4 or 5 man rotations

[Syd Thrift]

Yeah, I am a bit skeptical of the "you can pitch all you want as long as you keep your mechanics" thing because in so many cases it's not terribly predictive. Adherents of this belief do seem to engage in some major confirmation bias - a guy goes down and they say, "hey look, he had bad mechanics", rather than say "this guy here, his mechanics are getting sloppy, he's about to get hurt".

That being said, the mechanism does make sense, as does the idea that tired pitchers will get sloppy and hurt themselves. I just don't think that pitching coaches and managers can be trusted to notice this before the injury actually happens. Hence, the variable we can control is pitch counts or, if you're not into pitch counts, the level of fatigue itself.

[PhillieFever]

Given the fact that medicine, especially sports medicine is so far advanced than it was even 25 years ago let alone 100 years ago, and the athletes themselves are in much better physical condition why is it do you think that the pitchers of today seemingly cannot handle the load that the pitchers of the past did? I for one do not buy the argument that old time pitchers "coasted" until key spots in the game and then turned it on, athletes are just not hardwired that way, they are super-competitive and "most" give 100% all of the time. If you think about it, it's one of the lone examples in sports where performance has decreased over time.

[Syd Thrift]

Originally Posted by PhillieFever

Given the fact that medicine, especially sports medicine is so far advanced than it was even 25 years ago let alone 100 years ago, and the athletes themselves are in much better physical condition why is it do you think that the pitchers of today seemingly cannot handle the load that the pitchers of the past did? I for one do not buy the argument that old time pitchers "coasted" until key spots in the game and then turned it on, athletes are just not hardwired that way, they are super-competitive and "most" give 100% all of the time. If you think about it, it's one of the lone examples in sports where performance has decreased over time.

Well, I think that's true depending on how far back you go. Christy Mathewson outright admitted as much in Pitching In A Pinch. That said, I agree with you that pitchers at least as far back as the 50s did not have the freedom to coast against weaker hitters, at least not from the few games I've seen from then. One difference between then and now is the advent of pitching from the stretch. In the 50s in particular there wasn't a lot of stealing bases so pitchers often didn't just throw from a full windup, they threw from big windmilly windups which often look comically exaggerated compared to the windups of today. IIRC this has not been shown to increase velocity but I'd be very surprised if using your shoulders and legs this much to generate your delivery doesn't reduce the strain on one's arms.

The other issue is that there are a lot of different kinds of pitches thrown now compared to back then. The slider, which may or may not have been similar to the nickel curve thrown by guys going back to the 1920s, was in its heyday in the 1970s and allegedly produces a lot more strain on the arm than most pitches do. The same goes for the splitter, thrown with a grip somewhat similar to a forkball but with the speed of a fastball. In the meantime, you really aren't seeing a lot of pitches that were a strain on the arm going away. I guess the screwball isn't thrown quite as much as it was back in the day (although it always was, I think, a novelty pitch as much as anything else). The knuckler isn't used as one pitch among many much at all anymore, although from what I gather that's about an easy a ball to throw as one exists.

Perhaps the lack of stamina is just an outgrowth of the continuing evolution of the game in other areas. Strikeout rates continue to climb from one generation to the next, as do BABIPs. The level of K/9 a pitcher has to achieve at a minimum to be of major league quality is a lot higher than it was even 30 years ago. Perhaps this requirement forces people to throw more pitches than they used to and also to throw harder on a pitch-to-pitch basis than they used to as well.

[RchW]

There are those who feel that current day pitchers throw too much between starts because of the extra day and that they would be better off throwing those pitches in games. One should also remember that back in the 4-man rotations pitchers got quite a few 4-day rests due to the schedule.

I know its only a snapshot but I looked at a pitcher in 1974 and one in 2010 and compared days rest for each start plus the total days rest in the season.

Pitcher 1 1974

41 Starts - note the total is 40 because the first start does not have days rest from ST.
26 on 3-days rest
10 on 4-days rest
3 on 5-days rest
1 on 6-days rest
14 of 41 or 34% of the starts were 4-days or more rest.

139 days rest 41 starts = 180 days.

