PDA

View Full Version : Thoughts about "Winning Pitcher" stat



durl
06-11-2009, 11:44 AM
Last night's game (5/10) made me start thinking again about what constitutes a "Win" for a pitcher. This may have been discussed years before I arrived here, so feel free to tell me if this is worthless rambling.

Harang went 7 2/3 with 5 hits and ZERO earned runs and left with the lead. The closer blows the win and a pitcher who just happened to be on the mound before the Reds took the lead gets the win.

So what would be the ramifications if the pitcher who went 5+ innings and left with the lead were to get credit for the win? While I realize this could create a very convoluted system, it just seems to be little consolation for a guy who pitches a shutout and has the lead to only get a "Quality Start" on his record.

redsfandan
06-11-2009, 11:50 AM
This may have been discussed years before I arrived here, so feel free to tell me if this is worthless rambling.

It's not.

Imo, wins is one of the most overrated stats. Yes if a pitcher manages to get 20+ wins chances are he really is one of the better pitchers that year. But there have also been pitchers that pitched great but their season win-loss record didn't show it.

reds1869
06-11-2009, 11:50 AM
I don't put too much stock in wins and losses. After all, a guy could have an ERA of 1.00, a WHIP of 0.111, throw nothing but complete game one hitters and still lose every single ball game. On the other hand, a pitcher could have ridiculously inflated numbers but still win because he pitches for the Yankees. Relievers with a lot of wins often blew saves to get those wins. I long ago decided to ignore W/L as a measure of value.

hebroncougar
06-11-2009, 11:54 AM
I don't put too much stock in wins and losses. After all, a guy could have an ERA of 1.00, a WHIP of 0.111, throw nothing but complete game one hitters and still lose every single ball game. On the other hand, a pitcher could have ridiculously inflated numbers but still win because he pitches for the Yankees. Relievers with a lot of wins often blew saves to get those wins. I long ago decided to ignore W/L as a measure of value.

Me too. I wish Dusty would have, instead of bringing Harang back to try and get him the all important win. Wins and saves are two of the most (if not the most) meaningless stats in baseballl.

TheNext44
06-11-2009, 11:57 AM
Bob Gibson has a 1.12 ERA and went 19-9. That says it all.

The rule about having to be on the mound when the winning run scores is what throws it all out of wack. Drop that rule, and let the official scorer decide on his own, and it would be much fairer.

RedsManRick
06-11-2009, 12:44 PM
Or just get rid of the win entirely. Seriously, at the most basic level, no single player can win or lose a game. It's silly to use a "stat" that infers such. Even the somewhat convoluted "quality start" is much better assessment of pitcher performance than a "win". The same can be said about the save. Imagine if for hitters used "game winning RBI" or game winning run scored" as a primary stat for them -- it's silly on the face of it because it describes the context in which a performance occurred rather than the performance itself.

It's really not hard to say "100 IP, 3.50 ERA" and be done with it. That certainly tells me more than 11-4 does.

nate
06-11-2009, 01:28 PM
I think this example fully illustrates the usefulness of "wins" in evaluating a pitcher:



Rk Pos Age W L W-L% ERA G GS GF CG SHO SV IP H R ER HR BB IBB SO HBP BK WP BF ERA+ WHIP H/9 HR/9 BB/9 SO/9 SO/BB
1 SP Bronson Arroyo 32 7 4 .636 5.00 12 12 0 1 0 0 77.1 79 45 43 13 26 1 39 4 0 1 327 89 1.358 9.2 1.5 3.0 4.5 1.50
2 SP Johnny Cueto 23 6 3 .667 2.33 12 12 0 0 0 0 81.0 61 22 21 8 24 0 58 3 0 2 320 191 1.049 6.8 0.9 2.7 6.4 2.42


To me, the only entity that should get credit for a "win" or "loss" is the team.

RedlegJake
06-11-2009, 01:47 PM
A more meaningful stat could be derived from the QS if it included being ahead or tied with 4 or less or behind with, say 2 allowed after 6. I agree, though, that wins for a pitcher are meaningless. They should at the least be the discretion of the scorer.

bucksfan2
06-11-2009, 01:55 PM
Seriously, at the most basic level, no single player can win or lose a game. It's silly to use a "stat" that infers such. Even the somewhat convoluted "quality start" is much better assessment of pitcher performance than a "win".

