I think that’s a bigger reason why he doesn’t go deep into games…it’s going to be hard to shut him down when the Cards are going to be in the World Series Race through September. Seems to be a deliberate attempt to keep him under 100 pitchers/game
Do you know who else was also a 22nd round pick in 2005? Tommy Hanson. In fact, Tommy Hanson and Jaime Garcia were THREE PICKS away, with Hanson taken at #677 and Garcia taken at #680.
If you look at WAR and FIP or xFIP… Halladay is the obvious Cy Young so far and will get Neyer’s vote… heck Jiminez is 4th in FIP in the NL (Halladay, Wainwright, Lincecum) and on XFIP he’s way down the list, he’s also 2nd to WAR to Halladay.
While the SABR stats are useful for predicting future performance, there is a growing need to assign luck or external factors (defense behind you) to using these stats as a measurement of actual past performance.
A model like FIP (and make no mistake it is a model) is useful and powerful for predicting future performance/trends… a model that is solely based on K’s, BB’s and HR’s should not be used as a substitute for actual performance though.
To be clear – I have no issue with this article which is forward looking and a question of whether Garcia will keep it up, just a misuse of statistics to re-evaluate past performance when looking at things like Cy Young
The idea behind FIP is that a pitcher can’t control whether his defence plays well or poorly behind him and it’s unfair to penalize a guy because of that.
If you’re ready to include luck and “external factors” (defence, aside from the pitcher’s personal contribution, IS luck) into your discussion for the Cy Young Award, you might as well include Wins. An NL pitcher probably has as much control over his run-support as he does his defence.
FIP, xFIP, tRA, etc. are not simply models for predicting future performance, but also gauges for evaluating past performance. Should we give extra credit to batters who accrue RBI because “external factors” (having a good lineup around them) dictate that they will bat with men in scoring position more than other players?
I understand the theory behind it… and why it is important for predictive purpose, but the theory has lots of holes in it when using it to measure actual performance. If you want to evaluate the defense behind the pitcher instead of choosing to ignore singles, doubles and triples, how about using the defensive metrics of the players?
A pitcher that gives up 4 runs via RBI singles and doubles is not more successful than a pitcher that gives up 1 solo HR…but if strikeouts and walks are similar the pitcher that gives up 4 runs will be considered better in context of FIP. Now you can say the pitcher that doesn’t give up HR’s is more likely to be successful in the future, but in terms of ACTUAL performance to ignore runs not scored via the HR is to ignore the actual game.
So we are to believe Jiminez ERA being 2.5X better than say Halladay is a luck thing because Halladay’s FIP is better? Or is JIminez’ defense behind him really that much better than Halladay (or Wainwright or Linceucm who also have better FIPs than Jiminez) and that is the reason behind giving up fewer runs?
If you use FIP like it is intended, I think the FIP #’s suggest we should expect Linceum/Halladay/Jiminez to be similar in performance going forward, but to take that metric and say they have performed the same because the FIPs are similar is a misuse of that statistic/model. The other major flaw I have with FIP is park factor – look at Lincecum’s road vs home HR’s last year… any chance his home park helps his FIP? (which Neyer chose to ignore when voting for Cy last year)
It would be interesting to see WHO Garcia is walking and WHEN. A 2-out pitch around walk is certainly less damaging than say a lead-off walk.
Walks to set up good matchups for the pitcher are also sometimes a good idea. Been my experience that the timing of the walk is a key factor. We had a stud pitcher a few years ago (D2 All-Wrican in 2010) that had this annoying habit of walking the 8 hitter with 2 outs. But those runners never scored, as oppossed to say walking the 2 hitter to start an inning.
“So we are to believe Jiminez ERA being 2.5X better than say Halladay is a luck thing because Halladay’s FIP is better?”
Isn’t it a much bigger leap of faith to believe that Ubaldo Jimenez has actually displayed the ability to keep 98.2% of flyballs in the park, strand 91.7% of runners and hold hitters to a BABIP of .226?