Pitcher 2 2010
35 Starts - note the same data missing on the first start
28 on 4-days rest
4 on 5-days rest
2 on 6-days rest
6 of 35 or 17% of the starts were 5-days or more rest.

144 days rest 35 starts = 179 days.

Obviously no conclusion can be reached but it does seem to indicate that SP in current day baseball seem to be underused considering the availability/quality of relief pitching in the two eras.

[Layton is my Homeboy]

I just want to thank RonCo for the outstanding and thorough research in this area. This is really shedding some light on the how the game models pitcher fatigue, performance and injuries. Great work!

[Le Grande Orange]

Originally Posted by RchW

One should also remember that back in the 4-man rotations pitchers got quite a few 4-day rests due to the schedule.

I know its only a snapshot but I looked at a pitcher in 1974...

To amplify on the above, in 1974 there were a total of 80 doubleheaders scheduled in the original schedule. That's 8.2% of all games set to be played as part of doubleheaders. And this is how many there'd be in an OOTP historical league since it uses the original schedules.

A total of 140 doubleheaders were actually played during the season. That's 14.4% of all played games that were played as part of doubleheaders. (In other words, one game in seven during the 1974 season was part of a doubleheader.)

I don't have the 1974 season done, but I can say that in 1975 the postponement rate of originally scheduled games was 4%. That is, 4% of scheduled games—77 out of 1,944—were postponed for one reason or another. In 2010, the postponement rate was just 0.8%—20 games out of 2,430. (In fairness 2010 was unusually low in terms of the number of postponements. The year prior 37 games out of 2,430 were postponed, a 1.5% postponement rate.)

[Curve Ball Dave]

Originally Posted by Syd Thrift

In the 50s in particular there wasn't a lot of stealing bases so pitchers often didn't just throw from a full windup, they threw from big windmilly windups which often look comically exaggerated compared to the windups of today. IIRC this has not been shown to increase velocity but I'd be very surprised if using your shoulders and legs this much to generate your delivery doesn't reduce the strain on one's arms.

Three pitchers I can think of off the top of my head, although they did not pitch until after the 1950s, come to mind when one talks about using legs to drive the ball. Maybe you've heard of them: Nolan Ryan, Tom Seaver, and Mickey Lolich. If you take a look at how many innings they pitched year after year without developing any serious arm trouble should answer the question about the benefits of driving with the legs as opposed to standing upright and using the arm and shoulder. Why pitching coaches teach the "tall and fall" technique rather than the "drop and drive" is beyond me.

Also keep in mind that pitchers from a generation ago and earlier pitched to contact. Knowing they were expected to finish what they started, they tried to get batters out with the fewest amount of pitches and that meant forgetting about strikeouts (if you throw your best pitch with two strikes and the guy misses-great, if he grounds out it's just as well) and forced the batters to put the ball in play. Greg Maddux is the best example of a modern pitcher who pitched to contact.

As for the debate between the four man and five man rotation, it's all about how the pitchers are conditioned. MLB teams used four man rotations for decades. Pitchers were conditioned from rookie ball to the majors to pitch every fourth day. If you train your pitchers properly to pitch every fourth day, teach them to use their legs instead of just their arms, and get them in the habit of pitching to contact so they can get outs early in the count, 40 starts per year as opposed to 32 or 33 won't be an undue strain on their arms. It also saves the manager the burden of putting a replacement level pitcher on the mound every fifth game to get hammered and in the showers by the sixth inning.

[Biggio509]

To some extent I think part of the difference may be weight regimes. When atheletes first started using weights as part of conditioning most of the art was learned from bodybuilding which has a completely different purpose than sports condition. A lot of these guys lifted in the wrong way for what they were trying to do and stretching was not emphasized as much. In the older days stamina condition and stretching was emphasized over strength training. I think this has led to more injuries overall not just in pitching.

Today, the art is changing and you have many more trainers who are learning better stretching and lifting for sports. It still has not fully gotten into a science tailored to each sport but it is getting there. I remember reading about college level how college level weight trainers were more likely to have body builder type programs than football or baseball programs. The better trainers were former players who have been able to adapt lifting programs to the sport.