You ever seen Pujols take over a game?

I agree with you that no single player can win a game, I just think pitchers W-L record is a bench line for measurement. I think "quality start" has some value but it can be somewhat vague as well. A pitcher pitches a complete game in which he gives up 4 runs in a 10-4 victory. That wouldn't register as a quality start or even a good ERA but it sure helps a team out. W-L record is like any stat taken by it self, it is just way too vague. When you combine that with various other measures you can get a more accurate picture of the pitcher as a whole. I do think that most executives and CY Young voters have begun to use other stats as a better measuring stick than W-L. WHIP, ERA, K, and BB/9 all have become almost common when measuring pitchers.

Sea Ray
06-11-2009, 01:55 PM
By far the most meaningless stat is the Hold. I looked at a box score the other day and the same pitcher got a Hold and a loss. Apparently he left with a lead but his runners scored after he left resulting in a loss. I've also seen LOOGYs come in, walk the only batter they face and get credited with a hold. Throwout the Hold stat altogether.

traderumor
06-11-2009, 01:56 PM
The "Winning Pitcher" doesn't make any more sense than awarding a win to the QB in football. The save is just as irrelevant, IMO. The ridiculous thing is that managers seem to make decisions at times on these arbitrary statistical designations. For example, Harang wanting to get the last out of the fifth against the Astros so he could "get the win," and Dusty buying into it. Many managers also buy into the "save situation" mentality in bringing in their "closer" also.

durl
06-11-2009, 02:14 PM
I agree with those who say that Win/Loss is basically a useless stat. Even so, it's still a very dominant stat in the game. And I believe that having some type of quick stat is good to help us see how well a pitcher is doing.

We've seen a lot of growth in the acceptance of OBP and SLG, so perhaps it's time to start promoting a new way to grade pitchers.

nate
06-11-2009, 02:15 PM
You ever seen Pujols take over a game?

I agree with you that no single player can win a game, I just think pitchers W-L record is a bench line for measurement.

OK, so what exactly does it measure about an individual pitcher?

nate
06-11-2009, 02:17 PM
I agree with those who say that Win/Loss is basically a useless stat. Even so, it's still a very dominant stat in the game. And I believe that having some type of quick stat is good to help us see how well a pitcher is doing.

I kind of like FIP (http://www.hardballtimes.com/main/statpages/glossary/#fip) for that.


Fielding Independent Pitching, a measure of all those things for which a pitcher is specifically responsible. The formula is (HR*13+(BB+HBP-IBB)*3-K*2)/IP, plus a league-specific factor (usually around 3.2) to round out the number to an equivalent ERA number. FIP helps you understand how well a pitcher pitched, regardless of how well his fielders fielded. FIP was invented by Tangotiger (http://www.tangotiger.net/).

Danny Serafini
06-11-2009, 02:23 PM
(HR*13+(BB+HBP-IBB)*3-K*2)/IP

There's no way most fans are going to relate to that at all. It looks like some sort of garble you'd see on the chalkboard of a math class. It may spit out a more meaningful number than wins, but there's no ease of use to that, so it won't catch on.

Hoosier Red
06-11-2009, 02:27 PM
I kind of like FIP (http://www.hardballtimes.com/main/statpages/glossary/#fip) for that.


Yeah, that'll catch on quickly.

I think many are too quick to dismiss it. Its a stat that is useful at the margins in that there are few 300 game winners who were stiffs and its hard to go 0 and 23 and have a good year. While luck does play into it a lot, the stat also tends to favor pitchers who "give themselves a chance to win."

As an analogy, when I was selling loans, I was only the first person in the process, it could have been a tough customer(bad offense), or there may have just been a better offer somewhere else(better pitcher), or the processors could have screwed it up for me(bullpen.) I was howling mad when these happened, especially the last one, but in the end the total number of loans reflected how often I put myself in a good position to make a sale.

jojo
06-11-2009, 03:18 PM
(HR*13+(BB+HBP-IBB)*3-K*2)/IP

There's no way most fans are going to relate to that at all. It looks like some sort of garble you'd see on the chalkboard of a math class. It may spit out a more meaningful number than wins, but there's no ease of use to that, so it won't catch on.

