The Indians Did Have Maybe the Best Pitching Staff Ever
The Dodgers snapped out of it. After an extended slump, the Dodgers did manage to snap out of it, and so they finished with baseball’s best record. The Indians ultimately had to settle for simply having the best record in the American League. And yet, one might consider the Indians to be baseball’s best team. They had easily baseball’s best second half. They had easily baseball’s best run differential. They had easily baseball’s best BaseRuns record. You could say it’s all just recency bias, but, recently, the Indians have been almost literally unbeatable. It’s quite a difficult thing to exaggerate.
The Indians this season could field, and the Indians this season could hit. More than anything else, however, the Indians this season could pitch. They could pitch when games were beginning, and they could pitch when games were concluding. It was a pitching staff without a clear weakness. According to WAR, Indians pitchers were better than the next-best staff by about seven wins. According to the other version of WAR, the one based on actual runs allowed, the Indians pitchers were again better than the next-best staff by about seven wins. Turns out these Indians weren’t just the best in 2017. They’re the best we’ve seen in a long, long while.
I just wrote about this a few weeks ago. My apologies if you don’t like when a subject is repeated. Yet I’m coming back to this for two reasons. One, the regular season is now actually over, so all the numbers are final. No more extrapolating or projecting out. And two, this is kind of important, at least as fun facts are concerned. This isn’t something just about how, like, the Nationals were pretty good at running the bases. This is about the Indians pitching staff being perhaps the best one ever. In baseball, “ever” is a long time. I mean, I guess in existence, “ever” is an even longer time, but there’s been a lot of baseball, is the point. And what is baseball if not the sum of its own history?
I went back to 1900. I looked at all the individual team-seasons. Here are the top 10 pitching staffs, sorted by WAR per 162 games.
| Team | Year | WAR/162 |
|---|---|---|
| Indians | 2017 | 31.7 |
| Braves | 1996 | 29.5 |
| Braves | 1994 | 28.8 |
| Braves | 1997 | 28.5 |
| Yankees | 2003 | 28.4 |
| Yankees | 2002 | 28.4 |
| Phillies | 2011 | 28.1 |
| Braves | 1999 | 27.8 |
| Braves | 1995 | 27.2 |
| Cubs | 1970 | 27.1 |
Indians, first place. This is out of a sample of 2,490. Let’s do the familiar thing with the numbers: the difference between first and second is 2.2 wins. The difference between second and ninth is 2.3. What’s driving this? The Indians finished with a team FIP- of 75, meaning, by the peripherals, they were 25% better than average. The previous best FIP- for any team was 80. That’s a fairly enormous gap, as these things go.
The WAR shown above is somewhat theoretical. It strips out, for example, matters of timing. There’s also RA9-WAR, which considers runs that have actually happened. Here is the same table, only this time showing RA9-WAR per 162 games.
| Team | Year | RA9-WAR/162 |
|---|---|---|
| Cubs | 1905 | 39.2 |
| Cubs | 1906 | 39.0 |
| Cubs | 1909 | 38.0 |
| Cardinals | 1944 | 37.1 |
| Yankees | 1939 | 37.1 |
| Indians | 2017 | 35.4 |
| Braves | 1974 | 35.2 |
| Braves | 1993 | 34.2 |
| Cardinals | 1943 | 33.9 |
| Cubs | 1907 | 33.5 |
Now we find the Indians in sixth, instead of first. That might be less sexy, but then, the most recent team with a better mark played baseball in 1944. Jackie Robinson debuted in the majors in 1947, and so began the so-called Integration Era. By this version of WAR, the Indians had the best pitching staff since the Paris Peace Treaties. It’s been more than 70 years. That’s a lot of baseball history being overwritten. See, the Indians finished with an ERA- of 74, which also stands as the sixth-best, and the best since 1910. For the record, last year’s Cubs finished at 75. That run-prevention unit was also absurd.
As a compromise, this table goes 50/50. So here’s one more top-10, taking the midpoint of each team’s WARs.
| Team | Year | Blend/162 |
|---|---|---|
| Indians | 2017 | 33.6 |
| Braves | 1997 | 30.9 |
| Cubs | 1909 | 30.7 |
| Phillies | 2011 | 30.4 |
| Cardinals | 1944 | 30.2 |
| Braves | 1974 | 30.2 |
| Braves | 1993 | 30.0 |
| Braves | 1999 | 29.9 |
| Dodgers | 2003 | 29.5 |
| Braves | 1995 | 29.4 |
The Indians go back to first, by a healthy margin. I’m not saying it’s true beyond any reasonable doubt the Indians had the best pitching staff ever. I’m just saying, if you have a counter-argument, the argument for the Indians is fairly convincing. They’re the best ever by our flagship pitching statistic. If you put more weight on actual runs allowed, they’re the best of at least the Integration Era. By ERA, there have been better teams, but those teams played under dramatically different conditions. Since baseball more inclusively opened its doors, no team has done what the Indians just finished doing.
The numbers are right there. There’s no hiding from them. And yet, I suspect this is a somewhat unpopular opinion. You don’t see many people calling this the best pitching staff ever. How could it be better than, say, those Braves staffs from the 90s? Between Greg Maddux, Tom Glavine, and John Smoltz, the Braves had three first-ballot Hall-of-Famers. The Indians can’t compete with that. But, well, allow me to break things down. Let’s look at all the Braves seasons from 1993 through 1999. They were all great! In this table, a comparison between those Braves and these Indians.
| Team | Split | ERA- | FIP- | IP |
|---|---|---|---|---|
| 1993 – 1999 Braves | Overall | 80 | 83 | 9617.1 |
| 2017 Indians | Overall | 74 | 75 | 1440.2 |
| 1993 – 1999 Braves | Starters | 77 | 81 | 7035.2 |
| 2017 Indians | Starters | 79 | 77 | 951.1 |
| 1993 – 1999 Braves | Relievers | 87 | 88 | 2581.2 |
| 2017 Indians | Relievers | 64 | 72 | 489.1 |
The Braves had amazing starters, and they threw 73% of all the innings. Indians starters, meanwhile, picked up just 66% of the workload. But look at the relievers. By ERA-, the Indians are better by 23 points. By FIP-, the gap is still an enormous 16. And Indians relievers threw the other 34% of the innings. Indians relievers were used more heavily, and, as a collective, the bullpen was spectacular. Look at that again — Indians relievers combined for an ERA- of 64. Pedro Martinez’s career ERA- was 66. The Indians simply didn’t use bad pitchers. They didn’t have any.
