Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Tuesday, January 16, 2018

Andy Pettitte

Eligible in 2019

DL from MN Posted: January 16, 2018 at 10:10 AM | 51 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. DL from MN Posted: January 16, 2018 at 10:12 AM (#5607156)
Chuck Finley plus a lot of postseason credit
   2. . . . . . . Posted: January 16, 2018 at 10:35 AM (#5607168)
Mmmmmm I think that understates Pettitte. Pettitte played in front of some absolutely ghastly defenses in NY, and did so while being coached (by guess who!?) to be a pitch-to-contact GB pitcher. There's a 0.20 run difference between Finley and Pettitte in career FIP, and that doesn't even take into account the coaching issues that I've always suspected cost Pettitte a real peak and flattened out his career into the odd Sutton-ian shape.
   3. Carl Goetz Posted: January 16, 2018 at 10:37 AM (#5607169)
Not a bad comp actually. Pettitte's 1997 is better than any of Finley's years but Finley has 2 years better than Pettitte's 2nd best. Somewhat borderline in my mind, but definitely worthy of consideration.
   4. OCF Posted: January 16, 2018 at 12:45 PM (#5607309)
RA-based equivalent record: 208-161, big years bonus 16. Best equivalent seasons: 18-7, 18-9, 15-10, 13-11, 13-11. Lots and lots of seasons that come up with equivalent records like 13-11, 12-11, 11-10.

Unusually low number of decisions for his IP (9.58 innings per decision), unusual extent to which his actual W-L of 219-127 is better than his equivalent record. Essentially a Jack Morris-sized discrepancy there.

Some useful comparisons:

Pettitte 208-161 [16] (number in brackets is big years bonus)

Drysdale 209-157 [31]
Reuschel 221-174 [14]
Koosman 233-193 [21]
Tanana 245-220 [21]
Finley 199-156 [8]

Trying to put a single number value on a career W-L record, I have this virtually tied with

Moyer 241-211 [13]
Santana 146-79 [38]

(very different career shapes) and pretty close to Appier 172-116 [27]

I would put him ahead of Finley. The biggest weakness of the just-RA method is that it doesn't take into account defensive support, which is the main point of post #2.
   5. Carl Goetz Posted: January 16, 2018 at 01:13 PM (#5607333)
"The biggest weakness of the just-RA method is that it doesn't take into account defensive support, which is the main point of post #2."

Is the RA-based equivalent record something you calculate on your own? What is that exactly?

"Pettitte played in front of some absolutely ghastly defenses in NY, and did so while being coached (by guess who!?) to be a pitch-to-contact GB pitcher. "

I'm not saying I disagree with the idea that Pettitte was poorly coached; I don't honestly know enough about that situation to judge. But, in general, I have a hard time giving a guy too much credit for what he might have done under a better coach. I would consider giving some FIP credit if I feel RA-9 doesn't tell the whole story however.
   6. . . . . . . Posted: January 16, 2018 at 01:45 PM (#5607376)
I'm not saying I disagree with the idea that Pettitte was poorly coached; I don't honestly know enough about that situation to judge. But, in general, I have a hard time giving a guy too much credit for what he might have done under a better coach. I would consider giving some FIP credit if I feel RA-9 doesn't tell the whole story however.


There's two separate issues here. The first is that Pettitte played in front of what was likely the worst defense in the AL during much of his peak. The second is that he was coached to pitched to contact in front of that defense by the atrocious Mel Stottlemyre. Taking aside the coaching issue, Pettitte still needs to be considered in light of his defensive support. The coaching aspect is merely relevant b/c Pettitte's career is curiously flat, and in this case, it's because his pitch selection was highly suboptimal during the peak of his talents. I don't think you credit that but it makes his career make much more 'sense', so to speak.
   7. snapper (history's 42nd greatest monster) Posted: January 16, 2018 at 01:45 PM (#5607377)
Pettitte has a really large discrepancy between WARs. fWAR 689.9, bWAR 60.9. FIP doesn't seem to explain it: 3.74 career FIP, vs 3.85 ERA.

Anyone with any insight?
   8. Kiko Sakata Posted: January 16, 2018 at 02:04 PM (#5607393)
Unusually low number of decisions for his IP (9.58 innings per decision), unusual extent to which his actual W-L of 219-127 is better than his equivalent record.


I don't know if this is a typo or if you're missing his three seasons in Houston, but Pettitte had a career W-L record of 219-127 as a Yankee, but 256-153 overall (regular season - he was 19-11 in the postseason).

