Nicklaus used to say that he might have 2-4 perfectly hit shots (non putts) in a round.
Assuming he had 34-36 full swings, the was "flushing" something like 1/9-1/18 of them.
That's with the ball sitting still. But, of course, Jack's "non-perfect" shots still looks pretty "flushed" to the layman viewer.
Line drives are like that. Some of them aren't ripping rockets, but they still find a hole. They look perfectly nice. Back in 2002 Dan Wilson nearly hit .300 (.295). I said at the time that he was the only guy to hit .300 and not damage a blade of grass, because he hit these floating-soft liners that fell (nestled) in. But they were line drives all the same.
It stands to reason that the harder the ball is hit, the less time fielders will have to get to it. That's why BABIP is tied to liners.
Of all the great ball strikers in golf's past, all except Moe Norman (no relation) hit a ball that curved. Nicklaus hit a power fade, Trevino a WAY big fade, Hogan a baby fade (once he got great by conquering the rollicking hot hook), Nelson a draw, etc. The old ball spun to much to hit it very straight very consistently. Nobody even tried.
Today's players (excepting Bubba) generally hit a pretty straight ball. The current ball doesn't spin nearly as much.....and is designed so essentially the harder you hit it, the less it spins.
moe
Why is it that 30% of all batted balls go for base hits, and it is even true that no individual hitter sustains a BABIP outside the 25-35% range?
What is it in the nature of a baseball game that causes 25-35% of batted balls to hit the ground and fall for base hits? If we could grasp that, we could grasp why a 25% line drive rate, with a .250 BABIP, mean bad luck, even over the course of 30 at-bats. Here's a hint: would as many balls fall into the sides of this Pachinko machine as would fall into the middle? Why or why not?
............
If you dropped 30 balls down a Pascal board, and they all fell into the two left receptacles, you'd say "ah, that's bad luck." And any other explanation would be incorrect.
The thing is, you have to first thoroughly absorb WHY a ball, coming off a hitter's bat, is a Pascal's Triangle situation. Does a hitter's bat make contact 1/8 inch higher, or lower, because he meant for it to do so?
Does BABIP actually work?
.
=== Crystal Ball Dept. ===
BABIP actually does work, and it works always. (Within a certain range.) BABIP always works just like it always works that Mark Reynolds never leads the league in strikeouts and batting average at the same time. It's not a statistical trick; it's the fabric of the reality out there.
One week ago, we published this article on the M's line drive rate vs. their BABIP. We concluded:
There is one correct thing to tell the Mariners' hitters. GREAT job, men! Keep doing what you're doing. The breaks will even out pretty quick here.
"Pretty quick" turned out to mean, "One week later, everybody will be smiling about the Mariners' offense again." Their BABIP - LD was 22 a week ago; right now it is already up to 51. It will certainly go much higher. Just like Alex Liddi's batting average will certainly go much lower.
..........
The earth is 70% covered by water, and the defense -- IN ITS BEST FORM -- maxes out at 70% coverage of the baseball field. You have one acre of grass, you have 8 men with gloves, and you have batted balls averaging 70 MPH. Ichiro's have averaged more.
When Chris Sale throws a slider low and away, and a left hand hitter reaches out miraculously to throw the barrel of the bat at it, the direction of the ball is random. It's semi-random as to the direction it goes, and it's almost completely random as to whether the bat hits the ball 1/16 of an inch higher or lower.
Jack Nicklaus said (help me out here Moe) that the perfectly flush golf strike is so difficult as to be nearly a physical impossibility. That's why the pro's choose to fade or draw a bit - margin for error. The ball is not even moving, they can swing at the moment they choose, and .... as human beings they still have no way to guarantee a flush strike.
BABIP isn't a statistician's trick. It's a way to quantify 1 acre, 8 gloves, and 70 MPH.
.