The Braves are remembered fondly in part because they stayed together for so long. They were dominant year after year after year, and the Indians are still working on that. But in terms of individual dominant seasons, the Braves are held in such high esteem because the best pitchers on the roster were some of the best pitchers on the planet. Those Braves were top-heavy, but that’s great for their perception, because staffs are often evaluated by their best weapons. The 2017 Indians didn’t function like the Braves of the 90s. There’s terrific top-end talent, sure, but there’s also just so much more depth, and more reliance on the relievers. Less was asked of the starters. Part of it’s just because of the era. Part of it’s because of how the team’s been constructed.
The Indians’ pitching staff doesn’t have three first-ballot Hall-of-Famers. It might not have a single Hall-of-Famer at all. In that sense, they have nothing on the best recent years from the Braves. Yet pitching staffs aren’t evaluated by the number of Hall-of-Famers they include. You don’t win games by having Hall-of-Famers. You win games by preventing runs from scoring. Top to bottom, the Indians did that better than the Braves did. The Indians did that better than anyone’s done, for an extremely long time. Pitching staffs nowadays are more complicated than ever, because of how starters have lessened in importance. But also, everything’s still wonderfully simple. You either keep runs off the board, or you don’t. The Indians did. You couldn’t ask for any more.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
The Indians have the staff which embodies the current pitcher WAR formula more than any in history. This makes sense as WAR is built around modern eras, not those of the past. The Indians are a low walk / high K staff which is why they are in 1st place for WAR. A handful of years ago, teams wanted their starters to pitch deeper and Ks don’t fit that mold. The Indians didn’t even lead the league in quality starts, which seems like a good enough stand-alone measure of a starting rotation. Without a special rotation, I am going to look elsewhere for the best staff in history. I think most discussions about historic staffs are exactly about the hall-of-famers on the staff – otherwise we are just looking at a print-out of numbers and that’s only really good for writing headlines. The fact is that “best ever” conversations are supposed to be fun – there are no right answers. If you are lucky you might learn something you didn’t know or re-live memories from the past. Sabermetrics often miss that mark as it is too often simply, insert_player/team_name had the most insert_stat_here in insert_year.
I would argue that WAR gives way too much value to relievers, which is what I think we see here. They do have the WAR-centric skillset of K-BB. Anything beyond that is hard for a machine to evaluate. I am not saying that K-BB is bad, but there is a lot more to pitching than striking guys out (see Braves staffs). Numbers folks want to find ways to call that “luck” or luck’s cousin, “sequencing”. Maddux and Glavine were good at being lucky. I will easily buy the Indians as the best staff this year, but I don’t buy them as a historic group. I don’t worship WAR and I encourage others to think outside of the box that we are creating, just to break – maybe its more like a pedestal…
Hence the inclusion of RA9-WAR, which is based on ERA, and which shows the Indians as having had the most effective staff since the end of the second world war.
I don’t know much about that stat. It is obviously not the same thing as ERA as the Cubs beat them last year and they are not even on the list, nor are the Dodgers who finished just behind them this year. Which means (maybe) that this stat penalized the Cubs staff for having a historic defense last year (laughable, given this year) or it is just relative to the mediocrity of the rest of the league in 2017, or it isn’t really ERA-based. I know I don’t have much business critiquing a stat that I didn’t do my homework on… I am just saying that new-age stats aren’t going to convince me of anything historic. Why not just use ERA and at least use a stat that existed in previous eras that players were actually trying to accomplish? I know it would not produce any headlines, but that is kind of my point. I thought your article was good, I am not trying to belittle it, just trying to serve up some discussion… I expect the downvotes to outnumber the upvotes, but I would think that some people would want to read a somewhat thought out alternative comment.
> “It is obviously not the same thing as ERA as the Cubs beat them last year and they are not even on the list”
Where are you getting this from?
From the article: “See, the Indians finished with an ERA- of 74, which also stands as the sixth-best, and the best since 1910. For the record, last year’s Cubs finished at 75. “
ERA is not ERA-. Not sure what your point is in regards to mine.
You’re right. ERA- is meaningful (at least when compared to ERA) when comparing players of different eras, and ERA is not.
Why use ERA? It’s lacking context. A 3.50 ERA in 2017 isn’t the same as a 3.50 ERA in 1960. Similarly, a 3.50 ERA in the National League is not the same as a 3.50 ERA in the American League. This information is important and that’s why ERA- is a far superior metric.
Just because something is thought out doesn’t mean it was thought out well.
I don’t care much about your “ERA” stat. Why not just use the average height of the pitchers, and multiply that by the number of letters in the manager’s last name? I know that this won’t produce headlines, but I think at least that some people would like an “alternative” way of looking at things.
Exactly. Why not?
I think most discussions about historic staffs are exactly about the hall-of-famers on the staff
I think you are generally correct about how people think.
That being said, a pitching staff is a lot more than the three best guys. Quibble with the numbers or the methodology, sure, but to say that you only care about the top of the staff is to say that you’re not answering the same question.
Sure, the back-end definitely counts! I am sure the shaky back-end of the Indians staff helps them here as the bullpen gets more innings as racks up WAR. It is probably pretty clear that I enjoy the history of the game, it’s great players and their achievements – I would imagine that many hardcore fans do think in those terms. I agree that I have a different viewpoint. This number that Jeff presents is not wrong, it just isn’t an accurate way to look at history for me. For me, we can talk about right now, which we are already doing, or we can reflect on 50 years of history – I find that to be much more interesting.
