Four Factors: Brennan Boesch
It seems like there was a warm response to my four factors for evaluating hitting that I presented in this post on Martin Prado last Friday. As such, I’d like to use it to take a look at a few more players. One player about whom I think these stats tell a particularly interesting story is Tigers rookie Brennan Boesch.
As a reminder, the four factors of hitting are BB%, K%, POW (XB/H), and BABIP. That is, walk rate, strikeout rate, power (as measured by extra bases per hit), and success on balls in play. I feel that these four statistics sum up extremely well four facets of hitting under which the batter has control.
Let’s take a look at how Brennan Boesch fares under these four measures.
Boesch, to date, has excelled in both power and success on balls in play, performing at a level nearly 130% of the league average in both factors. He’s slightly above average in strikeout rate and a bit below average in terms of walk rate. There’s no doubt that he has been excellent so far this season; his .426 wOBA is right between Joey Votto and Kevin Youkilis on the leaderboard – certainly impressive company for a rookie.
Of course, when a young player with no MLB track record posts these kinds of numbers over any significant period of time, we have to wonder if it’s sustainable. Simply due to the nature of regression, even if a player has ridiculous minor league numbers (which Boesch doesn’t), we would never expect a .426 wOBA to stay that high. What can we reasonably expect out of this extremely hot hitter as the season winds down?
Unfortunately, ZiPS doesn’t have a projection for him in the system. CHONE, however, in its July update, projected Boesch for a .259/.300/.438 line, worth about -7 runs per 150 games. That’s a marginally over replacement level player with average defense in the corners.
To see why CHONE is so down on Boesch, we need to look no further than the four factors. Take a look at how quickly each of the statistics involved here stabilize:
K%: 150 PA
BB%: 200 PA
ISO (presumably, POW would stabilize at a similar time): 550 PA
AVG: >650 PA
Boesch is only 267 plate appearances into his MLB career. For that reason, we can be relatively confident on his below-average walking ability and his above-average ability to avoid the strikeout. However, the two aspects of his game that have him at the top of the MLB leaderboards – power and balls in play – just simply don’t show themselves this early in a player’s career. Boesch showed solid power in AA (0.84 POW) but hasn’t ever showed the kind of BIP skills that would necessarily lead to a .384 or even a .320 BABIP in the majors.
Right now, it’s simply too early for us to conclude that Boesch is so much better than his minor league performance would suggest to even say that he projects as an average hitter going forward. He has certainly impressed in his time on the field in 2010, but his power and success on balls in play are likely to drop. Given the rest of his skills, he simply can’t maintain the elite hitting if this drop occurs. It’s certainly not impossible that this is just a breakout year for Boesch, but the Tigers shouldn’t act as if Boesch is another Miguel Cabrera in their lineup as they plan for the second half.
28
This is a great idea for getting a snapshot of players. I went back and reread your column on Prado and then tried applying the stat to several others who are sell-high and buy-low candidates. Please continue this series. It makes good sense and good reading.
Why POW over ISO?
Is the K% the one that Fangraphs carries or the better K/PA?
why does fangraphs hate Boesch so much?!!!
This is the rare article here that has used sound logic and maths against Boesch. There have been some really questionable things written, but this article is well-reasoned and logically sound.
because of teh EAST COAST BIAS!
and Seattle bias.
and sabermetric bias.
And the “they don’t think exactly like *I* do” bias.
These sites try to install made up, non-traditional, computer geek silliness to baseball. The game is played between the lines…”you throw the ball, you hit the ball, you catch the ball,” it’s that simple.
Brennan Boesch isn’t perfect, no such ball player exists.
When you watch this kid you see a young ballplayer trying to live out his dream. He hustles out every ground ball, every fly ball, every base hit.
Baseball is a game to be enjoyed, to be savored for it’s singular beauty. I don’t care if Brennan K’s 140 times this year, because that’s what free swinging power hitters do. They also hit HRs, and drive in runs, and when Brennan has 25+ HRs and over 90+ RBIs and has hit .330, all the rest of the numbers, including scoring runs, and slugging %, all legitimate baseball numbers, is all that I, as a baseball fan, care about.
Fangraphs doesn’t hate Brennan Boesch. They’re just playing the odds; if somebody hasn’t shown the skill or potential in the minors, rarely do they bust out in the big leagues. If you’ve never been good at organic or inorganic chemistry, for example, its highly unlikely that you will get a PhD in Chemistry in the future. Same way in baseball. What percent (not number; rate stats are more demostrative) of major league stars never hit very well in the minors?
Oh, and I’m a Tigers fan.
It may be presumptuous to claim that there are “no hatters here”. Fangraphs could be quite popular among milliners.
Well played, sir.