I have a feeling I might end up being Pettitte's best friend among HOM voters. He's the kind of pitcher my system really likes (e.g., Tommy John). Basically, a solid #2 starter for the equivalent of about 17 full seasons, counting his postseason. A team is going to win a lot of games - and, in the modern game, make a lot of playoff appearances - with that at or near the top of its rotation.
   9. OCF Posted: January 16, 2018 at 02:33 PM (#5607441)
Oops, that was a mistake on my part, copying down the wrong thing from bb-ref. So in fact he has one of the highest number of decisions for his IP that I've seen, with 8.11 innings per decision.

And the extent to which is actual record is better than his equivalent record is the greatest I've seen of any pitcher, far beyond Morris. (2nd biggest such discrepancy since the dead ball is Wells). Actual record minus equivalent record is 48-(-8).

And one of the largest such discrepancies the other way is Mariano Rivera. Which seems fitting.
   10. Jaack Posted: January 16, 2018 at 02:57 PM (#5607473)
Pettitte has a really large discrepancy between WARs. fWAR 689.9, bWAR 60.9. FIP doesn't seem to explain it: 3.74 career FIP, vs 3.85 ERA.

Anyone with any insight?


Well bWAR has its own defensive adjustments - Pettitte's RA9 WAR per fangraphs is 62.3. The rest of the difference is probably because Pettitte allowed a large number of unearned runs - including them, his (not E)RA is 4.27, which is pretty substantial.

I like Pettitte as a candidate, but a lot of that is postseason credit. The inning count is enormous, but he was basically the same guy in the postseason as the regular season. There wasn't really any year his performances were indespensible to the Yankees - his great performances were often balanced out by poor ones in the same series.

Purely looking at the postseason, Pettitte clearly ranks him lower than Christy Mathewson or Curt Schilling, as they rose to the occasion and pitched better than usual. But how does he compare to Koufax or Lolich, who had 50ish brilliant innings of postseason work? I feel like those guys willing teams to championships is more valuable than Pettitte's large number of good, not great games in the LDS and LCS, but I'm not sure on that.
   11. 6 - 4 - 3 Posted: January 16, 2018 at 02:58 PM (#5607474)
Totally subjective, but he never seemed like the ace. The Yankees were always trying (and often getting) to get someone better than him: Cone, Clemens, Mussina, Sabathia, etc. His highest CYA finish came in his second year, so he also had the feel of a guy who never quite lived up to his potential.

Off the top of my head, he's borderline (right at 30 WAA). I like the Chuck Finley plus postseason comp. I'm not sure what, if any, credit he should get for having a supposedly bad pitching coach or even a subpar defense. His career ERA was right around his career FIP, so hard to say how much it really hurt him.

My gut says that he should be on the outside looking in, but a closer look at some advanced metrics might convince me otherwise as well as further consideration about how much weight to give postseason performance, which was basically over a season's worth of IP. He's got to have pitched the most postseason innings, right? Or maybe one of the Braves?

And still need to read Kiko's book. I'm curious to better understand why John and apparently Petitte rate so highly.
   12. Carl Goetz Posted: January 16, 2018 at 03:04 PM (#5607485)
"Taking aside the coaching issue, Pettitte still needs to be considered in light of his defensive support."

Agreed, but BBRef WAR (based on RA/9) does adjust for defense. Even so, he's a guy like Lolich where I will take FIP WAR into consideration. I feel like he's a pretty easy HoMer by fWAR and more borderline by BBRef WAR.
   13. . . . . . . Posted: January 16, 2018 at 03:10 PM (#5607492)
Agreed, but BBRef WAR (based on RA/9) does adjust for defense. Even so, he's a guy like Lolich where I will take FIP WAR into consideration. I feel like he's a pretty easy HoMer by fWAR and more borderline by BBRef WAR.


But that adjustment is significantly less accurate (and, I've always suspected, tends to regress defense to the mean) as you go back in time.

I don't get folks acknowledging the role of defense and then saying he's Finley+. Crudely, his FIP is 0.2 better than Finley for a career. That's a meaningful difference. They're peers, sure, but Pettitte was clear cut above even ignoring the postseason.

