King Felix and the power of teh internets

[dawgfan]

So as most serious baseball fans know, Felix Hernandez has been struggling some this season. After a phenomenal start to the season that showed the nation why there's been so much hype over Felix, he developed elbow tightness and since he's returned from that injury has been mostly mediocre.

What most don't know is that a couple of well-written, well-researched Mariners blogs (USSMariner.com and LookoutLanding.com) have both done a lot of work looking at exactly what's been going on with Felix, with 2 main results:

1. Based off of the Enhanced Gameday data available from certain stadiums through MLB.com coverage, we have hard data showing pitch speed and pitch break for a number of his starts. As of a few weeks ago, this data shows a dropoff in both velocity and break on his pitches since the start of the season. This suggested that either Felix was still hurting, or he wasn't fully cutting loose for fear of re-injuring himself. Further, video analysis of his pitching motion suggested that his arm slot had lowered slightly, which further reinforced this theory.

2. Felix has been extremely predictable in his pitching patterns to start games - almost nothing but fastballs to start off games. While his 4-seam fastball touches the high-90's and his 2-seamer the mid-90's with excellent movement, his slider and curveball are actually his best pitches. When he's mixing his pitches, he's at his most effective. This isn't just visual observation - this is borne out by facts.

The M's, up until very recently, have blamed Felix's struggles on a lack of command of his fastball. And that is true - he doesn't have excellent command of his fastball. If he can improve his command of his fastball, he'll have more success and be better able to get away with being predictable in his pitch selection. But he doesn't have better command, he may never have excellent command, he is more effective when he mixes his pitches and the M's are in playoff contention now - which is why Dave Cameron at USSMariner.com posted this open letter to M's pitching coach Raffy Chaves: http://ussmariner.com/2007/06/27/an-...rafael-chaves/

And lo and behold, that letter got to Chaves, he showed it to Felix, and after getting in trouble in his first couple of batters throwing all fastballs in his start against KC on Monday, he started mixing his pitches and did very well, and followed that up with an 8 inning/2 hit performance today vs. Oakland where he mixed his pitches from the very start. And he credited reading that letter and talking with Chaves with changing his mindset. Chaves himself, while saying that the info in the letter wasn't anything he didn't already know and was telling Felix, gave credit to it showing Felix that everyone knew how predictable his pitch selection was.

What's interesting was the reaction of the post-game radio hosts - somewhat predictably, one of the hosts was dismissive, saying to the reporter who was conveying this story that he hoped Felix wasn't paying attention to anything else on the internet that he read about himself. You could feel the derision dripping from him, the notion that anything someone on the internet might say might actually have some real insight and value (as opposed to a "professional journalist" such as radio sports jock like himself). On the other hand, the woman doing the report seemed to have no issue with the story and didn't appear to ridicule it at all.

The author of that open letter has taken pains to deflect any credit, saying it's up to Felix to make use of information given to him and to Chaves for driving those points made in the letter home.

But still, it's gratifying to see that well-informed data like this can make it from a good sports blog and have a positive impact on a professional player. Slowly but surely, I think some progress is being made on viewing blogs as having some real value in certain cases.

It should also be noted that it sounds like Felix is regaining velocity and sharpness on his pitches, which obviously helps his cause (and at the very least reduces concerns that he may still be hurting).

[dawgfan]

Dola - it should also be noted that, going into his start today, Felix's BABiP was .368, which is high and likely to return to the mean over the rest of the season (even factoring in that Seattle's defense has been below average). Today was evidence of regression to the mean as he only gave up 2 hits and yet only struck out 2, meaning a lot of balls in play were turned into outs.

[Ksyrup]

There are a couple of great blogs out there that Rob Neyer always links to that do a wonderful job of analyzing video of pitchers' arm angles and such. Amazing stuff.

[JonInMiddleGA]

What kind of strikes me as, I dunno, "odd" for lack of a better word is the impression that Hernandez only listened to his pitching coach after some random person (as far as Felix is concerned) told him the same stuff that the coach says he had been telling him.

Seems to me like that raises a question of how much respect Hernandez has for Chaves, in spite of the fact that he was apparently his coach in his pitcher of the year AAA season (before being called up in August)

[ISiddiqui]

Originally Posted by JonInMiddleGA

What kind of strikes me as, I dunno, "odd" for lack of a better word is the impression that Hernandez only listened to his pitching coach after some random person (as far as Felix is concerned) told him the same stuff that the coach says he had been telling him.