“The other major flaw I have with FIP is park factor – look at Lincecum’s road vs home HR’s last year… any chance his home park helps his FIP? (which Neyer chose to ignore when voting for Cy last year)”
Lincecum FIP
2010
home: 2.99; away: 2.26
2009
home: 1.69; away: 3.15
2008
home: 2.78; away: 2.46
2007
home: 3.67; away: 3.59
2008 was the only season in which Lincecum’s home FIP was lower than his road FIP. The actual reason that his home FIP was so low has nothing to do with park factors, it has to do with the fact that he didn’t give up a single homerun in AT&T Park that season.
“Isn’t it a much bigger leap of faith to believe that Ubaldo Jimenez has actually displayed the ability to keep 98.2% of flyballs in the park, strand 91.7% of runners and hold hitters to a BABIP of .226?”
That’s EXACTLY my point… what you are talking about is what you expect to l happen in the future and that is exactly what these stats/models are intended for, but it doesn’t change what he HAS done. The bottom line is he has stranded those runners, kept the ball in the park and held batters to a low BABIP… call it luck, call it unsustainable (hint – that what these #’s are used for), call it team defense – but bottom line is he has achieved it.
If you believe FIP can accurately discount defense and luck for past results, Jiminez has actually 3-4X more lucky/defensively enabled (via ERA) then Lincecum/Wainwright/Hallady. FIP suggest all 3 of these pitchers have been better so far this year. I choose to think FIP is suggesting that they will perform similarly going forward, but can’t adequately capture how they have actually performed.
You don’t use these stats to diminish or discount what he has accomplished, you use these stats to predict what will happen going forward and how sustainable the performance might be (but the Cy Young voting and past performance is not about what should have happened, it’s about what did happen).
“what you are talking about is what you expect to happen in the future”
No, it isn’t. It’s what I expect to have actually happened in the past. I think Ubaldo Jimenez is great and I have no problem believing that over a start or two he can be as electric as he seems to have been. However, no, I don’t think he’s done it all himself (and he hasn’t) . . . he’s had considerable help from his team’s defence and from luck with balls in play and stranding runners. I don’t believe in crediting a pitcher for his team’s defence or for benefiting from good luck.
“The bottom line is he has stranded those runners, kept the ball in the park and held batters to a low BABIP… call it luck, call it unsustainable (hint – that what these #’s are used for), call it team defense – but bottom line is he has achieved it.”
You have an interesting definition of the word achieve. In my opinion, FIP and xFIP are more accurate gauges of what the pitcher, himself, has achieved. Something like ERA is greatly influenced by a pitcher’s defence and the timing of hits with respect to stranding runners. Has Livan Hernandez also pitched higher quality innings than Roy Halladay because his ERA is a bit lower?
Though this is oversimplifying a bit, pitchers are generally trying to “achieve” three things when they face batters — either induce a grounder, induce a pop-up or strike a batter out — and are trying to avoid two things — long flyballs or linedrives and walks. The whole point of xFIP, tRA and (to a certain extent) FIP is that they credit a pitcher for his successes (Ks and groundballs) and dock him for his failures (flyballs/liners and BBs).
For the four pitchers you named:
GB-rate
Halladay has a very slightly higher GB-rate than Ubaldo. Wainwright’s is a little worse but still very good and Lincecum’s is behind Wainwright but still fine.
K-rate
Lincecum is king. His K-rate is by far better than any of the other three’s. Wainwright’s is very good, Ubaldo’s is good and Halladay’s is average.
Infield Flyball-rate (as a proportion of flyballs)
Ubaldo is very good. Halladay is good. Lincecum and Wainwright are about average.
FB/LD-rate
This is just the inverse of GB-rate . . . Halladay best with Jimenez very close behind and Wainwright and Lincecum trailing by a bit.
BB-rate
Halladay’s is amazing. Wainwright’s is very good, Ubaldo’s is good and Lincecum’s is average.
The point of xFIP and FIP is that they combine all these together to tell you both how frequently the pitcher “succeeded” — that is how frequently he got grounders and strikeouts — and how frequently he has “failed” — that is issued walks and induced either flyballs or linedrives. For our convenience these stats are scaled to our perception of a normal ERA. By FIP, Halladay is first but they’re all very close. By xFIP, Lincecum and Halladay are a bit better than Wainwright and fairly significantly better than Jimenez. By tERA, Jimenez and Wainwright are best with Lincecum slightly behind and Halladay trailing.