Then I guess something like OPS doesn't have a chance:

OPS = {AB*(H+BB+HBP)+TB*(AB+BB+SF+HBP)}/{AB*(AB+BB+SF+HBP)}

:cool:

nate
06-11-2009, 03:24 PM
I couldn't really be bothered less if "the fan" is "scared" by the numbers.

Those interested in expanding their statistical baseball knowledge beyond the back of the baseball card are going to be in for math that's a bit more complex than counting stats.

TheNext44
06-11-2009, 03:32 PM
I kind of like FIP (http://www.hardballtimes.com/main/statpages/glossary/#fip) for that.

The problem I have with FIP as a general stat to evaluate pitchers is that assumes that every pitcher has the same BABIP, which really isn't the case, or is at least debatable. Basically, is assumes that pitchers have little or no control over whether a batted ball becomes a hit or an error.

There is the whole McCracken/Tippett debate. I am sure many are aware of it. If not, the short story is that McCracken did a three year study and found that many pitcher's BABIP varied from year to year, which implied that limiting hits was not a skill, but dependent more on your team's defense. Tippett did another study looking at pitchers entire career and concluded that most did have a steady BABIP, but many would have one or two off years. Therefore, the volatility that McCracken found was due to off years instead of a lack of skill. Hope that made sense.

Even Tango, who invented them, uses FIP and xFIP in a more limited sense, after Tippett's findings.

It clearly does give us a lot of info on the talent and skill of a pitcher, but is not very predictive, and unfairly hurts "contact" pitchers who rely on a their defense (i.e. Bronson Arroyo) while unfairly rewarding "K" pitchers who get hit hard (i.e Matt Belisle).

Still, it a solid stat, maybe the best one out there, but it is a bit flawed.

nate
06-11-2009, 03:39 PM
The problem I have with FIP as a general stat to evaluate pitchers is that assumes that every pitcher has the same BABIP, which really isn't the case, or is at least debatable. Basically, is assumes that pitchers have little or no control over whether a batted ball becomes a hit or an error.

There is the whole McCracken/Tippett debate. I am sure many are aware of it. If not, the short story is that McCracken did a three year study and found that many pitcher's BABIP varied from year to year, which implied that limiting hits was not a skill, but dependent more on your team's defense. Tippett did another study looking at pitchers entire career and concluded that most did have a steady BABIP, but many would have one or two off years. Therefore, the volatility that McCracken found was due to off years instead of a lack of skill. Hope that made sense.

Even Tango, who invented them, uses FIP and xFIP in a more limited sense, after Tippett's findings.

It clearly does give us a lot of info on the talent and skill of a pitcher, but is not very predictive, and unfairly hurts "contact" pitchers who rely on a their defense (i.e. Bronson Arroyo) while unfairly rewarding "K" pitchers who get hit hard (i.e Matt Belisle).

Still, it a solid stat, maybe the best one out there, but it is a bit flawed.

I understand there are subtleties and disagreements as to which defensive independent pitching stat is most accurate. However, all of them are better at isolating individual pitcher effectiveness than "Wins."

jojo
06-11-2009, 03:40 PM
When did Matt Belisle start missing bats?

traderumor
06-11-2009, 03:46 PM
When did Matt Belisle start missing bats?Since he lost his major league job. Can't hit what he aint throwing ;)

Blitz Dorsey
06-11-2009, 03:54 PM
I tell you one thing that has always bothered me: A pitcher should NEVER be awarded a win IMO when he comes in, gives up 2-3 runs to tie the game, then the offense comes back in the bottom half to win the game. When closers come in with a 3-0 lead, give up three runs in the ninth and then get the "win" it's a complete joke. In situations like that, either the starting pitcher should get the win or there should be no win awarded at all.

reds1869
06-11-2009, 03:59 PM
I tell you one thing that has always bothered me: A pitcher should NEVER be awarded a win IMO when he comes in, gives up 2-3 runs to tie the game, then the offense comes back in the bottom half to win the game. When closers come in with a 3-0 lead, give up three runs in the ninth and then get the "win" it's a complete joke. In situations like that, either the starting pitcher should get the win or there should be no win awarded at all.

That's where there needs to be a rule change. currently, the official scorer can award the win to another reliever if the pitcher who technically picked up the win is deemed "ineffective" at the scorer's discretion. The rule should be modified to allow the starter to be awarded a win in this case.

