Over and Undervalued Power Hitters
Earlier this week, I spent some time explaining an alternative way of evaluating power: Weighted Extra Bases (wXB). It’s not a statistic you’ll find in the FanGraphs pages as I made it up — although I am not the first to do so — but I like to think it does a better job at capturing the value a player creates through power than Isolated Power or Slugging Percentage.
That’s not to say that wXB is better than ISO or SLG — simply that it’s answering a different question. While ISO and SLG concern themselves with a player’s extra bases and total bases, wXB focuses specifically on the value a player adds through their extra base hits. The formula uses the coefficients for each hit from wOBA, so doubles are given slightly more weight than in ISO, and home runs slightly less.
Using this stat, we’re able to see which players are over- or undervalued by ISO, compared with how much value they have actually contributed through their extra base hits. There are a few notable names on both sides of the coin, so let’s check them out.
Overvalued
In order to determine which players were over- and under-ranked, I calculated out the wXB ratings for all qualified hitters this season. I then ranked them based on their wXB score, and compared this ranking to where they fell when sorted by ISO. So here, I’m highlighting the top five players that were ranked higher when sorted by ISO than when sorted by wXB:
*Note: For the sake of simplicity, I’m presenting wXB in a form that looks close to ISO. This was called wXB/AB in my previous post, and it’s on a slightly lower scale than ISO.
It should be obvious, but all of these hitters are ones that have a large amount of home runs but few doubles or triples this season. Brian McCann and Corey Hart are especially interesting, as they both have Slugging Percentages above .500, putting them both around the top 30 in the league when ranked by Slugging. But when ranked based on the value of their extra base hits, both these players end up around 50th best.
Another interesting case here is Dan Uggla. Uggla has slugged over 30 home runs, yet his Slugging Percentage is a mere .453. Since 1990, there has only been one player to hit 30 home runs or more and finish with a lower Slugging than Uggla: Mark Reynolds in 2010 (.433). Uggla is currently tied for second place with Troy Glaus (2002) and Curtis Granderson (2009), so if his Slugging dips any lower this season, second place will be his alone.
Another fun note: Dave Kingman was the (excuse the pun) “king” of hitting lots of home runs and little else. In 1982 he led the league with 37 home runs, yet posted a .432 Slugging because he only had nine doubles and a .204 average. Home runs are great and all, but they alone don’t make a complete power hitter.
Undervalued
This list is a lot more exciting if we make it a Top 10. But so I don’t overwhelm you with data, I’m going to limit the other columns:
I’ve been aboard the “Ben Zobrist is horribly underrated” train for a long time, and I doubt FanGraphs readers need another reminder. But holy cow, Jeff Francoeur has over 40 doubles this season? I honestly had no idea. People love to snark about Francoeur, but he’s actually been a valuable player this season and he’s already surpassed his previous season-high of 59 extra base hits.* If this power flash is for real — he is 27-years-old — then there’s no reason why he can’t continue to be a valuable offensive piece for the Royals.
*The numbers I have above are a few days old. Frankie has 61 extra base hits so far this season.
And yes, I know that Reyes and Fowler aren’t “power hitters”, but they have still contributed a lot to their teams through their extra base hits. It’s way too easy to be blinded by home runs totals, but if that’s all you look at, you’re missing out on plenty of hidden value with these players.
If interested, here’s a full workbook on wXB. Enjoy!
Piper was the editor-in-chief of DRaysBay and the keeper of the FanGraphs Library.
The impressive thing about Fowler is that his 31XBH have come in 387 ABs and his .868 OPS since adding a leg kick in July hopefully point to his taking a step forward.
Another slap in the face for Vernon Wells.
I’m not sure if I would really use the phrasing “over/undervalued” since ISO is not a big piece of the value puzzle in the first place.
I do, however, appreciate wXB and subscribe to the notion that it is a better value indicator than ISO.
I may have missed this in the first article, but is wXB/AB normalized to mirror tradition ISO numbers? If so, how is this done?
Nah, wXB/AB isn’t quite on the same scale as ISO. I tried messing around with it, but couldn’t make it work…that’d be up for someone else to try.
On average, it looks as though wXB/AB rates are typically 30 pts. below the comparable ISO score. So to get an general gist, just add 30 pts to wXB/AB. It’s not perfect, but it’s a decent estimate.
“…since adding a leg kick in July hopefully point to his taking a step forward.”
*rimshot*
Heh, I heard that sentence in my head as I wrote that
I wouldn’t mind seeing a stat like this listed on FanGraphs. I’ve always thought it makes a lot more sense to develop a form of ISO that uses the more accurate weights from wOBA.
Not to mention Fowler’s abilities to hurdle over infielders
As good as wXB is, it probably slightly overvalues the speedsters who hit lots of doubles and few home runs, because more of their doubles are likely to be of the “hustle” variety instead of going to the wall. As a result, they’re going to score fewer runners from first base. If you wanted to get super-fancy with wXB, you could weight doubles slightly less when the player has only a few home runs.
