Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Primate Studies > Discussion
Primate Studies
— Where BTF's Members Investigate the Grand Old Game

Thursday, December 12, 2002

Quantifying Greatness

Jason looks back at the Pedro/Zito Cy Young debate.

Sometimes true greatness is difficult to comprehend. In some cases, we grow accustomed to it and, after awhile, fail to recognize it. At least, this is what I keep telling myself whenever award voters snub Pedro Martinez.

Pedro narrowly lost the 1999 AL MVP award because two voters left him completely off their ballots, claiming they believed pitchers didn’t deserve MVP votes. This was hypocritical, as one (George King) had voted for David Wells and Rick Helling the previous year. But I digress, that’s a story for a different column. As in ‘99, this year Oakland’s Barry Zito usurped the Cy Young Award from Pedro, depite being less qualified.

Both pitchers threw terrifically this season, as the chart below depicts.

    W-L Win%  ERA   G   IP
Zito 23-5 .821 2.75 35 229.1
Pedro 20-4 .833 2.26 30 199.1

Pedro led the majors in ERA, and AL in strikeouts. Zito, on the other hand, pitched pretty well too, and offered five more starts and 30 more innings than Pedro. His more durable contribution, in addition to pitching for a team in the pennant race, earned him the majority of votes. However, the performances of the two pitchers get a little more skewered under closer scrutiny.

A pitcher owes a great deal of his success to:

A) The defense behind him, and

B) Luck

Let me expand: As Voros McCracken exhaustively studied, pitchers have little effect on balls put into play. Once the ball touches the bat, it’s up to chance and the fielders to determine what happens. So, how did the fielders and Lady Luck affect our two Cy candidates this year?

DILP*

Zito .243

Pedro .273

*Defense Induced Luck Percentage (H-HR/BF-HR-K-BB-HBP)

So, we see that either Zito’s defense was a lot better, or Pedro was far unluckier than Zito. Zito is a flyball pitcher with one of the weaker defensive outfields in the game, and thus it’s truly amazing that his defense converted such a high percentage of balls in play into outs. Not meaning to discredit Zito’s season, but he ducked serious odds putting up the numbers he did. Luck played a healthy role in his success.

Pedro, despite having a DILP 30 points higher than Zito, still allowed far fewer runs per nine innings and let fewer opponents on base. Pedro accomplished this through several statistics that are totally independent of a pitcher’s defense.

Walks

Pedro only walked 40 batters this season, good for a miniscule average of 1.8 per 9 innings. Zito actually missed the plate a good portion this year, ranking sixth in the AL in walks surrendered and averaging 3.06 walks per nine.

K’s

Pedro struck out far more opponents than Zito, averaging 10.79 per 9 innings, while Zito only whiffed 7.14 per 9 innings.

Homers

Hitters also took Zito deep more frequently than Pedro, as he allowed 24 homers to Pedro’s 13. Zito actually let up one fewer home run than Pedro and Beantown counterpart Derek Lowe combined (24-25).

Even though Zito’s luck and defense overwhelmingly aided him this season, Pedro still posted superior numbers because he pitched considerably better.

Ah, but you already knew Pedro pitched better than Zito. Zito’s Cy qualifications rest on throwing 5 additional games and 30 more innings than Pedro the Great. The argument seems to be that Pedro pitched tremendously, but Zito pitched great too, and the extra innings make him more valuable than Pedro. But do they?

Here’s an exercise: Pedro, projected to 229.1 innings this season, the same as Zito. However, I’ve projected Pedro’s extra innings with double the BBs per nine innings, double the homers allowed per nine innings, half the Ks per nine and with Zito’s ERA added to his own:

      Zito Adjusted Pedro
ERA     2.75     2.63
BB/9   3.06     2.04
HR/9   1.08     .67
K/9     7.14     10.1

As you can see, even if pro-rated to an extra 30 innings of awful pitching, Pedro still clearly comes out on top. If the two pitchers had posted these numbers, Pedro probably would’ve won the award because it would’ve made the two much easier to compare, a comparison Zito clearly loses. Yet, these numbers represent Pedro getting absolutely Lima-ed over those 30 innings.