I don't know if I think Pettitte was HOM material - I lean no - but he's an interesting case and certainly a reflexive (oh, 60 BBRefWAR, meh) approach is insufficient given his context.
   14. Carl Goetz Posted: January 16, 2018 at 03:20 PM (#5607498)
As for postseason credit, he started 44 games & 276.2 IPs (or 1.33 seasons) at roughly his career average ERA & WHIP (slightly better actually). At 60.9 WAR (BBRef), that equates to 5.14 WAR/ 44 Starts or 5.08 WAR/276 IP. Just a quick back of envelope calc, but that seems like a reasonable starting point for a bonus. I'd consider bumping that a little for the better competition in the postseason and the greater importance of the games but not sure about that part. At a minimum though, he deserves credit an extra regular season plus.
   15. Kiko Sakata Posted: January 16, 2018 at 03:21 PM (#5607500)
And still need to read Kiko's book. I'm curious to better understand why John and apparently Petitte rate so highly.


Basically, player performance isn't really linear. It's somewhat multiplicative. If you have above average players everywhere, even if they're mostly just a little above average, they'll win a ton of games; if you have below-average players, they'll lose a ton of games. Living in Chicago, the 2005 White Sox strike me as a good example of the former - no stars, but no holes.

Because of this, in calculating expected wins (eWins), I incorporate an adjustment called an expected inter-game win adjustment, which basically says, a good player's impact on his team will be magnified (and likewise for a bad player in the other direction). This effect is most pronounced for starting pitchers because they concentrate their performance within fewer games. So an above-average starting pitcher - as Pettitte was literally every season of his career - will have a stronger impact on team wins than would be estimated from a linear system.
   16. Carl Goetz Posted: January 16, 2018 at 04:00 PM (#5607553)
So in essence, a team should want 2 3-WAR players rather than a 5-WAR guy and a 1-WAR guy for a given pair of positions to fill?

Do you think it would follow then that its better for a single player to have consecutive 3-WAR seasons rather than 5 WAR 1 season and 1 the next? Traditional thinking says the 5 WAR season helps the team more to win the 1 pennant. However, by your statement, I'm assuming being a solid above average player both years eliminates 1 "hole" on both teams. Not sure where I come down on this issue. I've traditionally been more of a "peak" guy; the type who prefers 5&1, but its definitely something I'd like to understand better.



   17. Kiko Sakata Posted: January 16, 2018 at 04:13 PM (#5607561)
So in essence, a team should want 2 3-WAR players rather than a 5-WAR guy and a 1-WAR guy for a given pair of positions to fill?


If 2 WAR is average (and I think it is, by construction), then yes. I think a team should try to minimize its "holes". Although for position players, I'm not sure that it makes a HUGE difference. Moreover, in MY system, this is already baked into the players' eWins (eWOPA, eWORL) - which gives above-average players a little boost and below-average players a little kick. Although, even with eWins, the adjustment assumes you're adding a player to an average team; if you add an above-average player to an above-average team, then you get into interaction effects.

And, of course, it also matters what your goal is. In the old pennant-only days, you might prefer the erratic player even if the consistent guy produced a higher expected win total if the erratic guy increased your odds of winning, say, 95 games or whatever it might take to win a pennant. In the modern playoff system (including the one that Pettitte played in - I'm not sure that the second wildcard changes things much) I suspect you're better off just trying to maximize your expected wins.

I'm not really arguing how one should value peak vs. prime vs. career for an individual player (aesthetically, I like peaks, although I try to be pretty open-minded about HOW players accumulate their value in putting my ballot together). I'm just explaining why the starting point for whatever weighting scheme you'd want to apply to my numbers - i.e., the raw pWins, eWins, pWOPA, eWORL, etc. - are more favorable to pitchers like Tommy John and Andy Pettitte than I think BB-Ref tends to be.
   18. Kiko Sakata Posted: January 16, 2018 at 04:20 PM (#5607570)
Here's a comparison that I stumbled upon toward the end of Andy Pettitte's career that I've always liked.

For this career, Pettitte pitched 3,316 regular-season innings with an ERA+ of 117 (ERA of 3.85).
He pitched 276.2 postseason innings with an ERA of 3.81.
Adding those together, that gives Pettitte 3,592.2 innings pitched with an ERA+ of 117, maybe 118

Jim Bunning, in his career, pitched 3,760.1 innings with an ERA+ of 115. Bunning never pitched in the postseason.
If you throw out the first and last seasons of Bunning's career - both of which were short and terrible, he pitched 3,599.1 innings with an ERA+ of 119.

So, if you include his postseason work - and I see no reason why you wouldn't include his postseason work - Andy Pettitte's career numbers are very similar to those of Jim Bunning.
   19. cookiedabookie Posted: January 16, 2018 at 04:30 PM (#5607578)
By bWAR/200 IP, Pettitte's 3.67 is tied with Robin Roberts and Amos Rusie, and practically tied with John Clarkson, Tom Glavine, and Clark Griffith.