=== Still On the Agenda ===
Ichiro's line drive rate stands at .312 and his BABIP at .316. He's still awesomely unlucky, despite a .301 / .322 / .410 slash line that has his OPS+ at 108, excellent for a "leadoff" hitter who creates plenty of extra offense when running around the bases.
If you've watched the games, you've seen one screaming line drive after another that has held up for the center fielder and right fielder to catch 275 feet away. Whether that ball goes between the outfielders, or to an outfielder, is nothing more than luck. An argument that these line drives will keep hitting OF's, is just like an argument that Alex Liddi will finish his career batting .375.
........
People retreat to a foxhole: well, that BABIP will go back up towards 100-140, sure. But the line drive rate will sink to the level of 2010-11, and it will even out. For one thing, this begs the question. The arguer gets to assume that Ichiro is no different than he was in 2011. That was the very thing to be debated in March: will Ichiro contract his strike zone and hit better?
But, secondly, notice that Ichiro's GB/FB ratio is radically changed. Career: 2.35. Last year this deteriorated: 2.84. This year ... 1.21! Ichiro's line drives were suppressed to 20% in the past because he was deliberately hitting the top half of the ball. And as Ichiro topped the ball more, as he pulled more 2-bouncers to second base, he became less effective.
What is Ichiro's line drive rate going to be, now that he has revolutionized his game and is uppercutting the pitch? You have no way to know for sure what his LD rate will be. It's a different Ichiro now.
I suspect that, if Ichiro chooses to hit this way all through 2012, that he will lead the major leagues in line drive percentage.
Why wouldn't he? Who can cut a mosquito with a sword better than Ichiro?
...............
There is another player who hits like Ichiro, 200 hits per year, and this player has a career 24.0% line drive rate. If you want to know what Ichiro's line would look like, after the BABIP luck evened out, check Michael Young.
You have to park adjust, of course. Michael Young with 40 SB's and a Gold Glove would be a pretty decent player.
BABVA,
Dr D
Comments
And, of course, the corollary here is that the barrel of the bat cannot be delivered to the baseball precisely as the batter intends. Where it strikes the ball has a large random factor, and that is part of what drives BABIP.
We might assume that a given hitter can BABIP .500 (or .100) by purposely striking a pitched baseball between the infielders. No can do.
Where the ball goes, off the hitter's bat, is chance, and this is proven by the consistency of BABIP.
The randomness of clean contact on upward trajectories yields random finishing distances assuming solid MLB typical contact. The HR/Fly stuff doesn't work for pitchers because some are better at getting less than MLB typical contact on upward trajectories and the ratio varies for hitters but in characteristic ways based on their power. But you can correct for the statistical randomness in the same way.
Like what Michael Saunders did almost single-handedly for the whole team tonight. :)
Please explain .... Richie Sexson 2007 BABIP = .217 -- (career .293)
Please explain ... Junior 2008/2009 BABIP = .220 -- (career .287)
Please explain ... CAREER BABIP - LD results for these players:
Mark McLemore - .295 -- 24.9 = 46
Dan Wilson .309 -- 24.1 = 68
David Bell .274 -- 23.2 = 42
Btw, I got these guys by sorting by highest career LD% since 1960.
These are not results of small sample from a single year. These three guys amassed those final tallies over a career worth of data. And while there are many guys who maintained a 100 skew and many who maintained skews well above 100 ... my point is that while the reality for the mass numbers is absolutely true - on an individual level, the link between LD% and BABIP is just not that tight or predictable.
For a decade, Ichiro maintained a +150 skew of BABIP over LD%. In 2011 Ichiro had (for him) a completely normal LD% 19.1% and a completely (for him) abnormal BABIP, .295. His BABIP was more than 50 points below norm and his skew was also way out his personal boundaries. He decoupled with his own previously establsied norms.
Based on previous BABIP crashes for aging players ... the odds are not in Ichiro's favor. Could he have successfully remade himself into a completely new hitter? It's possible. What would HIS previous LD/BABIP numbers tell us. Nothing really.
