Chicken or the egg?

by Home Run Tony Cogan on Jul 25, 2009 11:17 PM EDT

I was going to post this elsewhere, but in my small mind I thought it was important enough to generate separate discussion. In reading the "Open Letter to Dayton Moore", there was a discussion about OBP vs. BA, and it got me to thinking-what comes first?

For example, is a good OBP a residual effect of being a good hitter, and therefore secondary to the importance to the actual ability to hit the ball, or is OBP a skill in and of itself? I am assuming that pitch recognition (the "batter's eye", Dayton, if you're reading) is a skill, but that is also probably tied up in bat speed, reaction time, etc. that allows a batter to let the ball travel deeper and therefore get a good read on the pitch before triggering.

Perhaps there are two-types of "walkers": the good hitters who never see anything (Barry Bonds) and the great pitch-recognition types (I'm guessing somebody like Youkilis might fall into this category). Obviously, Bonds also has great pitch-recognition skills as well.

What confounds me a little bit, however are the Adam Dunn/Jack Cust types: those who strike out a ton and would therefore seem to have poor pitch-recognition skills, but also have a good ability to get on base.

I think where the disconnect may come from for the BA vs. OBP crowds are due to guys like these. If a guy is relatively lousy at making contact, how can he hit so many XBH? Why aren't pitchers going right at these guys more often, because most likely they are going to strike out or get themselves out? Do these guys just not swing unless they can absolutely square the ball up, and the walks and strikeouts are just a residual effect of that particular approach? I wonder if they actually think about this before going to bat, and if they change their approach for specific situations, like RISP w/ less than 2 outs?

A corollary to this (kind of a tangent and I'm winging it so bear with me) is if it would make sense to look at individual hitters like a "market", where the expectancy of a particular result could be compared with the expectancy of another particular result and a decision then made on what the best approach would be for that batter.

For example, take Albert Pujols. His career line is .334/427/629. Based on this line, is there a way to calculate the expectancy of him creating a run every time he comes up? Let's say that it's something like .25 runs created for every plate appearance. Is there a way, based on game situaton, to determine quantitatively what the best approach is (walk him, pitch carefully, go right after him)? Intuitively, if there's two outs, runners on 2nd and 3rd, and he comes up, you probably walk him. Two outs, nobody on, you go right after him. But what about runner on 2nd and 1 out? Is there an appropriate way according to numbers to determine the correct approach? Is any team using such an approach?

Then the last question would be, if all of this information were available, do pitchers over or under (or appropriately) "value" his at-bats from an approach perspective? Could this be done for every batter/game situation? Maybe I'm over-thinking it, but if anybody has any insight into this I would appreciate hearing it.

Tweet Comment 12 comments | 0 recs |

Do you like this story?

Comments

Let's of stuff in there

Not sure where it’s going. If it would helop, here’s some ideas about one part that may or may not be worth reading:

For example, take Albert Pujols. His career line is .334/427/629. Based on this line, is there a way to calculate the expectancy of him creating a run every time he comes up?

What you want here are his “linear weights” or RC per plate appearance. Now, a career line isn’t where I wouid start, especially for a guy around as long as Albert, but whatever. You do a “projection” however you want — a weighted average of his recent performance with the appropriate amount of regression. at least. The “three slash” line is a decent description, but what you really need is his wOBA or some other linear-weights based metric. Once you get that wOBA number, you can calculate something. I won’t do a projection at the moment. For the sake of ease, let’s take this Stat Corner wOBA (since it gives other handy info right there without me having to open Excel). Again, let’s pretend it’s a projection, not the current stats.

Here’s his hitter’s page.

the (basic) wOBA formula for Run Value per Plate Appearance is (wOBA-lgwOBA)/1.15

But what that’s giving you isn’t his RC/plate appearance, but his runs created above average per plate appearance. So You would simply take the league runs/league plate appearances for the run environment and add that number to the runs created above average to get his expected runs/plate appearance.

