The Statistical King
You guys seemed to enjoy the breakdowns I did on Gil Meche recently, so let’s writeup King Felix from a similar perspective. After another good outing last night, everyone seems to agree that the King is back on track and pitching like we expected from the outset. So let’s break him down by month, and see where the improvement has come from.
Month IP BB% K% GB% FB% LD% HR/FB FIP xFIP ERA April 26.2 10% 23% 56% 25% 19% 25% 4.89 3.52 5.06 May 35.2 7% 22% 53% 25% 22% 21% 4.35 3.29 6.31 June 34.2 4% 20% 59% 23% 18% 13% 3.17 3.04 3.37 July 26.1 12% 19% 59% 27% 15% 15% 4.79 4.40 3.42 August 13 10% 17% 70% 16% 14% 0% 2.97 3.63 1.38
Lots of numbers there, so, here’s a quick overview, by component, and what it’s meant to his run prevention by month:
BB% – He threw a lot of strikes in June, but other than that, he’s still struggled with his command. 10% walk rates aren’t death, but they’re still something that can be improved upon. He’s not really cutting down on his walks lately, so this isn’t the key to his improvement.
K% – He’s actually missing less bats as the year goes on, as he’s struck out fewer batters each month than the previous. His 18% strikeout rate the last 5 weeks is actually just a tick above average. This certainly isn’t the cause for the improvement.
GB%/FB%/LD% – He’s getting more groundballs, less fly balls, and less line drives lately, and that has certainly helped. This is definitely part of the improvement, but the difference isn’t big enough to explain the rebound entirely. The LD% is probably due for a spike going forward, as 14% line drives isn’t sustainable.
HR/FB% – And here’s 95% of the improvement, right here. A league average starting pitcher will allow 11-12% HR/FB rate, and significant variance from that is in almost every case unsustainable. Felix’s HR/FB rate was an absurd 25% in April, a still awful 21% in May, then an almost normal 13% in June, slightly high 15% in July, and he has yet to give up a home run in either start in August. The August number isn’t an abberation, as you might think from the percentage, since he’s only given up six fly balls total in his last two starts.
His Fielding Independent ERA basically tracks the difference in run prevention if his HR/FB rate was indeed a true skill, and you can see how big of a difference home runs make. xFIP translates HR/FB rate into a league average number, eliminating that “skill” from the analysis, and gives a pretty clear picture that Felix hasn’t really changed much at all. Outside of July, he’s posted consistent xFIPs in the 3.00-3.60 range, which is outstanding. For comparison, Johan Santana’s xFIP for the season is 3.42. Roy Halladay’s is 3.57. That’s the class of pitcher that Felix has been in most of the year, by xFIP.
This is about as clear a picture as one could paint for why ERA is not a good tool for predicting pitcher performance going forward. The only thing that has significantly changed for Felix from April to August is the amount of his flyballs that go over the wall. That entire change, mostly out of his control, has led to the mirage of improvement.
Felix has been pitching mostly well all year. ERA just didn’t do a good job of telling people that.
Interesting.
I would guess that part of that has to do with the flukish year-to-year correlation of the HR allowed stat. Some years a pitcher may make more mistake pitches but they don’t get hit out of the yard, some he may make a few less but they get hit more often.
The greats though, will consistently make less mistakes and I’d imagine it would show if you studied a group of years.
In other words, if you look at a pitcher’s 3-5 year prime I’d be willing to bet that you could find a bunch of pitchers that suppressed HR rate at an above-average rate despite being flyball pitchers.
Off the top of my head, Pedro in his prime supressed HRs despite Fenway Park.
In other words, if you look at a pitcher’s 3-5 year prime I’d be willing to bet that you could find a bunch of pitchers that suppressed HR rate at an above-average rate despite being flyball pitchers.
No, really, you can’t. We’ve studied this. We’ve tried. Its. not. true.
Off the top of my head, Pedro in his prime supressed HRs despite Fenway Park.
I’d imagine Fenway was a big part of Pedro suppressing HRs. The Monster turns a lot of homers into doubles.
Just wondering, almost hate to ask, but have your expectations for Felix’s career diminished a little bit since last year when he was posting high ground ball rates AND high strikeout rates?
No. He’s a 20 year old who is already one of the ten best pitchers in the American League.
Last year, we had Peak Felix for 12 starts. We’ll see those kinds of runs again.