The basic argument on Ichiro is that for the last decade he was a completely abnormal hitter ... but this year, he has turned himself into a 'normal' hitter ... who will run a 100 skew and hit a lot more LDs My counter argument is that if he is indeed becoming a "normal" hitter. He is becoming a normal 38 year old hitter.
Normally, 38 year old hitters are not very good.
comes into play, Sandy. I have over the last few years become an Ichiro apostate, questioning his game and his value, especially compared to his contract. I also was publicly skeptical of what we would get out of Ichiro this season. So I'm not saying what I say below as a fanboy.
Ichiro is hitting the snot out of the ball, consistently making hard contact. His hitting reminds me more of his early years. And I'll bet if we had HittingFx available we would see that even his ground balls are hit much harder this year than in recent years.
WIll he keep it up? I'm making no predictions, that's not what I do. I know you fervently believe that eyes on sometimes skews a person's judgment, and by far the most reliable judge is statistical. I understand that. But I think in this case it is the lack of eyes on that is skewing your judgment.
As always, I respect your work and your point of view.
Because I don't want to go 4/50+ to lock up RF for a guy about to turn 40. The only thing that can stop it is Ichiro falling off a cliff, and regardless whatever he does this year will be taken into consideration for his next deal with us.
So Sandy, I'm sorta pulling for you. :) I agree with you that BABIP tends to fall off significantly when a hitter is D U N. Ichiro's swings haven't looked like the swings of a fried turkey.
If he is, and it proves out that way as we keep running him out there with little to no rest, then fine, his goose is cooked.
But we're not there yet.
~G
On McLemore, for example, you're using 2002-2004 line drive data against his 1986-2004 career. His career line drive rate would have been about 20%, not 25%, leaving his BABIP - LD rate around 100.
Stats for anything will skew simply because of random deviation, and often over more than a single season. Bill James makes this point constantly: one YEAR's data is very seldom enough for breaks to even out.
The idea that a lousy hitter might run a poor BABIP, that has some traction, within limits.
It still is not possible to project a .210 type of BABIP even for a lousy hitter. It is not possible to "deserve" less than a .600 BABIP on the 20% line drives, and that leaves you needing an .090 BABIP on everything else.
...........
Whether 38-year-old hitters are usually lousy or not, catch a game, man. Ichiro is blasting the ball. Today again, a 100-MPH line drive that was hit so hard that it held UP for the center fielder.
It's like I was reading the box scores and guaranteeing you that Brian McCann was done. Gotta watch him play, dude.
You got serious game Sandy :- ) but on this one, you're swingin' at the wrong pitches amigo.
You're defending the position very ingeniously, LOL, but we're watching Ichiro hit ropes three times a night. His BABIP isn't because he's hitting two-bouncers to second.
G,
Thanks for the put. It's certainly lonely out here right now. And the fan in me WANTS to be wrong. And maybe I'm just allowing some of my recent reading to color my analysis. I recently finished "Thinking: Fast & Slow" by Daniel Kahneman ... which speaks on the vast and varied ways that humans, as a group, mis-perceive things.
In this particular case, I know that attempting to pick age-based collapse is a high-risk proposition. Absolutely nobody has a good methodology for doing so. Seems like every great player is great ... until he isn't anymore. Yet, I've seen precious little research into even attempting to separate the "down year" from the "end of career" changes in production.
In all honesty, 8 years ago, I would have been drafting Ichiro high in my fantasy draft, because I'd seen so many aging players bounce back after a sub-par season. So, I recognize that I could easily be wrong in this case. But, I was doing that during the height of the steroid era.
Ultimately, I think BABIP is vastly misunderstood and poorly applied by the majority of baseball analysts today. The "all batters" norm of BABIP - LD of 100 is being used (not just by Doc) as a "standard". But, it's NOT standard.