As you can see here, there’s a Catch-22 for Zito backers. How can Zito deserve the Cy Young over Pedro for throwing more innings, because even if Pedro pitched those 30 innings like a Devil Ray cast-off, he’d clearly be the better hurler? Yet, adding 30 extra innings of batting practice meatballs only tarnishes the quality of Pedro’s season.

The truth is that durability counts, but so does dominance. When you analyze it, Pedro was so clearly more dominant that Zito during the course of the season that a paltry 30 innings doesn’t nearly bridge the chasm between the two.

Some people may be tempted to believe that Zito’s presence in a pennant race, pitching against the AL West competition, meant that he posted his numbers against far better competition. Various writers addressed this issue, with mixed results. One ESPN.com column by Matt Szfec and David Schoenfield broke down Zito and Pedro’s performances against winning and losing teams.

    G, ERA > .500     G, ERA < .500
Zito   18, 3.66         14, 1.88
Pedro 13, 2.14         17, 2.36

Zito pitched well enough to get 11 wins against winning teams, but without a spectacular ERA. In fact, Zito sculpted his formidable 2.75 ERA by demolishing weak opponents, while Pedro actually pitched better against winning teams.

Barry Zito racked up a lot of wins and anchored a playoff team. He received a lot of positive press for pitching terrifically down the stretch for a team that won 20 straight games and battled their way into the playoffs. He also didn’t come close to matching Pedro Martinez, the best pitcher in the American League.

As I wrote above, sometimes greatness is difficult to quantify. Pedro didn’t quite match his divine status of previous seasons, and perhaps some voters mistakenly viewed this as a negative. Perhaps others tired of voting for Pedro, and since they actually had an alternative, opted for it. Regardless, I just hope these same voters know how lucky they are to be watching Pedro, even if they don’t bequeath him with the awards to prove it.

Jason Tuohey Posted: December 12, 2002 at 06:00 AM | 24 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Jason Posted: December 12, 2002 at 02:09 AM (#607553)
Yeah, that's fine and all, but when your ace smugly shuts himself down after he gets his 20th win, he loses points. I don't care what the numbers say. Sure he's great, but Pedro's a skunk.
   2. Brian Posted: December 12, 2002 at 02:09 AM (#607558)
Im glad this topic came up again. The voters have obviously lost all statistics, it seems more centered on hype and fanfare, and since when do players on playoff teams get special consideration?
   3. bob mong Posted: December 12, 2002 at 02:09 AM (#607559)
Consider me unconvinced.

Those 30 innings have value even if they are of crappy quality. And besides, that argument can be used to justify some nutty cases for the Cy Young. Like, for example, Arthur Rhodes.

Here are Arthur's 2002 stats:

69.7 IP, 2.32 ERA, 81 SO (10.5 per 9), 4 HR (0.52 per 9), 13 BB (1.7 per 9). He only allowed 45 hits in those 69+ innings, so we can probably assume that he was somewhat fortunate with regards to hits on balls in play.

But he only pitched 70 innings? No problem, lets just add 150 innings at half the strikeout rate, double the walk and homerun rate, and at Zito's ERA (2.75). What's the result?

219.7 IP, 2.61 ERA, 168 SO (6.9 per 9), 69 BB (2.83 per nine), 21 HR (0.87 per 9) - why, those numbers are better than Zito's! We should give Arthur Rhodes the Cy Young!
I am being a bit hyperbolic here; no one has (or would) argue that Arthur Rhodes is a better pitcher or a more valuable pitcher than Barry Zito.

But those extra innings have value, and we can't just add mythical innings to someone's total and then say, Presto! That's a Cy Young candidate! The whole point is that Zito's extra innings, even though at a lower quality than Pedro's, have value. If for no other reason than that they kept a worse pitcher off the mound for 4-5 starts, or that they kept the bullpen a little more healthy.

Nice analysis, though, and a good article. I just don't agree :)
   4. MattB Posted: December 12, 2002 at 02:09 AM (#607560)
"Let me expand: As Voros McCracken exhaustively studied, pitchers have little effect on balls put into play. Once the ball touches the bat, it's up to chance and the fielders to determine what happens."

Except that not what he said. He said that the corrolation between hits allowed on balls in play is a lot lower than it is for the other stats, so randomness within a season will be more likely to mask true ability, if any.