Thanks, Dave.
Well, my next question would be.
How are you calculating fly balls? Are infield fly outs, and pop ups counted as flyballs? Or are the flyballs you are calculating only of the outfield type?
If its the latter, then yes flyball rate will have some predictability with HRS. But then again so would infield outs and pop ups similarly correlate with fly outs, since that is essentially saying the same thing. The difference between popping it up as to hitting it to the outfield is small fraction of where you make contact, and the difference between a fly ball and a HR is also a small fraction.
If you are willing to seperate flyballs from infield pop ups, then logically you have to be willing to seperate flyballs from HRs.
Yes, if a pitcher gives up a bunch of outfield outs as opposed to infield outs and pop outs, he’ll have a more difficult time suppressing HRs (since guys are making decent contact with those flyball outs), but it also doesn’t mean that there aren’t pitches that do exactly that (like Clemens, Lackey as you’ve mentioned).
Well, you’d have to agree that Fenway is more homer prone than your average park. Despite this Pedro suppressed HRs during his prime. (Although I have no idea whether Pedro was statistically a flyball pitcher by your definition of flyballs.)
How are you calculating fly balls? Are infield fly outs, and pop ups counted as flyballs? Or are the flyballs you are calculating only of the outfield type?
Infield flies and popups are not included in the HR/FB calculation.
If you are willing to seperate flyballs from infield pop ups, then logically you have to be willing to seperate flyballs from HRs.
No you don’t.
Yes, if a pitcher gives up a bunch of outfield outs as opposed to infield outs and pop outs, he’ll have a more difficult time suppressing HRs (since guys are making decent contact with those flyball outs), but it also doesn’t mean that there aren’t pitches that do exactly that (like Clemens, Lackey as you’ve mentioned).
There’s Clemens. There might be Lackey, who we need more time to watch and see if he can sustain his recent performance. And there’s no one else.
I’m sorry, but 1.5 exceptions to the rule do not invalidate the rule.
Well, you’d have to agree that Fenway is more homer prone than your average park. Despite this Pedro suppressed HRs during his prime. (Although I have no idea whether Pedro was statistically a flyball pitcher by your definition of flyballs.)
If you don’t know what his HR/FB rate was during that time, and we know now that its certainly not below the average (2002-2006 data is available), then why should we assume that his HR/FB rate was lower than average for the periods we don’t have data for?
Of course you do.
If the argument is that the difference between a 250 foot fly ball (an outfield out) and a 400 foot flyball (a HR) is all due to “luck”. Then the difference between a 50 foot infield fly out or foul out and an outfield flyout, would logically also have to be due to “luck”.
In both cases it would just be a fraction of a difference in where you make contact.
#56: I don’t know about when Pedro was there, but the last couple of years, Fenway has been death on home runs.
Dave, is HR/FB truly random, or is it BABIP-like random, in which pitchers have a little control, but luck often drowns it out? Also, do park factors alter HR/FB rates?
If the argument is that the difference between a 250 foot fly ball (an outfield out) and a 400 foot flyball (a HR) is all due to “luckâ€. Then the difference between a 50 foot infield fly out or foul out and an outfield flyout, would logically also have to be due to “luckâ€.
If we had ball velocity, arc, and travel distance, I’d gladly do more analysis on the topic. We don’t.
What we do have is statistical proof that starting pitchers have very little control over the result of their outfield fly balls. That is up to the hitter, not the pitcher.
Dave, is HR/FB truly random, or is it BABIP-like random, in which pitchers have a little control, but luck often drowns it out? Also, do park factors alter HR/FB rates?
It’s more BABIP-like random. Like I mentioned earlier, extreme GB pitchers will give up slightly more HR/FB than extreme FB pitchers. But the differences aren’t huge, and most pitchers hover around the norm.
And yes, park factors have an impact. Safeco turns flyballs into outs far more often than most parks, so the M’s pitchers get a HR/FB boost.
Yes, but in seperating infield outs and foul outs from outfield outs the statistical community is also, strangely, admitting that pitchers have SOME control over the outcome of how a fly ball is hit.
So you seperate 50 foot flyball outs from 200 foot flyball outs, but 400 foot flyballs that go out of the park are completely disregarded as a reliable statistic? I just don’t follow the logic there. The diffence in quality of contact between a infield fly out and an outfield fly out is similar to that of the difference between a routine fly out and a HR.