These systems don’t agree with one another on whose been best because they assign different levels of importance to each event, however, they are alike in that they are grading a pitcher based on how frequently he’s been able to do what he’s sought to do, which is why I prefer to use them to something like ERA.
Sorry for the long post and it’s been fun discussing this.
Defense cannot simply be considered luck when the point of getting ground balls with low pitches is to use your defense. If you are trying to use it and are it is doing what you want, then it cannot be luck.
The article states that Garcia could be a mainstay in the rotation for years to come.
That would be refreshing. StL doesn’t exactly have a good track record with lefty prospects and acquisitions Who was our last lefty mainstay? Joe Magrane?
The concern with Garcia has always been injuries. I don’t believe he’s even come close to 170 IP in any season.
When you’re inducing that many grounders, you’re making the other team play in a situation where they likely need 3 runners in an inning to score. You’re also likely erasing a walk or hit with a DP a decent amount of time.
I don’t think it’s luck in how we traditionally use the word. If you’re changing speeds, living down in the zone, and throwing late breaking pitches and get a lot of grounders … That’s not luck, that’s design (As Duncan has continually demonstrated). It’s not as if hitters are choosing to hit grounders instead of lifting the ball for extra bases.
It’s like saying strikeouts are luck because the pitcher cannot control whether the batters make contact or not. Control? Probably not. Influence?definitely.
A pitcher succeeds by not giving up runs… once you account for the defensive quality behind the pitcher, I really don’t care if it’s 10K’s per game or 2K’s per game or a 50% GB rate or a 25% GB rate. Now if I’m a GM and am trying to figure out how much to pay this person going forward then I care a lot about it (which gets back to predictive quality for FIP vs its ability to measure existing performance).
Jiminez has allowed 7 runs to cross the plate this year… Halladay 19 (I’ll used ER and ignore IP differences)… FIP, if you use it as an indicator of current performance would indicate that Jiminez should have given up more runs than Halladay. Jiminez has just gotten more lucky or been playing behind a much better defense than Halladay (or the inverse – Halladay has gotten unlucky and/or has got less benefit from his defense than Jiminez)
Having watched these guys pitch (yeah I know the dreaded “eye test”) I don’t believe you can say that difference in runs allowed is luck and the defensive component (it’s not like the Phils are a poor defensive team), and that Jiminez would have given up MORE runs than Halladay (which is what FIP suggests). There is no doubt in my mind that Jiminez has performed significantly better than Halladay while FIP would tell me Halladay has performed marginally better.
To put a bow on this for this comment, FIP indicates both Josh Johnson and Tom Gorzelanny have performed nearly as well as Jiminez so far this season.and Jeremy Bonderman is not that far behind.
No one said that he’s lucky to get grounders. xFIP greatly values grounders over flyballs. Despite very similar K- and BB-rates, Hiroki Kuroda’s xFIP is 3.52 and Kevin Slowey’s is 4.63.
It has long been accepted that different pitchers can have varying levels of control over their BABIP. In fact, Voros McCracken himself noted that knuckeballers have lower BABIPs than others. It’s simply that the ability to control BABIP is weak and its affect on run scoring is substantially less in comparison to the divergence in skill to control BBs, Ks and HRs.
Cardinals645 says:
May 27, 2010 at 1:03 pm
Garcia was not a second round pick; he was a TWENTY second (22nd) round pick.
Makes his performance all the more impressive, right?
chongo says:
May 27, 2010 at 1:15 pm
He will be limited in innings anyway, right? The most he ever threw was in 08 and it was like 122 IP.
Ray says:
May 27, 2010 at 1:25 pm
I think that’s a bigger reason why he doesn’t go deep into games…it’s going to be hard to shut him down when the Cards are going to be in the World Series Race through September. Seems to be a deliberate attempt to keep him under 100 pitchers/game
dustin says:
May 27, 2010 at 1:59 pm
Holy crap! If you’re going to have flash ads, please don’t have them at max volume. My ears are ringing.
Shaun says:
May 27, 2010 at 2:01 pm
I know its just a typo, but I have a fun time thinking about how TLR would have to make a deliberate attempt to keep his team under 100 pitchers/game.