Having said all that, I don’t think that’s enough information to tell you how to pitch him in most situations. Like most people, not only do I hate teh intentional walk, but I think it’s one of the most overused strategies there is. The Book has a long and detailed chapter on the intentional walk… suffice it to say, it’s rarely worth it.

But you’re asking for more than that. I imagine it would take a lot of scouting, or, these days, a combination of the Hitter’s pitch f/x info and Hit f/x info. I don’t know if any teams are doing much of the latter, but I’d find it surprising if some weren’t (e.g., Tampa Bay).

So… I gave the easiest, least relevant part… anyone else?

I'm not a sabermetrician, but I do play one at Driveline Mechanics.

Can't get enough of me? Check out my Twitter feed.

by Matt Klaassen on Jul 25, 2009 11:30 PM EDT reply actions

I'm pretty sure the answer to this part is yes.

Intuitively, if there’s two outs, runners on 2nd and 3rd, and he comes up, you probably walk him. Two outs, nobody on, you go right after him. But what about runner on 2nd and 1 out? Is there an appropriate way according to numbers to determine the correct approach? Is any team using such an approach?

Doesn’t run expectancy from The Book answer a lot of this? If I remember correctly, it is pretty generic though.

He can get 4, NOT 5.

by Warden11 on Jul 25, 2009 11:32 PM EDT reply actions

i know about game states and run expectancy

but is anyone using this to influence pitching decisions? ironically after i posted this mellinger had an entry about runs created per PA which answered that part of the question, and i guess that if “the book” you’re referring to discounts intentional walks, then it’s maybe not as important to determine an approach (ie it sounds like you always “go after them”.

am i correct in deducing that there are two “types” of OBP guys, or does someone have a different view? are the “take it and rake it” guys willingly living with high Ks as a residual effect of plate discipline, or are pitchers “overvaluing” their potential to contribute and nibbling too much when what they should be doing is being more aggressive?

take somebody like mike jacobs-his walks are up i believe but his average and production have plummeted. it seems like these guys aren’t natural on-base types and they consistently find themselves in an 0-2 hole, then make an out because their hitting skills and plate recognition aren’t good enough to produce when down in the count? perhaps for some players, being a hacker is better because they will get better pitches to hit earlier in the count? maybe mike jacobs can have only a .295 OBP with the patient approach but would have a .310 OBP with more power with a free-swinging approach?

in other words, maybe taking pitches just for the sake of taking doesn’t work for everyone (players that aren’t as good at pitch recognition). i’m not defending one way or the other just genuinely curious if there should be a more tailored approach for different types of hitters.

Kansas City Royals - rearranging the deck chairs on the Titanic since 1994.

by Home Run Tony Cogan on Jul 26, 2009 12:19 AM EDT up reply actions

I think you've got a valid question there.

I think some hackers would actually be worse players if they tried to drastically alter their plate approach in that manner; god only knows what would happen to Olivo if he tried to start being selective.

However, I’m pretty sure that’s more an effect of years of training and habit than any concrete physical skill set; in an alternate reality where Miguel Olivo is taught selectivity at a young age, perhaps he’s a true all-star hitter.

This space for rent.

by jonfmorse on Jul 26, 2009 12:26 AM EDT up reply actions

To answer your question

Guys like Dunn and Cust don’t strikeout because they’re up there hacking away and can’t hit the ball; they get struck out a LOT as a result of balls which are barely strikes. They have lots of at-bats where they’ll be fouling off borderline pitches trying to induce a hittable pitch, then get rung up what they think is ball four.

Put another way, if guys like Dunn and Cust got fed a steady diet of meatballs, they’d have .380 BAs and hit about 90 HR a year. It’s perhaps ironic, but guys who are very selective about what they’ll go after strike out more. (I think Teahen’s got a little of this in him, too. How often have we seen him take strike three on a full count?)

This space for rent.

by jonfmorse on Jul 26, 2009 12:23 AM EDT reply actions

This makes sense.