Granted if that is the way your calculating “flyouts”, then yes it will be more difficult to find guys that suppress homeruns because an outfield flyout is decent contact in comparison to a foul out or an infield fly out. The exact opposite could be said as well, that high HR rates correlate to a high percentage of outfield flyouts.
But it is counter productive to completely disregard the “HRs allowed” category as luck, for the same reason that it is counter productive to disregard an infield pop up as luck. There are pitchers that suppress HRs despite high HR rates, you’ve even mentioned a couple outliers youself.
So you seperate 50 foot flyball outs from 200 foot flyball outs, but 400 foot flyballs that go out of the park are completely disregarded as a reliable statistic? I just don’t follow the logic there. The diffence in quality of contact between a infield fly out and an outfield fly out is similar to that of the difference between a routine fly out and a HR.
That’s a theory unsupported by evidence.
But it is counter productive to completely disregard the “HRs allowed†category as luck, for the same reason that it is counter productive to disregard an infield pop up as luck. There are pitchers that suppress HRs despite high HR rates, you’ve even mentioned a couple outliers youself.
Can I suggest you do a little more reading on the subject? Pick up a copy of the 2006 Hardball Times Annual and read through the analysis section.
Being able to statistically follow ball velocity, travel arc, and distance of batted balls would be truly revolutionary. Hopefully someday we get there.
It would be more reliable in predicting the future performances of established major league hitters and pitchers than any statiscian or scout out there.
Being able to statistically follow ball velocity, travel arc, and distance of batted balls would be truly revolutionary. Hopefully someday we get there.
Absolutely. I’ve said before that I think that’s the next great leap in evaluating players.
I’ve read Hardball. This isn’t a question of “evidence”, this is a question of “logic”.
If your willing to seperate infield outs from outfield outs, you also logically have to be willing to seperate outfields outs from ballpark outs (lol).
Like I said before its a useful stat to look at, just because its so difficult to suppress homeruns when your giving up a bunch of flyball outs (which is usually decent contact – when infield flyouts, pop outs, and fouls outs are taken out). Year to year you can identify certain flyball pitchers that have been getting “lucky” or “unlucky” with their HR rate.
But you also can’t take a quantum jump and use that to conclude that flyball pitchers have no control over their homerun rates.
But you also can’t take a quantum jump and use that to conclude that flyball pitchers have no control over their homerun rates.
I’m not making that statement. I already mentioned that extreme GB and FB pitchers will vary slightly from the norm. That was the second comment in this thread.
And your theory that infield flies and outfield flies are the result of microscoping differences in swing not only has no supporting evidence, but it doesn’t ring true to me. Infield popups are swings on an incorrect plane, whereas outfield outs are often balls that are off the of the end of the bat.
In my observation, outfield flies aren’t the result of an incorrect plane on a swing. So I disagree with your entire “logical premise”.
And you still have no evidence to support your position.
Also, if you’ve read the articles in the THT annual, and you still believe what you believe, I suggest you re-read them.
And Homeruns are the difference between guessing the right speed and/or location and hitting the sweet part of the bat, as opposed to guessing right and missing or guessing wrong and adjusting late when you hit a fly out.
Of all outs outfield outs are the “closest” to resulting in a Homerun. But the difference between guessing wrong and missing is just as different from guessing right and hitting. The quality of an atbat between a Homerun and a flyball out isn’t all that from that of an infield out and a flyball out.
After reading you last post though I think we’re agreeing more than disagreeing.
I guess the point where we may differ is that yes outfield flyouts correlate to HR rate. But unlike ERA where peripherals are more predictive of ERA than ERA, I’d bet that a pitcher’s established HR rate is more predictive of his HR rate than his flyball out % is predictive of his HR rate.
Not to sidetrack the discussion, but I want to go back to something Dave said:
Strikeout rate goes down for most pitchers as they age, not up. Stuff deteriorates, command improves, and they mature from higher BB/K guys to lower BB/K guys. The great ones can sustain their higher K rates, but almost no one strikes out more batters as they get older.
For some reason, this worries me. For most pitchers, strikeouts tend to start declining in their mid-20s. I worry that Felix is going to be an extreme GB pitcher like Chien-Ming Wang before his six years are up, and I would feel a little better if he was running at 7-8 K’s per 9 right now rather than 5.