Is the “shaky back-end” of the staff the three guys who each had more than 2.0 WAR and started a combined 66 games? (Clevinger, Tomlin, & Salazar).
Their rotation used only 7 guys all year, and one of those guys (Merritt) only started 4 times. That alone is stellar. They only had one reliever exceed any of the core 6 starters in WAR. Of the Indians 31.7 pitching WAR, 22.6 was from the starters.
So yes, the back-end counts, and it’s because they were very, very good.
“I am sure the shaky back-end of the Indians staff helps them here as the bullpen gets more innings…”
Actually, Cleveland relievers pitched the fewest innings in the AL, and 2nd fewest in the Majors.
My mistake. I am just looking at high level stats and making guesses.
The Indians don’t have a shaky back end of their staff. That’s sort of the point. And the top end, Kluber and Carrasco, is obviously every bit as good as peak Maddux and Glavine. Kluber should win the Cy Young, and Carrasco was one of the top 4 starters in the league by just about any measure. For any pitcher who threw over 100 innings in the AL, the Indians have 5 of the top 11 in k/9. Their top 5 starters are in the top 19 in the league in FIP. They have 3 of the top 10 in ERA.
I think part of the problem is you can’t wrap your head around the fact that someone like the Braves was an NL team. If you put the Indians pitchers in the NL where they faced a pitcher their strikeouts and ERA would look even more ridiculous.
Its shaky if you want to talk about a historic staff. All of those back-end guys were in some jeopardy of losing their jobs at points this season. Sure, it compares well to the rest of the league in 2018, which is what this “historic” staff is all about.
The only reason pitchers on their staff might have been in jeopardy of losing their jobs is because they were pitching for Cleveland, who had 6 good starting options when Salazar wasn’t on the DL. The Indians #6 starter had a 3.19 ERA and 3.41 FIP in the 2nd half. Their #7 starter threw 20.2 innings on the year with a 1.74 ERA and 3.06 FIP. Who are you even referring to, Bauer? Of the 5.24 ERA but 4.06 FIP in the 1st half? Bauer would be the best pitcher on multiple different major league teams, for Cleveland he’s really their #4 or #5 starter (Clevinger is better than Bauer, and Salazar is better than Bauer when he’s healthy)
That’s revisionist history. Salazar, Clevinger and the rest of them did lose their jobs. This is also the first point in history where Bauer’s job is safe. People have been clamoring for him move to the bullpen since he has been in Cleveland.
I don’t think I’ve ever heard anyone criticize WAR for overvaluing relievers. Most critics seem to think it undervalues them drastically because it doesn’t account for leverage.
Maybe the argument is that they allow inherited runners from SPs to score? Well, for RA-9 WAR that is. I dunno, kinda grasping at straws here.
RP have an easier job than starters. They also have a WAR-friendly job description – strike this guy out, leave game after/before any adversity. Using the same tool is problematic for me. If RP begin to tire, they are done. They don’t even have to face opposite-handed hitters in many cases. They don’t have to worry about ever throwing their third best pitch or the next time through the order. There is a good reason that busted position players recoup a shred of value through moving to the bullpen. Its the easiest job description in baseball. When you treat a staff as one big unit that pitches 9 innings, the team that deploys more RP should accumulate more WAR if you have the bullpen to do it – and this is a staff built around the bullpen. I do think that WAR overvalues RP for sure… relative to SP.
I think this is totally wrong. If I understand the definition of WAR correctly (https://www.fangraphs.com/library/war/calculating-war-pitchers/) relievers, if anything, are penalized for the fact that their “replacement” is often someone already at the major league level – that is, another pitcher already in that team’s bullpen. They are “rewarded” for often pitching in higher leverage situations though. All of this has been taken into account in the formula. This is why the stat allows us – even if not perfect – to compare players across positions.
Also, stats aside, I’d say that relievers and starters have the exact same job: getting outs, preventing runs, preventing baserunners.
I don’t know how WAR works with RP. You could certainly be right. If you say you are then I believe you. I just always figured that it must be kooky as every year lots of very replaceable guys accumulate value. I figured the only way they could have value would be to compare them to SP. I was just trying to guess at how this staff could be special by any metric.
I do not agree that RP and SP have much in common at all in terms of job description. One has to work through tough situations and one gets to pass them off to a guy with a better match-up. The minor leagues are full of players that could provide relief innings, but very few real starters.
Again, it’s scaled to league average, so when you’re looking at reliever WAR, you’re comparing them to OTHER RELIEVERS who have an equally “WAR-friendly” job (which means it can’t really be WAR-friendly). So a replacement-level reliever is going to be a guy who still does a pretty good job of striking players out, etc. Like Kelvin Herrera for the Royals was pretty much a replacement level reliever this year
>This makes sense as WAR is built around modern eras, not those of the past. The Indians are a low walk / high K staff which is why they are in 1st place for WAR. A handful of years ago, teams wanted their starters to pitch deeper and Ks don’t fit that mold.
Sure, but unpack that statement a bit. Teams use pitchers differently these days. But just why do you think they made that change? To be less effective?
For one, the league changed around them. Everybody strikes out more. That alone inflates WAR over previous eras. I am not sure pitching staffs are better these days – they are just different. It’s not like this staff had some historic ERA or anything like that. Newer isn’t better in every case. My two cents – radar guns have led to more arm injuries and forced bullpen usage. Perhaps guys don’t learn how to pitch the way they used to. When your ticket to MLB looks like a triple-digit radar gun reading, you know what you have to do.
I don’t think things are better these days, just different. I don’t buy this as the inevitable course for pitching staffs, just the current situation. In the future, when the game changes, people will act like that is the pinnacle of the evolution of the game. I am always saying that using WAR in a historic context is a bad idea – that’s all it boils down to for me.
“For one, the league changed around them. Everybody strikes out more. That alone inflates WAR over previous eras.”
No it doesn’t. There’s a fixed amount of WAR available every year (depending on the number of teams/games). And Fangraphs splits WAR between pitchers and hitters by the same percentage every year. Baseball Reference does the same.