Danny Serafini
06-11-2009, 03:59 PM
Then I guess something like OPS doesn't have a chance:

OPS = {AB*(H+BB+HBP)+TB*(AB+BB+SF+HBP)}/{AB*(AB+BB+SF+HBP)}

:cool:

That wouldn't have a chance, but here's something that would:

OPS = OBP+SLG

Much, much easier way of presenting things.

traderumor
06-11-2009, 04:21 PM
I tell you one thing that has always bothered me: A pitcher should NEVER be awarded a win IMO when he comes in, gives up 2-3 runs to tie the game, then the offense comes back in the bottom half to win the game. When closers come in with a 3-0 lead, give up three runs in the ninth and then get the "win" it's a complete joke. In situations like that, either the starting pitcher should get the win or there should be no win awarded at all.Yep, that type of win is about as relevant as the short-lived GWRBI. On the flip side, why does a starter get a win when he pitches bad for 5 innings, but is able to hang around that long because his team scored a lot of early runs? Of course, paying any attention to the stat is a choice, as is railing at those who still choose to pay any attention to a pitcher's W-L record.

blumj
06-12-2009, 07:07 AM
I'd prefer a starting pitcher's record to just be his team's record in games he started, really the simplest thing possible, and relievers don't need to have records. At least, it's something I find myself wanting to know more than pitcher's records the way they do it now.

BCubb2003
06-12-2009, 09:27 AM
It's interesting how Quality Start caught on, for all its flaws. Give your formula a friendly name and it might catch on, too.

traderumor
06-12-2009, 10:10 AM
I'd prefer a starting pitcher's record to just be his team's record in games he started, really the simplest thing possible, and relievers don't need to have records. At least, it's something I find myself wanting to know more than pitcher's records the way they do it now.I don't see how that provides any better information. That is still very dependent on run support and bullpen. It is probably not going anywhere as an ingrained tradition, but it would be nice if GMs, managers and media quit ever referring to it as proof that a pitcher is doing something right.

It is so ludicrous to see minor league reports on FSN where they show Travis Wood with a 1 something ERA, then say "he has pitched in bad luck, only being 1-3." Why not just leave it at "this guy is getting the job done?"

Then you see the sour look on the starter, who is disappointed when the reliever coughs up the lead, and you know it is secondarily because his team is now losing, and is primarily because they do not get the "W."

Then there is the myth that causes millions of dollars to be flushed down the toilet on the likes of Eric Milton, whose contract was justified by Reds leadership because "he is a proven winner," based on his historic W-L records.

bucksfan2
06-12-2009, 10:44 AM
It is so ludicrous to see minor league reports on FSN where they show Travis Wood with a 1 something ERA, then say "he has pitched in bad luck, only being 1-3." Why not just leave it at "this guy is getting the job done?"

Technically if he isn't winning the games he isn't getting the job done. I would imagine that over the course of 30-35 starts in a season the good losses as well as bad wins would even themselves out. Of course there will always be outliers, Harang comes to mind right off the bat, but if a pitcher has a sub 3 ERA bets are that he will have quite an impressive W-L record.

westofyou
06-12-2009, 10:50 AM
Technically if he isn't winning the games he isn't getting the job done.

Jim Abbott in 1992 wasn't getting the job done based on the above belief, I'd say otherwise myself, he was getting his job done.


JIM ABBOTT

1992

YEAR TEAM AGE W L PCT G GS CG SV GF IP H R ER BB SO ERA RSAA
1992 Angels 24 7 15 .318 29 29 7 0 0 211 208 73 65 68 130 2.77 32
TOTALS 7 15 .318 29 29 7 0 0 211 208 73 65 68 130 2.77 32
LG AVERAGE 12 12 .500 3 0 211 208 102 93 80 127 3.95 0

BCubb2003
06-12-2009, 11:09 AM
One guy gives up two runs in seven innings, loses 2-0. One guy gives up four runs in seven innings, wins 6-4. There has to be a way to measure how well each did, especially if you're thinking, "Who would I want on my team?"

RichRed
06-12-2009, 11:23 AM
Nolan Ryan's 1987 and Roger Clemens' 2004 seasons are almost mirror images of each other statistically, with Ryan actually coming out a bit ahead in some areas.

Yet Clemens' record was 18-4 while Ryan's was 8-16.

Did Nolan just not want it enough?