Great pitchers almost universally give up fewer hits than would be expected using team averages. They just fluctuate a lot year to year so it takes a bunch of years to know if there's a skill hiding in there or not.

Cy Young gave up 203 fewer hits on balls in play than would be expected based solely on his defense. Nolan Ryan, 132, Jim Palmer 169, Koufax 94.

Over his career, Pedro has saved about one hit on a ball in play every 41 innings. Zito has saved about one hit on a ball in play every 13 innings. Now, for Zito we obviously have a relatively small sample size, but that's hardly a reason to right it all off as luck.

Early returns show that preventing hits on balls in play is an important part of Zito's repertoire. Now, the low correlation means that Zito is more likely than most to have fluke high and low years, but that hardly is a reason to completely discount this ability as a real strength of his.

What I have noticed is that players who do not have this ability to prevent balls in play from being hits over a career tend to be those loved by the stats crowd, but not be the hall of fame voters (e.g., Blyleven, Kaat, T. John). Don't know if that's significant or not.
   5. Ned Garvin: Male Prostitute Posted: December 12, 2002 at 02:09 AM (#607563)
Seems to me Derek Lowe got lost in the Pedro-Zito argument this year. I haven't looked exhaustively at the stats, but I seem to recall that my voting went Pedro-Lowe-Zito. Any comment on that, or opinion to whether or not Lowe was the 2nd best pitcher in the AL this year?
   6. Michael Posted: December 12, 2002 at 02:09 AM (#607570)
What about Roy Halladay, largely overlooked.

Halladay had a DILP of .286 even worse than pedro. Doing the same adjustments for Halladay I get:

ERA Halladay 2.93, adjusted Pedro 2.75
BB/9 Halladay 2.33, adjusted Pedro 2.11
K/9 Halladay 6.32, adjusted Pedro 9.89
HR/9 Halladay 0.38, adjusted Pedro 0.69

So it is much close. In particular the ERA and BB is close, the K's go to Pedro but the HR's go to Roy. More over I'm not sure your quick and dirty projecting method to Pedro does full justice to the replacement value of the extra 206 batters that Halladay faced. That is more than 25% more batters than Pedro!

I mean that means we are defining replacement level to be:

ERA 5.19, BB/9 3.61, K/9 5.4, HR/9 1.17. That seems closer to league average stats than to true replacement level.

I agree Pedro was great, I agree Pedro was better than Zito, I'm not sure Pedro was better than Halladay, and if he was it was pretty close.
   7. Boileryard Posted: December 12, 2002 at 02:09 AM (#607571)
No, no, no!

A pitcher's value is strictly tied to the expected wins he gives his team (or a neutral team), depending on your preference.

The run distribution is VERY important in situations like this.

When you get to top-5 pitchers, 30 games of Zito-like pitching will usually lead to more wins than 25 games of Pedro-like pitching + 5 games of replacement-level pitching.

   8. Indy Posted: December 12, 2002 at 02:09 AM (#607572)
I think Ryan's got a good point with respect to run distribution. And I think that the Cy Young voters pay more attention to the difference in wins than anything else.

For giggles, though, we can work backwards, and see that to get to Barry's stats over 230 innings, Pedro would have to have closed out the season no worse than:
ERA 6 (2.74 season average)
K/9 0 (9.38)
BB/9 9 (2.73)
HR/9 4 (1.03)

Of course, at this rate, he wouldn't have gotten any more of those wins...
   9. Boileryard Posted: December 12, 2002 at 02:09 AM (#607573)
Thanks, Indy, for doing some quick calculations that illustrate my point.

If you believe in Support-Neutral Wins from Wolverton, Zito provided 1.4 more SNW than Pedro.

I'm fairly certain 30 innings of 6.00 ERA ball would not provide the additional 1.4 SNW needed to make Pedro as valuable (in expected wins) as Zito was.

Now, if you're going to make DIPS adjustments, Zito and Pedro will end up being much closer in expected wins. But then you're getting dangerously close to advocating the Cy Young be determined by formula as soon as the season's over. That's not very fun, now, is it?
   10. Boileryard Posted: December 13, 2002 at 02:09 AM (#607576)
Oops, my bad. Not sure what I was thinking. Thanks for the correction, Vinay.

Anyway, I'll still stand by my point that the distribution is important, but in the case of Zito v Pedro, the case is quite muddy indeed when you limit yourself to what actually happened.