Seems to me like that raises a question of how much respect Hernandez has for Chaves, in spite of the fact that he was apparently his coach in his pitcher of the year AAA season (before being called up in August)

Actually I think it says that Chaves was full of shit when he said he'd been telling him the same stuff. Either that or Chaves didn't put it all together into one concise format like the letter did and was feeding him all other sorts of BS which led to the message being lost to Felix.

[TroyF]

Originally Posted by ISiddiqui

Actually I think it says that Chaves was full of shit when he said he'd been telling him the same stuff. Either that or Chaves didn't put it all together into one concise format like the letter did and was feeding him all other sorts of BS which led to the message being lost to Felix.

That was my thought.

[EagleFan]

Originally Posted by ISiddiqui

Actually I think it says that Chaves was full of shit when he said he'd been telling him the same stuff. Either that or Chaves didn't put it all together into one concise format like the letter did and was feeding him all other sorts of BS which led to the message being lost to Felix.

Because internet blogs are much more reliable than pitching coaches...

[Huckleberry]

The coach might have been telling the truth. He had a good point, at least, when he pointed out that the blog shows that the kid's pitch selection really was obvious to everyone.

[Schmidty]

Originally Posted by dawgfan

So as most serious baseball fans know, Felix Hernandez has been struggling some this season. After a phenomenal start to the season that showed the nation why there's been so much hype over Felix, he developed elbow tightness and since he's returned from that injury has been mostly mediocre.

What most don't know is that a couple of well-written, well-researched Mariners blogs (USSMariner.com and LookoutLanding.com) have both done a lot of work looking at exactly what's been going on with Felix, with 2 main results:

1. Based off of the Enhanced Gameday data available from certain stadiums through MLB.com coverage, we have hard data showing pitch speed and pitch break for a number of his starts. As of a few weeks ago, this data shows a dropoff in both velocity and break on his pitches since the start of the season. This suggested that either Felix was still hurting, or he wasn't fully cutting loose for fear of re-injuring himself. Further, video analysis of his pitching motion suggested that his arm slot had lowered slightly, which further reinforced this theory.

2. Felix has been extremely predictable in his pitching patterns to start games - almost nothing but fastballs to start off games. While his 4-seam fastball touches the high-90's and his 2-seamer the mid-90's with excellent movement, his slider and curveball are actually his best pitches. When he's mixing his pitches, he's at his most effective. This isn't just visual observation - this is borne out by facts.

The M's, up until very recently, have blamed Felix's struggles on a lack of command of his fastball. And that is true - he doesn't have excellent command of his fastball. If he can improve his command of his fastball, he'll have more success and be better able to get away with being predictable in his pitch selection. But he doesn't have better command, he may never have excellent command, he is more effective when he mixes his pitches and the M's are in playoff contention now - which is why Dave Cameron at USSMariner.com posted this open letter to M's pitching coach Raffy Chaves: http://ussmariner.com/2007/06/27/an-...rafael-chaves/

And lo and behold, that letter got to Chaves, he showed it to Felix, and after getting in trouble in his first couple of batters throwing all fastballs in his start against KC on Monday, he started mixing his pitches and did very well, and followed that up with an 8 inning/2 hit performance today vs. Oakland where he mixed his pitches from the very start. And he credited reading that letter and talking with Chaves with changing his mindset. Chaves himself, while saying that the info in the letter wasn't anything he didn't already know and was telling Felix, gave credit to it showing Felix that everyone knew how predictable his pitch selection was.

What's interesting was the reaction of the post-game radio hosts - somewhat predictably, one of the hosts was dismissive, saying to the reporter who was conveying this story that he hoped Felix wasn't paying attention to anything else on the internet that he read about himself. You could feel the derision dripping from him, the notion that anything someone on the internet might say might actually have some real insight and value (as opposed to a "professional journalist" such as radio sports jock like himself). On the other hand, the woman doing the report seemed to have no issue with the story and didn't appear to ridicule it at all.

The author of that open letter has taken pains to deflect any credit, saying it's up to Felix to make use of information given to him and to Chaves for driving those points made in the letter home.

But still, it's gratifying to see that well-informed data like this can make it from a good sports blog and have a positive impact on a professional player. Slowly but surely, I think some progress is being made on viewing blogs as having some real value in certain cases.

It should also be noted that it sounds like Felix is regaining velocity and sharpness on his pitches, which obviously helps his cause (and at the very least reduces concerns that he may still be hurting).

This was reported by Shannon Drayer...errr...Sportsdigs 8 weeks ago.