I guess the point where we may differ is that yes outfield flyouts correlate to HR rate. But unlike ERA where peripherals are more predictive of ERA than ERA, I’d bet that a pitcher’s established HR rate is more predictive of his HR rate than his flyball out % is predictive of his HR rate.
I can tell you with certainty that this isn’t true. The data is in the THT annual.
In other words, if you look at a pitcher’s 3-5 year prime I’d be willing to bet that you could find a bunch of pitchers that suppressed HR rate at an above-average rate despite being flyball pitchers.
Dave,
What I read here from taro, in my opinion, actually does have a ring of truth, but not in the same way that I believe he expected it to be taken. It’s more of a fallacious ‘truth,’ so to speak. It’s begging the question.
When most people take a pitcher’s prime years, they would select those with the best results. The best results are most likely to occur when ‘semi-random’ or ‘uncontrollable’ stats such as HR/FB end up on the good side. Thus, more often than otherwise average, the selected years will have nice posted HR/FB ratios.
Unfortunately, that kind of begging the question happens a lot when looking at the result and assuming it is the reason.
In that sense it’s also affirmation of the consequent:
If he were in his prime, he would have better HR/FB rates.
He had better HR/FB rates.
Therefore, he must have been in his prime.
In either manner, it’s a subjective sort of reasoning that really has no legitimate explanation. As they’d say in logic; it can be a true statement, but the arguments are not sound.
Could be true, although I’m highly suspicious. Its an interesting enough debate that I think I’ll reread the article tonight or tommorow night and get back to you. I remember disagreeing with the process and identifying some holes in the analysis the first time I read through, but I may have missed something.
On a player by player basis though, I think there is too much “noise” in the flyball statistic (hard hit balls, weakly hit balls, good and bad contact, deep fly balls vs shallow fly balls) to rely on. I’m more prone to look at how many actaul HRs the guy gives up as opposed to theoretical HRs – and then from there its scouting – why or why hasn’t this guy been “unluckly” or “lucky”.
Felix for example was definetly not “unluckly” earlier in the year. He was a bit unlucky from a base hit standpoint, but those HRs were BLASTED. Opposing hitters were sitting dead fastball and LAUNCHING it. Maybe a couple HRs were unlucky, but Felix fully earned his gopheritis in the first half of the season.
Its one of the reasons I HATE the xFIP stat. According to xFIP Felix was an impact starter in the first half, but that couldn’t be further from the truth. Many times a pitcher is earning his HR rate good or bad, and you need to scout the pitcher to get a real idea whether its “luck” or not. (In Felix’s case, you came to the same conclusion yourself scouting him earlier in the year…but now its luck? or did I misread you?)
I also remember you guys targetting Weaver early in the year because of his supposed bad luck with flyball %. Honestly, thought you guys were joking till I read the comments… Weaver’s gopheritis is continuing in St. Louis and he will continue to get punished. Weaver isn’t giving up HRs because hes unlucky with his flyball%, hes giving up HRs because now that hitters are used to his awkward delivery, he simply doesn’t have the stuff to get by anytmore. Weaver doesn’t have a single plus pitch, his slider is mush, and theres no zip to his fastball anymore. Without any weapons, hitters are knocking each other over to get to the plate against Weaver.
I also remember one of you guys being interested in Green as a callup despite his medicore K/9 and BB/9, solely because of the interest in his GB%. Sean Green hasn’t really suppressed HRs and has been a marginal Major League reliever overall.
Flyball % is interesting and I keep track of it, but it personally takes a back seat to the HR stat (as well as BB/9 and K/9) for me.
The bottom line is, taro, you’re arguing what the reality ought to be, rather than what it is — you’re trying to trump data with logic — and that just won’t work. When the data disprove a logical argument, however impeccable the logic, the data win.
Oh, and btw, on Jeff Weaver: his HR/FB is 15.9%, not much out of normal. He was unlucky with his HR/FB before, and that has evened out to some degree; he’s still getting blasted, yes, but it’s because he’s giving up a huge number of fly balls. As such, he doesn’t constitute a counterpoint against Dave.
On a player by player basis though, I think there is too much “noise†in the flyball statistic (hard hit balls, weakly hit balls, good and bad contact, deep fly balls vs shallow fly balls) to rely on. I’m more prone to look at how many actaul HRs the guy gives up as opposed to theoretical HRs – and then from there its scouting – why or why hasn’t this guy been “unluckly†or “luckyâ€.
