When I first started watching baseball, what broadcasters, newspapermen and other fans drilled into my head was that you batted your fastest guy first, a guy who could hit to right field second, your best hitter third, your strongest guy fourth, pitching was 80 percent of the game, the most important stat for batters was AVG and for pitchers it was Wins.
Of course it all turned out to be false.
But it doesn’t mean that we are completely enlightened now and that today’s conventional wisdom is miles ahead of what it was in the 1970s. We don’t have to look very far to see areas where the majority of clubs are doing things in a less than ideal way.
Every team runs their bullpen to maximize the save chances for their closer, every team employs a LOOGY, seven man bullpens are standard, teams still bat their best hitter third, every team employs a five-man rotation.
Wait, wait – what?
There’s either very little or no proof that a five-man rotation is good for a club yet teams continue to use it, even as good starting pitching is scarcer than ever. Last year out of desperation, the Rockies went to a four-man rotation. While I’m not a Rockies fan, I thought this was fantastic news. Unfortunately, they have already announced plans to scrap it and move back to a five-man rotation for 2013.
But let’s examine the results of the Rockies’ experiment. Here are the stats for their starters when they went on three days rest (four-man rotation) and when they went on four days rest (five-man rotation):
3 days: .296/.356/.499 — .854 OPS, 5.49 ERA, 1.490 WHIP
4 days: .308/.371/.488 — .860 OPS, 5.57 ERA, 1.660 WHIP
Do you see one thing here from a quality standpoint that indicates that the pitchers did better with an extra day of rest? I would gladly trade 11 points of SLG for 15 fewer points of OBP.
It’s important to note that this does not “prove” that a four-man rotation is better. It’s just a single data point – the strongest one we have this century – that shows a four-man rotation does not provide worse results.
Let’s go back to the 1970s, the transition period when teams went from a four-man to a five-man rotation but it was still common to see pitchers go on three days of rest. Picking a year at random, I chose the bicentennial of 1976. Here is the same 3 days/4 days split we used above, this time for the entire NL:
3 days: .249/.313/.350 — .663, 3.35 ERA, 1.281 WHIP
4 days: .256/.313/.363 — .676, 3.47 ERA, 1.284 WHIP
Again, not much to choose from in regards to quality. Of course, if the quality is the same, a five-man rotation could be preferable if it helped prevent injuries. In 1976, 44 pitchers in the NL made 25 starts or more. In 2012, that number was 52. Of course there were only 12 teams in the NL in 1976, meaning an average of 3.7 per team. The average in 2012 was 3.3 per team.
If the five-man rotation kept pitchers healthier, we should have seen more pitchers on average making 25+ starts in 2012 but we found the exact opposite. Now, this could have been a quality issue as much as an injury issue – either way it’s not a ringing endorsement for the five-man rotation. Just looking at the Mets, we saw Dillon Gee, Mike Pelfrey and Johan Santana all go down with injuries last year.
How about the average length of an outing? Last year the Rockies had their starters on a very strict pitch count. But in 1976, the pitchers going on three days of rest averaged 6.5 IP per start while those going on four days averaged 6.6 IP per start.
Of course it should be pointed out that there were 371 starts on three days rest compared to 862 on four days rest in the NL in 1976. So, let’s look at an earlier year to get those numbers closer to even. Let’s try 1970:
3 days: .253/.319/.386 — .705 OPS, 3.90 ERA, 1.322 WHIP
4 days: .257/.322/.391 — .713 OPS, 4.07 ERA, 1.344 WHIP
There were 580 games on three days rest in the NL in 1970 compared to 729 on four days rest. Pitchers on three days averaged 6.7 IP while those on four days averaged 6.7 IP.
Only 39 pitchers made 25 or more starts in the NL in 1970, down from 44 in 1976. But it appears that may be an expansion/quality issue. This was just the second year in the league for Montreal and San Diego and those two only had five pitchers combined make 25 or more starts. Take away those second year expansion franchises and the rest of the NL averaged 3.4 per team, right in line with the 3.3 average of 2012.
So, if there is no improvement in the average quality of pitching, the average quantity of pitching and the average health of pitching with a five-man staff – wouldn’t clubs be better off employing a four-man rotation and having an extra hitter instead?