I would say the opposite about speedsters and doubles. They can wreck havoc in all out of proportion to another 90 feet on a station to station team. Elite Speed , used properly, forces the defense into extra errors, extra pitches, less pitches from the wind up, makes drawing the infield in more risky and in games without the DH can force teams to pull starting pitchers earlier.
Of course if the team with the speeder has crapping pitching what does the one run matter if there are a dozen outs left in the game?
Rather than try and figure all the actual game situations for each player in each extra base hit I would view leavening speedsters and double alone as an “honest cheat” or at least a more accurate picture given the time available to work it. Going the other way opens up a huge can of worms of how to weigh every extra base hit given run differential, pitcher etc. Should a 3b in a 3 run game count the same as 1 run game ? Should home run in Petco be counted as 5 bases? Etc.
Any chance the “overvalued” power hitters might be getting a bit unlucky in how few doubles they hit?
I think perhaps the more likely result is that they’re getting lucky with the number of HR they’re hitting (and HR are still more valuable). Yes, luck swings both ways, but I imagine a fair majority of hits “lost” to poor luck tend to be of the single variety.
This tells me “Overvalued” means you hit more HR than 2B and 3B combined and “Undervalued” means more 2B and 3B than HR. I hope that is not what it is suggesting. I know Slow and I am sure that is not the basis of this stat. Someone please correct me.
Hm, sorry, I don’t mean to be misleading. What it’s basically saying is that if a player hits lots of homers with few doubles or other XBHs, they’re likely overrated by ISO/SLG. And the inverse…if they hit lots of doubles and few homers, they’re going to come across as undervalued.
Pretty intuitive stuff, really. And it only really makes a big difference for players with extreme profiles.
Gotcha. Thanks for clarifying that. That makes much more sense. I wonder if there is a way to separate the doubles in the gap from doubles blooped and legged out. Would help the theory a bit.
I’m confused as to why Zobrist comes in second behind Headley? And why Gordon & Frenchy are so far below given their numbers are comparable to Zorilla’s? Is it PA’s? (Meaning, has Headley been more productive as a “power” hitter in significantly fewer PA’s?)
I compared rankings, so was #31 best in the majors in wXB but only #48th best in ISO…a difference of 17 spots. Headley came in at #101 in wXB but only #122 in ISO, a larger difference.
Imperfect way to rank them? Yeah, probably…my guess is hitters are grouped closer together toward the middle of the pack. But it seemed a simple way to get at the general idea.
And yeah, Frenchie and Gordon are both very similar to Zobrist, but they have a handful more at bats and haven’t put up quite as good power numbers. The difference between their rankings was each 11 spots.
The only other thing I see with your rankings is just a thought based on McCann.
Seems like the main reason he ranks so high in SLG is that his BA has hovered around or above .300. If his BA dropped to the levels of most of these other guys then I don’t think he’d be ranked so high by SLG. Just my impression, but this jives pretty well with the typical view of McCann, that he’s not really a ‘power hitter’ so much as he is a good hitter with power.
Yeah, but half the guys on the undervalued list have BAs higher than McCann’s this season.
Another thing to note is that McCann is one of the slowest players in the majors. Can easily see him turning 5-10 doubles into singles over the course of a season
good stuff!
Hmmm. I hate to say it, but I think that all this article explains is how two players with inverse amounts of 2B’s and HR’s can have a similar slugging %. Being a statistics graduate student, I am having a hard time seeing the meat and potatoes of this article, and I think this article just made me more comfortable with SLG %.
There are always players who hit lots of 2B’s and 3B’s, but give us no HR’s, and this can be caused by numerous factors: a) park environment, b) player speed, c) willingness to risk an out on gaining an extra base. Then there are the players that can rack up the HR’s, but fail to give us the 2B’s and 3B’s. This is often caused by a few factors: a) they are ‘Dunn’ or ‘McCann’ slow (the extra ‘n’ may weigh them down), b) they are just really strong. Balls these players hit in the air tend to carry.
There is a problem with this whole story though, if I am interpreting it correctly; a 2B from Reyes is about as likely to drive in a man on 2nd as a 1B is from Jose Molina. SLG % does a great job of giving us a “power profile” on a very basic level, but I propose a statistic to digest concurrently to get the whole picture. In order to determine ‘value’ of power hitters, I suggest looking at SLG % and ‘RBI success wRISP’. As a Jays fan, I hate to say it, but this would show that a Wells 2B is more valuable than a Reyes 2B when looking at ‘power’.
I’m not sure Jose Reyes has ever been undervalued as a major leaguer. I do believe he’s already penciled in as an early favorite for the 2012-14 MVP awards in both leagues.