As I have said a few times, they just match the arbitrary metric better than other teams and respective eras. This gets hard because I am replying to 10 different angles at one time. I was more responding to the implication that modern staffs are better than older staffs. The question was, why do teams K more guys… something like that. I think I could form a coherent, more relevant reply here, but I am stretched too thin. Even if WAR is fixed among pitchers, the weight of Ks would allow a team to horde more of the WAR than in previous generations. Things were obviously more balanced in the past, which is what this article is about. It does appear to be a huge input, too high I am sure. I will admit that I am learning a bit about WAR in these comments.
WAR is meant to be used in a historic context. What would you use, ERA? You think comparing pitchers before the mound was lowered versus a watered-down, pre-integration league by ERA is better than comparing them by WAR? The inputs in WAR are scaled to league average, so something like league strikeout rate shouldn’t actually inflate WAR, since everyone in the league is going against those same hitters who are striking out more.
I wouldn’t compare one generation to another. the whole thing has a very elitist feel to it. Enjoy the game. Follow it for decades and enjoy it all. Why would anyone want to declare a generation better than another, other than to write a headline?
“Sure, but unpack that statement a bit. Teams use pitchers differently these days. But just why do you think they made that change? To be less effective?”
This is the key point here. What RonnieDobbs wants to argue is that because there’s alignment between current best practices and what we understand to be good there will be some recent teams will look better.
This is technically correct, but it’s hard to understand why it’s worth getting worked up about it. The teams appear better because they *are* better, precisely because their practices align well with what we know works.
The only way this is an issue is if there is some radically different formula that gets value better than FIP-based WAR. So far there isn’t*
*SIERA is better, but SIERA and FIP mostly agree.
We are creating metrics to affirm current practices. I don’t think this is healthy for the game. This is the recipe for speeding down a path, potentially in the wrong direction. I don’t believe that 2017 MLB is better than 1997 MLB. The OBP revolution was real, but beyond that I think sabermetrics is mainly grasping for straws – which is how you find something eventually. I think if you had previous modern era teams play 2017 teams they would split the games, despite what anyone here thinks – I wouldn’t be shocked if the older teams won more than they lost. I am not sure what other readers think about that point. I have a lot of problems with calling one era better than another. What does this say about the players that built this game? I watch modern baseball and I don’t think its better, its just different. I would say that modern players are less baseball-skilled, but they have better measurables. I am sure that modern bullpens would give older players some fits, but the small-ball, professional at-bats would give those same hard-throwing poor-command arms a lot of problems. It would be fun to watch. Rather than just talk down on the greats from the past, I want to recognize that they were legitimately great by any standard.
“We are creating metrics to affirm current practices.” What? Dear god, no, I hope not. To influence or change current practices? Maybe. To understand and compare what seems otherwise incomparable? For sure. But not to affirm, in general, not when these metrics find so much to critique in current on-field and back-room practices.
If you created a metric and it wasn’t flattering to the current era, then it would be discarded. I see it all the time. When a metric is created, they look at the list of players that it floats to the top, that’s probably the most important test. The end result is just a bunch of alternative metrics that all tell the same story – which is the story that you want to tell. You would literally tweak the inputs and weights to make a metric match up with what is already out there or your narrative if it didn’t. I think that modern metrics rarely tell baseball people anything that they don’t already know. Statistics are used by people to support their own narrative all the time. That’s not unique to baseball! Statistics can be used to unearth things or to simply support an argument. Numbers are objective, the methodology and people who use them are not. I don’t accept that we are viewing some higher-order evolution of the game. Sabermetrics are just a tool for analysis that people can use or misuse as they choose – the idea that this is the path to baseball enlightenment is flawed IMO. If this was about 2017, then I would have nothing to say about it. I don’t like WAR, but that wouldn’t even warrant a comment. My issue is that any methodology which calls the Indians staff (same starters and closer as last year’s non-historic group) the best staff of all time is wrong. Go ahead and disagree, but citing arbitrary metrics ins’t terribly compelling to me. If you like them, then that is great for you! How will you feel about the next “best stat” that comes along?
You’re getting dangerously close to “fWar is a conspiracy to make the 2017 Indians look like the best staff ever” territory.
Who’s talking down the past? Our current practice affirming metrics as you call them still show Ruth as the best player to play the game.
Generational talent still looks like generational talent with these metrics. Heck you could say people in this generation are finally playing by the advise and approach used by Ted Williams. That is pick your pitches where you’re good at hitting them, not just any pitch, and add a bit of loft. Part of of his legendary talent is taking a walk.
Pitchers aren’t just throwing heat without command. Cleveland rained down and away on the Blue Jays last ALCS crushed offense historically hard. Teddy was only a .250 hitter in that region of the strike zone vs .344 overall. His career was murdering pitchers slightly above the middle 1/3 of the zone.
The article… If this is the best staff of all time, then what does that say about baseball history? That is a fine point about Ruth. I personally would never even consider comparing segregated baseball to today.
RonnieDobbs, I see where you’re coming from but this just isn’t correct in this case.
People do create statistics to validate current practices, but new metrics also push people to align their practices with what works. No one created FIP-based WAR to judge today’s pitchers the best in MLB history, especially since WAR is scaled for era.
“What RonnieDobbs wants to argue is that because there’s alignment between current best practices and what we understand to be good there will be some recent teams will look better.
This is technically correct, but it’s hard to understand why it’s worth getting worked up about it. The teams appear better because they *are* better, precisely because their practices align well with what we know works.”
I fundamentally disagree with this.
For example, do Carrasco’s stats look good because he only faced 227 batters the 3rd/4th time through the order, while Glavine in 2002 faced 298? How much worse would Carrasco be if he had the usage pattern of a pitcher from 15 years ago?
The goal of WAR is to normalize across eras, but if usage has changed, then you’re not really making fair comparisons.
That makes zero sense
Carrasco was dramatically better at keeping runners off base the first time he faced them, so the 3rd or 4th time he (Carlos) would have faced batters was pushed much later in the game than 2002 Glavine was.