M says:
May 27, 2010 at 2:30 pm
Woah.
Do you know who else was also a 22nd round pick in 2005? Tommy Hanson. In fact, Tommy Hanson and Jaime Garcia were THREE PICKS away, with Hanson taken at #677 and Garcia taken at #680.
TedWilson says:
May 27, 2010 at 2:49 pm
Remember when the stats community thought BABIP was .300 no matter what? HA!
frank says:
May 27, 2010 at 3:25 pm
If you look at WAR and FIP or xFIP… Halladay is the obvious Cy Young so far and will get Neyer’s vote… heck Jiminez is 4th in FIP in the NL (Halladay, Wainwright, Lincecum) and on XFIP he’s way down the list, he’s also 2nd to WAR to Halladay.
While the SABR stats are useful for predicting future performance, there is a growing need to assign luck or external factors (defense behind you) to using these stats as a measurement of actual past performance.
A model like FIP (and make no mistake it is a model) is useful and powerful for predicting future performance/trends… a model that is solely based on K’s, BB’s and HR’s should not be used as a substitute for actual performance though.
frank says:
May 27, 2010 at 3:29 pm
To be clear – I have no issue with this article which is forward looking and a question of whether Garcia will keep it up, just a misuse of statistics to re-evaluate past performance when looking at things like Cy Young
jessef says:
May 27, 2010 at 4:21 pm
The idea behind FIP is that a pitcher can’t control whether his defence plays well or poorly behind him and it’s unfair to penalize a guy because of that.
If you’re ready to include luck and “external factors” (defence, aside from the pitcher’s personal contribution, IS luck) into your discussion for the Cy Young Award, you might as well include Wins. An NL pitcher probably has as much control over his run-support as he does his defence.
FIP, xFIP, tRA, etc. are not simply models for predicting future performance, but also gauges for evaluating past performance. Should we give extra credit to batters who accrue RBI because “external factors” (having a good lineup around them) dictate that they will bat with men in scoring position more than other players?
frank says:
May 27, 2010 at 5:46 pm
I understand the theory behind it… and why it is important for predictive purpose, but the theory has lots of holes in it when using it to measure actual performance. If you want to evaluate the defense behind the pitcher instead of choosing to ignore singles, doubles and triples, how about using the defensive metrics of the players?
A pitcher that gives up 4 runs via RBI singles and doubles is not more successful than a pitcher that gives up 1 solo HR…but if strikeouts and walks are similar the pitcher that gives up 4 runs will be considered better in context of FIP. Now you can say the pitcher that doesn’t give up HR’s is more likely to be successful in the future, but in terms of ACTUAL performance to ignore runs not scored via the HR is to ignore the actual game.
So we are to believe Jiminez ERA being 2.5X better than say Halladay is a luck thing because Halladay’s FIP is better? Or is JIminez’ defense behind him really that much better than Halladay (or Wainwright or Linceucm who also have better FIPs than Jiminez) and that is the reason behind giving up fewer runs?
If you use FIP like it is intended, I think the FIP #’s suggest we should expect Linceum/Halladay/Jiminez to be similar in performance going forward, but to take that metric and say they have performed the same because the FIPs are similar is a misuse of that statistic/model. The other major flaw I have with FIP is park factor – look at Lincecum’s road vs home HR’s last year… any chance his home park helps his FIP? (which Neyer chose to ignore when voting for Cy last year)
Circlechange11 says:
May 27, 2010 at 6:00 pm
It would be interesting to see WHO Garcia is walking and WHEN. A 2-out pitch around walk is certainly less damaging than say a lead-off walk.
Walks to set up good matchups for the pitcher are also sometimes a good idea. Been my experience that the timing of the walk is a key factor. We had a stud pitcher a few years ago (D2 All-Wrican in 2010) that had this annoying habit of walking the 8 hitter with 2 outs. But those runners never scored, as oppossed to say walking the 2 hitter to start an inning.
jessef says:
May 27, 2010 at 6:23 pm
“So we are to believe Jiminez ERA being 2.5X better than say Halladay is a luck thing because Halladay’s FIP is better?”