[dawgfan]

Originally Posted by Schmidty

This was reported by Shannon Drayer...errr...Sportsdigs 8 weeks ago.

What was reported - the predictable pitching pattern? Dave at USSMariner.com has been bitching about Felix's pitch selection since the middle of last season.

[dervack]

Originally Posted by dawgfan

What was reported - the predictable pitching pattern? Dave at USSMariner.com has been bitching about Felix's pitch selection since the middle of last season.

It's a joke from months ago.

[DeToxRox]

Perhaps then the nickname King should be reserved for someone living up to said hype? Verlander perhaps?

[wade moore]

Originally Posted by DeToxRox

Perhaps then the nickname King should be reserved for someone living up to said hype? Verlander perhaps?

w00t, CAA, w00t!

[Passacaglia]

Yeah, but how many all-star votes per second did they get for him?

[MrBigglesworth]

As a Felix fantasy owner, I applaud the efforts of the Mariners bloggers.

[dawgfan]

Originally Posted by DeToxRox

Perhaps then the nickname King should be reserved for someone living up to said hype? Verlander perhaps?

Verlander is a very, very good young pitcher, one of the top young pitchers (and top pitchers period) in the game. He's also 3+ years older than Felix.

Let's take a look in 3 years and see how Felix's results compare with what Verlander does this season. If Felix remains healthy and follows a normal growth and maturity pattern, he'll be a Cy Young contender - I don't think there's any pitcher in baseball right now that has better stuff than Felix. As he gets more experience and learns how to work hitters and mix his stuff effectively, he should emerge as the top starter in the game.

That's if he stays healthy - if his elbow issues return (which is a definite possibility) all bets are off...

[DeToxRox]

Originally Posted by dawgfan

Verlander is a very, very good young pitcher, one of the top young pitchers (and top pitchers period) in the game. He's also 3+ years older than Felix.

Let's take a look in 3 years and see how Felix's results compare with what Verlander does this season. If Felix remains healthy and follows a normal growth and maturity pattern, he'll be a Cy Young contender - I don't think there's any pitcher in baseball right now that has better stuff than Felix. As he gets more experience and learns how to work hitters and mix his stuff effectively, he should emerge as the top starter in the game.

That's if he stays healthy - if his elbow issues return (which is a definite possibility) all bets are off...

While I love Felix and thing he is awesome to watch, the whole King thing just irks me, probably because I am a Verlander fan.

[dawgfan]

Originally Posted by DeToxRox

While I love Felix and thing he is awesome to watch, the whole King thing just irks me, probably because I am a Verlander fan.

Well, to be fair it was a nickname bestowed on him by an M's blog, partly because he's got so much potential and partly because he's come along during a really down period in M's history where there hasn't been much to get excited about.

FWIW, Felix's preferred nickname is "el Cartelua", roughly translated as "bad-ass".

[Schmidty]

Originally Posted by dawgfan

Well, to be fair it was a nickname bestowed on him by an M's blog, partly because he's got so much potential and partly because he's come along during a really down period in M's history where there hasn't been much to get excited about.

FWIW, Felix's preferred nickname is "el Cartelua", roughly translated as "bad-ass".

Potential means poop. Verlander had potential and has already fulfilled it. PLUS, he only just turned 24. He will only get better, which is scary.

Also, I think you are really, really stretching it when you say that Peasant Felix has the best stuff in the majors. Verlander already has one of the top 3 curveballs in baseball, and can top 100 mph whenever he wants to, even in the 9th (102 anyone?). Not only that, his changeup is already good, and getting better as he learns it. Even if Felix reaches his potential, he'd have a hard time matching that kind of stuff.

[MrBigglesworth]

Originally Posted by Schmidty

Potential means poop. Verlander had potential and has already fulfilled it. PLUS, he only just turned 24. He will only get better, which is scary.

Also, I think you are really, really stretching it when you say that Peasant Felix has the best stuff in the majors. Verlander already has one of the top 3 curveballs in baseball, and can top 100 mph whenever he wants to, even in the 9th (102 anyone?). Not only that, his changeup is already good, and getting better as he learns it. Even if Felix reaches his potential, he'd have a hard time matching that kind of stuff.

Felix has more K's per inning than Verlander, and an ERA only a half a run more despite pitching while hurt for a while. I'm a big Verlander fan, but when both are healthy I don't see a big difference between them. I could easily see them having similar stats at the end of the year, if both stay healthy.

