Calling Dr. B - LASIK worth 12 wRC+ ?!, part 1

I/O: Jeff Sullivan with a solid little quick-study on performance before and after LASIK.

Sully selected the 26 names with a reasonably non-prejudicial method - the 26 names that came up when he Googled "LASIK" and some other term ("MLB"? who knows).

Even for peer review, Googling can count as a random-enough selection method, actually. It's not like you're taking the names off of last year's leaderboards.

97 wRC+ before; 109 wRC+ after the surgery.

CRUNCH: Slap me silly. It's one thing to see performance influence reflected in a study. It's another thing to get a result like that.

The first question that hits me is, if this were true, why in the world wouldn't everybody be getting it. LASIK is outpatient, right? :- )

CRUNCH: And the first answer that was given, before the question was even asked, was by our resident optometrist. He cites a contradictory study by MLB doctors that "found no significant improvement in performance could be attributed directly to the LASIK procedure compared to wearing glasses or contact lenses. Subsequent studies by Laby and Kirschen have actually demonstrated a decrease in visual function after surgery..."

The method used by this hi-falutin' Harvard Med School study? Look at the before-and-after OPS's of 12 major leaguers -- found by "extensive search of public media." HEH! Well, Google unquestionably searches extensively there, lads.

Sully, consciously or not, selected precisely the same design and, um, kinda failed to replicate their data, I think you could say.

..............

I was going to quibble just a bit with the method of selecting players -- maybe players who get a lot of pub are better hitters, for example -- but hey, if it's good enough for Harvard, it's good enough for us, right?

Sully finding 26 guys, Harvard only 12, my knee-jerk opinion is that the Harvard guys coulda cherrypicked their subjects to cook the data. But far be it from me to be cynical.

We all know Sully well enough to know that he didn't cook the data, so what's he doing with twice the subjects?

Anyway, Jeffy probably didn't realize he was noodling up a Harvard-class study while watching TV over the laptop. Higher education these days ;- )

CRUNCH: Would love to hear Dr. B's remarks as to Sullivan's failure to replicate the data. We defer to the (far) greater expertise.

Typically when the studies are 1 vs 1, or 1 vs 2, it's time for more studies. But if you fail to replicate the data, isn't the default conclusion "No Conclusion Warranted"?

And the default assumption here has to be that better vision allows you to hit baseballs better, we would think...

Printer-friendly version

Categories:

Sports

Tags:

Rob Johnson

Replies

Couple things to note

The journal article (which you can read here: http://www.optometryjaoa.com/article/PIIS1529183905001569/fulltext ) is from November 2005. So they naturally are working with a smaller sample than me, because I have the benefit of a few extra years.

Also worth noting is that they used raw, unadjusted BA/OBP/SLG/OPS. So that could be a contributing factor in our finding different results.

But ultimately, we're dealing with such a small sample size here that we can't arrive at a conclusion either way. Have to figure that improved vision helps, but there's no evidence that it helps a whole lot.

...

didn't notice the 2005 date. Okay, yeah.

...........

Not sure about the sample size concern; Laby et al give an 0.997 coefficient for OPS and publish their study, right?

And Dr. B notes that the MLB panel of doctors posted a finding and recommendation (whether based on Laby, or their own study of presumably similar size).

.............

At any rate, thanks for the great spadework Jeff. Good stuff.

And the 12 wRC+ are much, much more than I'd have expected to see. In my book, that's a dramatic finding. Dr. B?

That's rather comical.

How the heck did a Harvard biostatistics study get published using RAW UNADJUSTED OPS??? That would be like me using un-normalized meteorology data to make a claim about long term climate variability...oh wait a second...those studies got published too...

*furrowed brow*

Academia just ain't what it used to be.

Oh incidentally....12 ballplayers is a study?? Since when is 12 subjects enough of a sample to confidently blast a deeply counter-intuitive conclusion like "Don't bother with LASIK, boys! It won't help you hit!" Better vision = better hitting. Period.

Good stuff Matt

Maybe you and Jeff can help me out here, since am a bit confused... both of you wondered about the size of the group of players studied; I'm not sure what the problemo is with it...

In this case, Googled players (for once, in sabermetrics! ...) actually are a "sample" since they are reasonably construed to randomly represent a larger universe. Neither of you disputed that.

If I toss a 1d6 twelve times, and it comes up six each time, the odds are 1:6^12 against the straight sixes (or 1:6^11 against any nonspecified number twelve times)...

Six raised to the twelfth is astronomical, of course, so there is no problem with my using 12 throws as a "sample" to conclude the die is loaded... in fact, in only 3 casts I have pushed past the 99.5% level of confidence...

Whether a sample be 3 players or 750, if the ol' binomial calculator pushes us past 95%, our sample is large enough, correct? Harvard-amigos asserted 99.7%...

or what are you amigos talking about? :- )

As to the substance Matty, agreed

It's not that there's anything wrong with LL's cool mini-study, of course, but was a bit taken aback to see such a simple design used by the Harvard amigos ... Am guessing Jeff did his offhand, in an hour or so, whereas the Harvard paper might have been a good piece of a quarter's grade or research time...

As with you and Jeff, was puzzled at their choice of OPS (and actually even OBP and AVG!) when it would have been just as easy to take better stats for the purpose... wRC+ occurs as an alternative.

:- )

.............

We are a super-specialized audience in sabermetrics, of course, so to see the optometry students (?) run a slightly unsophisticated approach on such an important topic ("should I get my eyes operated on?!") is bemusing...