Isn’t it a much bigger leap of faith to believe that Ubaldo Jimenez has actually displayed the ability to keep 98.2% of flyballs in the park, strand 91.7% of runners and hold hitters to a BABIP of .226?
“The other major flaw I have with FIP is park factor – look at Lincecum’s road vs home HR’s last year… any chance his home park helps his FIP? (which Neyer chose to ignore when voting for Cy last year)”
Lincecum FIP
2010
home: 2.99; away: 2.26
2009
home: 1.69; away: 3.15
2008
home: 2.78; away: 2.46
2007
home: 3.67; away: 3.59
2008 was the only season in which Lincecum’s home FIP was lower than his road FIP. The actual reason that his home FIP was so low has nothing to do with park factors, it has to do with the fact that he didn’t give up a single homerun in AT&T Park that season.
frank says:
May 27, 2010 at 7:04 pm
“Isn’t it a much bigger leap of faith to believe that Ubaldo Jimenez has actually displayed the ability to keep 98.2% of flyballs in the park, strand 91.7% of runners and hold hitters to a BABIP of .226?”
That’s EXACTLY my point… what you are talking about is what you expect to l happen in the future and that is exactly what these stats/models are intended for, but it doesn’t change what he HAS done. The bottom line is he has stranded those runners, kept the ball in the park and held batters to a low BABIP… call it luck, call it unsustainable (hint – that what these #’s are used for), call it team defense – but bottom line is he has achieved it.
If you believe FIP can accurately discount defense and luck for past results, Jiminez has actually 3-4X more lucky/defensively enabled (via ERA) then Lincecum/Wainwright/Hallady. FIP suggest all 3 of these pitchers have been better so far this year. I choose to think FIP is suggesting that they will perform similarly going forward, but can’t adequately capture how they have actually performed.
You don’t use these stats to diminish or discount what he has accomplished, you use these stats to predict what will happen going forward and how sustainable the performance might be (but the Cy Young voting and past performance is not about what should have happened, it’s about what did happen).
jessef says:
May 27, 2010 at 10:38 pm
“what you are talking about is what you expect to happen in the future”
No, it isn’t. It’s what I expect to have actually happened in the past. I think Ubaldo Jimenez is great and I have no problem believing that over a start or two he can be as electric as he seems to have been. However, no, I don’t think he’s done it all himself (and he hasn’t) . . . he’s had considerable help from his team’s defence and from luck with balls in play and stranding runners. I don’t believe in crediting a pitcher for his team’s defence or for benefiting from good luck.
“The bottom line is he has stranded those runners, kept the ball in the park and held batters to a low BABIP… call it luck, call it unsustainable (hint – that what these #’s are used for), call it team defense – but bottom line is he has achieved it.”
You have an interesting definition of the word achieve. In my opinion, FIP and xFIP are more accurate gauges of what the pitcher, himself, has achieved. Something like ERA is greatly influenced by a pitcher’s defence and the timing of hits with respect to stranding runners. Has Livan Hernandez also pitched higher quality innings than Roy Halladay because his ERA is a bit lower?
Though this is oversimplifying a bit, pitchers are generally trying to “achieve” three things when they face batters — either induce a grounder, induce a pop-up or strike a batter out — and are trying to avoid two things — long flyballs or linedrives and walks. The whole point of xFIP, tRA and (to a certain extent) FIP is that they credit a pitcher for his successes (Ks and groundballs) and dock him for his failures (flyballs/liners and BBs).
For the four pitchers you named:
GB-rate
Halladay has a very slightly higher GB-rate than Ubaldo. Wainwright’s is a little worse but still very good and Lincecum’s is behind Wainwright but still fine.
K-rate
Lincecum is king. His K-rate is by far better than any of the other three’s. Wainwright’s is very good, Ubaldo’s is good and Halladay’s is average.
Infield Flyball-rate (as a proportion of flyballs)
Ubaldo is very good. Halladay is good. Lincecum and Wainwright are about average.
FB/LD-rate
This is just the inverse of GB-rate . . . Halladay best with Jimenez very close behind and Wainwright and Lincecum trailing by a bit.
BB-rate
Halladay’s is amazing. Wainwright’s is very good, Ubaldo’s is good and Lincecum’s is average.