[Schmidty]

Originally Posted by MrBigglesworth

Felix has more K's per inning than Verlander, and an ERA only a half a run more despite pitching while hurt for a while. I'm a big Verlander fan, but when both are healthy I don't see a big difference between them. I could easily see them having similar stats at the end of the year, if both stay healthy.

More K's per inning = a lot more pitches = higher chance of injury. While strikeouts, and a good strikeout pitch, are a good thing (especially in jams), I'd much rather have a guy who can pitch to contact with movement (getting groundballs and popups) instead of constantly trying to strike guys out. If Felix was a few years older and seasoned, the high k's wouldn't bother me so much, but as young he is, the potential for injury is very high.

[ISiddiqui]

Hmmm... strikeouts are a great evaluator of a pitcher's future performance. I don't think many folks these days say they'd rather have the pitcher who strikes out less folks everything else being equal. Now if he walks and gives up hits a ton more (Felix slightly does at this age, but defense may play a role), then you have a legitimate point, but if he doesn't, then take the strikeout guy. Less dependant on BABIP for one.

Saying the high number of K's bothers you is absolutely ridiculous and shows a bias, I think. K/BB and K/9 are two very important measures scouts use in predicting the future success of pitchers and for good reason.

[dawgfan]

Originally Posted by Schmidty

Potential means poop. Verlander had potential and has already fulfilled it.

Verlander, as I've said, is one of the best young pitchers and among the better pitchers period in the game right now.

PLUS, he only just turned 24. He will only get better, which is scary.

Yep, he should improve with age (barring injury). The big point I think you missed in my comparison though is that Felix is 3 years younger than Verlander - if he follows normal improvement patterns, by the time he's 24 in 3 years (Verlander's current age) I suspect he will be posting even better results than Verlander is currently.

One could also argue that, in the things a pitcher has the most control over (K rate, BB rate, groundball rate) Felix is already pitching better than Verlander. He strikes out more guys (8.6 vs. 8.2 per 9), walks less guys (2.6 vs. 3.3), gets a lot more grounders (59.5% vs. 44.3%) which means less chances of surrendering homeruns and more chances for double-plays to be turned behind him.

So why is Felix's ERA higher? Verlander has gotten way more help from his defense - of balls put in play against Verlander, 74.6% have been turned into outs, while only 66.0% have been turned into outs off of Hernandez. While pitchers have a small measure of control over this, it's mostly a function of the quality of defense played behind the pitcher plus random variation (i.e. "luck"). Seattle's defense has been below average this year, so while we can expect Felix's BABiP rate to regress to the mean some more, it probably will still be below average due to the M's inconsistent defense.

The other reason is that a much higher number of flyballs allowed by Felix have gone over the fence than is typical (16.3%, vs. a normal rate of around 10%). Again, this is a stat that is primarily out of the control of the pitcher and mostly due to luck and ballpark configuration.

Also, I think you are really, really stretching it when you say that Peasant Felix has the best stuff in the majors. Verlander already has one of the top 3 curveballs in baseball, and can top 100 mph whenever he wants to, even in the 9th (102 anyone?). Not only that, his changeup is already good, and getting better as he learns it. Even if Felix reaches his potential, he'd have a hard time matching that kind of stuff.

First off, I said I don't think anyone in baseball has better stuff than Felix. I do think a small number of pitchers have roughly equal stuff to Felix, Verlander being one of them (I'd rate Rich Harden, when healthy, up there as well, and you obviously have to consider Johan Santana in this group).

Felix, when he's healthy, throws his 4-seamer in the high 90's, occasionally hitting triple digits. His 2-seamer sits in the mid 90's with excellent run and sink. His curveball is outstanding - it's a power-curveball in the Doc Gooden mode and I'd match it with just about any other pitcher's curve. His slider is also filthy, and his changeup is pretty good.

Felix already has stuff as good as anyone in the game (and you'd have a real hard time finding a scout that disagreed with this statement), and with experience his changeup should improve from being good to outstanding, and he may yet gain very good command of his fastball.

The only things holding Felix back right now have been his elbow injury, his predictable pitching patterns and poor defense and bad luck behind him.

[cthomer5000]

I dont even like baseball, and this is a great story. It reinforces a lot of stuff talked about in moneyball about just how resistant many in the game are to new ways of looking at things, regardless of how much info can be plainly presented.

[MrBigglesworth]

Originally Posted by dawgfan

So why is Felix's ERA higher? Verlander has gotten way more help from his defense - of balls put in play against Verlander, 74.6% have been turned into outs, while only 66.0% have been turned into outs off of Hernandez. While pitchers have a small measure of control over this, it's mostly a function of the quality of defense played behind the pitcher plus random variation (i.e. "luck").

