320 innings in a season

[Gee Walker]

People are making one fundamental error in this comparison. Most of the difference in innings between Lolich and Greinke occur because Lolich started 12 more games than Greinke, not because Greinke was being replaced earlier in the game.

Lolich started 45 games, so his 376 innings amount to 8.36 innings per game - numbers that look like they come from the 19th century. Greinke averaged a very respectable 6.95 innings per start in his 229 2/3 innings. Giving Lolich his 8.36 innings per start, you'd have to find a starting pitcher who could throw 100.32 innings in 12 starts to add to the Royals' rotation in order to match Lolich's contributions as a starter. Can you find one - remember this is adding innings to your fifth starter, or worse, bringing up a guy to add to the rotation when your fifth starter fails? In 2009, the Royals 3, 4, and 5 starters (in terms of innings pitched) had ERA's of 6.55, 5.09, and 5.27. Naturally, you can't be happy with this, so the next guys who had 9 starts each had ERA's of 5.78 and 7.36. Sure, it's fairly easy to find relievers with decent ERA's, but a lot of good that'll do you if they're coming in after the starter has given up 5 or 6 runs.

I can't recommend using any pitcher like Billy Martin used Lolich, but if your purpose is to win at all costs, it seems like Lolich must be more valuable than Greinke.

[filihok]

People are making one fundamental error in this comparison. Most of the difference in innings between Lolich and Greinke occur because Lolich started 12 more games than Greinke, not because Greinke was being replaced earlier in the game.

Lolich started 45 games, so his 376 innings amount to 8.36 innings per game - numbers that look like they come from the 19th century. Greinke averaged a very respectable 6.95 innings per start in his 229 2/3 innings. Giving Lolich his 8.36 innings per start, you'd have to find a starting pitcher who could throw 100.32 innings in 12 starts to add to the Royals' rotation in order to match Lolich's contributions as a starter.

Well, that's exactly what WAR does.

http://www.fangraphs.com/library/index.php/misc/war/

[Matthew C.]

Well, that's exactly what WAR does.

http://www.fangraphs.com/library/index.php/misc/war/

Exactly - WAR gives full credit for all of the extra innings pitched...almost a whopping 1.5 WAR in this case. And Grienke allowed runners at such an incredible clip less than Lolich, he was STILL more valuable. That is what happens when your ERA+ is over 200.

Nobody is saying that racking up innings isn't very valuable. WAR is huge on racking up playing time - hence the "R" part in WAR.

[GiambiJuice]

Curt Schilling pitched over 300 innings in 2001 if we include the postseason. His teammate Randy Johnson just missed with 291.

[drstrangelove]

Looking at it from another perspective, not all these single season workhorses were among the leaders in career innings pitched. Of all pitchers who started their careers in 1926 or later (to scoop up Dean, Farrell, and Crowder), five from the above list were in the top ten career innings pitched: P Niekro, Ryan, Perry, Carlton, Blyleven.

Roberts and Jenkins round out the top twenty. These are probably who we think of when the term "workhorse" comes up. But already, 13 out of the top twenty never pitched 320 innings/year.

Number 10 on our list is Lolich, in 31st place overall, number 15 is Trout, who is 93rd overall. *

Next is the hundredth overall, Wilbur Wood, with "only" 2684. Then the final 7 dribble off from 155, Koufax, to Crowder at 426.

So it depends what you mean by a stud. Spahn, Sutton, Maddux, Clemens, Seaver, John, Wynn, Kaat, R Johnson, Glavine . . . Bob Friend, . . . Luis Tiant . . . . None of these guys put in 320 innings in a season. They all had more career innings (along with a lot of other guys) than did the average 320 inning hurler.

Not to take away anything from the guys on our list. What they did deserves every word of praise heaped on them. But for some, at least, that's equaled by the frustration over what was lost.

I'm not going to weigh in on Lolich-Grienke, but I can say I'd much rather have a Koufax treated like Grienke than treated like Koufax. I think the five-man rotation was one of the greatest contributions to the game, and annoying as the constant pitching changes are, the pitch count and deep bullpen may be another.


Palmer, Feller, Lolich, Marichal, Osteen, Hunter, and Drysdale the top 42--or top 92, since the next is Trout, number 93.