So are you arguing we should penalize Carrasco for pitching so much better than Glavine to start the game?
First time thru order:
.202/.267/.357/.625 – Carrassco ’17
.268/.333/.432/.765 – Glavine ’02
Second time:
.244/.295/.368/.664 – Carrasco
.236/.313/.351/.664 – Glavine
Third time:
.271/.307/.452/.759 – Carrasco
.253/.325/.419/.745 – Glavine
Forth time:
.200/.250/.333/.583 – Carrasco
.344/.476/.563/1.039 – Glavine
That is a .066 point difference between their respective OBPs for the first time they faced the order. Had Tom been able to do what Carlos did the first time thru his number of 3rd & 4th times faced would have dropped dramatically. And as you can see, Carrasco was ironically just as good to better than Glavine the 2nd, 3rd and 4th times he faced guys, so…
Also ironically;
6.25 – Carrasco ’17
6.24 – Glavine ’02
That is their respective number of Innings Pitched per game
…and all of this ignores the fact that Carrasco did it in the AL with a DH instead of facing an automatic out in the 9th slot. Think that might make it easier to get thru the order for Glavine? Despite that advantage, he didn’t – he faced hitters a 3rd and 4th time more often (and performed worse doing it) despite averaging the same exact number of innings per start
…and it’s off-base to begin with as a pitcher is heavily rewarded for keeping a replacement level guy off the Field. That is why relievers have such a hard time getting a high WAR; they leave too much time on the mound to be filled by the theoretical replacement guy. So a starter pitching 200 innings of even league average will have a much higher WAR by default. Similarly, a shutdown starter not pitching very often/long will have a decreased WAR to account for the replacement guy he would theoretically be turning the game over to. Want an example, compare:
Geo Gonzalez – 201 IP, 3.93 FIP, 3.3 WAR
Brad Peacock – 132 IP, 3.07 FIP, 3.4 WAR
The two were no where close to equal in performance yet ended up equal in value solely because Geo pitched 70 innings more than Brad. Coming out of the game/not staying healthy dramatically decreases your WAR, not improves it
…and all of that is completely besides the point anyway considering we are talking about the entire teams performances (“current best practices {of the team}”), not the performance of one individual guy. Even if the starter comes out of the game earlier, the rest of the team still needs to get the lineup out in the same order the starter left it
For someone who is so steadfast in Cleveland NOT being the best staff of all time, you’re using words/phrases like guess, I don’t know much about/I don’t know how, my mistake, I’m sure, etc to describe your “analysis” a little too much.
So instead of constantly replying with something you pulled out of your you-know-where and having to acknowledge you don’t understand a stat or aren’t familiar enough with the Indians staff after, maybe just chalk this one up as an L and agree to disagree.
I’m a big fan of the Yankees, but I’m really surprised by the 2002-2003 Yankees being on the WAR/162 list. I remember having a good staff, but not historically good. Goes to show you how quietly good Mussina was.
Wow, it wasn’t that long ago that the Dodgers had a massive lead over everyone else in team ERA. That didn’t take long to dry up.
Losing the best pitcher in baseball hurts a lot. Even with the complete collapse of the rotation, they finished within a tenth of the Indians in ERA. To me that doesn’t sound like a historic staff.
The Dodgers are in the NL. An AL pitching staff with the same ERA as an NL staff is a much better staff
While I agree, the Dodgers have a tougher division and they play games in AZ and COL. As with everything, its all more complicated than it seems. I’ll take the Dodgers staff over the Indians staff if they are both reasonably healthy.
I think modern scholarship tends to regard the Tinker-Evers-Chance infield defense as overrated. Does the RA9-WAR table above justify a higher reputation than they currently have?
WAR is a somewhat economic measure.
Many sub-sections of the game,are hard to quantify.
Examples,catchers,Def metric in Colorados outfield,aka The Grand Canyon…etc.
Still,I would argue,Clevelands staff is better overall,than any dead ball eras.
Everyone knows the rules,todays game is much more challenging.
Thats like comparing Pedros 1998 season with W.Johnsons 1918.
The Indians are No.1,and all the dead ball teams should be re-evaluated.
In case this notion seems hyperbole to anyone, just check the plain old runs allowed:
— Cleveland’s 3.48 RA/G was 1.27 runs less than the other 14 AL teams (4.75).
— That’s the 6th-best figure in raw terms since 1901, behind the ’39 Yanks, three versions of Tinker-Evers-Chance, and the ’44 Cards. It’s 8th-best as a percentage.
Is that not an equal reflection of the mediocrity of the rest of the league? That’s how you get to the top of a list – a good performance and the right surrounding circumstances.
Also, 2017 Cleveland is just the 3rd staff since at least 1913 to lead its league in all “three true outcomes” per 9 innings — highest SO/9, lowest BB/9 and lowest HR/9. The other teams had big park assists (’99 Astros and ’88 Mets).
I think today’s league has totally awful pitching talent. Even with juiced balls and players trying to hit fly balls (because they know the balls are juiced), I really think it’s true. Great pitchers are now at a serious premium and I think we’re at a little valley in terms of the level of talent that is in the league right now compared to just a few years ago. So I think this Indian’s staff looks great not because they would compare favorably if actually pitted against a historical team, but instead because the pitchers surrounding them are simply historically poor and/or injury ridden.
I also don’t think there’s a point to making a “wins above replacement” statistic based upon peripherals. It should instead be called “expected wins above replacement,” because in reality that’s all it is. A peripheral-based WAR does not denote actual wins that a player caused.
Also, of course a WAR based upon K-BB% is going to say that a modern team is the best team ever. That’s because it’s in style to maximize K-BB%. RonnieDobbs has some decent points.
Your #3 is proving you don’t actually understand what WAR is at all as it is based on Balls In Play and Base-runners, not K/9 and BB/9
Your #2 is kind of (extremely poorly) describing what the intent of WAR is, while amazingly claiming it shouldn’t be used because it does as intended. Then you throw out the truth you stumbled upon to instead go back to showing your lack of understanding of the purpose. Here’s a hint; add “in a completely equal, bubble environment” to your “expected wins above replacement” and see if you can come up with the value in that
Your #1 is just nonsensical rambling as you look for an excuse not to have to believe stats you don’t want to believe.