I think a large part of the problem with injuries is much less emphasis on stamina and more emphasis on strength building. Without proper stretching lifting results in tight muscles and more chance of injury. To some extent that is why Bagwell's career was shortened. The less emphasis on cardio and stamina, I believe, is why pitchers just can't seem to pitch as much as in the old days. You have stronger pitchers who on average pitch faster than the old days but fewer pitchers who have the stamina to complete games on a regular basis. The addition of lifting if done during the season also means more recovery time. So if say a pitcher lifts on his second day off, it takes 48 hours to recover maybe longer since the body is fatigued. I am not sure what in season lifting habits are put they can contribute.

[Syd Thrift]

Stretching is overrated but training to stamina might not be.

[Curtis]

Originally Posted by RchW

One should also remember that back in the 4-man rotations pitchers got quite a few 4-day rests due to the schedule.

I know its only a snapshot but I looked at a pitcher in 1974 and one in 2010 and compared days rest for each start plus the total days rest in the season.

Pitcher 1 1974

14 of 41 or 34% of the starts were 4-days or more rest.

Pitcher 2 2010

6 of 35 or 17% of the starts were 5-days or more rest.

Someone on this board was kind enough to dig up this information for one of Tom Seaver's seasons, and he was four or more days of rest more than half the time, back in the 'four man rotation' era.

Originally Posted by PhillieFever

I for one do not buy the argument that old time pitchers "coasted" until key spots in the game and then turned it on, athletes are just not hardwired that way, they are super-competitive and "most" give 100% all of the time.

In addition to Christie Matthewson, Bob Feller made this claim in his autobiography, at least for his pre-WWII seasons. In an interview on SNY a couple of years ago, so did Tom Seaver, though he claimed to go about it in reverse. He said that he 'bore down' on the 8/9 hitters and let up just a little on the 3/4 guys. His rationale was that you should be ashamed of yourself to let yourself get beaten by some guy who can't hit his weight. On the other hand, every team has one or two guys who will get their hits no matter what you do, so don't waste effort on them.

[RonCo]

Every era's different. No surprise there.

In the days of 154 games there were very few off-days and many more double headers. Both of these would favor 5 starters, and in fact I think the double headers are the origin of the spot starter (though I could be wrong there).

In the 70s and 80s there were actually fewer days off over a longer, 162-game schedule. One could argue that's a case for the 5-man rotation, but then double-headers were becoming more rare.

In today's game there are almost no double-headers and nearly twice the number of days off (162 games in 182 days = 20 days off, assuming zero double-headers). Removing double-headers and increasing actual days off are additional reasons to argue for 4-man rotations, so a decent argument can be made for this.

There are lots of other things you can take into account. Mound height, steroid era, big parks/little parks, whatever. But at the end of the day it all boils down to your opinion of whether throwing every 4 games or every 5 games is the root cause of greater injury risk, or if it's throwing more than 100 (or 110, or 120, or whatever) pitches in any single outing.

The data that I've seen makes a more compelling argument that starting a baseball game every 4-days or every 5-days makes limited or no difference, but that if a pitcher throws more than 120 pitches in any of those outings (regardless of how long he rested prior to it), he's under some greater risk of future injury. Throwing 140 pitches or more is a much greater risk. I'm sure that at the end of the day every pitcher's number is a little different, but as rules of thumb go, the 120/140 lines of demarcation seem practical.

[RonCo]

Just for kicks, I went to retrosheet and pulled the 1965 Dodgers schedule and compiled data on their 4-man starters -- Drysdale, Koufax, Osteen, and Podres:

	Drys	Koufax	Osteen	Podres
Starts	41	40	38	23
Days Rest				
3	1	2	5	0
4	29	29	19	7
5	10	9	10	6
6	0	0	2	1
7	0	0	0	3
8	0	0	1	2
9	0	0	1	0
10+	0	0	0	4

You can see they were brought back on even 3-days rest on occasion--though many of those were after short-stints the start before. The schedule resulted in about 25-30% of their starts being on 5 days' rest. There is, of course, no record of the number of pitches these guys threw in any particular game, but it's probably fair to say they experienced more than 120 on more than afew occasions.

Note that Podres had several 10-day+ layoffs, which I assume are due to nagging injuries. I admit to being blissfully ignorant on that front right now, but his usage pattern suggests as much.