After you eliminate the dead ball era pitchers Young(1), Galvin (2), Johnson(3), Alexander(10), Nichols (11), Keefe (12), here is the list of all-time for innings pitched in a career:

Niekro
Ryan
Perry
Sutton
Spahn
Carlton
Maddux
Blyleven

Of these 8, 7 started their careers in the 4 man rotation. (Spahn, by the way, pitched 7 seasons of 289.2 - 310.2 innings or more, but in the 154 game schedule. Under today's schedule, he would be more like 7 seasons of 304 - 325, with another 2 at just under 300.) Each of these pitchers (excluding Maddux) pitched well into their 40's, despite huge pitch counts and massive innings.

Afaik, there is no evidence that reduced pitch counts in a game or pitching fewer innings a season is allowing pitchers to pitch longer careers or more innings over the course of their career. Or even to reduce the number of injuries they suffer. They are simply pitching less.

[Matthew C.]

I'd be happy too. But if I could get 220 innings with a 200 ERA+ from a starter and a bunch of specialists filling the other 140 innings with 120+ ERA (about the norm for relievers), I'd take that easily.

But that really isn't the question here. The question is, was Grienke with historic run prevention numbers (200+ ERA+) and a good IP total better than a guy with historic IP totals but just a good run prevention season? Yes, Lolich had 140 more innings...but he also gave up 69 more runs. Which is more valuable to a team? 140 innings or 69 runs? Run values suggest 69 runs is more valuable. WAR and I would agree. WAA says yes too. Cy voters agreed as well.

But I will say again, if big IP totals are your thing, than WAR is the stat for you.

jr - I forgot, if you want to weigh IP even more heavily, look at Fangraph's WAR which has a lower replacement level. So Lolich would likely be closer to Grienke in their variation.

[Jackaroo Dave]

After you eliminate the dead ball era pitchers Young(1), Galvin (2), Johnson(3), Alexander(10), Nichols (11), Keefe (12), here is the list of all-time for innings pitched in a career:
"..........................
Of these 8, 7 started their careers in the 4 man rotation. (Spahn, by the way, pitched 7 seasons of 289.2 - 310.2 innings or more, but in the 154 game schedule. Under today's schedule, he would be more like 7 seasons of 304 - 325, with another 2 at just under 300.) Each of these pitchers (excluding Maddux) pitched well into their 40's, despite huge pitch counts and massive innings.

Afaik, there is no evidence that reduced pitch counts in a game or pitching fewer innings a season is allowing pitchers to pitch longer careers or more innings over the course of their career. Or even to reduce the number of injuries they suffer. They are simply pitching less.

There is a step or two missing between "Afaik there is no evidence . . . " and "They are simply pitching less."

Those pitchers were on my list, too, except Spahn and Sutton, but mine didn't stop at 8. Obviously no one is going to be among the top 8 unless they pitch a lot of innings per season along the way. But my concern is for those who pitch a lot of innings in a season and subsequently (I would say consequently) lose a good chunk of their careers.

To my mind, there is no question that it is better to pitch 200 innings a year for 15 years than to pitch 300 for 10. Thus for me a five-man rotation would be preferable if

a. Pitching injury is a function of tissue fatigue over a season
OR
b. Pitching injury is a function of the total number of pitches thrown, and variation is just a genetic lottery
OR
c. Pitching injury is completely random: it can come at any time, after any pitch.

Case c would still favor the 5-man rotation in my view simply because the probability of pitching more seasons would be greater.

To convince me that heavier pitching loads per season are equivalent would entail showing that pitchers' career pitch counts were shortened by pitching fewer innings a season.

[Jackaroo Dave]

Actually, thinking it over, I wonder whether advances in sports medicine have increased number of innings thrown over a career. (Well, Tommy John surgery, of course, but generally?) One of the best things you can do for the health of an arm is not pitch with it. It may be that improved diagnoses and more and longer prescriptions of rest (or retirement) have lightened the pitching load. That wouldn't mean that advances in diagnosis and rehabilitation are a bad thiing, though.

[drstrangelove]

There is a step or two missing between "Afaik there is no evidence . . . " and "They are simply pitching less."

Those pitchers were on my list, too, except Spahn and Sutton, but mine didn't stop at 8. Obviously no one is going to be among the top 8 unless they pitch a lot of innings per season along the way. But my concern is for those who pitch a lot of innings in a season and subsequently (I would say consequently) lose a good chunk of their careers.