So I might suggest spending less time coming up with your personal thoughts on things you have absolutely no understanding of in the first place.
Aren’t Ks and balls in play the same thing expressed inversely more-or-less?
Your #1, I am looking at the FIP equation and Ks and BBs are both part of the equation. In fact Ks are multiplied by 2 and BB by 3! To say that they don’t matter (implied), looks wrong to me. Ks, BBs and HRA, IP look like the only things that matter. One thing that is abundantly clear from this discussion is that very few people know what they are talking about – me included! My comments are just intended to be baseball comments, but some times they cross into sabermetric world around here.
tell you when the ball was not put in play, if players were getting on base without the ball in play, and indicating when batters would or would not be moving around the bases with the ball not in play. In other words, accounting for all the stuff the pitcher controlled (his performance) thereby isolating for normalization of the things he didn’t.
It is completely based off Balls In Play and base-runners; it is separating the controllable performance from the non-controllable (or “luck”) to normalize “luck” and therefore level the playing field. Ks stand out solely because they themselves are eliminating the luck (striking out the side eliminates all possibility for luck; there is no question you have a 0.00 ERA so “luck” doesn’t need to be normalized)
Here, I’ll see if I can put it in a Real Life situation to see if we can help you better understand though (and this will be long, but I’ll try to keep it as simple as possible)
Cy Young struck out merely 3.43 per 9, meaning about 23.5 of his Outs per 9 innings were made by his Fielders.
Now take a look at the disconnect between Cy’s Earned Runs and Runs Allowed though – he had a 2.63 ERA and 3.87 RA/9
So in a normal game would Cy Young be expected to give up merely about 2.5 runs over 9 innings his ERA shows, or the nearly 4 he actually did?
That 1.24 is an extreme run difference between his Earned Runs and the hard number of Runs that actually crossed the plate while he was on the mound. He was allowing more than a run per game over what was showing up in his ERA because of the volume of balls in play that the fielders were responsible for taking care of …and they obviously didn’t do that great of a job handling the extreme workload he gave them
That is why the bubble is created; isolating and normalizing the things he cant control to then show and better value that which a pitcher can. Based off what Cy could control coupled with a normalized, bubble around him responsible for his Balls in Play, we know Cy Young should have been giving up about 2.89 RA/9 (or FIP)
Since the 2.89 FIP is higher than his 2.63 ERA, we see that Cy Young was actually a bit lucky in the ERA department. Apparently the poor defense behind him was masking some of what should have otherwise been Earned Runs with their many errors. He was dramatically better than how many Runs actually crossed the plate while he was on the mound though.
And once you have isolated and eliminated “luck” and gotten your FIP, you can then compare it to what the baseline Replacement Level player would have been expected to be, giving you an end Runs Above Replacement (and in turn, Wins Above Replacement) value. Everyone goes thru the same process so everyone is on completely level ground, allowing for a comparison of pitchers bubbled controllable performance outside variables not in their control
First of all, I understand these things very well, because they are easy to understand, because statistics is generally a fairly easy school of mathematics to understand. Applying them is not always, unfortunately. Sorry if that bothers you, but it’s true.
From my reading, Fangraphs WAR is based upon FIP which is very much based upon the three true outcomes and discounts balls in play besides home runs. RA9 WAR was included in this article which calculates WAR based upon what actually happened. The Indians were impressive in this statistic but were not the best staff of all time considering it.
Maybe what I’m criticizing is WAR itself, because if that is the true purpose of WAR (I don’t think what you’re saying is ACTUALLY true but I’m entertaining your dumb and nearly impossible to read argument), then it does not properly accomplish that purpose. WAR presumes that the overall talent level of a given league in each individual year is roughly the same, so therefore comparing player WAR in 1970 and 2011 is totally understandable!
Wrong. There is obviously going to be a substantial difference in league talent on a year to year basis no matter how inconvenient that is for statistic makers. This has been argued over and over and over again and no one really has an answer for it besides, “Well, the difference can’t be THAT substantial if we don’t go back too far.”
My first argument is simply to stop calling an FIP-based calculation, “Wins Above Replacement” and instead begin calling it “Expected Wins Above Replacement” because that would be more accurate. My second argument was that the Indians staff looking so good from an FIP standpoint is more the result of a league-wide valley in pitching talent.
Arms are getting blown out all over the place, dying in boat accidents, etc. Talent is getting old. My point was that even discounting the juiced ball surge and subsequent popularity surge of hitting fly balls because of it, I sincerely feel as though this years pitching was especially weak, which makes the Indians’ staff look especially good from the standpoint of WAR. What makes it look even better is that it is built around the three true outcomes, when maybe pitching for the three true outcomes wasn’t always the best decision like it is now. Maybe the article should’ve been titled, “The Indians Did Have Maybe the Best Pitching Staff Ever Compared to League-Wide Performance In A Given Year.” That’s just me.
“From my reading, Fangraphs WAR is based upon FIP which is very much based upon the three true outcomes and discounts balls in play besides home runs”
Home Runs are not Balls In Play, and it doesn’t “discount” Balls in Play what so ever, it isolates them so they can be normalized.
The entire point of WAR is to normalize for “Luck,” creating an equal playing ground for all. FIP normalizes the luck that is inherent with Balls In Play, leaving the controllable performance abilities as they were. The basis for FIP is Balls In Play
“Arms are getting blown out all over the place, dying in boat accidents, etc. Talent is getting old.”
Arms have always been blown out, and pitchers have always died early off the field, or even been taken off the mound by things like batted balls – the difference is, ALL pitchers are also now able to play in the ML, and they now being developed all over the world starting at the age of around 5 most of the time, with the best physical training abilities and ideas in the history of the world in mind, and scouted constantly all over said world by more people than used to be involved in the game in total. Those players are then coming from all over the world to fill out the same tiny ML rosters that have always existed. There are a few more roster spots to fill (dozens, not thousands), but there are millions and millions more in a beyond primed talent pool to draw from
…if that somehow results in less talent present in the ML to you then I really don’t know what to say.