If you run a true four-man rotation and your top pitcher stays healthy, he should get 41 starts. No Mets pitcher has ever gotten 41 starts in a season. In fact, the record for starts for a Met is 36, done three times by Tom Seaver (‘70,’ 73, ’75) and once by Jack Fisher (’65). You could argue that the Mets never ran a four-man rotation for an entire year in franchise history.
The 1966 Dodgers had a rotation of Koufax, Drysdale, Osteen and Sutton. Those four combined to make 154 starts. That’s what a four-man rotation can look like. And this wasn’t a one-time thing, as Drysdale made 40+ starts in five consecutive seasons. The 1971 Orioles had a rotation of Palmer, Cuellar, McNally and Dobson. Those four combined for 142 starts.
I checked every year through 1978 and the most starts made by four members of the Mets rotation came in 1976, when Seaver, Jerry Koosman, Jon Matlack and Mickey Lolich combined for 131 starts. But it’s hard to say they used a four-man rotation when Craig Swan got 22 starts that year.
You might be thinking: This is all very interesting but what does it have to do with the 2013 Mets? With the team’s trouble affording an impact bat in the outfield, they should be open to trading a SP for an OF with the idea of going – at least some of the time – with a four-man rotation.
This seems crazy on the surface, especially with Gee and Santana coming back from injuries. But that thinking is still clinging to the current conventional wisdom that the extra day of rest keeps pitchers healthier and we just have not seen evidence of that. It’s similar to thinking that Frank Taveras was a good leadoff hitter because he stole a lot of bases and that his .301 OBP was meaningless.
Let’s monitor how pitchers are used in Spring Training. Do they start out on four days rest? I don’t know the answer but somehow it doesn’t feel right to me. My guess is that they start pitching on shorter rest and then switch to four days. If that’s the case, then what if they just never switched to four days and kept on their early Spring Training schedule?
It’s my opinion that the Rockies were on the right path last year in going to a four-man rotation and adhering to pitch counts. Of course, the Rockies’ pitch count was extreme. But there should be some way to have a sensible pitch count combined with a four-man rotation.
Of course it was done in pursuit of the club’s first no-hitter, but in 2013 there’s no reason for the Mets to ever allow a pitcher to throw 134 pitches in a game in which the team wins, 8-0. What’s the point of allowing a hurler to toss 118 pitches in a game the Mets win, 17-1? If a pitcher has a five-run lead after he’s qualified for a win, take him out before he reaches 100 pitches. As bad as the Mets’ pen was last year, it didn’t come close to allowing an average of five runs in four innings. And four innings would be the absolute most the pen would have to deliver in this scenario.
There’s no doubt that in the 1960s and 1970s that pitchers broke down for good earlier than they do here in the 21st Century. Part of today’s improved health can be explained by new surgical techniques that could have saved many of the pitchers from 40 years ago. And it’s my belief that sensible innings (pitch counts) limits have saved even more.
In 1964, Jim Bouton pitched 271.1 innings as a 25 year old. He never threw more than 151.1 innings in a year for the rest of his career. This century, no pitcher has reached Bouton’s innings total. The closest was 26-year-old Roy Halladay in 2003 when he threw 266 innings. And I’m sure it’s just a coincidence, but Halladay got hurt the following season and only pitched 133 innings.
Let’s look at Bouton’s game logs in 1964. There was a three-game stretch in July where Bouton turned in complete games – 27 IP – in games that the Yankees won by a combined 26-3 margin. There’s nine innings right there that Bouton absolutely did not need to pitch. He made 38 starts in 1964. It’s my opinion that he could have made the same number of starts and pitched 30 fewer innings without any harm done to the Yankees whatsoever.
Current wisdom is that the best way to keep pitchers healthy is to have them in a five-man rotation with a (roughly) 110-pitch limit per outing. But no one in MLB has ever tried having a four-man rotation with this same pitch limit. Back in 1964, Bouton threw 13 innings one game. While we don’t have exact pitch counts in this era, Tom Tango came up with a pitch count estimator and using that, we come up with 190 pitches for Bouton in this outing. In 2012 we lost our minds over a 134-pitch outing. Can you imagine the outcry today if a manager left his pitcher in for 190 pitches?