To my mind, there is no question that it is better to pitch 200 innings a year for 15 years than to pitch 300 for 10. Thus for me a five-man rotation would be preferable if

a. Pitching injury is a function of tissue fatigue over a season
OR
b. Pitching injury is a function of the total number of pitches thrown, and variation is just a genetic lottery
OR
c. Pitching injury is completely random: it can come at any time, after any pitch.

Case c would still favor the 5-man rotation in my view simply because the probability of pitching more seasons would be greater.

To convince me that heavier pitching loads per season are equivalent would entail showing that pitchers' career pitch counts were shortened by pitching fewer innings a season.

I think you've crafted the analysis.

a. Pitching injury is a function of tissue fatigue over a season
OR
b. Pitching injury is a function of the total number of pitches thrown, and variation is just a genetic lottery
OR
c. Pitching injury is completely random: it can come at any time, after any pitch
OR
d. Pitching injury is a result of pitching improperly because pitchers are encouraged to throw as hard as possible because they are expected to throw only a limited amount of pitches every 5 days.


Pitching injuries are related to using more physical effort than the body was designed to withstand. It's a threshold injury, not a usage injury. Koufax, Spahn, Marichal, Perry, Ryan, have games with pitch counts of 160-220. Throwing 70 pitches as hard as humanly possible with as much leverage as possible is not the same as throwing 115 pitches at 85-90% with 15 at 100%.

None of the pitchers in the 60's and 70's, barring a few relievers, were asked or expected to use absolutely 100% effort on each pitch or to throw as hard for as long as possible. Even hard throwers like Radatz were expected to go 2 or 3 innings if needed. And not surprisingly, the only guy in the top 8 from the 5-man rotation era was known to staunchly support the "throw less hard but with more control" method of pitching.

My point in my list is that:
1) pitching injuries have not declined, they've gone up
2) while pitching loads have declined
3) while pitch counts have been lowered and controlled

It's not really a comparison of pitching 300 innings for 10 years versus 200 innings for 15 years. (If it were, we'd see people on the top of the list from 5-man rotations.)

The reality is that it's a comparison of pitching 280 innings for 15 years while NOT throwing with absolute effort on each pitch versus pitching 200 innings for 15 years while trying to make every pitch 100%. The 280 innings a year from a number 1 starter is far more valuable at 90% than 200 innings at 100% with the extra 80 made up by a replacement pitcher. Pitchers since 1900 have known that you can pitch successfully without 100% effort. In fact, it's better for the team if you do.

Greinke was very good at 229 innings with a 205 ERA+. He would have been great for his team with 299 innings at 185 ERA+.

The list I posted proves my point. No one is getting close to the top of that list.

[White Knight]

The funny part is pitching injuries are not down one bit from the 1970's. While I wouldn't call it a conspiracy, innings limits, the pitch count, and the 5 man rotation are a farce supported by the uneducated baseball higher-ups. They all need to go.

[Ben Grimm]

If I was a manager and knew going in that I was going to get this: 45 starts 376.1 innings, 25-14 2.92 ERA+124, I'd be pretty happy. I wouldn't be worry about mythical replacement players. I'd know that I had a pony.

Just looking at it in general, there are a few things to like.

Those extra 12 or so starts would mean less games pitched by my #5 or call-ups. There'd be no need to waste money and roster spots on FA signings like Sidney Ponson and Bruce Chen that year. The 29 CGs would lead to less games needed by the bullpen who's ERA was near 6 not counting Soria.

[willshad]

Just looking at it in general, there are a few things to like.

Those extra 12 or so starts would mean less games pitched by my #5 or call-ups. There'd be no need to waste money and roster spots on FA signings like Sidney Ponson and Bruce Chen that year. The 29 CGs would lead to less games needed by the bullpen who's ERA was near 6 not counting Soria.

I don't think there's any inherent value in pitching more, if the additional level you are pitching at is not good. That's why i think that pitcher's WAR is unreliable, especially the high innings pitched guys.

For instance, imagine a hypothetical situation: a guy who was so good at playing shortstop, that he could play short AND third base at the same time. The manager only used 8 players, and the guy was inserted into the lineup 2 times. Imagine this guy played the field (both positions) and hit at an overall league average level.

What would his WAR be? Twice as much as the average player who played short and hit at a league average level. Even more than a star player. But would you rather have THIS guy, or have a star third baseman, along with a league average shortstop? Which situation would be better for your team?

[filihok]

I don't think there's any inherent value in pitching more, if the level you are pitching at is not good. That's why i think that pitcher's WAR is unreliable, especially the high innings pitched guys.