There are probably thousands of examples like this that show the fallacy of looking at W/L records as something especially meaningful.

bucksfan2
06-12-2009, 12:00 PM
One guy gives up two runs in seven innings, loses 2-0. One guy gives up four runs in seven innings, wins 6-4. There has to be a way to measure how well each did, especially if you're thinking, "Who would I want on my team?"

There are ways to measure "Who would I want on my team" other than W-L. As I stated before the W-L may just be used as a benchmark guide. I have never heard a GM sign a player strictly based upon W-L record.

I guess my question is do the good losses and bad wins balance. Are there many outliers? WOY are there many seasons like Jim Abbot's? I would look to guys presently like Brandon Webb, Lincecum, Cain, and Jake Peavy who are great pitchers but have putrid offenses.

RedsManRick
06-12-2009, 01:58 PM
Technically if he isn't winning the games he isn't getting the job done. I would imagine that over the course of 30-35 starts in a season the good losses as well as bad wins would even themselves out. Of course there will always be outliers, Harang comes to mind right off the bat, but if a pitcher has a sub 3 ERA bets are that he will have quite an impressive W-L record.

It is not a pitcher's job to "win games" any more than it is a hitter's job to "win games". It's his job to do what he does, namely pitch, as well as he possibly can. Imagine if a team's cleanup hitter went 4-4 with 2 HR and 6 RBI but we gave him the loss because the pitcher and defense gave up 10 runs. It's completely ludicrous.

Wins are something that happen as the result of the combined efforts of the guys taking the field that day. A pitcher should not be credited with a "win" or "loss" based on the efforts of those other players any more than any other player on the field should be. Should we say that Brandon Phillips has only done a mediocre job because the Reds have a mediocre record in the games he's started? Since it's everybody's job to try and win baseball games, maybe every player who played in a given day should get a win or a loss based on the outcome of the game....

We can measure how well a guy has pitched by looking at how well the guy has pitched. There's no need to add how well everybody else played in to the equation -- it does nothing to help us understand how well that pitcher performed and in many times obscures it. Yes, wins will tend to correlate with good pitching. But they don't add any information about the pitcher in to the evaluation of how well he performed. In the absence of other information, they're better than nothing. But if you have that information, wins add nothing to the conversation about the pitcher's performance.

blumj
06-12-2009, 03:51 PM
I don't see how that provides any better information. That is still very dependent on run support and bullpen. It is probably not going anywhere as an ingrained tradition, but it would be nice if GMs, managers and media quit ever referring to it as proof that a pitcher is doing something right.

It is so ludicrous to see minor league reports on FSN where they show Travis Wood with a 1 something ERA, then say "he has pitched in bad luck, only being 1-3." Why not just leave it at "this guy is getting the job done?"

Then you see the sour look on the starter, who is disappointed when the reliever coughs up the lead, and you know it is secondarily because his team is now losing, and is primarily because they do not get the "W."

Then there is the myth that causes millions of dollars to be flushed down the toilet on the likes of Eric Milton, whose contract was justified by Reds leadership because "he is a proven winner," based on his historic W-L records.

Well, there's a lot of things it doesn't tell you, but there are just times I find myself wondering if a specific team is winning more of a particular pitcher's starts than they're losing, so I wouldn't mind if that's what the pitcher's record told me vs. what it tells me now, which is even less than that.

traderumor
06-12-2009, 03:59 PM
I have never heard a GM sign a player strictly based upon W-L record.
Eric Milton by Dan O'Brien.

traderumor
06-12-2009, 04:02 PM
In the absence of other information, they're better than nothing.[/I]I would argue that you would be better off with nothing, that pitcher's W/L has absolutely nothing to offer for any evaluation of a pitcher for any reason.

PuffyPig
06-12-2009, 05:35 PM
If you start giving pitchers wins based on the start they had (like in Harnag's game), where do you draw the line? If the Reds had subsequently lost that game in extras, you couldn't give him the win. Would that be fair? Should he get credited with a win even if the Reds lost? He still pitched the same game each time.

Do you start giving a hitter credit for a hit when he hits a scorching line drive right at someone? A pitcher credit for an out when he gives up a bleader for a hit?

Baseball is full of close plays and "unfair" results. Why pick on the "wins" stat? It's no more deceiving than most other historical stats. And it's why a large number of other type of stats have been created.