Ichiro ran +150 for a decade. That is certainly NOT luck. Doc doesn't think so, neither do I. I completely believe there was a specific reason for Ichiro running BABIP rates at the extreme end of the envelope. His .351 BABIP is inherently tied to his speed, and his infield hit abilities. Derek Jeter has a similarly high (.355) career BABIP.
On the other extreme, you've got a guy like Tony Batista. Career BABIP of .257. His 5 years of LD data show a normal 18% LD rate. You look at what he did in '02-'04, three full seasons of data, and what do you see?
Year - Babip - LD% - OPS
2002 - .246 -- 16.0 -- .766
2003 - .240 -- 20.7 -- .663
2004 - .225 -- 16.5 -- .728
Clearly there are dozens of different hitter types. And with those different hitter profiles, there are a LOT of different BABIP/LD "standard and repeatable" skews. From my view, as soon as BABIP is mentioned these days, the standard response is to view it as a one-size fits all interpretation of the leage-wide norm. I don't.
Batista is a completely different hitter than Ichiro. But, the above is *ANOTHER* case of a guy able to maintain an "impossibly low" BABIP for three years. This is incredibly rare because hitters with such low BABIPs will nearly always have such horrid slashlines that they don't last. In addition, his BEST LD% year happens to correspond to his WORST OPS year.
My explanation for the disconnect between the guys who can watch Ichiro and myself is simple. Ichiro *IS* hitting the ball harder ... harder than Ichiro of previous years. But, is he hitting his LDs harder than Kotchman? How about Beltre? Or Branyan?
Back in 2009, was it possible to "tell from looking" that Ichiro had a BABIP-LD of 202 and Branyan was at 125? Was it possible to tell from looking that Ichiro was running an 18.2% LD rate and Branyan only a 17.5?
Was Doc dismissing Smoak as a hitter in 2010, when he was running a .298 - 21.3 skew of only 80? Was he pointing out how ridiculously "lucky" Smoak was in 2011 when his numbers were .273 - 13.8 ... and his skew leapt up to 135?
It gets even worse once we plant an idea in our head that a specific event is skewing against us. Because we then "notice" every event that we're looking for. The reason we HAVE to count these things is because as humans, we are lousy at perceiving them correctly.
In the end, the same data can often lead to completely contradictory conclusions. Doc sees Ichiro visually as being the most selective at the plate he's ever been. I look at a walk total of 3 and do the math and come up with 21 walks for the entire season. Selective isn't the first word that leaps to my mind.
When I look at Ichiro's Fangraph's Plate Discipline - O-Swing (swings outside of zone) is better than his last 3 seasons. But at 30.5%, this is still below his career 27.5%. In point of fact, Ichiro didn't go over 25.2% in his first five years. As a rookie, Ichiro had a 20% O-swing%.
However, his Z-contact and O-contact rates are BOTH career highs at the moment. He is "connecting" with everything. In fact, he's connecting with 96.2% of pitches in the zone and 88% outside. In the last 3 weeks, Ichiro is destroying his established ability to lay wood on ball, (which was always excellent). With a career 5% swinging strike % suddenly at 2.9% - (previous bests were 4.0 and 4.3) - I see a hitter who is "on fire", in a completely unsustainable manner. His LD% is unsustainable, (by ANY previous standard). His contact rate is unsustainable. So, I see a hitter who at some point will cool off and suffer accordingly.
Matt and Doc both project a hitter who has at the age of 38 made himself "better" in multiple facets of his game, that they fully expect to continue for the next 5 months.
In the end, the next 5 months will tell the story.
Interesting trade-off for Ichiro this year. Hes abandoned the Ackley swing, but stepping further back in the box.
I'm not suprised that Ichiro is hitting the ball harder. Hes gained an extra fraction of a second to react. His IH% should further decline (just a hair longer to run+age), but keeping things in perspective it should still be well above-average.