I don't think you understand WAR that well.

http://www.fangraphs.com/library/index.php/misc/war/
http://www.beyondtheboxscore.com/201...to-explain-war


War is dependent on both the quality of the playing time and the quantity of the playing time.

If a pitcher is replacement level and pitches 1 inning his WAR will be 0. If he pitches 100 innings his WAR will be 0. If he pitches 200 innings his WAR will be 0.

If a pitcher is a 2 WAR player at 200 innings and pitches 1 inning his WAR will be 0. If he pitches 100 innings his WAR will be 1. If the pitches 400 innings his WAR will be 4.

[willshad]

I don't think you understand WAR that well.

http://www.fangraphs.com/library/index.php/misc/war/
http://www.beyondtheboxscore.com/201...to-explain-war


War is dependent on both the quality of the playing time and the quantity of the playing time.

If a pitcher is replacement level and pitches 1 inning his WAR will be 0. If he pitches 100 innings his WAR will be 0. If he pitches 200 innings his WAR will be 0.

If a pitcher is a 2 WAR player at 200 innings and pitches 1 inning his WAR will be 0. If he pitches 100 innings his WAR will be 1. If the pitches 400 innings his WAR will be 4.

That's not how I understand it. I thought there was an additional value added to WAR, simply related to playing time. I know that batters have RRep, which gives them credit just for showing up. For example, if Wilbur Wood and Zack Grienke both pitch 200 innings at a 200 ERA+, they will have similar WAR, but if Wood pitches an extra 150 innings at replacement level, he will have a much higher WAR total than Grienke for the season.

Anyway, you are misunderstanding my point. I am saying that if a pitcher pitches as much as TWO starters would pitch, and pitches at a lower level than those two starters put together, then he really is not as more valuable than they are. He is accumulating more 'value' due to pitching more, but the team overall is suffering because overall the pitching is worse.

[filihok]

That's not how I understand it. I thought there was an additional value added to WAR, simply related to playing time. I know that batters have RRep, which gives them credit just for showing up. For example, if Wilbur Wood and Zack Grienke both pitch 200 innings at a 200 ERA+, they will have similar WAR, but if Wood pitches an extra 150 innings at replacement level, he will have a much higher WAR total than Grienke for the season.

That's not correct. If Wood and Greinke pitch 200 innings at a 4 WAR level, they will each have 4 WAR. If Wood then pitches an extra 150 inning at replacement level he'll have 200 IP @ 4 WAR + 150 IP @ 0 WAR = 4 WAR.

Anyway, you are misunderstanding my point. I am saying that if a pitcher pitches as much as TWO starters would pitch, and pitches at a lower level than those two starters put together, then he really is not as more valuable than they are.

You're right.
Take Wood from above. 350 innings of 4 WAR production.
If you have 2 starters, 1 who pitches 200 innings of 3 WAR production and another who pitches 150 innings of 2 WAR production that's 350 innings of 5 WAR production.
The 2 starters together are more valuable.

He is accumulating more 'value' due to pitching more, but the team overall is suffering because overall the pitching is worse.

I don't understand why you say this

[drstrangelove]

Anyway, you are misunderstanding my point. I am saying that if a pitcher pitches as much as TWO starters would pitch, and pitches at a lower level than those two starters put together, then he really is not as more valuable than they are. He is accumulating more 'value' due to pitching more, but the team overall is suffering because overall the pitching is worse.

Yes, this is correct. Having said that, however, what does it mean?

If a pitcher is pitching x number of innings at below a replacement level, than 100% of his innings should be replaced.
If a pitcher is pitching x number of innings at replacement level, but if he were pitching fewer innings he would be above replacement level, then he should pitch fewer innings with the extra made up by a replacement level pitcher.

If a pitcher is pitching x number of innings ABOVE replacement level, than the team should FIRST look to replace any other pitcher it has that are pitching at below replacement level.

Thus, no, it doesn't hurt the team to have Lolich at 124 ERA+ as long as the rest of the staff that would have replaced him was at 71-83 ERA+. It helps them a ton that he handles all that work load. It wouldn't be an issue unless the rest of the staff was Koufax, Gibson, Maddux and Martinez, in which case, yes, Lolich's starts need to be reduced.

I think it's confusing to discuss value versus quality. That's why WAR is a useful stat. A pitcher who pitches 376 innings and gets 9 WAR is just as valuable as a pitcher who pitches 229 innings and gets 9 WAR, even though he's not better (or quite as good).