Pitchers are always going to get hurt, no matter how much we try to protect them. No one knows the optimal approach to handling pitchers. Instead, what’s developed is a CYA approach where every team essentially uses the exact same method in handling its pitching staff. It’s going to take a manager/GM with a lot of juice or a team desperate enough to say, “The emperor has no clothes!” for things to change in this regard.
The 2012 Rockies were in the second category. After two decades of using high draft picks on pitchers and overpaying in the free agent market and continually producing poor staffs – they finally decided to try something new. But the pull of conventional wisdom is so strong that even though there was no performance benefit, they decided to go back to a five-man staff.
It’s reminiscent of the experiment with the monkeys, the bananas and the water spray. Researchers placed five monkeys together in a room, hung a banana from the ceiling and placed a ladder where a monkey could climb and reach the food. Eventually a monkey would try to use the ladder but when he did, he – and the other four monkeys – each got sprayed with ice cold water.
Eventually the monkeys got the message that climbing the ladder was bad for everyone and no one did it. Then they swapped out one monkey and brought a new one into the group. The new monkey would immediately go to the ladder to try and get the banana – but the other four monkeys would kick the crap out of him. New monkey did not get sprayed with ice water but got the message that you don’t climb the ladder.
Once the new monkey got the message, they would swap out another monkey. And they did this until all five of the original monkeys were replaced. Even though no monkey currently in the group had been sprayed with ice water, it was a learned behavior that the ladder was bad and that you beat up any monkey who went to the ladder.
In MLB, the “learned behavior” is that four-man pitching staffs are bad. But we have no idea if this is true. My take is that high-pitch outings are bad, that it wasn’t the 38-40 starts per year that did pitchers in but rather the consistent 125-pitch outings along with the occasional 190-pitch ones that felled a bunch of hurlers in the 1960s and 1970s.
A trade of a SP might make the 2013 Mets a candidate to run a four-man staff. Because they have a respected GM and they are not expected to challenge for the division. So they actually fit both categories mentioned above in terms of what’s needed to buck conventional wisdom and try something new.
Maybe because I grew up in the decade where things like the Herb Washington experience happened makes me more open to trying new things. Perhaps it’s because I saw first hand how the conventional wisdom of one era turned out to be so completely wrong that I’m not married to the CW of this time frame. Regardless, I’m confident that some team in my lifetime will buck the five-man pitching staff trend and be proven right.
And wouldn’t it be wonderful if it turned out to be the Mets as trailblazers in this regard?
Interesting read Brian. A lot of things to digest.
It remains to be seen whether 4 man rotations could ever work again. My first instinct to why it won’t work is because of the increased performance of the hitters. In the Manny Parra post a couple of days ago, you listed some reasons how hitters have become much more advanced and how lineups are devoid of “easy outs” now. Baseball is a zero-sum game, so as hitters get better, it means pitchers must suffer.
Another objection to converting to 4 man rotation is the huge amount of risk invovled. Not only does the GM/Manager have to take the risk (of being ridiculed), but also the fans, and most importantly, the pitchers themselves. Since we don’t know the aftereffects of 4 man rotations, who knows what damage (in terms of development) could have been done to the Rockies pitchers who participated in the 4 man rotation experiement.
Similar to risk is the reward part. I’m not seeing the huge benefits of the rewards for the one that decides to switch to a 4 man rotation. The team that does it cannot “patent” the idea, so at most i could see a team getting an advantage over other teams for maybe 2? years before other teams will just copy them.
The payoff of the reward does not justify the risk in my opinion. If it were ever to happen, i would think that it would have to start in the independent leagues.
Thanks Name.
I’d like to throw out my reactions to some of the things you mentioned. In the first graph you make the zero-sum point. The one thing I would add to that is that guys who were normally fifth starters are now in the bullpen, which means that the worst bullpen pitcher is no longer being deployed. So, there’s some gain to pitchers here, too.
There’s certainly risk anytime you buck conventional wisdom, which is why I said it would have to come from a team with a strong GM/manager or a team with nothing to lose. If the 4-man rotation blows up on Sandy Alderson – he’s still Sandy Alderson. If last year’s bullpen fiasco didn’t sully his reputation… And if the 4-man doesn’t work for the Rockies, how will we tell compared to any other pitching staff they’ve trotted out there the entire time they’ve existed?
