Brandon Allen, Strikeouts, and Definitions
Sometimes baseball analysis is similar to semantics. Define Brandon Allen one way, and he’s got a bright future despite some early struggles. Define the 25-year-old first baseman from a different direction, and the odds are incredibly stacked against him.
Let’s de-emphasize Allen’s 209 plate appearances so far. Even if those plate appearances had all happened in one season — which they haven’t — we’d only barely be looking at a reliable sample for his in-season strikeout rate, and we’d know next to nothing about his power.
It’s obvious that his strikeout rate is the big ‘problem.’ So far in the major leagues, he’s had a 34.9% strikeout rate, which simply is not enough contact. But Allen’s minor league strikeout rate was only 23%. What happens if we define Allen as a minor leaguer with a 23% strikeout rate?
He looks a lot better.
Minor leaguers with at least 600 MiLB PAs and a 23-25% strikeout rate averaged 27.9% strikeout rates in the major leagues. A 27-28% strikeout rate in the major leagues is not great either, but Josh Willingham, Carlos Pena, Mike Stanton, Austin Jackson, Kelly Johnson, Mark Reynolds, Miguel Olivo and Drew Stubbs have all managed to provide positive value to their teams while showing similar rates this year. It’s no death knell, and a player with enough power and patience can make it work. Since Allen had a double-digit walk rate in both the majors and minors, and prodigious power in the minor leagues, this definition is friendly to Allen, especially since we included him in a bucket with players with worse minor league strikeout rates.
But say we do combine Allen’s partial seasons. And we do find a reliable sample for his strikeout rate therein. And we do define Brandon Allen as a slugger with a strikeout rate over 30%. What happens to his prognosis then?
He looks a lot worse.
Sort the career leaderboard by strikeout percentage with a minimum of 1000 PAs in order to find players that put up 30+% strikeout rates over their careers. Right away, you’ll notice that the sample is small: 29. Now take out the pitchers, and we’re down to 14. Take out the catchers. We’re down to 12. That’s right, over the history of baseball, only 12 non-catchers have managed to accrue 1000 PAs with a strikeout rate over 30%. It’s a small group. It’s hard to stick around with a strikeout rate that high.
Obviously, not all minor leaguers with 23-25% strikeout rates are created equal. Nelson Cruz had a 23% strikeout rate in the minors and has shown a 22.4% strikeout rate in the majors. Chris Davis had a 24.5% strikeout rate in the minors, in some of the same parks, and he’s closing in 1000 PAs with 31.5% strikeout rate in the majors.
But if you define Brandon Allen by his minor league brethren, he’s got a fighter’s chance. If you choose to define him by his major league work so far, he’s got quite the uphill battle. It’s a question of definition as much as it is a question of contact.
Thanks to Jeff Zimmerman for researching some of the numbers in this piece.
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
Since 200 isn’t 1000, how about looking at people who had 30%+ K-rates over their first 200 PA?
Here’s the list of firstbasemen, +30% K-rate, 200 AB min.
Jake Stahl 50.0%
Hal O’Hagan 50.0%
Tim Jordan 40.0%
Brad Eldred 36.5%
Brandon Allen 34.9%
J.R. Phillips 33.0%
Russell Branyan 32.9%
Chris Davis 31.7%
Mike Hessman 31.6%
Andy Tracy 31.5%
Rusty Ryal 30.3%
Jeff Larish 30.1%
DH’s:
Jason Botts 33.1%
Jack Cust 31.7%
Alejandro Sanchez 30.7%
Andy Tomberlin 30.5%
Jeff Larish 30.1%
OF’s:
Myron Allen 42.9%
Jim Fuller 38.1%
Rooney Sweeney 36.4%
Jason Dubois 35.7%
Chris Latham 35.4%
Kyle Blanks 35.3%
Melvin Nieves 34.7%
Dave Nicholson 34.5%
Billy Ashley 34.3%
Bill Wise 33.3%
Jason Botts 33.1%
Phil Hiatt 32.2%
Bo Jackson 32.0%
Justin Maxwell 31.9%
Jack Cust 31.7%
Harry Wheeler 31.7%
Cliff Cook 31.3%
Rob Deer 31.2%
Chad Hermansen 31.1%
Alejandro Sanchez 30.7%
Rolando Roomes 30.5%
Andy Tomberlin 30.5%
Wily Mo Pena 30.4%
Rusty Ryal 30.3%
Hensley Meulens 30.1%
Nixey Callahan 30.0%
Edgard Clemente 30.0%
Kit Pellow 30.0%
The point of using 1000 PAs, though, was to talk about his chances of making it to 1000 PAs. Obviously a ton of these guys flamed out early.