This isn't like batting where getting 200 hits in 500 at bats is BOTH better than getting 200 hits in 650 at bats, and more valuable. The extra 150 at bats get spread over extra hitters who will get added hits and added runs to the team.

The pitcher who gets 9 WAR in 376 innings is by definition better than replacement level, just like the one who is better but does the same in 229 innings. The extra 147 innings (at above replacement level) are still valuable because without them the team will have had to use pitchers who ERA+ of 71-83. The team would give up more runs. It's not as if Lolich stopped the Tigers from using better pitchers.

By the same token, Greinke was certainly better than Lolich, but he also sat on the bench doing nothing for 147 innings while the Royals used pitchers with ERA+ of 68-94. He didn't help them by letting those guys pitch the way a batter helps the team by getting 200 hits with 150 fewer outs.

To use an extreme example, a team ERA+ of 124 is an amazingly good stat. Last year, the 2 highest was Tampa was at 120 and Washington at 119. 124 is really excellent. The Dodgers in the 60's would win at below 120.

If Lolich could have been at 124 ERA+ for 162 complete games, he would be the most valuable player who ever stepped on a field. Volume at a pretty good level can be just as valuable at high quality at a pretty good volume. That's what pitching WAR tells us.

[leewileyfan]

I don't like WAR, purs and simple; so I came at the Grienke-Lolich from another perspective, pretty well convinced, up front, that I'd come down in favor of Lolich.

Nope. That was a surprise to be.

I took Earned Runs for Lolich 122], then didvided that by BFP, Lolich [1,538] and got a quotient of
.0793. The League pitchers that season yielded 6,650 ER with 72,931 = .0912. That was my first red flag: The entire League had ER/BFP below .1000 which is unusually low.

I applied the League rate [.0912] tp Lolich's BFP [1,538] = .0912*1,538 = 140.27. That gave me what I wanted to know: the number of earned runs Lolich saved in his 376 workhorse innings pitched. It surprised me that it was only 140.27-122 = 18.27 runs saved.

Doing the same with Grienke, his 55 ER came with 915 BFP = .0601, astounding, like HoF stuff. Moreover, his League numbers were .1142, a tougher pitching climate, 25.2% higher than in Lolich's season.

Applying the League rate [.1142] to Grienke's BFP I get 104.49 runs. Grienke saved [104.49-55] = 49.49 runs and did it in 146.9 fewer IP [or 623 fewer BF].

Another way to look at this is a Pythago W-L projection based on pitcher runs to League runs:

Lolich: .0793*5209 [Team AVG BFP] = 413 ER vs LG at 475 ER projects a .569 W-L Pct. Lolich was 25-14, or .641 vs. 22-17 projected by Pythago.

Grienke: .0601*6245 = 375.32 ER vs. LG at 713 ER projects a .783 W-L Pct. Grienke was 16-8, or .667 when Pythago would have projected 18.8 W and 5.2 L in 24 decisions.

Any way I look at it Grienke saved more runs with less physical effort in IP ... man, is that ever efficient!

[willshad]

Apparently I was mistaken in how WAR related to innings pitched.

I still believe that it is misleading, and the high innings pitched reaches a point of diminishing returns.

For instance, in 1972 Wilbur Wood pitched 376 innings, and had a 126 ERA+. His WAR for the season was 10.3, which is not just a good total, but an all time great total...as good as any season Tom Seaver had in his career. Seaver actually had LESS WAR in 1971, when he had a 194 ERA+. If the White Sox did not have Wood that season had used a 5 man rotation instead of 4, and the two pitchers pitching in place of Wood had a 140 ERA+ each, then his WAR would be MUCH higher than either guy, even though his team is worse off due to him pitching in place of the two men.

To me, being 'good' for a ton of innings should not translate to an all time great season.

If some guy managed to pitch every inning of every game for his team, and pitched to a league average level, he would be have had the greatest season of all time, according to WAR.

In reality, he would be just an average pitcher.

[filihok]

Apparently I was mistaken in how WAR related to innings pitched.

I still believe that it is misleading, and the high innings pitched reaches a point of diminishing returns.

For instance, in 1972 Wilbur Wood pitched 376 innings, and had a 126 ERA+. His WAR for the season was 10.3, which is not just a good total, but an all time great total...as good as any season Tom Seaver had in his career. Seaver actually had LESS WAR in 1971, when he had a 194 ERA+. If the White Sox did not have Wood that season had used a 5 man rotation instead of 4, and the two pitchers pitching in place of Wood had a 140 ERA+ each, then his WAR would be MUCH higher than either guy,

His WAR would be higher than either guy, but it wouldn't be higher than both guys' totals added together.