Look, neither HR/9 or HR/FB are perfect. Both have flaws – I could put out a laundry list of problems with HR/9 and show its poor year to year correlation, but I doubt that would sway you. HR/FB is better than HR/9 – that’s not really a disputable fact. If you want to use HR/9, feel free – you’ll be wrong more often than I’m willing to be, though.
Its one of the reasons I HATE the xFIP stat. According to xFIP Felix was an impact starter in the first half, but that couldn’t be further from the truth. Many times a pitcher is earning his HR rate good or bad, and you need to scout the pitcher to get a real idea whether its “luck†or not. (In Felix’s case, you came to the same conclusion yourself scouting him earlier in the year…but now its luck? or did I misread you?)
According to xFIP, Felix was likely to be an impact starter going forward. xFIP is a predictive stat, not a value stat. There is a massive difference. No one uses xFIP to hand out Cy Young awards – we use it to project the future. If you don’t like xFIP because it doesn’t evaluate past results perfectly, then you’re complaining that your car doesn’t make you breakfast. Use it to do what it was designed to do, and it does great.
I also remember you guys targetting Weaver early in the year because of his supposed bad luck with flyball %. Honestly, thought you guys were joking till I read the comments… Weaver’s gopheritis is continuing in St. Louis and he will continue to get punished. Weaver isn’t giving up HRs because hes unlucky with his flyball%, hes giving up HRs because now that hitters are used to his awkward delivery, he simply doesn’t have the stuff to get by anytmore. Weaver doesn’t have a single plus pitch, his slider is mush, and theres no zip to his fastball anymore. Without any weapons, hitters are knocking each other over to get to the plate against Weaver.
Actually, the argument I made for Weaver was that he wasn’t stranding any runners. His LOB% in Anaheim was just 62.4%, a ridiculously low total that just wasn’t going to continue. Sure enough, his LOB% in St. Louis is 75.9% – it regressed to the mean, just like I said it would.
And I’ve consistently stated that HR/FB is only valid for major league quality pitchers. If you’re right (I don’t think you are) and Jeff Weaver is no longer a major league quality pitcher because his stuff has deteriorated past the point of return, then all bets are off. Of course, there’s pretty much no other indicators that his stuff has fallen off as much as you claim. In fact, his 2006 looks remarkably like his 2003, the last time everyone wrote him off as a major league pitcher, before he proved them wrong in ’04 and ’05.
I also remember one of you guys being interested in Green as a callup despite his medicore K/9 and BB/9, solely because of the interest in his GB%. Sean Green hasn’t really suppressed HRs and has been a marginal Major League reliever overall.
Honestly, it’s a little annoying having people try to restate my opinions to me when they don’t remember what they were. A little googling could turn up everything I’ve written about Sean Green. The evaluation was, and still is, that his extreme groundball tendencies will compensate somewhat for the low strikeout rate, and for the major league minimum, he could be an effective enough replacement level 6th reliever. That opinion is still true.
This whole argument reminds me of the people who keep sending me emails asking if I’m finally ready to admit I was wrong about Jose Lopez, despite the fact that pretty much everything I’ve ever written about him has turned out to be true. The archives of the site are easy to search – rather than trying to argue against what you vaguely remember me maybe saying, it’d be great if you could find out what I actually said before you tried to refute it.
Flyball % is interesting and I keep track of it, but it personally takes a back seat to the HR stat (as well as BB/9 and K/9) for me.
That’s a flaw in your analysis, then. HR/9 is less predictive of future home run rate than HR/FB rates (once adjusted for ballpark, anyways). It just is.
Felix has the heavy-moving-sinking fastball that is extremely difficult to hit. And this is the point that I think people are missing, it’s not if Felix can get good location (for example down and away) with this pitch, its whether he can get it across the plate, AT ALL. This is a pitch that just takes off on him and he has zero control of it at times. But I’ve rarely seen a clean hit off of this pitch, pretty much just groundballs, choppers, broken bats, etc.
The four-seam fastball that Felix throws is extremely straight and that is the pitch he used to go to when behind in the count. During the last start it almost seemed like Chaves told him to stop throwing it (basically telling him to actually work on the command of the other fastball he can’t locate). And I think this is a wise move. He may have games where he walks 3-6 hitters but he has to improve sometime. I think he got away with four-seam fastball at 96 on 2-0 counts in the minor leagues and it’s just not working anymore.