The group surrounding Pettitte, minimum 50 bWAR and 3000 IP:

Name bWAR/200IP
Fergie Jenkins HOF 3.77
Carl Hubbell HOF 3.76
Tim Hudson 3.75
CC Sabathia 3.71
John Clarkson HOF 3.70
Tom Glavine HOF 3.69
Robin Roberts HOF 3.67
Amos Rusie HOF 3.67
Andy Pettitte 3.67
Chuck Finley 3.65
Clark Griffith HOF 3.65
Charlie Buffinton 3.65
Orel Hershiser 3.63
Whitey Ford HOF 3.61
Juan Marichal HOF 3.60
Eddie Cicotte 3.60
George Uhle 3.59
Phil Niekro HOF 3.58

Look at that, he's just ahead of Chuck Finley. Among the group of 50+ bWAR and 3000 IP, he's tied for 29th out of 83 matches. That seems to be a pretty strong HoF argument to me.
   20. Carl Goetz Posted: January 16, 2018 at 04:33 PM (#5607579)
"So, if you include his postseason work - and I see no reason why you wouldn't include his postseason work - Andy Pettitte's career numbers are very similar to those of Jim Bunning."
Interesting comparison. That does come out to greater WAR for Pettitte (BBRef). I assume Bunning had better defenses behind him on average and that would account for the difference.
   21. Sweatpants Posted: January 16, 2018 at 09:21 PM (#5607750)
For this career, Pettitte pitched 3,316 regular-season innings with an ERA+ of 117 (ERA of 3.85).
He pitched 276.2 postseason innings with an ERA of 3.81.
Adding those together, that gives Pettitte 3,592.2 innings pitched with an ERA+ of 117, maybe 118

Jim Bunning, in his career, pitched 3,760.1 innings with an ERA+ of 115. Bunning never pitched in the postseason.
If you throw out the first and last seasons of Bunning's career - both of which were short and terrible, he pitched 3,599.1 innings with an ERA+ of 119.

So, if you include his postseason work - and I see no reason why you wouldn't include his postseason work - Andy Pettitte's career numbers are very similar to those of Jim Bunning.
By the same method, Pettitte's career numbers are also very similar to those of Wilbur Cooper, which isn't as beneficial a comparison.
   22. Carl Goetz Posted: January 17, 2018 at 08:17 AM (#5607836)
"By the same method, Pettitte's career numbers are also very similar to those of Wilbur Cooper, which isn't as beneficial a comparison."

Which shows up in Wilbur coopers lesser WAR numbers. Cooper was a deadball pitcher half his career, so those numbers aren't quite as impressive from him.
   23. Carl Goetz Posted: January 17, 2018 at 09:45 AM (#5607874)
My bad; I shouldn't post before my coffee :)

Obviously ERA+ is unaffected by his era since its already adjusted for that. The issue with Cooper is that he allowed a lot more unearned runs than Pettitte or Bunning.

Per BBRef 162 Gm Avg
Cooper 104 Runs 82 ER 22 Unearned 21% Unearned
Bunning 94 Runs 84 ER 10 Unearned 10.6% Unearned
Pettitte 102 Runs 92 ER 10 Unearned 9.8% Unearned

That a big chunk of runs that a pitcher is at least somewhat responsible for on which ERA+ lets him off scot free.
   24. Kiko Sakata Posted: January 17, 2018 at 11:06 AM (#5607970)
That a big chunk of runs that a pitcher is at least somewhat responsible for on which ERA+ lets him off scot free.


Some of that - maybe all of that - is era. A much larger percentage of Deadball Era runs were unearned. That said, I'd love to see BB-Ref or Fangraphs show an RA+ (or RA- in the case of Fangraphs) as unearned runs definitely vary between pitchers within the same era in ways that are properly attributed to the pitchers themselves (Kevin Brown - extreme groundball pitcher - vs. Curt Schilling - extreme flyball pitcher - is probably the best example of this).