A pitcher signifies that a pitcher was the pitcher of record in a game when his team took the lead which they never gave up. Nothing more, nothing less. Nick Massett pitched two great innings in that game and was very deserving of the win. Why would you take it from him and give it to Harang?

RedsManRick
06-12-2009, 05:43 PM
If you start giving pitchers wins based on the start they had (like in Harnag's game), where do you draw the line? If the Reds had subsequently lost that game in extras, you couldn't give him the win. Would that be fair? Should he get credited with a win even if the Reds lost? He still pitched the same game each time.

Do you start giving a hitter credit for a hit when he hits a scorching line drive right at someone? A pitcher credit for an out when he gives up a bleader for a hit?

Baseball is full of close plays and "unfair" results. Why pick on the "wins" stat? It's no more deceiving than most other historical stats. And it's why a large number of other type of stats have been created.

A pitcher signifies that a pitcher was the pitcher of record in a game when his team took the lead which they never gave up. Nothing more, nothing less. Nick Massett pitched two great innings in that game and was very deserving of the win. Why would you take it from him and give it to Harang?

I think this pretty clearly points out that assigning an entire win to any individual player is a pretty silly idea on the face of it.

You're right, most stats involve the performances of multiple people. Even stats like OPS and ERA are based in part by what other people do. The stats we should look for are the ones that are most directly tied to the performance of the individual players to whom they are assigned. Few are perfect, but some are better than others. This is essentially what sabermetric stats attempt to do - give credit where it's due.

westofyou
06-13-2009, 11:40 AM
Outliers exist on the winning side and the losing side, but I'm prone to believe that there are more bad pitchers piling up wins with good clubs then good pitchers piling them up with bad clubs, but look at the lists below and you'd know who you would want on your staff. You'll also see that there have been more bad ERA's getting W's as of late.


ERA YEAR DIFF PLAYER LEAGUE GS W L N_W N_L
1 Mark Hendrickson 2005 -1.54 5.90 4.36 31 11 8 7 12
2 Shawn Estes 2004 -1.52 5.84 4.31 34 15 8 10 13
3 Wes Ferrell 1938 -1.49 6.28 4.79 26 15 10 9 16
4 Wilbur Wood 1978 -1.42 5.20 3.78 27 10 10 7 13
5 Rick Sutcliffe 1993 -1.41 5.75 4.34 28 10 10 8 12
6 Joe Niekro 1968 -1.34 4.32 2.98 29 14 10 10 14
7 Bob Walk 1982 -1.27 4.87 3.61 27 11 9 8 12
8 Guy Bush 1930 -1.23 6.20 4.97 25 15 10 11 14
9 Larry Gura 1984 -1.18 5.18 4.00 25 12 9 8 13
10 Randy Lerch 1977 -1.16 5.07 3.91 28 10 6 6 10
11 Mark Redman 2006 -1.15 5.71 4.56 29 11 10 9 12
12 Shane Reynolds 2003 -1.15 5.43 4.29 29 11 9 8 12
13 Red Causey 1919 -1.12 4.03 2.91 26 13 8 8 13
14 Bill Bernhard 1900 -1.08 4.77 3.69 27 15 10 10 15
15 Esteban Loaiza 2004 -1.07 5.70 4.64 27 10 7 8 9
16 Dave Roberts 1972 -1.04 4.50 3.46 28 12 7 8 11
17 Rich Dotson 1988 -1.02 5.00 3.98 29 12 9 8 13
18 Josh Fogg 2006 -1.00 5.49 4.49 31 11 9 9 11
19 Brett Tomko 2003 -1.00 5.28 4.29 32 13 9 9 13
20 Jesse Haines 1929 -.99 5.70 4.71 25 13 10 9 14
21 Ben Rivera 1993 -.98 5.02 4.05 28 13 9 9 13
22 Bob Walk 1989 -.91 4.41 3.50 31 13 10 8 15
23 Mike Moore 1993 -.89 5.22 4.34 36 13 9 9 13
24 Hal Schumacher 1939 -.88 4.80 3.91 27 13 10 9 14
25 Joe Nuxhall 1957 -.88 4.76 3.88 28 10 10 9 11