I'm kind of new to the BABiP way of looking at things, but is this really true? Sure defense plays a big part, but isn't it also a function of how hard balls are hit, something that pitchers have a lot of control over?

[Karlifornia]

lol, I just have to say Schmidty's "Peasant Felix" crack didn't go unappreciated.

[Chief Rum]

Originally Posted by MrBigglesworth

I'm kind of new to the BABiP way of looking at things, but is this really true? Sure defense plays a big part, but isn't it also a function of how hard balls are hit, something that pitchers have a lot of control over?

That sorta thinking is where I am usually headed, too. These guys like to rip on Jered Weaver and his flyball tendencies, calling him lucky. I personally think the way his ball moves and how well he places it, he gets a lot of flyballs, but few that hitters can get a good bat on. The result is more "luck" than their numebrs suggest he should have. Luck my ass.

Too much luck, IMO, goes into those numbers when they don't get the results they want or expect.

[Chief Rum]

dola,

Not to suggets the math is bad. I'm fairly familiar with the math and the reasonings behind it for much these non-traditional statistics and they are very sound and intriguing. I just don't like that luck suddenly becomes a major factor whenever results don't match expectations.

[dawgfan]

Originally Posted by Chief Rum

dola,

Not to suggets the math is bad. I'm fairly familiar with the math and the reasonings behind it for much these non-traditional statistics and they are very sound and intriguing. I just don't like that luck suddenly becomes a major factor whenever results don't match expectations.

Here's the thing - DIPs theory, when it was first postulated by Voros McCracken, held that pitchers had no control over the results of balls in play, that it was simply a function of defense and random variation (which can also be called "luck").

Tom Tippett had a hard time believing this could be true (like almost everyone else when first presented with this theory) and did some more extensive research. What he found was that, in fact, some pitchers did exert some measure of control over the results of balls put in play. But the amount of influence in these cases is still very, very low in comparison to things like K rate, BB rate and groundball rate.

So the end result is that on average, pitchers yield about a .300 average on balls in play. That rate can vary of course, with the prime influence being the play of the defense behind the pitcher. Less of a factor, but still a significant one is random variation (or "luck"). The least factor is the pitcher himself.

When you consider the tremendous variety of outcomes that can happen to a ball put in play, and the fact that sometimes when a batter barely makes contact and the ball dribbles not far in front of the plate it can act like a perfect bunt and the player reaches on an infield single, while the next time at the plate he can hit the ball right on the sweet spot and squarely and send it directly at a fielder for a line out, it becomes easier to accept the notion that DIPs theory is essentially correct - that the pitcher has very little influence on whether balls in play go for hits or not.

Like it or not, random variation, i.e. luck really does play a role in outcomes in sports, and often times luck doesn't even out over the course of a season for an individual player.

[MrBigglesworth]

Originally Posted by dawgfan

When you consider the tremendous variety of outcomes that can happen to a ball put in play, and the fact that sometimes when a batter barely makes contact and the ball dribbles not far in front of the plate it can act like a perfect bunt and the player reaches on an infield single, while the next time at the plate he can hit the ball right on the sweet spot and squarely and send it directly at a fielder for a line out, it becomes easier to accept the notion that DIPs theory is essentially correct - that the pitcher has very little influence on whether balls in play go for hits or not.

Using that logic, a batter has very little to do with whether or not he gets a hit besides making contact, ie, controlling for running speed, all batters should have the same BA for balls put into play. That doesn't intuitively seem correct to me, I would think that a batter who consistently hits the ball harder will have a higher batting average.

[BigPapi]

Originally Posted by MrBigglesworth

Using that logic, a batter has very little to do with whether or not he gets a hit besides making contact, ie, controlling for running speed, all batters should have the same BA for balls put into play. That doesn't intuitively seem correct to me, I would think that a batter who consistently hits the ball harder will have a higher batting average.

Not sure how you're drawing that conclusion- as the batter actually swings the bat- not the pitcher. Of course the batter has some control of how hard a ball is hit and where it is hit- the pitcher has no control of how hard a ball is hit although he may have some control over where it is hit- but Jeter goes the other way on inside pitches all the time- so it depends on the batter.

The key to DIPS theory- as many have pointed out- is the seeming indifference the theory has to the particular pitcher; That is, most pitches hover around that .300 BABIP- or pretty close too it.