Right, the problem is there could be hundreds of guys who struck out 30% of the time in their first 200 PA, and then didn’t keep that rate up for the next 800.
Those are all career #’s though… NOT just 1st 200 PA’s.
But looking at those names, Allen is not gonna last long if he can’t get the K-rate down.
Eno, I totally agree with you!
I gotta admit, I was surprised (in a bad way) when I looked at these lists of names.
Wow guys, thanks for doing the leg work on that. Is that from b-ref? I can’t make it do things like that: gotta take some lessons from you. So could we make a “success” list by looking for people with 30% k-rates in the first 200AB and then going on to post 2 or more seasons of 3 or more WAR? I was wondering if many other folks had had early bad stretches and had gone on to do better, but the career list sure looks grim. I’m an A’s fan but I live in Reno and I’ve sure enjoyed watching Allen poke pitch after pitch out of the park! I hope he can be the exception to this ugly trend, but I’m sure not holding my breath now.
@A’s fan:
No… got the stats right here at Fangraphs!
At the top of this page… mouse over “Leaders” box. Then in the drop down box that appears, click “career” in the “batters” section.
A page will open to sortable stats.
“Advanced” stats page has K% as the second column. (mine opens to this page)
On this page you can set minimum PAs, filter by position, etc… then click on a stat column to sort by that column. Click it again to reverse sort it.
The problem with the stats I posted here, is that they are all career stats. So, these guys all ended their careers with +30% Krates.
A much more interesting list would be all guys that had a +30% Krate in their first 200 PAs. Then, look at how many improved it, by how much, and then how those players performed.
I do NOT know how to find Krates for all players’ first 200PAs.
Care to post the list of those 12 players?
Sure:
Jake Stahl, 1B
Tim Jordan, 1B
Melvin Nieves, OF
Dave Nicholson, OF
(Kelly Shoppach, C)
Russell Branyan, 1B
Mark Reynolds, 1b/3b
Bo Jackson, OF
Jack Cust, DH
Harry Wheeler, OF
Rob Deer, OF
(Tom Egan, C)
Wily MO Pena, OF
Nixey Callahan, OF
Does it make sense to compare K% to league average?
For example Cust K’d 32% of the time in a league that K’d about 18%-19% of the time. Bobby Bonds K’d 22% of the time in a league with about a 12% K-rate, and Babe Ruth had a 12% or so K-rate in a league with about a 7% K-rate.
I suspect the upper limit would still be around 28% in today’s game, but you’d get a much longer list of success stories that way.
Jack Cust is probably a good comparison and discussion piece. When his ISO is ~.245, he’s league average (on average). When it’s below .~200, he below replacement level.
Seems weird that the team that DFA’d Cust would trade for Allen.
So, it would seem, the question with Allen is how much power does he have?
The A’s are also the main reason why Cust had any significant major league career to begin with; he was in several different teams’ farms and raking the minors and getting cups of tea in the MLB before being kept up with the A’s in 2007.
So, not really.
This is just a wrong summary of Cust’s abilities and skills as a hitter. (If you’re counting the fact that he can’t field and is necessarily a DH, then I more or less agree. But the question begin raised here isn’t about defensive value.)
In seasons in which Cust has an ISO of .245ish, he’s been a wRC+ 130-140. That’s one of the best hitters in the league. His career ISO is 197 and his career wRC+ is 120. This year, his ISO is .116 and his wRC+ is 96. This is the only year that Cust has hit like a replacement level DH since he started with Oakland in 2007.
I wasn’t clear. I meant “average” in terms of WAR.
That’s misleading because he was an ~average (2.5, 1.9, 2.3) DH during his 3 “full” seasons. But an “average” DH is a good hitter.
If Allen duplicates that while playing decent 1B, it’ll be great for OAK. But we already knew that. We just don;t know if he can.