“The big thing last night was that the fastball was consistently down in the strikezone.” ” . . . Six flyballs in his last two starts.” “GB 70%” “He’s actually missing less bats as the year goes on.”
This is the difference in Felix, last year, to this Spring, to now: burying the fastball for the GB out. The guy who came up last year had his monster sinker just buried in the bottom of the zone, with movement, at 95+. He gave up ridiculously few HRs because, basically, nobody could lift his pitch. His K rate wasn’t extraordinary because he was getting so many GBs that he didn’t get to 3-2 that much. And even then, a lot of his Ks came off the change-up or curve.
Felix this Spring wasn’t commanding his sinker; it was up, out, in, sometimes at lower velocity, seldom with that vicious, Kevin Brown-like running movement down in the zone. When he came up in the zone with the pitch with the hitters sitting on dead red—and too often they were sitting on it because of bad pitch selection by the Ms to ‘establish the fastball’—ridiculous numbers of the FBs hit off him went over the fence. But that fits the ‘GB pitcher profile’ as you say: pitches up by a sinkerballer are mistakes by definiton, and mistakes tend to end up in the seats, especially mistakes in a zone where the hitter is looking for them.
Now, come July, Felix has finally gotten his arm loose, and his natural motion is giving him that wicked running sink again on the fastball. Yes, better pitch selection is making an impact, and will certainly be important in his development as a pitcher. The return of his ‘unliftable’ sinker is all the difference. I’m looking for Felix to run off another stretch through August like he did a year ago, maybe not quite as dominant, but with a very similar game result.
Felix’s offspeed pitches are, obviously, quite good enough for him to win games with them alone. When he’s more experienced, he’ll do this on days when his command or velocity is off on the sinker; for example, by throwing curve early in the count to get the called strike and keep the hitters off balance as was discussed here early in the year. For now, though, it’s the monster sinker which takes Felix from an outstanding prospect to an other-worldly prospect. His whole package keys off this pitch, and allows him to ‘out-stuff’ the other guys rather than outpitch them, which mentally he’s not quite there with yet.
Let’s see: Brown’s sinker + Pedro’s change-up and curve = what? Helluva package. But what’s missing is the mental part. The nasty, “Mere mortals cannot beat me” part. Brown had it; Pedro has a mountain of it. Randy Johnson for years _didn’t_ have it. All his time in Montreal, and his first years here, RJ lacked confidence on the mound. Then, he had a (what’s the nasty synonym, not antonym, of an epiphany?) change, where he took his always privately nasty attitude with him to the mound—and dominated.
Felix Hernandez is still a 20-year-old kid having the time of his life making more money than he ever dreamed of playing a game. Next year, or the next one, there will be a time when the mental part comes together for Felix; probably after a crisis of some kind, but I don’t know. Now, when things aren’t going well, he falls back on pride; he’s made remarks after some games that add up to that for me. When it comes together, he’ll be looking to _crush_ the other guy before it ever gets to the point of pride. He’ll start out hard. But that part of his game isn’t there yet. That’s OK: I’m enjoying watching the parts of his game that are here, now.
re: #31, I also don’t think Bavasi is going anywhere else this offseason, unfortunately, which is one reason why I keep piping up for him to go. Lincoln is likely to extend him, ’cause Lincoln doesn’t like disruptions, the budget is going to be met, and there will be no major collapse. To me, that’s how Lincoln thinks. Unbalanced skillsets and too few wins: these are distant things to Howard. 40K+ in the stands for the Oakland series most days: these are big things to Howard. If Felix breaks off a hot run while AB keeps hitting and the Ms creep a few games over .500 by 1 Oct, that’s plenty good enough for Lincoln, and Bavasi will mind the store for another year or two.
Well xFIP is fine if your loosely analyzing the future performance of a group of players. Flyball% may indeed be more predictive of homerun for a majority of MLB pitchers (still haven’t been convinced-but its still a subject under study).
The problem with the stat comes when its used dogmatically on a player by player basis.