The same cannot be said for hitters. If you doubt it- check out Ryan Howard's batting average on balls in play vs. say Juan Pierre or Ichiro for example. There are literally dozens of examples like this. Jim Thome and Adam Dunn are two other examples off the top of my head that are members the seemingly growing category of "The three true outcomes" (those being the walk, strikeout and homerun).

[dawgfan]

Originally Posted by MrBigglesworth

Using that logic, a batter has very little to do with whether or not he gets a hit besides making contact, ie, controlling for running speed, all batters should have the same BA for balls put into play. That doesn't intuitively seem correct to me, I would think that a batter who consistently hits the ball harder will have a higher batting average.

Yeah, it's a tough one to wrap your head around. Data seems to show this is true though, that batters have more control over whether balls put in play go for hits than pitchers do. I believe a lot of that difference is due to what you point out, the ability of an individual batter to consistently hit the ball hard (line drives, well-hit fly balls). But even then, as I was mentioning above, sometimes you hit the ball hard but right at people. That was part of Richie Sexson's struggles to start the season - he was hitting the ball hard, but right at people, and his BABiP was very low. As random variation regressed him to the mean, his average started to climb (though he's slumped again lately).

I think one of the overlooked factors of DIPs theory is this - that the theory is based off of an assumption of "major-league level ability". In other words, it's not really true that pitchers can't control the outcome of balls put in play - it's more accurate to say that the ones that tend to be best at this rise to the majors, and this sub-section of all pitchers has reached a level of ability that tends to center around a .300 BABiP, with some pitchers slightly better and some slightly worse. Pitchers that consistently surrender hard hit line drives and fly balls find themselves no longer in the majors.

[st.cronin]

Originally Posted by dawgfan

Here's the thing - DIPs theory, when it was first postulated by Voros McCracken, held that pitchers had no control over the results of balls in play, that it was simply a function of defense and random variation (which can also be called "luck").

Tom Tippett had a hard time believing this could be true (like almost everyone else when first presented with this theory) and did some more extensive research. What he found was that, in fact, some pitchers did exert some measure of control over the results of balls put in play. But the amount of influence in these cases is still very, very low in comparison to things like K rate, BB rate and groundball rate.

So the end result is that on average, pitchers yield about a .300 average on balls in play. That rate can vary of course, with the prime influence being the play of the defense behind the pitcher. Less of a factor, but still a significant one is random variation (or "luck"). The least factor is the pitcher himself.

When you consider the tremendous variety of outcomes that can happen to a ball put in play, and the fact that sometimes when a batter barely makes contact and the ball dribbles not far in front of the plate it can act like a perfect bunt and the player reaches on an infield single, while the next time at the plate he can hit the ball right on the sweet spot and squarely and send it directly at a fielder for a line out, it becomes easier to accept the notion that DIPs theory is essentially correct - that the pitcher has very little influence on whether balls in play go for hits or not.

Like it or not, random variation, i.e. luck really does play a role in outcomes in sports, and often times luck doesn't even out over the course of a season for an individual player.

Of course, a more scientific way of interpreting the data is that "the pitchers ability to control the result of balls in play is not predictable." Of course pitchers have control over this. Its just that, in most cases, its not a "skill" that you can measure with stats or forecast, as opposed to something like batting average. I haven't looked at Tippet's article in some time, but I think he identified several classes of pitchers where it is in fact predictable - knuckleball pitchers, and pitchers who throw a lot of changeups. It might have been somebody other than Tippett that I'm remembering, though.

[dawgfan]

Originally Posted by st.cronin

Of course, a more scientific way of interpreting the data is that "the pitchers ability to control the result of balls in play is not predictable." Of course pitchers have control over this. Its just that, in most cases, its not a "skill" that you can measure with stats or forecast, as opposed to something like batting average. I haven't looked at Tippet's article in some time, but I think he identified several classes of pitchers where it is in fact predictable - knuckleball pitchers, and pitchers who throw a lot of changeups. It might have been somebody other than Tippett that I'm remembering, though.

Yeah, Tippet's research showed knuckleballers tend to have lower BABiP. However, it's not accurate to say they are more predictable - they still have random variation in their results. It's just that the baseline for BABiP for knuckleballers is usually a bit lower than for other types of pitchers.