Osteen's pattern is interesting. His longer spans appear to generally be related to shuffling of rotation to take advantage of the schedule and get Drysdale or Koufax more starts...or maybe due to match-ups. It doesn't appear he was really replaced (as if he were injured, or just pushed back).

The point being that a 4-man rotation results in several 5-day stints.

[RonCo]

The Dodgers used three other pitchers to start games that season. All of them came either:

(1) Immeditately before or during double-headers, or
(2) During Podres' down-time periods, or
(3) The last day of the season when the title was secure

[Le Grande Orange]

Originally Posted by RonCo

In the days of 154 games there were very few off-days and many more double headers.

I think you mean there were more off days, not fewer.

For much of the 154-game schedule period it was played over 168 days (24 weeks)—that's only 14 days off assuming no twin bills. But there were plenty of scheduled doubleheaders which added days off. In 1955, for example, there were an average of 16 scheduled doubleheaders per club (it ranged for a low of 9 for the Cardinals to a high of 23 for the White Sox). That means the actual number of off days scheduled was an average of 30: the 14 initial days plus the additional 16 days freed up by the doubleheaders.

In 1955 the postponement rate was 6.3%, so that added even more doubleheaders to the mix—a total of 188 were actually played compared to the 128 which were originally scheduled. That works out to an average of over 23 per club, which translates into 37 off days on average during the season. (Baltimore played the most doubleheaders: 34. It had been scheduled for 20. Milwaukee played the fewest: 15. It had been scheduled to play 12.)

The upshot of all this is that as the years progressed the schedules became more consistent and predictable. Doubleheaders (both scheduled and as played) dropped in number meaning off days were known with more certainty in advance and unplanned days off occurred less often. That increasing predictability and consistency must surely have had an impact on pitching usage and rotations.


To show the large differences between doubleheaders as scheduled and as played, the following graph should be helpful. The dark blue bars are the scheduled number of doubleheaders for that season while the light blue bars show the number that were actually played.

[Curtis]

Was data missing from 1981, or were there really no unscheduled doubleheaders that year? Seems like an anomaly.

[Curve Ball Dave]

Originally Posted by Curtis

Was data missing from 1981, or were there really no unscheduled doubleheaders that year? Seems like an anomaly.

Players went on strike in the middle of the season.

[Curtis]

Originally Posted by Curve Ball Dave

Players went on strike in the middle of the season.

Good catch, Dave. But, I was under the impression that most rainouts took place in the spring.

[Syd Thrift]

Well, what they did that year was they split the season into two parts: before the strike and after. Since this wasn't a planned thing, it doesn't surprise me that they didn't bother to reschedule rainouts later in the season.

[Le Grande Orange]

Originally Posted by Curtis

But, I was under the impression that most rainouts took place in the spring.

They can take place throughout the season, though to what extent April and May have more postponements than other months I can't say. I don't have detailed data to hand, but here's a few samples to consider.

Postponements by month

1930

April: 31
May: 38
June: 28
July: 7
August: 17
September: 6

Total: 127 (of these, 4 were postponements of make up games)


1940

April: 36
May: 56
June: 25
July: 12
August: 35
September: 17

Total: 181 (of these, 12 were postponements of make up games)


1953

April: 30
May: 43
June: 6
July: 7
August: 9
September: 9

Total: 104 (of these, 5 were postponements of make up games)


1975

April: 23
May: 17
June: 9
July: 7
August: 17
September: 17

Total: 90 (of these, 12 were postponements of make up games)


2009

April: 12
May: 8
June: 6
July: 3
August: 2
Sept/Oct: 6

Total: 37 (of these, none were postponements of make up games)


2010

April: 2
May: 7
June: 3
July: 1
August: 3
Sept/Oct: 5

Total: 21 (of these, 1 was a postponement of a make up game)

[Le Grande Orange]

Originally Posted by Curtis

Was data missing from 1981, or were there really no unscheduled doubleheaders that year? Seems like an anomaly.

The number of doubleheaders played was less than the number that were originally scheduled due to the players' strike. There were 47 doubleheaders played in the abbreviated 1981 schedule as compared to the 48 that were present in the original 162-game schedule.