ERA YEAR DIFF PLAYER LEAGUE GS W L N_W N_L
1 Ismael Valdes 1997 1.55 2.65 4.21 30 10 11 14 7
2 Kevin Millwood 2005 1.50 2.86 4.36 30 9 11 13 7
3 Jake Peavy 2008 1.45 2.85 4.30 27 10 11 13 8
4 Roger Clemens 1996 1.36 3.63 5.00 34 10 13 15 8
5 Nolan Ryan 1987 1.32 2.76 4.09 34 8 16 15 9
6 Ned Garvin 1900 1.28 2.41 3.69 28 10 18 18 10
7 Ted Lyons 1932 1.21 3.27 4.48 26 10 15 15 10
8 Jim Abbott 1992 1.18 2.77 3.95 29 7 15 15 7
9 Kevin Appier 1997 1.17 3.40 4.57 34 9 13 14 8
10 Jack Taylor 1900 1.14 2.55 3.69 26 10 17 17 10
11 Curt Schilling 1996 1.03 3.19 4.22 26 9 10 12 7
12 Jose Rosado 1999 1.02 3.85 4.87 33 10 14 15 9
13 Joe Horlen 1966 1.00 2.43 3.44 29 10 13 14 9
14 Bill Wight 1950 0.99 3.58 4.58 28 10 16 16 10
15 Chris Short 1967 0.98 2.40 3.38 26 9 11 13 7
16 Ewell Blackwell 1946 0.96 2.46 3.42 25 9 13 14 8
17 Virgil Barnes 1926 0.95 2.87 3.82 25 8 13 13 8
18 Tom Candiotti 1993 0.93 3.12 4.05 32 8 10 11 7
19 Sad Sam Jones 1933 0.93 3.36 4.28 25 10 12 14 8
20 Turk Farrell 1962 0.93 3.01 3.94 29 10 20 18 12
21 Curt Simmons 1961 0.91 3.12 4.03 29 9 10 13 6
22 Felix Hernandez 2008 0.90 3.45 4.36 31 9 11 12 8
23 George Brunet 1965 0.90 2.56 3.46 26 9 11 12 8
24 Dennis Martinez 1990 0.85 2.95 3.80 32 10 11 12 9
25 Sonny Siebert 1967 0.85 2.38 3.23 26 10 12 14 8

Blitz Dorsey
06-16-2009, 12:08 AM
Eric Milton by Dan O'Brien.

That gritty Eric Milton just had a way of giving up five or six HRs in a game, but still finding a way to hang in there. Quite the bulldog, he was.

(I just threw up on my laptop. Excuse me.)

dsmith421
06-16-2009, 12:55 AM
I think the bottom line is that the W-L statistics made a decent amount of sense when pitchers generally threw complete games. In those instances, saying that Hod Eller was 16-12 gave a pretty good indication of what happened when Eller started. Nowadays the statistic proves nothing and provides little indication about the quality of a pitcher.

VR
06-16-2009, 01:11 AM
Kent Bottenfield's 1999....18-7. A great example of over inflated value...which StL knew and quickly flipped him for Jim Edmonds. Angels....frisked.

PuffyPig
06-16-2009, 07:49 AM
One guy gives up two runs in seven innings, loses 2-0. One guy gives up four runs in seven innings, wins 6-4. There has to be a way to measure how well each did, especially if you're thinking, "Who would I want on my team?"


There is, it's called ERA, and now FIP ERA and xFIP ERA.

Since the dawn of time, most fans have used ERA as the best method of judging starting pitchers.

nate
06-16-2009, 09:00 AM
Since the dawn of time, most fans have used ERA as the best method of judging starting pitchers.

It might be the most convenient/nostalgic.

But it's not the best.

RED VAN HOT
06-17-2009, 10:21 PM
When you get down to it, wins are only significant for starting pitchers. Further, in a given year, it is only one dimension of excellence. High lifetime win totals, however, are indicative of pitchers who has been good for an extended period of time. Over a long career, there is an opportunity for luck to even out. As a fan, I like to see lifetime career wins and would not be in favor of a change in how they are awarded. Allowing official scorers too much freedom in awarding wins is a controversy waiting to happen.

I like the five inning rule. It precludes cheap wins for the starter and makes lifetime wins more meaningful. There is one change I would make. If the starter goes fewer than five and leaves the game because of injury or game delay with a lead, then the official scorer should be allowed to award him the win if the lead holds up and no relief pitcher performs better. This would remove the temptation a manager has to support his starter by having him resume pitching after a lengthy rain delay.