I have to disagree with your reward scenario. The purpose in running your team is to maximize the talent for your team — not to get caught up in what other teams may or may not do. A team should employ a 4-man rotation if it makes sense for them to do so with the talent on hand. If you have five great pitchers and no obvious bat to promote – then there’s not a whole lot of point in going to a 4-man staff. But it’s not too difficult to imagine a scenario where the Mets going to a 4-man means eight extra starts for RA Dickey and a platoon bat for Baxter or Nieuwenhuis that they may not otherwise have. That seems like a good thing to me.
An indy league would be a great place to try this. It might even be better in Low-A where the major league club could better control how it was instituted.
The one major consideration i think you are leaving out is the human aspect. For the team, yes, it may be worth the try out a 4-man rotation if the team is going no where, but players are not like chess pieces that you can command at will. I would gather most would not want to risk their career to try it out. They are used to 5-man rotations, know that it works, know the relative payout/rewards. Winning is great, but making money is(and should be) the most important thing to them.
That’s why i brought up the independent leagues is because frankly the pitchers/owners/GM’s have very little to lose in trying it out unlike the MLB where millions of dollars are at stake.
You’re right – I am leaving this out. Because the tail shouldn’t wag the dog. I don’t want anything bad to happen to anyone who plays in MLB. But injuries are an occupational hazard. Every batter steps into the plate knowing that his career can end in one pitch. Every pitcher knows that one line drive back thru the box can mean the end.
But they still go out and pitch and hit. And I think they’ll still go out and pitch every four days instead of every five days. Because they’re competitors and they want to go out and give their best.
And you’re still operating under the assumption that there’s any risk here at all. I’m not saying that there isn’t – I’m just saying it’s not a given. And when 60% of the Mets planned rotation went down with injuries under the “safety” of the five-man rotation…
I just realized we are on two different planes. When I’m using the term “risk”, i don’t mean “injury risk” but more of “uncertaintity”.
Why did RA decide that he didn’t want to pitch every 4th day when the idea was brought up? He probably could have done it, but he’s been pitching every 5th day for a long time and it’s been working, so why should he risk it?
The reply “for the benefit of the team” may be enough to convince fringe players who are fighting to stay in the MLB, but it won’t be able to convince well-established players.
So i guess the pitchers themselves have to decide to want it or else it’s never going to happen.
The five man didn’t become such a necessity till expansion. If you think about it, the influx of pitchers – whether they were flamethrowers or mediocre – were valuable commodities due to the parity in baseball. However, I don’t think that four man rotations are evil. In fact, I think a problem with pitching today is the babying of them (see: Strasburg, Stephen). But I could see a team like the Texas Ranger reinstitute the four man rotation if given the chance. However, good luck with it. Pitchers need a job, so five man rotations aren’t just necessary they are necessary evils.
I wouldn’t mind seeing the Mets go to a 4 man rotation with the “5th” starter used as long relief. Their innings would probably come out the same. Nolan Ryan once said pitchers would get stronger if they could pitch more often. Isn’t that the system they employ now in Texas? Build arm strength first. In comparison I seem to recall back in the 70’s pitchers pitching into the 8th inning was not unheard of.No set-up man for the 8th inning. But to me the biggest difference is pitchers knew how to pitch before they where brought up to the major league level where as today teams are looking for a pitcher to blow their fastball by every hitter. I’ll take a Jonathan Niese type pitcher anytime over Bobby Parnell prototype.
Very interesting and well presented. A four man rotation makes a lot of sense. I always was annoyed when the Mets had an off day and didn’t skip the 5th starter so that a Niese or Dickey could get a few more starts. It also plays well for me with the Parra post I made the other day. I agree, relief pitchers should pitch more than one inning, and this idea would embrace that concept. One more bench player is the icing on the cake. The additional benefit is if the team makes the playoffs, the pitching staff is already in the 4 man routine. Thanks for the read but I doubt we will ever see it happen.
If I were an agent for a starting pitcher, and I knew that a team planned to institute a 4-man rotation, I’d seek more money for my client because of an increase in his workload and the danger of possibly shortening his career. Agents play a much larger role than they did back when 4-men rotations were last in vogue. None of this means that I disagree with Brian’s premise, but changes often have unintended consequences.