If two pitchers combined to pitch 376 innings with a 126 ERA+ their combined WAR total would be 10.3. 5.15 for each. So, of course, if one pitcher pitches all 376 innings his WAR should be 10.3 since the production is the same.

If the two 140 ERA+ pitchers pitched 376 innings and had, say, 13 WAR then Wood would have 13 WAR if he pitched 376 innings with a 126 ERA+

If some guy managed to pitch every inning of every game for his team, and pitched to a league average level, he would be have had the greatest season of all time, according to WAR.

In reality, he would be just an average pitcher.

Really? A guy pitching 1450 innings you would consider average?

You're still confusing things.

There were about 43355.3 innings pitched in the major leagues last season. A total of 460.6 WAR were credited to pitchers.

That's 1445 innings and 15.4 WAR per team

Whether those innings and WAR were produced by 1 pitcher, 12 pitchers, 20 pitchers, or 1445 pitchers, the level of production is the same.

Suppose an entire team had only one player. He pitches, he hits, he fields balls. He does everything on the field. The team finishes at .500 (an average team).

How much credit should the sole player get? All of it right? How much is that?

A replacement level team is expected to win about 47 games. An average team would win 81 games. 81-47=34 wins. The sole player playing every position and doing everything is solely responsible for those 34 wins, correct? Even though his overall production was exactly average.

[dominik]

No.

Average WAR (using FanGraphs FIP based WAR and RA-9 WAR and B-R WAR) for Greinke = 9.7
Average WAR for Lolich 8.5

yes but I do not think it is that easy. it is both quality and quantity that is important. remember that a game still has 9 innings and the other innings have to be pitched always. so those innings have to be pitched by a reliever who is likely weaker. so by leaving earlier the pitcher actually does cost a team runs (unless the have 5 mo riveras in the pen). WAR does factor playing time (rightfully) and the top20 WAR seasons by pitchers all happened before WW1 (I'm not sure if that a good POV just like the fact that all clean huge OPS+ seasons happened till about 1950).

[willshad]

His WAR would be higher than either guy, but it wouldn't be higher than both guys' totals added together.

If two pitchers combined to pitch 376 innings with a 126 ERA+ their combined WAR total would be 10.3. 5.15 for each. So, of course, if one pitcher pitches all 376 innings his WAR should be 10.3 since the production is the same.

If the two 140 ERA+ pitchers pitched 376 innings and had, say, 13 WAR then Wood would have 13 WAR if he pitched 376 innings with a 126 ERA+


Really? A guy pitching 1450 innings you would consider average?

You're still confusing things.

There were about 43355.3 innings pitched in the major leagues last season. A total of 460.6 WAR were credited to pitchers.

That's 1445 innings and 15.4 WAR per team

Whether those innings and WAR were produced by 1 pitcher, 12 pitchers, 20 pitchers, or 1445 pitchers, the level of production is the same.

Suppose an entire team had only one player. He pitches, he hits, he fields balls. He does everything on the field. The team finishes at .500 (an average team).

How much credit should the sole player get? All of it right? How much is that?

A replacement level team is expected to win about 47 games. An average team would win 81 games. 81-47=34 wins. The sole player playing every position and doing everything is solely responsible for those 34 wins, correct? Even though his overall production was exactly average.

I do understand how it works. My point is not that their WAR value should CHANGE, but that it is misleading. To me if you are an average pitcher, then you are an average pitcher, whether you pitch 200 innings or pitch 1000 innings. Being 'average', or even 'good' for a massive amount of time should not somehow translate into greatness.

I do see the value in pitching more innings, but I just think that you should have to perform 'great' in order to end up in the realm of greatness. So, I am much more impressed with Seaver's 1971 season than I am with Wood's 1972 season.

[filihok]

I do understand how it works. My point is not that their WAR value should CHANGE, but that it is misleading. To me if you are an average pitcher, then you are an average pitcher, whether you pitch 200 innings or pitch 1000 innings. Being 'average', or even 'good' for a massive amount of time should not somehow translate into greatness.

I do see the value in pitching more innings, but I just think that you should have to perform 'great' in order to end up in the realm of greatness. So, I am much more impressed with Seaver's 1971 season than I am with Wood's 1972 season.