If you trust BB-Ref's fielding numbers - and, in terms of direction at least, they're as good as you're likely to find - the difference is in fielding support. Per BB-Ref, Pettitte mostly pitched in front of bad defenses (outside of Houston, when he pitched in front of very good defenses) (career RA9def of -0.08; -0.14 in NY, +0.27 in HOU), Bunning pitched in front of average defenses (career RA9def of -0.01) and Cooper pitched in front of very good defenses (career RA9def of 0.11). One can certainly find fault with the specific magnitude of any of those numbers, but for three guys as closely matched on IP/ERA+, all that really matters is the signs and/or rank order of the three numbers and those are probably pretty solid in this case. (Bunning also has a somewhat different shape to his career than Pettitte and Cooper which could affect where one chooses to rank him among the three of them.)
   25. Carl Goetz Posted: January 17, 2018 at 11:44 AM (#5607999)
Thanks Kiko. I certainly suspected that was the case with respect to the defenses behind the 3 pitchers in question.
   26. Carl Goetz Posted: January 17, 2018 at 11:55 AM (#5608019)
Quick question: Do you know if RA9Def makes any adjustment for the pitcher himself? Example, if a team has a great SS-3B combo and a crappy 1B-2B combo, the pitcher himself (specifically his handedness) has a great deal of effect on how good his defense is. Also, if he's got a great infield and a poor outfield, his GB% would matter quite a bit.
   27. Kiko Sakata Posted: January 17, 2018 at 12:10 PM (#5608047)
Quick question: Do you know if RA9Def makes any adjustment for the pitcher himself? Example, if a team has a great SS-3B combo and a crappy 1B-2B combo, the pitcher himself (specifically his handedness) has a great deal of effect on how good his defense is. Also, if he's got a great infield and a poor outfield, his GB% would matter quite a bit.


Here's BB-Ref's article explaining WAR. This looks a bit different than when last I read it, and it includes references to the recent interaction w/ Bill James re: Judge and Altuve - but jumping to the explanation of "xRA_def, Adjusting for Team Defense", this is what I thought. They adjust for the number of BIP allowed by a pitcher - so, for example, Roger Clemens is less affected by his defense than Pettitte, because Clemens strikes more batters out - but they do not appear to adjust to handedness or the groundball tendency of the pitcher.

Their final formula (see last link in the previous paragraph) is

xRA_def = (BIP_pitcher)/(BIP_team) * TeamDefensiveRunsSaved

which looks like it makes no such adjustments.

To be honest, I'm not really in love with BB-Ref's fielding numbers (I think you and I have discussed this before) and I'm not crazy about BB-Ref's Pitching WAR, which is a bit too opaque for my taste. But, in this specific example, I think it's reasonable to say that Pettitte's Yankee teams had a lot of below-average fielders and, without doing any real research myself, Cooper's teams were probably pretty good defensively. Such that Pettitte was probably a bit better than his raw RA numbers and Cooper was probably a bit worse, and the two pitchers were close enough in terms of raw IP/RA+ numbers that this is enough to put Pettitte definitively ahead of Cooper. How far ahead of Cooper? I'm less confident about that.
   28. Carl Goetz Posted: January 17, 2018 at 12:39 PM (#5608101)
Fair enough. Thanks for the explanation.
   29. Ithaca2323 Posted: January 17, 2018 at 01:01 PM (#5608120)
I'm a 95-year old man when it comes to Pettitte and the postseason: You know what Andy Pettite was in the postseason? A #2 starter. In his career, he pitched in 16 ALDS. He was the Game 2 starter in 14 of them. Yeah, sometimes, the G2 starter pitches G5—as was the case for him in 1997 and 2000. But mostly, he wasn't "the guy" in the playoffs

As for the rest of his career, here's what I'll say:

4 WAR seasons
Jack Morris: 6
Pettitte: 3

Not my cup of tea
   30. Carl Goetz Posted: January 17, 2018 at 01:12 PM (#5608130)
Well, Pettitte did play on a lot of teams that could afford to go out and get the best pitcher in the league. And on average, game 2 of a playoff series is just as important as game 1.

Less than 2 WAR seasons
Morris: 8
Pettitte: 2

I will agree that Pettitte's case is primarily a career case , which will hurt him with me once I'm actually ranking him with other pitchers on the ballot. That said, the career case should be bolstered by at least 1.33 additional seasons worth of stats or a minimum of 5 WAR (as I explained above) over his career. And that's just counting the IPs as the same value as regular season IPs (which is my preference). If you value postseason IPs as more valuable, you'd give him more credit. In most non-Yankee cases, this is too small a sample size for me to do this so I rate them as of equal importance to the regular season. I won't try to divide it among his seasons where he played in the postseason. I'll just add the total to his career value.
   31. OCF Posted: January 17, 2018 at 01:34 PM (#5608151)
The issue with Cooper is that he allowed a lot more unearned runs

I have Cooper in my spreadsheet, with an RA+ equivalent record of 220-166. Which puts the difference between him and Pettitte as 12-5, which is definitely in Cooper's favor. And because I'm using RA, I'm charging him for the unearned runs. You have to understand that everyone in Cooper's time allowed more UER than anyone does today; it tends to wash out. But you'll notice that the comparisons I've been making have generally been with post-WWII pitchers, and usually on the recent side of that. The era differences loom large, especially when it comes to simply accumulating innings in a career.
   32. Ithaca2323 Posted: January 17, 2018 at 01:35 PM (#5608152)
Well, Pettitte did play on a lot of teams that could afford to go out and get the best pitcher in the league.