“…you batted your fastest guy first, a guy who could hit to right field second, your best hitter third, your strongest guy fourth, pitching was 80 percent of the game, the most important stat for batters was AVG and for pitchers it was Wins.
Of course it all turned out to be false.”
“Of course” means acceptance without question. Can you point me to some articles, links, to help me understand why these are all false? I have been out of things for quite some time and were raised on the same stats as you. Thanks, Peter
And here’s the tipsheet:
1. This is the most OBP-centric spot in the lineup. Your hitter here might very well be your best hitter, IF his best attribute is his OBP. A hitter with a .425 OBP and a .500 SLG would fit in here well, provided that there’s not a better OBP threat elsewhere on the roster.
2. The 2-hitter should be the lineup’s most balanced hitter, a good combination of OBP and SLG.
3. This was the biggest surprise: the 3 hitter should be the player that doesn’t fit into any of the other spots. Every other spot has some significance, but if I were building a lineup, I would just put the leftover player in the 3 hole. This seemed very counterintuitive to me when I first heard it, but David Pinto noted, “Part of what it’s telling us is that you need to spread out your easy outs.” I still struggled to get this, but I’m starting to, now. Marc said something to the effect of “the worst players have to go somewhere.” I guess this is really it; the other spots just have greater needs. If you can get a good hitter here, it means that your lineup is very deep.
4. This is the bopper. This guy’s best attribute should be his power, with OBP being of secondary importance. He should be the foil to the leadoff hitter, in a way; both players could be similar if they’re both very complete.
5. Picking the 5 hitter is simple: it’s the second choice for the two slot.
6. The 6 hitter shows the biggest difference between SLG and OBP on the roster. This is because you’re going to want to have guys driving in the leftovers. The 6 hitter is the most exclusively power-dependent hitter of the bunch. His OBP is VERY unimportant.
*****
“6Since you brought it up, the old saying that baseball is 90 percent pitching actually arose in this way – in spring of 1906 a pitcher named Addie (Joss) made the comment that baseball is 90 percent pitching and was immediately assailed by experts on all sides for saying this and apologized for saying it and explained that he meant saying pitching was 90 percent of the game from the defensive standpoint and later tried to back off of that. John McGraw gave a very analysis at that time in which he said that baseball is one-third pitching, one-sixth fielding and one-half batting and baserunning – very thoughtful analysis. Over time, this was corrupted to where people were saying ‘some say it was John McGraw said was 90 percent pitching,’ when what he clearly, definitively said wasn’t true. The one one-third for pitching and one-sixth for defense for that time was very reason able. If I made an estimate of baseball in 1906, I’d say the same thing. Baseball in 2010, pitching is more than that and defense is smaller. But that’s not to say it’s not important. It’s still a big deal, it’s not as important as it was 100 years ago.
“No pitcher allows home runs as often as Dale Murphy hits home runs. No pitcher allows home runs as seldom as Bob Dernier hits home runs. . .No pitcher allows hits as often as Wade Boggs gets hits. No pitcher, not even Dwight Gooden, allows hits as infrequently as Steve Lake will get a hit. . .No pitcher strikes out hitters as often as Rob Deer strikes out. No pitcher strikes out hitters as rarely as Bill Buckner strikes out.
“This is true of every significant area of performance, including those things like walks and hit batsmen, which are usually considered to be controlled by the pitcher. And what does that mean? It means that in order to create a working model or simulation of a baseball game, you must allow the hitters to be the dominant, shaping force in the game. And if baseball were 75% pitching, one would not expect that to be true.”
*****
Correl RMSE
Batting Average .828 39.52
On-base Percentage .866 34.16
Slugging Percentage .890 31.56
OPS has a higher correlation and OBP*1.8 + SLG has an even higher one.
*****
Wins are an attempt to attribute a team outcome to an individual, with rules governing who gets one almost arbitrary. Let’s say Dickey has 8 IP and allows 0 ER and hands over a 1-0 lead to Francisco in the top of the ninth. Frank-Frank allows two runs but the Mets come back and score two in the bottom of the ninth. Francisco gets the win. And that’s not some far-fetched scenario — things of that type happen every week. And I bet you can think of a half dozen other situations where a pitcher gets credited with a win without really “deserving” one.
Brian,
thank you.
Since we began with the same belief, when did you begin the change, and what caused it? Peter