From your piece it would seem that when statistic stabilizes, it reaches true talent level. Actually, it means that that statistic needs to be regressed 50% towards the league average. So, if somebody hit for 30% better power than league average over 550 PA, you should regress him to 15% better for future projections.
This. No offense to Jack, as I love the articles, but the saberblogging community is going crazy with these stabilizing PA numbers. It means they have reached a certain threshold of significance, but it doesn’t mean that they are true talent numbers. A 106 K%+ regressed to 103 is not what I’d describe as “above-average ability to avoid the strikeout.” Particularly if his minor league numbers were less than stellar.
This makes a lot of sense! Boesch is over performing but he is hitting between Guillen and Cabrera. Of course he isn’t going to keep hitting like this, he isn’t Pujols or Cabrera. But he is actually hitting for less power (the raw POW), according to your stats, then he did in the minors. So he may not be in line for an across the board regression.
The article is well reasoned but the conclusion sets up the same straw man as the rest of the articles written on Fangraphs about Boesch. That conclusion is that he isn’t an MVP-type hitter so therefore won’t perform like an MVP-type hitter. The Fangraphs community certainly knows this and one would hope Tiger brass does too. The question no one seems to be able to answer is what he WILL hit like. One article on this site that was based exclusively on numbers suggested Jeff Francouer. I’m skeptical. He sure as hell isn’t Miguel Cabrera, but continually forwarding the argument that he’ll regress to his A-level numbers I still think is kind of lazy.
I’m not sure if it’s been said already but Boesch reminds me a lot of Matt Holliday. If you look at Holliday’s minor league numbers, they don’t project to MVP caliber numbers either. Granted much of Holliday’s major league career numbers were inflated by Coors Field. Regardless, he never had a full season in the minors with a batting average above .280 or SLG above .450. His career minor triple slash was .276/.353/.427. Also he showed about a 9.8% BB rate and a 16.7% K rate in the minors, neither of which are far off from his major league career rates. Boesch’s rates are not as stable as Holliday’s but the comparision hopefully does illustrate that it’s possible for players to break out in the majors beyond their minor league track record.
So which is it: (1) be careful about drawing conclusions from small samples; or (2) confidently draw conclusions from small sample sizes, but draw conclusions that contradict the actual small sample?
Boesch has a gorgeous swing and great physical tools. He also performed much better last year than any year before, and much better this year than last. I respect small sample sizes and high BABIP enough to agree that he likely will not sustain his current level of production, but there is better evidence that over time he will be a very good player than a below average one.
His BABIP will obviously take a dive, but the power is unlikely to shrink to Ichiro-levels as CHONE predicts. His last full season in the minors was AA, in which he ISO’ed .235. His BB rate has improved noticeably since then (and stabilized at an adequate, if not really good level), but the power has been there for well over 550 PA. You noted all of this, but I’m wondering why you choose to assume that Boesch is suddenly going to be a below average hitter when given this data. The only thing likely to regress significantly is his BABIP. A .280/.330/.510 line going forward isn’t really that unreasonable given his track record of power hitting. That is definitely better than average, even in LF.
He’s no Cabrera and never will be, but it’s a bit reactionary to say that Boesch is a scrub, when the data can only definitively project a regression in his BABIP.
As a Tigers fan I have no problem with this article because of the way it was written, particularly the final quote “It’s certainly not impossible that this is just a breakout year for Boesch, but the Tigers shouldn’t act as if Boesch is another Miguel Cabrera in their lineup as they plan for the second half.”
I think that’s what has bothered most Tigers fans about all the Boesch articles is that the writers don’t say things like “it’s certainly not impossible” they basically say “He will collapse”, so it does make you a little more defensive as a fan when you say your player or team is going to suck and not even give them a chance of doing well. Atleast this one offers a glimmer of hope and isn’t an all out end of the world article about him.
And for the record I do think he’s going to fall off but I still think he’ll be productive and useful, which is more than what alot of “experts” have been predicting.
Guy’s a ballplayer. Wish I had him on my club.
I posted this in the What the Tigers Should Do thread re: Boesch –
As for the Boesch analysis, it’s been beaten to death a bit, but it illustrates how stats-only analysis can be a bit misleading or at least lead to an obvious conclusion. If I didn’t see Boesch every single day, I’m sure I would come to the same conclusions, but watching him play makes it hard to believe he will fall off the table & hit sub-.250 the rest of way.
I do believe there will be some regression from a .340 batting average, but not 90 points worth. And the power is legitimate. And though his composite minor league numbers aren’t overwhelming, there was clearly a development step from 2008 to 2009 that CHONE figures *seem* to wholly discount based on the analysis.
Of course it’s easier and safer to project that someone riding high without the minor league pedigree to do so will fade because if you say he will remain very productive and he doesn’t, the naysayers come back with, “well his minor league number said this fade was coming, you idiot!” so I understand going the safe route, but it isn’t without flaws, either.