That is NOT the entire point of WAR, and I consider a home run a ball that is technically put into play, and actually not totally fielding independent. There are many home runs that MAY have been saved by a good enough fielder.. we just don’t really account for them. The entire point of WAR is defined by Fangraphs here:
“Wins Above Replacement (WAR) is an attempt by the sabermetric baseball community to summarize a player’s total contributions to their team in one statistic.”
So no, what a pitchers EXPECTED and/or FIELDING INDEPENDENT metrics look like should not contribute to WAR because WAR is about what HAPPENED, not “what should’ve happened if not for luck.” This is a distinction that you do not seem to understand.
Secondly, please stop pretending like you’re educating me. You aren’t. If you still think that pitchers can’t control balls that are put into play, you should probably start reading more literature and get updated. It isn’t 2003. We know that pitchers like Dallas Keuchel aren’t totally lucky.
Luck is a word that is abused by statisticians, because it masks their limitations. Something that they may have a hard time forming a mathematical relationship for using broad-stroke catch-all methods (which is literally what statistics as a school is lol) will be considered a product of “luck” and/or “randomness” even where there is RATIONAL evidence (as opposed to quantified empirical) for a clear pattern. Because without “luck” or “randomness” or “noise” statistics is limited. And it is. Statistical proof is inductive proof, and no one here wants to admit it. Anyways, it’s been proven that some pitchers can sustain lower BABIPs than others. I think FanGraphs tends to do a pretty good job admitting this though. You seem to think differently.
Thirdly, it is simply my opinion that via LUCK (lol), we have simply entered a time where pitchers are more injury prone now than they ever were. ESPECIALLY ones who have strong throwing arms, etc (who tend to be the best pitchers). Pitchers are being taught to go “all-out” for 5-6 innings and to go for the K instead of pacing their arms, pitching for bad contact (which, yes, is possible, sorry to say) and not throwing as hard as they can all the time.
What’s crazy about what you’re saying is that we ARE in a time where young, promising arms are going down at a rate that they never have before in the history of the sport whether from UCL injuries or from blisters. This is an objective truth. I don’t think it, hard data thinks it. And don’t give me that crap about the world talent pool, because I’m not really comparing today’s game to the game of the 60’s and before (because in that case, you’re very correct about the limited talent pool). I’m more talking about post-60’s or 70’s and after baseball. I think if all other factors are TRULY equalized, that mid-90’s Braves staff is SORELY superior to the Indians’ today. I don’t think it’s close. I think that WAR is limited in this instance because it has to compare this Cleveland staff to a league with an injured talent pool.
No matter your opinion, you must consider that you are making a PRESUMPTION when stating that the pitching talent pool in 2017 is as great as it was in 1995. That if you took the best of then, the best of now and got enough samples, they’d be about even against the same hitters. In fact, not only is that a presumption, but that is a completely RIDICULOUS presumption. Of COURSE overall league talent changes from year to year. All you need to do to figure that out is look at your every day life experiences, look at other sports, look at schools of academics, arts like music and painting, to see that talent pools in all human practices wane and wax.
I don’t have any real preference for some old version of baseball or baseball now, whatever. I don’t have any biases or things I want to believe. I’m simply skeptical after watching this Cleveland staff pitch all year that it would measure up to other staffs of even RECENT history. That’s just my opinion.
No one cares what you personally think should be the case or the silly reasons you think it; a HR is not a Ball In Play
Yes, the entire point of WAR is to credit a pitcher for what he is responsible for and normalize that which he is not – and what he is not responsible for is the outcome on Balls In Play. He can control his Ks, he can control his BB, he is responsible for his HBP and such …but the outcome on a grounder to short he has little to no control on.
You quote the definition as “summarize a player’s total contributions to their team” …and that specific pitchers Short Stop being the worst fielder in the history of the game is not part of the Pitchers “contributions to the team” – the Balls In Play the SS cant get to would be “the Short Stops contribution”
– Quote – “So no, what a pitchers EXPECTED and/or FIELDING INDEPENDENT metrics look like should not contribute to WAR because WAR is about what HAPPENED, not “what should’ve happened if not for luck.” This is a distinction that you do not seem to understand.”
That is insanity as it removes both the definition and point of the stat to instead replace it with your personal feelings – feelings which would make this asinine statement your argument if such a situation presented itself: “it doesn’t matter that the A’s are fielded nothing but guys literally right out of High School, all that matters is how many hits that pitcher gave up”
– Quote – “It isn’t 2003. We know that pitchers like Dallas Keuchel aren’t totally lucky.”
…Keuchel has a career .291 BAbip and ERA within 10 points of his FIP, so very interesting example
– Quote – “What’s crazy about what you’re saying is that we ARE in a time where young, promising arms are going down at a rate that they never have before in the history of the sport whether from UCL injuries or from blisters. This is an objective truth”
It is not a “truth” that is depleting talent – it’s just some crud you’re spouting trying to justify what you want to believe.
We truly have zero idea how may would-be greats of the game never made it even out of High School or thru the minors 80 years ago because of injury anyway. We don’t even really know how many ML guys about to become greats blew out their arms in the first couple years they played.
What you, me and everyone else DOES know is that they never, ever, *EVER* could have come back from the injuries back then, so if you did destroy your arm early, you were forever gone and forgotten. And that’s true up until even about 40 years ago. Now though, the best pitcher in the history of the game can have an otherwise career ending injury and be back on the mound 12 months later without issue. That doesn’t deplete talent levels – it increases them!