It offers an interesting chance to sit in the seat of the "wise," if only for a moment, and get a feel for just how rigorous you really need to be to pass peer review in the year 2010...

...............

Not trying to bust anybody's chops over there at Harvard :- ). Laymen tend to think of any scientific consensus as Scripture From On High, but after all, most scientists are really just grad students with a coupla more years in.

I have long viewed "the scientific consensus" as something less than Truth.

That's not an insult to them; it's an empowerment of the people in the Internet Age.

Anybody can voice an argument in the year 2010. That's a good thing.

What is definitely amusing

Is that they used AVG, also, across 12 players. :- )

For all that

Am not saying it's a worthless study over there at Harvard Med -- it's a useful first cut -- but I'll predict that it's a pioneer study that's going to be contradicted by the next six studies in.

I think that .997 coefficient is going to develop some seal leaks as time goes on, first of all with the unadjusted OPS but also probably with other moving parts... could be wrong...

I can't even believe that

I can't even believe that qualifies as a study.

Even Jeff's post isn't anything more than a quick-handed one (and hes aware of it), but its still better than that nonsense.

As I get a little older I find I care less and less about a person's reputation/status, and more on just the content itself.

Surprisingly, a lot of

Surprisingly, a lot of medical studies like this one are poorly designed (mainly because well designed studies are difficult and expensive).

Improved eyesight will almost

Improved eyesight will almost certainly help you at the plate.. I mean, thats just logical.

What probably messes with the data are the varying degrees of success people had with the surgery, whether or not they wore contacts when hitting prior to the surgery, how much their vision improved, etc.

It is not clear that the

It is not clear that the ability to read nonmoving text better than average people translates directly to being able to hit a swiftly moving, tumbling round object with a round bat.

Lasik has been known to introduce artifacts (disclaimer... I'm not an opthalomologist)... something akin to glare.

http://www.streetdirectory.com/travel_guide/12780/lasik_surgery/lasik_su...

It is possible that this might actually interfere with tracking. Altitude shifts effect vision too after lasik.

Just because something is intuitive doesn't make it true or even likely. What would be needed are strong randomized controlled prospective trials... something unlikely to ever be done.

In this case

I'm real comfortable skipping the experimentation. :- )

Take the reductio ad absurdum on the other end. I assert that Stevie Wonder can't hit a pitched ball, and that this is due to his eyesight. You want a study first?

It's axiomatic that a hitter must read a pitch EARLY, and it's axiomatic that he does so by reading spin. He ain't doing it with his sense of smell.

Somebody else wants to withhold judgment, hey that's cool by me. My own firm conviction is that the sharper a hitter's vision, the better he'll hit -- in the same sense that the physically stronger he is, the better he'll hit.

Other things being equal, of course.

....................

Rob Johnson suddenly has 20/10 vision? It's no guarantee, but it's a big advantage going into 2010. He's a buy stock order for SSI. :- )

Like I said, it is probably

Like I said, it is probably true that better vision helps you hit, but what happens when vision is already good with contacts or goggles? Lasik wouldn't be expected to add as much as it might appear.

But medicine is littered with theses which seemed intuitive and plausible to the point of certainty but were later shown to be incorrect.

I am looking for a good year from Johnson both because of natural improvement and presumably from the Lasik.

However, the last player I recall getting Lasik was Brett Boone. Right before he fell off the table.

Right, which is what Dr. B addressed

That Lasik was held to lead to no performance improvement *vs. normal corrective lenses.*

The last Mariner player...

I remember before that who got LASIK was Al Martin, Prior to his first season here. He absolutely raved about his improved ability to see the baseball. Dr. D might have bought. Intuitively, and in a vaccuum, he might have hit better than he did in his time here, but with the change in league, home ballpark, coaching/hitting instruction, whether he was already in age decline, etc. etc. etc, who is really to say why he didn't hit better and actually hit worse? It is difficult to isolate the effects of LASIK on the different aspects of visual function as it is (studies are being performed in my field all the time), let alone how those changes in visual function, for better or worse, affect actual baseball performance. Have some more research by a few collegues on that exact subject I home I believe, and I will check once I am off duty.

Date and Sample Size...

All valid points on the faults of the study. Agree that study was meant to be a pilot study to provoke further, better designed and better funded studies into the same inquiry. Alot of the more recent research is not accessable to the public until years down the road, which is likely why 2005 is the most recent study found. Even to the American Optometric Association, which I am a part of, the latest research is made availible through subscription fees to each professional journal, which is easy for an academic institution, but difficult to scrounge up for a single doc like me. I will see about access through our medical research library, as I practice in the military at a large research hospital.

Dr. B

One guy I never bought into, chief

was Al Martin. :- )

BTW, the reason I am a *little* perked up about Johnson, is because Bill James pegged Johnson as a future impact hitter in the bigs. *intersecting* the puzzling lack of development, with Johnson's claim about poor eyesight, is what has my interest.

As well, there are things about Johnson that I like at the plate, such as his little bursts of 1:1 eye ratio.

Good stuff. :daps:

Isolating the effects

Is one problemo that I can definitely relate to. I hear you.

Your specialization in the field makes your commentary most, most welcome. Gracias.

Anytime you can connect...

what you love to what you do for a living is a blessed experience. I thank you all for the oppurtunity.

Dr. B

Seattle Sports Insider

You are here