Dunn is a guy that looks for his pitch, if he doesn’t get it he either tries to spoil it or just not swing. I’m too tired to look it up, but I wonder what Dunn’s swinging strike % is?

On the other hand, you have Olivo/Jacobs who swing at pitches that might be strikes. Not necessairly always good pitches to drive.

He can get 4, NOT 5.

by Warden11 on Jul 26, 2009 12:41 AM EDT up reply actions

For his career

30% of Dunn’s Ks have been looking. 37% for Cust. Butler’s at 30%, by the way.

There’s a middle ground, though, where the K-looking percentage may not be instructive at all. Data is only available for the final six seasons of Brett’s career, but 22% of his Ks were of the looking variety. Both Teahen and Guillen are at 23%, and Jacobs is at 21%, so you can see it’s kinda fluffy.

The league average is 26%.

Oh, Miguel Olivo? 12%. Chew on that.

More: noted high-average slap-hitter Luis Castillo, who Joe Morgan said might have trouble adjusting to the National League after the Twins traded him? 51%. And Barry Bonds, whose B-R page is available for you to sponsor for a mere $380, is at 45%. Yeah, I think there’s a clear pattern here among the outliers.

This space for rent.

by jonfmorse on Jul 26, 2009 1:04 AM EDT up reply actions

Denard Span has OBP skills

His career line in 789 PAs over 2 years has his OBP – BA = 91 points. And his career SLG is barely .400. He has just 43 XBHs, with just 11 homers. Pitchers are not afraid of this guy, yet he can still work the walk.

So I’d say the chicken or the egg can come first. Obviously, a batter like Pujols has power to go along with OBP skills making him a monster that gets even more walks because pitchers are afraid and nibble against him.

On the other hand, there are players like Sammy Sosa that never walked until pitchers couldn’t pitch to them without fear of a homer. Sosa had a .309 OBP the first big year he had with 33 homers. The year after he spiked to .339 and then .340 the next year. After a mild regression in the OBP department (but not the homer department), he went on to have his biggest homer years and OBP numbers above .400.

Having a strong sense of the strike zone can only make you better. But having a mighty powerful bat makes you even better too.

by AxDxMx on Jul 26, 2009 12:25 AM EDT reply actions

Gotta point out

that while barely .400 isn’t a slugger, it’s not like Span is TPJ or Bloomquist. He IS a threat if you give him a meatball. In fact, there have been lots of players whose first 1.3 seasons worth of PA looked a lot like Span’s who later turned into regular 20-25 (or more) homer types, simply due to physical development and the adjustment to big-league pitching.

This space for rent.

by jonfmorse on Jul 26, 2009 12:29 AM EDT up reply actions

Ok

But the point is, physically, he’s not there now. 11 homers in 789 PAs translates to 1 every 72 PAs, or 1 homer every 13-16 games depending on PAs per game. He gets 1 XBH about every 4 games. He’s about as far from Albert Pujols as you can get, yet he puts up good walk rates.

Span OBP – BA = .091
Pujols OBP – BA = .093

Yet the difference in their SLG is over 200 points. That’s all I’m pointing out. Not saying Span won’t develop into a higher power threat. Just saying right now, he seems to be a low power, high OBP guy. So in this case, OBP skills came first. In others, like Sosa, SLG skills came first, leading to OBP.

by AxDxMx on Jul 26, 2009 1:19 AM EDT up reply actions

Ah, I see where you're coming from.

The confusion stems from the fact that the original post wasn’t really positing a developmental difference, but a difference between guys who just can’t hit and guys who can.

I mean, sure, if there’s two out in the bottom of the ninth, runners on first and second, and I’m clinging to a two-run lead, I’d rather face Span than Pujols — but that doesn’t imply I particularly wanna face Span either. Can’t I have Nick Punto instead?

This space for rent.

by jonfmorse on Jul 26, 2009 2:55 AM EDT up reply actions