Just out of curiosity, if Pedro had made his final start as scheduled (and won), and Zito skipped his final start to lead off the playoffs, does anyone think that Pedro would have won the Cy Young?
   11. Jason Tuohey Posted: December 13, 2002 at 02:09 AM (#607581)
Hey all, thanks for the great posts, glad I could spur such active discussion.

A few points:

- I purposely didn't use the "replacement value" stats when averaging Pedro's 30 extra innings. You see, most of the articles I read triumphing Zito argued that Zito pitched very nearly as well as Pedro, and he pitched more innings, making his season better (Shaughnassy comes to mind). It seemed like most mainstream sportswriters merely looked at the 2.26-2.75 ERAs and thought "same ballpark." So, I doubled all of Pedro's stats, and more than doubled his ERA for those 30 innings to show that the two weren't really that close. The fact that Pedro still looks better after 30 innings of impossibly bad (for Pedro) hurling just helps illustrate this.

Some people noted that the BB/9 and K's/9 are better than average, but then again, that's just another indication of how supremely dominant Pedro is, no? You couldn't halve Zito's K's and double his BB's and end up with anything remotely useful.

Bruce: I guess my point is that the difference between Pedro and Zito, as I see it, is more like $50,000 per inning to $35,000 per inning, i.e. not close at all. Pedro did everything a lot better than Zito, and received much less assistance from his teammates. Additionally, you say that "Pedro was more dominant, but Zito made a more valuable contribution," but I thought the Cy Young was supposed to be for the most dominant, not a Tejada-like interpretation of value the way the MVP is.

- I'm a huge fan of Derek Lowe, but I thought Pedro was the better Boston pitcher, and I wanted to keep the argument strictly to "winner vs. most deserved." Additionally, Lowe's K's dropped monstrously this year (lowest ever) and his DILP was the astronomically low. The improved Sox infield defense greatly aided him (as I wrote in another column here). I believe Pedro, on his own, independent of defense, was the better pitcher.

- I'm not sure that good pitchers allow fewer hits on balls in play than bad ones; Pedro's '99 season he had a DILP of over .320, and he was as dominant as any pitcher I've ever seen . When I said Zito got lucky, it's because you'd expect a flyball pitcher with an outfield consisting of two corner outfielders and a DH to let up a large amount of hits on balls put into play. I don't buy that he somehow pitches in such a way that hitters hit flyballs within the limited range of his outfielders. If that was the case, why'd he let 24 balls clear the fence? I'm not trying to diss him, I still think he's great, I just think he could have a significantly higher amount of earned runs and hits allowed next year.

- I didn't really address how Pedro skipping the last start, which was only an opportunity to pad his stats against his own personal chew toy (the Devil Rays), might have affected his Cy chances. Obviously, any voter who used that as a method to evaluate a season's performance wouldn't be concerned with stats.

Thanks again for reading.

   12. bob mong Posted: December 13, 2002 at 02:09 AM (#607585)
From Vinay:
Bob Mong, your Rhodes example doesn't work. You used Zito's ERA as the replacement-level ERA, which doesn't make any sense at all. Jason Tuohey used the sum of Pedro's and Zito's ERAs, or 5.01.
...
And from the author:
So, I doubled all of Pedro's stats, and more than doubled his ERA for those 30 innings to show that the two weren't really that close.
(Bold mine.)

Vinay, thanks for catching my error. I misread the article. Though my silly example was just an attempt to show how the argument, when taken to extremes, was ridiculous, I still remain unconvinced that the proper way to evaluate a pitcher (for purposes of determining the Cy Young winner) is to play "what-if" games with his stats. Still, and again, an excellent article.
   13. MattB Posted: December 13, 2002 at 02:09 AM (#607587)
I wrote
Great pitchers almost universally give up fewer hits than would be expected using team averages.

Vinay Kumar wrote:
Evidence, please? MattB, I can replace all your references to defensive support with the corresponding references to run support, and I think the entire post will still be just as true. So, by extension, you should credit the pitchers for their run support also, and go primarily by W-L record.

My thesis is that great pitchers give up fewer hits on balls in play over a career than expected based on team averages.

Let's take as our universe of pitchers, Top 25 leaders in ERA+ who are in the Hall of Fame.