Overall, its likely that Ichiro runs a high BABIP with his skillset, but probably not as high as his career marks once the LD% starts to regress. His power should recover (if it doesn't, its a concern).
A quick rundown based on career averages for outcomes (with a markdown to IH%) could figure out a rough xBABIP for Ichiro this year. It will almost certainly be well above .300.
Taro,
If Ichiro were 31 going on 32, I would almost certainly agree with you.
In the end, my analysis leans heavily on 2011 data. Why? Because he was 37 ... not 27.
Every year forward, from about age 30 onward, the odds that an abberant year is in fact age-related goes up. It NEVER goes down, (unless influenced by pharmacueticals).
There are two interpretations of 2011 Ichiro. Age was catching up to him. This "caused" his massive swoon in BABIP, which obviously influences lots of other stats. The alternative explanation is that after running 350 BABIP numbers for a decade, Ichiro just had a really, really, really run of bad luck, which lasted for 161 games, (or more accurately, for the last 5 months of the 2011 season.
Note: In 2011, Ichiro had a .345 (normal for him) BABIP in April. His BABIP months in 2011:
Apr: .350
May .227
Jun .323
July .265
Aug .301
Sep .301
That's five consecutive months "signficantly" below his previously established normal level, including two months that are 85+ points below his norm. If that's one random month, it's random. If it's 5 consecutive months, that's no longer random.
I have argued previously that the first thing to go for star players is not "ability", but rather "consistency".
At this instant, heading into the final game for April, Ichiro has a BABIP of .292 for the season. If Matt or Doc were looking solely at the stats of ANY player in MLB over the last six months in regards to BABIP (absent all other data), and knew that they were 38 years old, they would both accept without question that the current .292 was a perfectly reasonable BABIP and that there would be little reason to expect improvement.
They are contending that he is BOTH hitting many more LDs, (supported by data), and hitting them much harder, (supported by direct observation -- but not supported through BABIP, HRs or ISO).
They view the BABIP as unlucky ... I view the LD% as lucky. We all agree that the miniscule skew cannot possibly continue, (though I contend that a skew of only 50ish for the year is completely possible for an aging player).
Except that his April BABIP and slash-lines are nearly identical to his August and September of 2011 numbers: Matt and Doc *see* a completely new Ichiro. His stats for latest 3 months of play:
Aug: .292/.313/.366 (.678) -- .301 BABIP
Sep: .268/.303/.374 (.678) -- .301 BABIP
Apr: .281/.300/.375 (.675) -- .292 BABIP
I will gladly concede that it is possible that Ichiro "should have" posted a .780 OPS for the month of April. That wouldn't change my analysis one whit, because I expected before the season began that he's more likely to start quick and fade. And even if he has been unlucky, you do not get that luck "back". Going forward, you assume neutral luck.
In truth, I believe a .750 month of May is completely plausible. But, a .538 month of June or July is not only plausible, but increasingly likely.
At this point, my growing fear is that we end up with no resolution to this issue, because the longer they play him non-stop, the greater the risk of a 15 of 60-day trip to the DL.
In the end, Ichiro's value is tied up almost exclusively in his BABIP. He has no power, (even this season, it's barely above 100, while he's "raking"). He has no walk rate, (on pace for only 21 walks), and the improvement to the rest of the team, change in batting order, and general decline in "fear" of Ichiro means he is unlikely to continue getting his baker's dozen intentional passes.
In the end, unfortunately, even if I could watch him, I suspect it wouldn't make much difference. My template for line drive hitting was formed by Chipper and Andruw and Sheffield. I expect if I was watching Ichiro I'd shrug ... and probably be unimpressed, because I wasn't watching Ichiro guide ground balls through the infield for the last decade. I have no baseline to compare him to ... him.
But, for all the talk of luck or unluck, in the end, it will come down to production. By year end, Ichiro's BABIP will be back up to .330+ ... or it won't. His OPS will be above .700, or it won't. In fact, there is only one thing we know with certainty. By October, Ichiro will be 5 months older than he is today.