Felix isn’t preventing homeruns now because xFIP said he was lucky in the the first half. Saying theres potential for more, and noting xFIP as “one” of your arguments is fine. But dismissing homeruns allowed as “unsustainable” fogs the analysis. Had Felix pitched the exact same way with the exact same stuff for the rest of his career he would have continued to have goperitis despite being a groudball pitcher. He hasn’t for reasons stated by you and Bela above. The hard sinking fastball is key to preventing HRs now, pitch selection will be key when he doesn’t have that pitch. But Felix’s improvement isn’t because xFIP predicted it – this is where scouting analysis has to come in play.
And as I said before I’m unconvinced that there aren’t exceptions to the rule like there were when McCracken discovered that pitchers had no control over whether balls in play fell for hits. Now of course they’ve discovered that some pitchers DO indeed have some control over balls in play. I imagine the same process is going to happen with HRs allowed.
Just from a local standpoint, how do you explain Jarrod Washburn? Gives up a ton of flyballs, but has a career league average HR rate despite it. The reason for this is because Jarrod Washburn has a moving fastball that cuts in different directions unpredictably. When he is effective he is throwing his fastball the majority of the time, and HIGH in the zone. Judging from his flyball% and pitching style, he SHOULD be projected for gopheritis EVERY year according to xFIP, but he never does give up a ton of HRS because its relatively hard for hitters to center his fastball. Would you expect 1.3+ HR/9 rates from Washburn for the next three years because xFIP predicts it, or would you just expect the same old league average HR rate hes always had (adjusted slightly for Safeco)?
My bad on Weaver. I’m not trying to prove you wrong (I’m not going to dig through the archives), but I’d still have to same objections if you we’re analyzing Weaver through LOB%.
Was Weaver unlucky? Yes. Take away the luck and is he still a terrible pitcher? Yup.
Its proven unrelated to the discussion, but Weaver is finished as an effective starter until he gains a plus pitch, suddenly develops Maddux command, or starts sipping some funky milkshakes. We’re both on record for that one.
Had Felix pitched the exact same way with the exact same stuff for the rest of his career he would have continued to have goperitis despite being a groudball pitcher.
But the point was that he wasn’t going to continude to do that. That’s the point here. A guy with that good of stuff (which xFIP identifies) isn’t going to continue to throw that many meatballs that get whacked over the fence. It just isn’t going to happen.
He hasn’t for reasons stated by you and Bela above. The hard sinking fastball is key to preventing HRs now, pitch selection will be key when he doesn’t have that pitch. But Felix’s improvement isn’t because xFIP predicted it – this is where scouting analysis has to come in play.
xFIP identified that he had the skills that made it very, very likely that he would improve. You didn’t need scouting analysis for that.
And remember, I’m one of the big proponants of the value of scouting. If you’re trying to paint me into a scouts vs stats debate, you’re barking up the wrong tree.
And as I said before I’m unconvinced that there aren’t exceptions to the rule like there were when McCracken discovered that pitchers had no control over whether balls in play fell for hits. Now of course they’ve discovered that some pitchers DO indeed have some control over balls in play. I imagine the same process is going to happen with HRs allowed.
We’ve already found exceptions. Clemens is the main one. And we’ve noted that extreme FB pitchers tend to give up slightly less HR/FB than extreme GB pitchers, due to the mistake-pitch issue. Why you continue to ignore the fact that we’re conceding those exceptions is beyond me.
Just from a local standpoint, how do you explain Jarrod Washburn?
Park factors. The HR/OF park factor for Edison Field is 90 and its a 95 for Safeco. He’s spent his entire career in parks that depress home runs, and he’s still barely beating the average.
The reason for this is because Jarrod Washburn has a moving fastball that cuts in different directions unpredictably.
Again, unsubstantiated theory. Why do you insist on believing things dogmatically that you have no possible way of knowing whether they’re true or not?
Was Weaver unlucky? Yes. Take away the luck and is he still a terrible pitcher? Yup.
I imagine you called Jeff Weaver terrible in 2003, too, when he posted a 5.99 ERA, correct? Did 2004 and 2005 teach you nothing?
The main issue of our disagreements is that you’re willing to believe things that you cannot factually substantiate, and you’re willing to ignore data that goes counter to those beliefs. That, to me, is mind boggling.