Here are their career hits allowed on balls in play below what would be expected compared to their team averages:

Lefty Grove -- 44
Walter Johnson -- 320
Hoyt Wilhelm -- 102
Ed Walsh -- 116
Addie Joss -- 170
Al Spalding -- 33
Kid Nichols -- 93
Mordecai Brown -- 46
Cy Young -- 203
Pete Alexander -- 211
Christy Mathewson -- 119
John Clarkson -- 55
Rube Waddell -- (39)
Whitey Ford -- (3)
Sandy Koufax -- 94
Dizzy Dean -- 52
Carl Hubbell -- 116
Hal Newhouser -- 38
Amos Rusie -- 164
Stan Coveleski -- 165
Bob Gibson -- 73
Tom Seaver -- 199
Lefty Gomez -- 66
Tim Keefe -- 105
Jim Palmer -- 169

Out of 25 pitchers who have been designated "great" both objectively (top 50 in ERA+) and subjectively (in the Hall of Famer), 23 (all except Rube Waddell and Whitey Ford) gave up fewer hits on balls in play than their team average. Over half (13) gave up fewer than 100 less than would be expected if this was not a skill than averaged out over a career.

If you add in the top ten win leaders who are in the Hall but not on the above list, you get:

Pud Galvin -- 106
Warren Spahn -- 182
STeve Carlton -- 31
Eddie Plank -- 62
Nolan Ryan -- 132
Don Sutton -- 137
Phil Niekro -- 170
Gaylord Perry -- (5)
Charlie Radbourne -- 50
Mickey Welch -- 43

So that's 9 out of 10 here, for a total of 32 out of 35 (91.4%).

Is that evidence sufficient to show that great pitchers give up fewer hits on balls in play?

There is no question that Voros is correct and the numbers jump around wildly year after year. The best in the stat one year could be the worst the next.

There is also no doubt, however, that among the universe of great pitchers, an overwhelming percentage give up fewer hits than their team average on balls in play. It is therefore pretty much nonsense to say that -- over the course of a career -- there is not a measurable skill here.

Am I willing to bet that Zito gives up fewer hits on balls in play next year than Martinez (or even than the team average)? Of course not. The variance is huge.

But, if it is a skill and not just luck (as the premise of this essay assumes), then it should be recognized as such and included in calculations of greatness.
   14. Voros McCracken Posted: December 13, 2002 at 02:09 AM (#607593)
"Let's take as our universe of pitchers, Top 25 leaders in ERA+ who are in the Hall of Fame."

And you've just begged the question (not intentionally). Since having a low H/BIP rate would lead to a low ERA+, naturally if you use ERA+ as the barometer of who the "best" pitchers are, the "best pitchers" will be somewhat selected to have low H/BIP.

If you used a DIPS measure to select the "best" pitchers, then you would have an argument.

In other words, you're including what you're trying to prove within the arguments itself, therefore leading to a simple but not very meaningful proof.

I think you will find that the best DIPS guys will be a touch better in H/BIP than the average pitcher because I found a relationship between more strikeouts and fewer hits, and more strikeouts would lead to better DIPS numbers as well.

In short, if you want to see whether the "best" pitchers have low H/BIP rates, one of the criteria for determining "best" can't be influenced by H/BIP, otherwise you've begged the question to an extent. (Note: begging a question is when you use your arguments as proof of themselves. E.G. if someone says "All good drivers keep their hands at 10 and 2" and you say "I know plenty of good drivers who keep their hands at 11 and 1," if that person responds "well those people aren't good drivers" he has just "begged the question.")
   15. MattB Posted: December 14, 2002 at 02:10 AM (#607597)
"Let's take as our universe of pitchers, Top 25 leaders in ERA+ who are in the Hall of Fame."

Voros wrote:
And you've just begged the question (not intentionally). Since having a low H/BIP rate would lead to a low ERA+, naturally if you use ERA+ as the barometer of who the "best" pitchers are, the "best pitchers" will be somewhat selected to have low H/BIP.

If you used a DIPS measure to select the "best" pitchers, then you would have an argument."


To an extent, you are absolutely right. Players with low ERA+ should have low h/bip, but I'm not sure if the way to correct for that is as easy as to just use DIPS ERA.