Sure. But Pettitte was also, for most of his career, worth between 2-4 WAR a season. That just feels like the epitome of a good pitcher to me. A #2 starter.

I actually think #1 nails it. He's Finley with postseason bulk
   33. Carl Goetz Posted: January 17, 2018 at 01:38 PM (#5608155)
I agree with that. I just didn't like the Morris comp since he's clearly better than Morris.
   34. . . . . . . Posted: January 17, 2018 at 01:38 PM (#5608156)
Ah the rigorous “he feels like” analysis.
   35. Ithaca2323 Posted: January 17, 2018 at 02:00 PM (#5608182)
Ah the rigorous “he feels like” analysis.


Fair. But of course, it's all factually backed up.

Three seasons of 4+ WAR, three seasons in Top 10 in pitcher WAR.
117 ERA+ (And just three full seasons better than 112)
Rarely led the league in anything significant
60.8 Career WAR, 34.8 WAR7

This is all fringy HOFer at best. He was a good pitcher who was great for two years and very good in a third. He lasted for a long time, and he pitched for some really good teams.

   36. Bleed the Freak Posted: January 17, 2018 at 06:15 PM (#5608410)
Fair. But of course, it's all factually backed up.

Three seasons of 4+ WAR, three seasons in Top 10 in pitcher WAR.
117 ERA+ (And just three full seasons better than 112)
Rarely led the league in anything significant
60.8 Career WAR, 34.8 WAR7

This is all fringy HOFer at best. He was a good pitcher who was great for two years and very good in a third. He lasted for a long time, and he pitched for some really good teams.


This is his baseline assessment.

His Baseball Gauge WAR is 64.5.
His Fangraphs WAR is 68.9
His Win-Loss WORL is 44.2p, 40.2e, Kiko's recommendation is ~2/3 e, ~1/3 p, so 41.6 p/e - ~73 translated to a WAR context.

If you include post-season work, which were ~277 high stress innings, Pettitte adds ~5-6 WAR to his resume.
   37. Ithaca2323 Posted: January 18, 2018 at 10:33 AM (#5608653)
Even by fangraphs, his year by year WAR among pitchers tells the exact same story. Good, very occasionally very good, for a long time.

1995: T-31st
1996: T-17th
1997: 5th
1998: 32nd
1999: 50th
2000: T-24th
2001: 7th
2002: Did not meet IP requirement
2003: 11th
2004: Did not meet IP requirement
2005: 8th
2006: 36th
2007: 20th
2008: 28th
2009: 33rd
2010: Did not meet IP requirement
2012: Did not meet IP requirement
2013: 40th
   38. DanG Posted: January 18, 2018 at 10:35 AM (#5608657)
Here are Pettitte's sims per BB-Ref. Players within 7 pitching WAR, 7 ERA+ and 500 IP:

Player         WAR ERA+    IP   W  WAAFrom   To
Luis Tiant    66.1  114 3486.1 229 34.5 1964 1982
D
Eckersley  62.5  116 3285.2 197 30.6 1975 1998 HOM
Juan Marichal 61.9  123 3507.0 243 30.0 1960 1975 HOM
David Cone    61.7  121 2898.2 194 35.6 1986 2003 HOM
Don Drysdale  61.2  121 3432.0 209 28.6 1956 1969 HOM
Andy Pettitte 60.9  117 3316.0 256 30.2 1995 2013
CC Sabathia   60.7  117 3317.0 237 29.6 2001 2017
Jim Bunning   60.3  115 3760.1 224 28.3 1955 1971 HOM
Mark Buehrle  59.2  117 3283.1 214 28.4 2000 2015
Chuck Finley  58.5  115 3197.1 200 28.6 1986 2002
Tim Hudson    57.2  120 3126.2 222 30.4 1999 2015
Dave Stieb    57.0  122 2895.1 176 31.1 1979 1998 HOM 
   39. . . . . . . Posted: January 18, 2018 at 12:12 PM (#5608794)
Fair. But of course, it's all factually backed up.

