?
Well, yeah. Of course.
It may be lacking in its power to project ultimate results, but what happens in April does count. And though the odds tell you not to risk it, sometimes you do find some solid excuses to hop on the bandwagons of certain teams and certain players.
We should all be wary of trusting April. But if it's a choice between trusting April and ignoring April, I'll take Door No. 1 any day.
If you want to talk baseball, hit me up on Twitter.
Like the new article format? Send us feedback!