What my argument was, basically, is that while the article used only "defense independent" stats to promote Martinez over Zito, that analysis did not capture all the ways in which a pitcher could contribute.

To look at the player with the best DIPS ERA and then look at h/bip would be begging the question as well, however (just a different question). Who says that the DIPS ERA leaders is the proper category to look at (over the course of a career) when defining great pitchers? That's the conclusion, not the assumption.

The initial question was, in effect, "Do the best players have only lower DIPS ERA, or do they have lower h/bip as well." Take only the 35 players with the lowest DIPS ERA as your universe of best, and you've potentially stacked the deck as to what defines a "great pitcher."

I wanted to capture in my definition of great pitchers those who were the best in any pitching category (h/bip, wins, k, bb, hr). Assume a mythical pitcher who struck no one out, walked many, gave up lots of homers, hit lots of batters, but never gave up a single hit on a ball in play. This pitcher would likely have had a very low ERA, and I'd want him in my universe of "great pitchers" as well.

I recognize that, since most of the greatest pitchers were great at many things, it's difficult to tease out which are not related. I thought looking at ERA+, Hall of Fame status, and Wins would be a broad enough list that the pitchers caught in the net would be undeniably "great" in any combination of ways that greatness is defined.

I thought that it was valid enough to make my original point, which is that it is entirely likely that there is a skill there that, inside all of the noise, Barry Zito has to greater degree than Pedro Martinez (although it's likely too soon to tell for Zito). One should certainly look at the h/bip numbers with a grain of salt, but if the available data show that, over a long enough period, Zito has a cognizable skill that exceeds Pedro's, it is unfair to Zito to completely discount that particular skill in Cy Young voting.
   16. Mike Emeigh Posted: December 14, 2002 at 02:10 AM (#607598)
Zito's *ability* to prevent hits on BIP might be due to nothing more than the fact that he is a flyball pitcher, and flyballs are turned into outs at a higher rate than are groundballs. Generally, a group of flyball pitchers will have a lower H/BIP rate than a similar group of groundball pitchers.

-- MWE
   17. Marc Posted: December 15, 2002 at 02:10 AM (#607600)
If the writers can make the MVP choice that they made, then we can hardly be surprised at their Cy choice. Zito was the better choice of the two. Face it, the A's and their winning streak were "our top story" this year. Writers don't know who the best players are, but they know who made the best copy.
   18. zach Posted: December 15, 2002 at 02:10 AM (#607604)
Boston Newspaper editor,When did pedro become the best pitcher any of us have ever seen?? Greg Maddox is clearly.
   19. zach Posted: December 15, 2002 at 02:10 AM (#607605)
Boston Newspaper editor,When did pedro become the best pitcher any of us have ever seen?? Greg Maddox is clearly.
   20. MattB Posted: December 16, 2002 at 02:10 AM (#607614)
FJ,

I think you are misinterpreting what I've said.

You say:

"And last I looked Martinez has been much better when he's pitched (i.e., a greater pitcher) than Zito has been so then, according to your argument wouldn't it be more likely that MARTINEZ had this ability more than Zito? This shows why Voros was saying that your argument begs the question."

H/BIP is a small part of the overall pitching package. Would it be shocking if a pitcher who general consensus felt was much better hit more batters than a lower quality pitcher? Of course not, because he'd make up for it with other parts of his game. Better pitchers are not all better than worse pitchers in every category.

"You have Johnson at +320 and Waddell at -39."

These numbers are compared to his teammates. While it is possible that a pitcher will get more or less help from the same defense over a small period, it is unlikely that over a career Walter Johnson was helped by his defense that the other pitchers on his team.

That leaves "luck" or "skill". Defense should not be an issue.

"So, you're basically saying that Zito is like the best and that Martinez is like the worst of all great pitchers at turning BIP into outs."

That's not what I'm saying at all.

Johnson and Waddell are at the extremes in total, but may not be in rate. You'd have to divide the total by total balls in play to see how they compare to the other greats. I don't see why it's so unbelievable, though, that one pitcher who is better at strikeouts and walks would not be better at some other "skill", especially one that is significantly less important than the others.

"What should probably be discussed is whether H/IPs was mainly due to defense or just luck. If luck, hey, Zito got them out somehow, so he shouldn't be penalized for this. If defense, you have to factor in that Zito had better defense than Martinez when you're discussing Cy Young considerations."