Three seasons of 4+ WAR, three seasons in Top 10 in pitcher WAR.
117 ERA+ (And just three full seasons better than 112)
Rarely led the league in anything significant
60.8 Career WAR, 34.8 WAR7

This is all fringy HOFer at best. He was a good pitcher who was great for two years and very good in a third. He lasted for a long time, and he pitched for some really good teams


This is, to be fair, lame and stupid 'factual' backup, for the reasons being discussed upthread:
- the WAR calculation depends on BBRef's defensive adjustments, which are (I assume) black box to you and less precise during Pettitte's era than they are today;
- the ERA+ completely disregards the importance of defense to ERA, which in Pettitte's case matters because he played in front of unusually bad defenses outside of his few years in Houston
- the black ink test is like, the dumbest 'factual backup' you could cite, though I guess technically not barred as a criteria for the HoM, though I've rarely seen long time voters cite it as anything other than a curiosity.
- see above re black box and defensive adjustments, and here we can also add in the postseason credit

Pettitte is a great litmus test for whether you think rigorously, because on a first look he's basically unqualified but he gets about as much non-straight-WAR credit as any near-modern player will get, which puts him right near (but probably slightly under) the line. So his case separates the men who do their own math from the boys who accept someone else's model (that they cant replicate) as the one-number authoritative word on merit.
   40. jmurph Posted: January 18, 2018 at 12:22 PM (#5608803)
This is, to be fair, lame and stupid

Ha!
   41. Carl Goetz Posted: January 18, 2018 at 01:40 PM (#5608876)
"- the WAR calculation depends on BBRef's defensive adjustments, which are (I assume) black box to you and less precise during Pettitte's era than they are today;"
"So his case separates the men who do their own math from the boys who accept someone else's model (that they cant replicate) as the one-number authoritative word on merit."

That's quite an assumption. An outline of TZ is available online and the Fielding Bibles tell you what goes into DRS. I'm sure Ithaca doesn't have the raw data in front of him, but that doesn't mean he quoting numbers from thin air with no understanding of them. Not everyone has time to create an uber-stat of their own from scratch; myself included. So I learn as much as I can about what goes into each stat; pick my favorite (BBRef in my case), make a few adjustments for the issues I have and then look at other numbers which I believe shed light at a slightly different angle (Fangraphs and Kiko's are my most common). Even if I (or someone else) did create my own model from scratch, its still a model, which generates an estimate; just like BBRef, Fangraphs etc.
Ultimately, Pettitte does require more than just looking at JAWS or career WAR. His case as I see it comes down to 3 primary factors 1) How much postseason credit do you give, 2)Which WAR (or other system) do you use because FIP vs RA/9 (w/Adjustments for Def) matters here and 3) Where do you stand on Career vs. Peak; ie if you love Dizzy Dean, you may not like Pettitte.
   42. . . . . . . Posted: January 18, 2018 at 02:39 PM (#5608930)
Ok, but Carl, where are you getting info on Pettitte's BABIP distribution pre-2002 from? Pettitte substantially changed his pitching style from his early career to his middle-late career, and I don't have a source that gives me any info on what his batted balls looked like when he was throwing all those cutters inside.

Now, you might say, this is only relevant if the Yankees had really hideous infield defense such that the GB/FB distinction matters. Let me introduce you to the 1999-2003 New York Yankees middle infield. I believe these are all TRS-based since they're from BBref, so make of them what you will:


1999
Jeter -15
Knoblauch -11

2000
Jeter -23 (still stunning!)
Knoblauch -10

2001
Jeter -17
Soriano -19

I don't know how often you play with TRS, but these are awful, awful numbers. Everyone remembers how bad Jeter's defensive stats were, but incredibly, the Yanks were trotting out second basemen who were just as bad!

2002, now we have data showing that Pettitte was an FB pitcher in this season. Wonder why?
Jeter -18
Soriano -13

2003, the infield defense finally improves a bit, everyone but Jeter is within a few runs of average - the team defense still sucks but it's not so concentrated in the middle infield.
   43. Carl Goetz Posted: January 18, 2018 at 03:00 PM (#5608948)
"Ok, but Carl, where are you getting info on Pettitte's BABIP distribution pre-2002 from?"
I'm not. I'm examining his FIP (and other metrics) to see if there's more to the story than RA/9 is telling. And you definitely didn't need to prove to me that the 99-03 Yankees were bad up the middle :) Now RA/9 is making an estimation of how much that defense hurt Pettitte. Its not perfect but neither is any system you or I might create on our own. That said, I am certainly open to any arguments you may have to make regarding the size of adjustments to to Pettitte for his defense or lack thereof as I tend to agree that he needs some adjustment. I can't say I disagree much with any of your specific arguments regarding Pettitte.
I just didn't like the calling Ithaca's argument "lame and stupid" or the notion that you need to create your own model to have anything valuable to say on the topic. He introduced himself as a 95-year-old man and he was referencing ERA+ and WAR (and I don't recall him citing Pettitte's Win total once). He sounds like a reasonably progressive baseball mind. He may be inclined to listen to your arguments if he's not being insulted.
   44. Mike Emeigh Posted: January 23, 2018 at 11:30 AM (#5611673)
And you definitely didn't need to prove to me that the 99-03 Yankees were bad up the middle :)