[Huckleberry]

Originally Posted by dawgfan

I think one of the overlooked factors of DIPs theory is this - that the theory is based off of an assumption of "major-league level ability". In other words, it's not really true that pitchers can't control the outcome of balls put in play - it's more accurate to say that the ones that tend to be best at this rise to the majors, and this sub-section of all pitchers has reached a level of ability that tends to center around a .300 BABiP, with some pitchers slightly better and some slightly worse. Pitchers that consistently surrender hard hit line drives and fly balls find themselves no longer in the majors.

This is, IMO, the key that needs to be repeated more often.

The theory doesn't, or shouldn't, state that pitchers have little or no control over BABiP. It's that they're all bunched together closely. The argument that there is no control is absurd. Taking that logic to its natural conclusion would mean that major league hitters would have a BABiP of about .300 against me if I took the mound tomorrow.

It also doesn't jive with the data that states that pitcher do have control over the number of home runs they allow. This statistical fact indicates that pitchers do have control over how hard the ball is hit to at least some degree. It just seems intuitive that higher strikeout, lower home runs allowed pitchers will have lower BABiP. The higher strikeouts and lower home runs would indicate that they are allowing less solid contact.

But somehow the numbers don't add up. So I guess we'll just have to go with W/L record as the ultimate judge of a pitcher's performance.

[dawgfan]

Originally Posted by Huckleberry

It also doesn't jive with the data that states that pitcher do have control over the number of home runs they allow.

I don't have a Baseball Prospectus account, and I haven't followed all the studies as closely as others so I'm curious:

When McCracken first proposed DIPs theory, it proposed that pitchers have some degree of control over home runs allowed (though less control than over K rate and BB rate). Since then, there seems to have been an evolution of this thought that says the control over HR rates is really a function of how many fly balls a pitcher gives up, with ballpark factors and random variation having significant influence but the general rate being around 10% of flyballs going over the fence.

Has there been more detailed analysis of this? Obviously, the more flyballs you surrender, the more likely you would be to surrender some homeruns, but I do admit to some skepticism that there isn't some additional degree of influence pitchers have over HR rates above and beyond how many flyballs they surrender. I'm not suggesting this influence would be a large one, it could be quite small. I know that so far in his career, Felix has given up a fairly high percentage of homeruns given his low flyball rate - is that simply a function of random variation that will equalize over his career, or is more prone to give up no-doubt blasts when he makes a mistake up in the zone?

[oykib]

Originally Posted by Huckleberry

This is, IMO, the key that needs to be repeated more often.

The theory doesn't, or shouldn't, state that pitchers have little or no control over BABiP. It's that they're all bunched together closely. The argument that there is no control is absurd. Taking that logic to its natural conclusion would mean that major league hitters would have a BABiP of about .300 against me if I took the mound tomorrow.

It also doesn't jive with the data that states that pitcher do have control over the number of home runs they allow. This statistical fact indicates that pitchers do have control over how hard the ball is hit to at least some degree. It just seems intuitive that higher strikeout, lower home runs allowed pitchers will have lower BABiP. The higher strikeouts and lower home runs would indicate that they are allowing less solid contact.

But somehow the numbers don't add up. So I guess we'll just have to go with W/L record as the ultimate judge of a pitcher's performance.

Isn't BABiP for balls that don't go over the fence? I think if you had an incompetent pitcher, his BABiP might still be around .300. It's just that it would be "Lima Time" all the time.

[John Galt]

Just to address some confusions in this thread.

The original DIPS theory (and all subsequent versions) held that pitchers had control over the 3 true outcomes (K, BB, and HR), but not of balls in play. Excluding knuckleball pitchers, GB/FB ratio helps account for most of the other pitchers out there in a modified DIPS theory (Weaver or Young on one extreme, Wang and Webb on the other).

DIPS has NO relation to a hitter's ability to control outcomes. All hitters do have the ability to control the outcome beyond the 3 true outcomes. DIPS is only a theory for pitchers.

[bhlloy]

Originally Posted by John Galt

DIPS has NO relation to a hitter's ability to control outcomes. All hitters do have the ability to control the outcome beyond the 3 true outcomes. DIPS is only a theory for pitchers.

Hitters don't control the number of HR's they hit? I'm a complete newbie to this kind of thing, but that doesn't seem right to me...

[MrBigglesworth]

I still can't wrap my head around the notion that pitchers have no control over how hard a ball is hit, but batters do. Just from general observance, it seems that a batter makes better contact against a random pitcher than a cy young quality pitcher. That's not the case when the stats are analyzed?

[st.cronin]

Originally Posted by John Galt

The original DIPS theory (and all subsequent versions) held that pitchers had control over the 3 true outcomes (K, BB, and HR), but not of balls in play.