So not only are careers no longer ended by what would have previously been career ending injuries, the talent pool for the game today is also millions upon millions upon millions of people deeper …and kids are better taught how to fully develop all thru the world …and there’s previously unimaginable levels of scouting across said world so they will not go unnoticed …and they are without archaic beliefs keeping them from playing in the US, and on and on …and yet you somehow argue that means there is less talent. Why? Well like above, it just fits what you want to believe; facts be damned
– Quote – “I’m simply skeptical after watching this Cleveland staff pitch all year that it would measure up to other staffs of even RECENT history. That’s just my opinion.”
An “opinion” which is not supported by facts – be them hard data or advanced metrics. But hey, people are free to believe in Santa too if they wish
You didn’t come back. I’m surprised. I anticipated you doing so; being forced to cede the fact that the talent pool today is clearly dramatically bigger than it has been in history, but doubling down on this silly notion that the one exception is this small window 20 years ago when the team you personally want to think was better played
I was eager to ask why that team you insist was superior was apparently unable to find the mythical better talent depth you claim clearly existed. Afterall:
3.30 ERA, 3.33 FIP – 2017 Indians
3.18 ERA, 3.45 FIP – 1997 Braves
3.25 ERA, 3.53 FIP – 1998 Braves
3.44 ERA, 3.58 FIP – 1995 Braves
3.54 ERA, 3.50 FIP – 1996 Braves
3.57 ERA, 3.58 FIP – 1994 Braves
3.65 ERA, 3.85 FIP – 1999 Braves
4.06 ERA, 4.19 FIP – 2000 Braves
Despite the Braves facing a pitcher in roughly 11% of their PA (as opposed to what is assumed to be the best hitter on the team the Indians saw in that spot) …well as you can see, despite that massive advantage, the Braves only bettered the Indians in ERA twice, neither time by a large margin.
And that’s just the hard ERA for the team facing a pitcher roughly 11% of the time – the Braves never had a FIP within even 10 points of the Indians
– Quote – “that mid-90’s Braves staff is SORELY superior to the Indians’ today. I don’t think it’s close”
Your claim is that its not even close, that the Braves teams were always so far above the Indians staff.
Yet I see, at best, an argument that could *maybe* be attempted on merely 1 season (1997) and otherwise the Indians pretty easily beating to flat out blowing away the Braves every other year …and that’s when using merely the archaic ERA stat, with FIP showing there was hardly a contest at all
Comparing team rate to rest of league, among 2,274 MLB teams since 1913, Cleveland ranks:
— 6th in net SO/9, at +1.92 (after 2003 Cubs, 1960-61 Dodgers, ’90 Met and ’32 Yanks).
— 20th in net HR/9.
— 80th in net BB/9.
No other team is in the top 130 of all three.
(Data via B-R’s team pitching splits; Federal League not included.)
First off, while it was stated in the article, you just have to marvel at the consistency of those Braves staffs. 20% above league average for 5.5 years-ish? That’s dang impressive.
Secondly, the 1970 Cubs threw me for a loop. Looked that team up and they were woefully unlucky that season. Such used to be life for a Cubs fan.
This is oh so AL-centric of you. What about the offense? If you add the Indians offensive contributions this year (zero) and the Braves in 1996 (2.1 WAR), you end up with a virtual tie (31.7-31.6). Those Braves had a robust 32 wRC+ and 3 dingers (which are actually good numbers!).
Edit: The ’96 Braves were actually the best offensive value pitching staff of any team from 1993-1999 by WAR. The 1997 Braves were the 4th best over that stretch. The 2010 Brewers are the only team to top them since (2.3 WAR)
Why on earth does this have down-votes?
I’m just fine with giving the 2017 Indians “The best pitching season ever” award. But not best pitching staff ever. Geez, those 90’s Braves did it for most of a decade. And it seems the 19-aughts Cubs defense was not over-rated. Hopefully Indians can finish the job this year. Really strong group of teams for this postseason, should be fun.
Any real baseball conversation would be about a bullpen or a rotation, both is weird… and I am sure that is why they win. I have to remind myself that this is a numbers talk first, and a baseball talk second. As the main commenter here, I have gotten too hung up on the rotation more often than not. Two firemen clean up nearly every mess. Not many teams have ever deployed two elite RP, which is probably what floats them to the top of the list.
“Any real baseball conversation would be about a bullpen or a rotation, both is weird…”
Do you similarly take out all the part-time and fill-in players when counting the number of runs a team has scored this season? That’s what seems beyond weird to me, but that is what is eventually being done when a person says something like “best pitchers ever” with regards to merely 3 guys at the top of a starting rotation.
What “floats (the Indians) to the top” is not “two elite RP,” it is an entire pitching staff, including almost every single person they called up from the minors to fill in, that has outperformed to vastly outperformed Replacement Level expectations.
Of the 30 pitchers they used, only 1 of them under-performed Replacement Level – and his WAR was just -0.1
No team in history has ever seen that – and THAT is largely why the team “floats to the top” as “the best pitching staff ever;” they had no holes anywhere
As much as I hate to admit it, it does look like the Indians might have the best pitching staff of all-time. I used fWAR for these results.
Based on the fWAR by offenses other than theirs, I divided each pitching staff’s fWAR into the average offensive fWAR of that season. It’s not perfect but the 2017 Indians are by far have the best. The next 2 closest staffs pitched before 1885. So removing them the next 3 are the 96, 97 and 94 Braves. The top 10 rounds out with 03 and 02 Yankees, the 11 Phillies, then 98, 99 and 95 Braves. Here is the top 12 with data:
Season Team____________WAR____OWAR____PWAR/OWAR
2017____Indians_________31.7____18.72___1.69
1875____Dark_Blues______9.6_____5.78____1.66
1884____Beaneaters______17.1____10.67___1.60
1996____Braves__________29.5____18.76___1.57
1997____Braves__________28.5____18.51___1.54
1994____Braves__________20.3____13.24___1.53
2003____Yankees_________28.6____18.74___1.53
2002____Yankees_________28.2____18.64___1.51
2011____Phillies__________28.1____18.93___1.48
1998____Braves__________27______18.39___1.47
1999____Braves__________27.8____19.02___1.46
1995____Braves__________24.2____16.87___1.43