They really weren't as terrible as the metrics suggest (although they weren't good either); it's in large part an illusion of the ball-in-play distribution against them, that metrics don't pick up.

I looked at a lot of video clips online back when I wrote the Jeter series in 2002. The Yankees of that era had a pronounced right-side skew on balls in play. Because of that, they positioned their fielders very differently than did most teams. The 3B played well off the line, over toward the hole. Jeter shaded up the middle, and the 2Bs were shifted toward 1B. As a result, the balls that did get through against them were in zones where other fielders would make plays, and the balls on which they did make plays didn't make up for the difference.

-- MWE
   45. Carl Goetz Posted: January 23, 2018 at 01:09 PM (#5611752)
So you're saying that the Yankees positioning made Brosius look really good at the expense of Jeter and Knoblauch/Soriano? I guess what's important to Pettitte's case is where his BABIPs stood compared to the league at the time; adjusted for park of course.
   46. Chicago Joe Posted: January 28, 2018 at 10:18 PM (#5614771)
I don't think Ithaca was saying he's actually a 95 year old man, just that when it comes to Pettite he's a crank: get off my lawn, yelling at clouds, etc. A more Conservative approach, if you will.
   47. . . . . . . Posted: February 02, 2018 at 03:01 PM (#5618190)
They really weren't as terrible as the metrics suggest (although they weren't good either); it's in large part an illusion of the ball-in-play distribution against them, that metrics don't pick up.

I looked at a lot of video clips online back when I wrote the Jeter series in 2002. The Yankees of that era had a pronounced right-side skew on balls in play. Because of that, they positioned their fielders very differently than did most teams. The 3B played well off the line, over toward the hole. Jeter shaded up the middle, and the 2Bs were shifted toward 1B. As a result, the balls that did get through against them were in zones where other fielders would make plays, and the balls on which they did make plays didn't make up for the difference.


FWIW, I think this misstates the issue. You and others (including me!) argued that the defensive ability of the Yankees infield wasn't as bad as the metrics suggest. The defensive value, however, was. It was never clear whether that was coaching or Jeter just positioning himself wrong or whatever, but more balls did in fact get through the Yankee infield than nearly any other infield in the contemporary era over a multi-year period.
   48. Blastin Posted: February 02, 2018 at 03:54 PM (#5618228)
So... the Yankee dynasty could have been considerably BETTER?
   49. . . . . . . Posted: February 02, 2018 at 04:17 PM (#5618251)
So... the Yankee dynasty could have been considerably BETTER?


This is a theadjack, but: yes yes, a thousand times yes. Incredibly strong teams undermined by glitch roster construction and bad coaching. 2003 could easily have been a 105 win team. Soriano was wasted as a 2B - based on how good he was as an OF in his early 30s, he probably would've been worth a win or two more as an CF, and a better 2B could've helped cover for defensive limitations at SS and 1B. Williams still had good range but his arm was just brutal in CF, and probably should've been moved to LF that season (and 2004). Matsui was a good signing in the sense that he produced more WAR than he was paid for over his contract but Kent or Durham would've been better signings for that team TACTICALLY and also for 2004.

Mussina-Clemens-Pettite-Wells-Weaver was just an incredible rotation, and backed by Rivera. Damn. Both the Red Sox and Yankees were so ####### strong in 2003 and 2004, we went more than a decade until we saw teams like that again in the last couple of seasons.
   50. Carl Goetz Posted: February 02, 2018 at 04:23 PM (#5618256)
"So... the Yankee dynasty could have been considerably BETTER?"

Of course they could. They've been paying for a good 120 wins a year for years. They're losing wins somewhere :)
   51. Carl Goetz Posted: February 02, 2018 at 04:27 PM (#5618260)
"Incredibly strong teams undermined by glitch roster construction and bad coaching."

I tend to agree. Ironically, Joe Torre is in the HOF for the wrong job. He's as overrated as a manager as he was underrated as a player.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
Dock Ellis
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 1.6765 seconds
41 querie(s) executed