I think this is an overstatement. I think the theory says that a pitchers ability to control the outcome of balls in play is not something that shows up in the stats. It is something they can control - this is just not a skill that you can predict.

[dawgfan]

Originally Posted by st.cronin

I think this is an overstatement. I think the theory says that a pitchers ability to control the outcome of balls in play is not something that shows up in the stats. It is something they can control - this is just not a skill that you can predict.

McCracken's original study was too limited, and as a result was his postulation that pitchers (at a major-league level it should be noted) didn't seem to have a consistent ability to prevent hits on balls in play. Tippet's subsequent study (which went into much more depth) showed that some pitchers did seem to show an ability to prevent hits on balls in play relative to the average major-league pitcher, and that knuckleball pitchers seemed to usually fall in the group of pitchers that showed an ability to prevent hits.

With all of the study that's happened since, I think the most accurate way to describe the data is to say that, if a pitcher reaches the major leagues, his ability to prevent hits on balls in play seems to sit at around a .300 BABiP figure. A few pitchers are a little better at preventing hits on balls in play, and many are worse (and those that are usually find themselves back in the minors before long).

However, due to the vagaries of defense behind a pitcher and random variation, the actual results for an individual pitcher can vary a fair amount around that .300 BABiP figure from year to year - it takes more than just a single season's worth of results for random variation to normalize enough for a true baseline to appear.

Even with the revamping of the DIPs theory that shows some pitchers do have some ability to prevent hits, that ability pales in significance to the ability of a pitcher to control strikeout rates, walk rates and groundball rates.

A pitcher can get away with average or even slightly below average ability to prevent hits if he limits his walks, induces a lot of groundballs and has a high strikeout rate. Conversely, even if a pitcher shows an ability prevent more hits than the average major league pitcher (as a rate), he can still suck if he doesn't strike out many batters, walks a lot and gives up a ton of fly balls (and thus more home runs).

[dawgfan]

Originally Posted by MrBigglesworth

I still can't wrap my head around the notion that pitchers have no control over how hard a ball is hit, but batters do. Just from general observance, it seems that a batter makes better contact against a random pitcher than a cy young quality pitcher. That's not the case when the stats are analyzed?

No. To help wrap your head around this, consider that no pitcher allows as many home runs per plate appearance as Barry Bonds hits, and no pitcher allows as few home runs per plate appearance as Willie Bloomquist hits. No pitcher strikes out as many batters per plate appearance as Adam Dunn strikes out, or strikes out as few guys as Placido Polanco strikes out per plate appearance, and no pitcher walks as many batters per plate appearance as Barry Bonds or as few as Ivan Rodriguez walks.*

* Note that this is limiting results to those players with a significant number of plate appearances

The data shows that there is not much difference, if analyzed long-term, between the ability of Jeff Weaver to prevent hits on balls in play as compared to Pedro Martinez. There is some difference, but it is quite a bit smaller than the ability of pitchers to control strikeout rates, walk rates and groundball rates.

[MrBigglesworth]

Doing some reading at this site:

http://www.baseballthinkfactory.org/..._2004-02-29_0/

To summarize the implications of the above chart, even though the y-t-y $H correlation of a 500 BIP pitcher is very small, a pitcher may have a fair amount of control over certain components of those BIP. The regression results suggest that good pitchers give up slightly fewer line drives and slightly more pop flies, as a percentage of their total BIP, and that their line drives hit to the outfield (and perhaps their ground balls) may be softer and therefore easier to catch.

...which is along the lines of what has been said here if I am understanding things correctly, that pitchers have a tiny but noticeable difference of what type of balls are hit, ie, better pitchers will have more pop-ups than line drives.

[John Galt]

Originally Posted by bhlloy

Hitters don't control the number of HR's they hit? I'm a complete newbie to this kind of thing, but that doesn't seem right to me...

That was my bad phrasing. Hitters control the 3 true outcomes AND all balls in play.

[John Galt]

Originally Posted by st.cronin

I think this is an overstatement. I think the theory says that a pitchers ability to control the outcome of balls in play is not something that shows up in the stats. It is something they can control - this is just not a skill that you can predict.

That was my bad phrasing as well. My sentence was only true of the original DIPS theory and not all subsequent ones.

I'm not sure what you mean by "not something that shows up in the stats." If pitchers could control the outcome of balls hit in play on a regular basis, it would show up in BABIP. However, very few pitchers consistently outperform league average BABIP. Those that do are almost always knuckleball pitchers or have a strong GB/FB skew.