I'm siding partially with Sandy here. The .300 BABiP for Ichiro is not unlucky, not because the quality of his line drives is lacking, but because his IFFB% is at 25%. Break it down this way, say you have 100 balls in play:
28 LD you bat 21/28
47 FB/GB you bat 9/47
25 IFFB you bat 1/25
Your grand total BABiP:
.310, and that's assuming good averages on GB/FB, and to an extent IFFB.
When you swing hard and miss slightly, there's high chance of hitting straight down or straight up, and those are 0 chance of hit balls.
If you didn't know, the Mariners are 6th in IFFB% in addition to being 5th in LD%. Why is this? I think it's the side effect of the 'killer' swing, you lose bat control, the bonus in force is subtracted from agility, keeping the batter (and particularly Ichiro) from making a swing more appropriate to the angle of the balls flight and instead swinging purely with intent to harm. In addition, every one has noted Ichiro standing at the back of the box instead of the front so that he has the extra millisecond, but the pitch gets the extra millisecond as well, and pitchers with nastier stuff have that extra break space. In effect, meatballs, Ichiro now has all the time in the world to react to, but fastballs, splitters, cutters and sliders with late break get an opportunity to break harder on him.
Mal, your numbers aren't correct, though your thought process seems valid to me. The infield fly ball percentage is not of all the ball in play, but rather only the fly balls in play. So it is 6 pop-ups, 19 outfield flyballs, 39 ground balls, and 26 line drives. Using the average BABIP for the distinct hit types found here, http://www.brewcrewball.com/2008/4/22/447297/all-about-babip, Ichiro would have a BABIP of 0.342. This would give him a BA of 0.323, rather than his current 0.281.
I would also like to point out that Ichiro has used a more agressive approach in the past, in particular in 2005 when he set his personal ISO record, set his career high for homeruns, set a career low for infield hits, and prior to 2011 had his lowest BA of 0.303. Anyway using his batted ball data from 2005 would have 'predicted' a BABIP of 0.322 versus his actual of 0.316. Given that Ichiro gets a high rate of infield hits and therefore has a higher than average BABIP for ground balls, it seems likely he runs a lower than average BABIP for line drives. How much lower, I don't really know.
BABIP is pretty flukish on a year-to-year basis, much less month-to-month. Its likely Ichiro's true-talent BABIP has declined given the change in skillset (particularly IH%), but unlikely that it declined to the level it did last year overall.
I did an analysis of Griffey's outcomes a couple years back and concluded that his BABIP was only 10-15 points 'unlucky' based on a career BABIP outcomes (probably within the range of physical decline). His BABIP was not that unlucky considering low LD, pop ups, etc.
With Ichiro's current outcomes we'd be talking 50+ points below his career rate outcomes, at which point we'd be talking extreme physical decline.
Career BABIP outcomes for Ichiro and difference over league average:
Groundballs: .299 (+.054)
Flyballs: .124 (-.010)
Linedrives: .693 (-.026)
Bunts: .642 (+.266)
I wonder why Ichiro doesn't bunt more often, especially now to keep infielders honest.
While the high Pop up rate is a bit worrying, he should be BABiPing .340 something based on his career averages, after tonight he's pretty close.
Thanks Malcontent ! :- )
Taking a couple out of dozens of possible points...
1. No doubt Ichiro's popups HAVE been the cause of some of his BABIP deflation, looking backward. His BABIP is as low as it is - partially - because of 3 or 4 (?) extra popups to date.
2. I wouldn't call that predictive.
3. Ichiro's BABIP after tonight is .305, rather than the .342 that Dr. Kelly gives - still only 10 or 20 points ahead of his LD%.
4. I'll bet you a dollar that Ichiro's BABIP ain't no 40 points above his LD rate at the end of the year.
Heh -