No average pitcher has come close to 1000 innings. Nothing average about that.

There are two kinds of statistics; counting stats and rate stats.
Counting stats are things like: hits, runs, hr, RBI, Wins, saves, strikeouts. They count what happens and players who play more get more of these stats.
Rate stats are things like: batting average, ERA, on base percentage, slugging percentage, strike our rate. The measure how often something happens and don't depend on playing time.

WAR is a counting stat. If you play more, you can accumulate more WAR.

Take the example of the one-man team. That player would have an average ERA, but 1450 innings, 81 wins, an unheard of amount of strikeouts. Batting, he'd have a league average batting average, a league average OBP and SLG. But over a hundred of home runs and hundreds of RBI.

By his rate stats he'd be average in every way.
By his counting stats he'd have unheard of numbers.

When we use WAR to say one player is 'better' than another it's a bit of a misnomer. What we are actually saying is that one player produced more wins (above replacement) than another. Typically, we are referring to full time players who have a similar number of plate appearances or innings pitched.

In this example, Seaver and Wood, we're not dealing with two players who played equally. We can then make WAR into a rate stat. We do this by figuring WAR per some amount of innings pitched. We can use Seaver's 286 innings, for example. Seaver had 9.7 WAR in 286 innings, or 9.7 WAR / 286 IP. How about Wood. 10.3 WAR in 386 innings. That's 7.9 WAR / 286 IP. This shows that Seaver was better on a per inning basis.

This is also why stats like RBI don't tell the whole story. In 1986 Willie McGee had 48 RBI. In 1987 McGee had 105 RBI, in 1988 McGee had 50 RBI. What happened to McGee in 1987? Why was he so good?

Well, he wasn't. He just had more opportunity.
in 1986 McGee drove in 16% of the 291 runners on base when he batted
in 1988 McGee drove in 14% of the 369 runners on base when he batted
in 1987 McGee drove in 17% of the 544 runners on base when he batted

Maybe an easier way to look at it would be hits.
Derek Jeter had the most hits in 2012. 216 of them. He also had the most at bats and plate appearances.
Miguel Cabrera 'lead the league in hitting'. He had a 205 hits in 622 at bats. A .330 batting average (hits/at bat).

WAR is like hits. The more you play, the more you get.
WAR/IP or WAR/PA is like batting average. The better you play. The higher your stat.

[filihok]

One last example.

Last season Fernando Rodney had a .60 ERA. Dominant. He accomplished this over 74 innings.
This season was worth 3.7 WAR (baseball-reference WAR).

Last season Justin Verlander led all major league pitchers in WAR with 7.6 (over double what Rodney had). He had a 2.64 ERA in 238 innings.

Obviously Rodney was more dominant. Just as obviously (I hope), Verlander had a bigger hand in helping his team win games.

[drstrangelove]

WAR is not a counting stat. It's like calling MVP awards won a counting stat.

WAR is not a measure of how many innings a pitcher pitches. It's a measure of his adjusted pitching quality (to replacement level) times the quantity of innings. The average pitcher is NOT replacement level. Replacement level (which could have it's own thread) is variously measured at 15-25% BELOW the average MLB pitching level.

A pitcher who is average is 15-25% better than whoever that team would need to go get to replace him. If one does not follow this critical point, then I can see that it's confusing. So, you really must understand this part.

Please note that: there are TONS of innings tossed each year by pitchers well below average (e.g., ERA+ of 60-90). If one thinks that there are 'average' pitchers in Triple A or Double A just toiling away because average pitchers are in the majors racking up counting stat WAR, then that is a gross error.

You don't get WAR just because you pitch. You don't get it just because you pitch a lot.

Since WAR compares a pitcher to replacement level, a pitcher with positive WAR is helping his team win games. You can accumulate a lot of WAR in 2 ways:

1) pitching a LOT at a decent amount better than replacement level or
2) pitching a decent amount at a lot better than replacement level.

Both methods help a team equally IF the WAR amount is equal. The "Why" of this is what seems to be confusing.

Teams must pitch all the innings they pitch regardless of how well any individual pitches his innings. If a team plays a full game and the starter throws 5 innings of perfect baseball and leaves, the team STILL has to pitch at least 4 more innings. There is no bonus for pitching perfect baseball.