As I said, defense should not be an issue here. If you want to call these large differences in totals over a career "luck", you'll need to show me some statistics to show me that these things aren't supposed to even out over 15 or 20 years.

So, if you count out defense and luck, that leaves a skill buried under a lot of noise. And if it is a skill, then it is unfair not to look at it when comparing two pitchers.
   21. MattB Posted: December 16, 2002 at 02:10 AM (#607623)
John M. --

Two reasons:

First, if the skill exists, it is a cop-out to say that since we cannot measure it exactly we should completely discard it. If, at the end of Addie Joss's career, we see that he had a great skill than almost none of his peers had, we can't go back and retroactively credit him with more points. Cy Young awards are awarded at the end of the year based on then-current knowledge. You only see it as a "fluke" later on. Absent true career-length knowledge, we have to make reasonable assumptions, giving players the benefit of the doubt where there is doubt.

It would be like saying in 1980 that there was no good way to measure defense (even though we know its a skill), so we'll just discount it and only measure offensive performance. If you believe it's there, you've got to try to measure it as best you can. You're more likely to come to the right answer than just discounting it 100%.

Second point: We know that flyball pitchers, as a whole, have a lower h/bip rate. We also know that flyball pitchers give up more homers. The lower h/bip helps make up for the homers. Leaving aside the advantage in h/bip penalizes Zito with the natural disadvantage for flyball pitchers without giving him to natural advantage for flyball pitchers.

Basically, we don't know if Zito has a skill or not, but the evidence so far points modestly to "yes." When determining who is the better pitcher in 2002, I don't think it's fair to Zito not look at h/bip at all.
   22. Danny Posted: December 18, 2002 at 02:11 AM (#607664)
Didn't the author violate the sabrmetric principle of not searching for statistics just to prove a conclusion one has already come to? This is what Gammons does. The author already believed that Martinez was the better pitcher, and then set out to find the statistics to prove it. He should have set out to find which statistics would prove who the better pitcher was, and then let the numbers decide. Also, how can one say Zito, a drastic flyball pitcher, got a lot of defensive help when is outfield was Giambi/Justice/Piatt, Long, and a hobbled Dye? All of them were below average fielders last year. If he was lucky, give him credit for that. It was part of his value.
   23. Dan Turkenkopf Posted: January 01, 2003 at 02:15 AM (#607949)
"Zito's *ability* to prevent hits on BIP might be due to nothing more than the fact that he is a flyball pitcher, and flyballs are turned into outs at a higher rate than are groundballs. Generally, a group of flyball pitchers will have a lower H/BIP rate than a similar group of groundball pitchers. "

This sounds like you consider this to be a random factor. But if fly balls are more likely to be outs than ground balls, then, all else equal, isn't a fly ball pitcher better, by definition, than a ground ball pitcher?


This is a little late so I don't know if anyone (Bernard) is still reading this - but I think the lower H/BIP rate is balanced by (if not outweighed by) the increased HR rate for FB pitchers.
   24. Boomer Posted: February 07, 2003 at 02:30 AM (#608744)
I hate to throw the "r" word (racism) around, but it's hard for me to account for the snubbings Pedro gets otherwise. If Clemens or Maddox had come back from an injury plagued season like Pedro had in 2001, there is no question at all that either would have won the Cy Young. The fact that two people completely left Pedro off their ballots for MVP in 1999 is equally absurd. I can see how somebody would never put a pitcher 1 or 2 or 3, but not even in the top ten? And, as was noted, that one voter had previously included TWO pitchers on his ballot.

Pedro is a tiny guy, for a power pitcher (he's barely average-sized for a regular Joe on the street). His arm needs to be nursed a little bit, and I don't see any reason why he should have made a meaningless start against Tampa Bay, who he has a history of dominating even more than other teams. If you don't think Pedro is tough, you obviously missed game five of the 2000 Division playoffs. He couldn't even lift his arm high enough to throw overhand, but he gutted out five no-hit innings. And stumped the mighty Yankees days later (and we would have won that year, if not for terrible umping--of course, I'm a sox die-hard, though, and complaining of umps is all we really ever have.)

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Adam M
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.6813 seconds
70 querie(s) executed