If you’re calling Allen Jack Cust with the ability to play a little defense (because Allen is absolutely a more useful defensive player), you’ve answered your own question about why the A’s would trade for him. That’s a darn good player to get for a reliever.
Absolutely. Allen has defensive value. If he hits like Jack Cust, he’s a damn good player.
(cust in his prime, of course)
He made some highlight plays in LF last year, but maybe they were highlights due to his size, speed, or lack of instincts?
It was more of a commentary on why they DFA’d Cust.
http://www.fangraphs.com/blogs/index.php/cust-cut-loose-by-oakland/
But, Allen and Cust seem like comparable hitters. If they hit for good/great power, they’re valuable … if they don’t, they’re not (or just average).
I don’t consider Allen’s D at 1B to be so great that it makes up for a league average or below bat. Having an “average 1B bat” and having a league average bat are two different issues. Ryan Howard has a “average bat” for 1B’s, but is better than a league average hitter.
Really, all I said was that Allen will need to hit for power to be valuable as a 1B. Isn’t that accurate?
The A’s DFA’d a guy that put up .371, .342, .371 wOBA’s (~%2M/y contract) and replaced him with Matsui (.326 wOBA, .356 in ’10) and paid Godzilla twice as much ($4.25M).
Daric Barton is having a horrible year, and acquiring Allen is not a risk … but now OAK seems to be in the same spot ARZ was in starting 2011 … 3 guys at 1B that project to put up similar wOBAs.
The tough thing to me is that OAK needs Allen to put up better than “career” numbers of .191 ISO and .306 BABIP to be valuable. In his partial seasons, he’s one extreme or the other (either very high ISO or very high BABIP).
Is Allen a .200+ ISO & .300+ BABIP guy over an entire season? If so, he may be valuable (1-1.5 WAR) to the A’s (provided average D)..
I think trying to project someone’s career based on their first 200 ABs is not a great idea. So I agree with de-emphasizing that. The problem is that I dont think that is stayed de-emphasized.
That is unless the the point was to say that a 34% K rate is too high to sustain. I think that even Brandon Allen knows that.
The above comment is the end of this entire discussion. I only regret that I have only one thumbs up to give it. I would like to add, though, that the miniscule 200+ plate appearance MLB sample size is over THREE (!!!) seasons.
For me the key is the difference in the AAA and MLB strikeout rates on multiple, albeit shortlived, opportunities. Think Chris Davis has figured it out based on his ridiculous numbers at AAA this year? Just watch a game. There’s just way too much swing and miss there. It’s hard to watch. Allen is Chris Davis, except that he will actually draw a lot of walks. But he’ll struggle to hit .200, which means really he’s Mark Reynolds who can’t play a really terrible 3B. But that’s just in the near term. Guys with that much swing and miss are obviously easy to pitch to. If he doesn’t correct the flaws, pitchers will exploit them over and over, like a bullfighter methodically taking down a wounded animal. And that will be that.
As much as Jack Cust was a three true outcomes guy, his hit tool was probably two grades better than Davis/Allen/Reynolds.
As a linguist who studies semantics, I can assure you that baseball analysis is nothing like semantics.
As a person with a real job and degree, I can assure you that your post accomplished nothing other than calling attention to your exaggerated sense of self and probable small penis.
Phew! I got this as an update in my inbox today and thought it was directed at me haha. Still, not so nice to our fellow reader.
I did take a few lingustics classes at Stanford (professor Rickford holla), so I know this was a stretch.
But! How do you define a “Brandon Allen”?
The D Backs coaches and scouts that Kevin Towers brought on determined that Allen has a hole in his swing and he was not able to take the coaching to close it. He’s 25. Tough to make a drastic change at that point.
A 4% jump from last year to this years K Rate in AAA is a pretty big indication that he wasn’t any closer to closing that hole.
I still think they should have just given him the job on opening day instead of trade for Miranda. At least we would have known for sure by now. But then again, if he had been given 250 PA’s over the first half of the season, it might have exposed that hole once and for all and Beane wouldn’t have even given up Ziegler for him. Maybe the best way to keep his trade value afloat was to leave him in Reno to put up inflated numbers and keep the dream alive.