I’m not ignoring the data – I’ve conceeded that in many cases flyball% can be a good indicator of HR rate. Still, the three outcomes ARE BB rate, K rate, and HR rate. Where I disagree with the rest of the saber community is in how xFIP is making us disregard the HR allowed stat. Flyball% is now widely accepted as a better measure of HRs allowed, than actual HRs allowed is. Until we really can record in detail every batted ball thats not a transition I’m willing to follow – because it can lead to foggy analysis on a player by player basis.
xFIP is fine when used as “part” of an argument, but it is far from the perfect formula. It is something that can’t be used alone as evidence for how a pitcher will perform from here on out. In fact I prefer regular old FIP (because it forces you to analyze homeruns allowed through scouting – and make a more educated decision yourself of whether or not it is “luck”).
The chances that a talent like Felix wouldn’t have figured it out are extremely low, yes. The point is that the HR rate that Felix had in the first half was earned, it was not luck. You seem to be conceeding that point. You also seem to be conceeding the fact that there are exceptions to the rule, and that you need analyze both the numbers and the player – and now I’m not even sure what we’re debating about anymore.
As for the smaller debates:
-Even adjusting for park, Washburn’s HR/9 should be higher considering his low BBs, low Ks, and high flyball%. Despite Angel Stadium he should have been experiencing gopheritis according to xFIP. Watching him pitch my reasoning for this is the moving fastball. Unless you see him running 1.3 HR/9 rates in Safeco from here on out as xFIP would predict?
-Have to admit I didn’t follow Weaver much until I saw him pitch this year, but sometimes one or two times is enough. THAT dude is toast unless he gains a weapon or two in his arsenal. He has NO plus pitches and has turned into the Cha Seung Beck/Ryan Franklin mode of here-it-is-hit-it type pitchers. I remember him having a little more zip on his stuff early on, and he had that funky delivery working for him that hitters have now adjusted to. I see very little chance of him bouncing back without a transformation (and in transformation I don’t mean a new release point).
We’ll see what happens but I’ve already written him off as an effective starter.
Still, the three outcomes ARE BB rate, K rate, and HR rate.
HR% has a year-to-year correlation of .28. That’s not good. By comparison, BB% has a correlation of .69, and K% has a correlation of .77. Even pretending like HR rate is anything close to predictive as walks and strikeouts are is foolish.
By the way, the correlation for HR% when FB% is removed is .08, which is about as close to random as you’re going to get. You’re not willing to use FB% because it’s not perfect, but the tool that you are using is even worse.
-Even adjusting for park, Washburn’s HR/9 should be higher considering his low BBs, low Ks, and high flyball%. Despite Angel Stadium he should have been experiencing gopheritis according to xFIP. Watching him pitch my reasoning for this is the moving fastball. Unless you see him running 1.3 HR/9 rates in Safeco from here on out as xFIP would predict?
In 2004, Washburn’s park adjusted HR/FB rate was 11.7%. He gave up 20 home runs, when we’d have expected him to give up 19. Guess he didn’t have his cut fastball working that year, even though all his other ratios were right in line with what we’d have expected.
In 2005, his park adjusted HR/FB was 9.8%. He actually gave up 19 home runs. We’d have expected him to give up 21.
In 2006, his park adjusted HR/FB is 9.4%. He’s giving up 16 home runs. We’d have expected him to give up 18.
So, over the course of three seasons, you’re arguing that HR/FB% is incorrect for Jarrod Washburn because it estimated that he would give up 58 home runs, and he actually gave up 55 home runs. 3 homers over the course of 3 years is enough for you to say “see, it was wrong!”. That’s sad.
Seriously, that’s what you are laying your claim too? We wouldn’t expect Washburn to have “gopheritis”, however you want to define that. Washburn is, just like I said, BARELY beating the league average HR/FB rate when adjusted for parks.
I’m sorry you can’t see that you’re wrong on this issue. The evidence is there if you want to see it. Hopefully you will.
Hmmm, looks like I was wrong on Washburn (only off by three). Regardless, there ARE exceptions which should make staticians question the validity of their approach.
I really think people are messing around too much with correlations and making dangerous presumptions while doing so. We need to think in terms of individual pitching “skills” (if a pitcher’s show a consistent ability over several years it needs to be considered an individual skill) and adjusting for each specific “environment (parks/defense/etc)” in which those skills are displayed. SABRMatt is doing some unbelievable stuff on this right now – hes approaching the problem the way I would if I had his math skills. I think its something you’d be interested in.
In any case I’d bet that HRs are evaluated differently in the next five years or so.