The point of xFIP and FIP is that they combine all these together to tell you both how frequently the pitcher “succeeded” — that is how frequently he got grounders and strikeouts — and how frequently he has “failed” — that is issued walks and induced either flyballs or linedrives. For our convenience these stats are scaled to our perception of a normal ERA. By FIP, Halladay is first but they’re all very close. By xFIP, Lincecum and Halladay are a bit better than Wainwright and fairly significantly better than Jimenez. By tERA, Jimenez and Wainwright are best with Lincecum slightly behind and Halladay trailing.
These systems don’t agree with one another on whose been best because they assign different levels of importance to each event, however, they are alike in that they are grading a pitcher based on how frequently he’s been able to do what he’s sought to do, which is why I prefer to use them to something like ERA.
Sorry for the long post and it’s been fun discussing this.
n_yount says:
May 27, 2010 at 10:59 pm
Defense cannot simply be considered luck when the point of getting ground balls with low pitches is to use your defense. If you are trying to use it and are it is doing what you want, then it cannot be luck.
Circlechange11 says:
May 27, 2010 at 11:26 pm
The article states that Garcia could be a mainstay in the rotation for years to come.
That would be refreshing. StL doesn’t exactly have a good track record with lefty prospects and acquisitions Who was our last lefty mainstay? Joe Magrane?
The concern with Garcia has always been injuries. I don’t believe he’s even come close to 170 IP in any season.
Circlechange11 says:
May 27, 2010 at 11:54 pm
When you’re inducing that many grounders, you’re making the other team play in a situation where they likely need 3 runners in an inning to score. You’re also likely erasing a walk or hit with a DP a decent amount of time.
I don’t think it’s luck in how we traditionally use the word. If you’re changing speeds, living down in the zone, and throwing late breaking pitches and get a lot of grounders … That’s not luck, that’s design (As Duncan has continually demonstrated). It’s not as if hitters are choosing to hit grounders instead of lifting the ball for extra bases.
It’s like saying strikeouts are luck because the pitcher cannot control whether the batters make contact or not. Control? Probably not. Influence?definitely.
frank says:
May 28, 2010 at 4:50 am
A pitcher succeeds by not giving up runs… once you account for the defensive quality behind the pitcher, I really don’t care if it’s 10K’s per game or 2K’s per game or a 50% GB rate or a 25% GB rate. Now if I’m a GM and am trying to figure out how much to pay this person going forward then I care a lot about it (which gets back to predictive quality for FIP vs its ability to measure existing performance).
Jiminez has allowed 7 runs to cross the plate this year… Halladay 19 (I’ll used ER and ignore IP differences)… FIP, if you use it as an indicator of current performance would indicate that Jiminez should have given up more runs than Halladay. Jiminez has just gotten more lucky or been playing behind a much better defense than Halladay (or the inverse – Halladay has gotten unlucky and/or has got less benefit from his defense than Jiminez)
Having watched these guys pitch (yeah I know the dreaded “eye test”) I don’t believe you can say that difference in runs allowed is luck and the defensive component (it’s not like the Phils are a poor defensive team), and that Jiminez would have given up MORE runs than Halladay (which is what FIP suggests). There is no doubt in my mind that Jiminez has performed significantly better than Halladay while FIP would tell me Halladay has performed marginally better.
To put a bow on this for this comment, FIP indicates both Josh Johnson and Tom Gorzelanny have performed nearly as well as Jiminez so far this season.and Jeremy Bonderman is not that far behind.
jessef says:
May 28, 2010 at 12:49 pm
No one said that he’s lucky to get grounders. xFIP greatly values grounders over flyballs. Despite very similar K- and BB-rates, Hiroki Kuroda’s xFIP is 3.52 and Kevin Slowey’s is 4.63.
Toffer Peak says:
May 28, 2010 at 3:47 pm
It has long been accepted that different pitchers can have varying levels of control over their BABIP. In fact, Voros McCracken himself noted that knuckeballers have lower BABIPs than others. It’s simply that the ability to control BABIP is weak and its affect on run scoring is substantially less in comparison to the divergence in skill to control BBs, Ks and HRs.