1) Every pitcher who is above replacement level helps his team win games but he ONLY does that while he pitches. He doesn't help the team by sitting on the bench, drinking beer in the locker room or napping in the bullpen.
2) Every inning that is NOT pitched by a pitcher above replacement level NEEDS to be pitched by a pitcher at or below replacement level.

If one were to assume that there are 140+ ERA pitchers sitting in AAA ball or not getting contracts, then would be erroneous. The 'leftover' pitchers not being used are close to or below replacement level. As a result, you can see from 1) and 2) above, that a pitcher above replacement level helps his team every time he pitches because he is better than anything else they have.

The more a pitcher is above replacement level, the more it helps.
The more a pitcher above replacement level pitches, the more it helps.

WAR is meant to measure total contribution to a team during a season or career, not contribution per inning or contribution per AB. It's (IMO) incorrect to confuse value with quality. A pitcher with the same WAR in a season as another pitcher has the same value.

[filihok]

WAR is not a counting stat.

Yes it is. It's certainly not a rate stat.

It's like calling MVP awards won a counting stat.

That'd also be a counting stat.

WAR is not a measure of how many innings a pitcher pitches. It's a measure of his adjusted pitching quality (to replacement level) times the quantity of innings. The average pitcher is NOT replacement level. Replacement level (which could have it's own thread) is variously measured at 15-25% BELOW the average MLB pitching level.

This is all true.

A pitcher who is average is 15-25% better than whoever that team would need to go get to replace him. If one does not follow this critical point, then I can see that it's confusing. So, you really must understand this part.

Except you misunderstood it; or explained it poorly. The word 'need' makes no sense.
Replacement level is a level of production where a player no longer has value.
Remember that value is related to scarcity.
How many players can be expected to produce 10 WAR in a season? Very few. They are very valuable.
How many players can be expected to produce 5-10 WAR per season? Also few. But there are some. Maybe 30? About 1 per team
How many players can be expected to produce 2-5 WAR per season? A lot more. About 130 last season. 4 or more per team.
How many players can be expected to produce 1-2 WAR per season? Hundreds. Enough fill out rosters for most every team in baseball.
How many players can be expected to produce 0 WAR per season? Even more than that. So, many that they have zero value.

A team always wants to replace a player with another productive player, however, there are a limited number of productive players. There are, basically, an unlimited amount of replacement players.

Please note that: there are TONS of innings tossed each year by pitchers well below average (e.g., ERA+ of 60-90)

.
By ERA+ any pitcher below 100 would be below average (not counting differences between leagues and starter/relieving)

If one thinks that there are 'average' pitchers in Triple A or Double A just toiling away because average pitchers are in the majors racking up counting stat WAR, then that is a gross error.

You don't get WAR just because you pitch. You don't get it just because you pitch a lot.

Since WAR compares a pitcher to replacement level, a pitcher with positive WAR is helping his team win games. You can accumulate a lot of WAR in 2 ways:

1) pitching a LOT at a decent amount better than replacement level or
2) pitching a decent amount at a lot better than replacement level.

3) Pitch a lot at a lot better than replacement level.

Both methods help a team equally IF the WAR amount is equal. The "Why" of this is what seems to be confusing.

Teams must pitch all the innings they pitch regardless of how well any individual pitches his innings. If a team plays a full game and the starter throws 5 innings of perfect baseball and leaves, the team STILL has to pitch at least 4 more innings. There is no bonus for pitching perfect baseball.

1) Every pitcher who is above replacement level helps his team win games but he ONLY does that while he pitches. He doesn't help the team by sitting on the bench, drinking beer in the locker room or napping in the bullpen.
2) Every inning that is NOT pitched by a pitcher above replacement level NEEDS to be pitched by a pitcher at or below replacement level.

If one were to assume that there are 140+ ERA pitchers sitting in AAA ball or not getting contracts, then would be erroneous. The 'leftover' pitchers not being used are close to or below replacement level. As a result, you can see from 1) and 2) above, that a pitcher above replacement level helps his team every time he pitches because he is better than anything else they have.

The more a pitcher is above replacement level, the more it helps.
The more a pitcher above replacement level pitches, the more it helps.

This is all pretty true.

WAR is meant to measure total contribution to a team during a season or career, not contribution per inning or contribution per AB. It's (IMO) incorrect to confuse value with quality. A pitcher with the same WAR in a season as another pitcher has the same value.

Of course we can use math to figure out that if a player produces 2 WAR in 200 innings that he produced WAR at a rate of 1 WAR per 100 innings.