I wish the Rockies' rotation all the luck in the world in their effort to throw a lot of strikes and not get lit up like a movie marquee.

if it reduces the number of pitches thrown by the Rockies by almost 25 percent over the course of a season.

How the hell can you do that? A 4-man gives you 40 starts a year and even at 5 IP/start, that's 200 innings when healthy. The 5-man gives you 32 and even at 6.5 IP/start, that's 208 innings.

Or ... 32 starts at 100 pitches is 3200 pitches; the same as 40 starts at 80 pitches each. To achieve that sort of reduction, you'd have to limit your guys to 60 pitches per start -- does he really think it's possible to average 12 pitches per inning? In Coors? Or are they seriously considering the 3-4 innings per start model?

Or ... is he arguing that it's pitches/start more than total pitches? I think Rany and/or Voros used to toss that idea out there.

Or ... is he arguing that it's pitches/start more than total pitches?

As with you, I'm not entirely sure what O'Dowd is arguing, but from a biomechanics perspective, the ratio would likely be the more important factor. For example, we've found in our concussion research that the average weekly loading over a window of multiple preceding weeks better predicts measures of cognitive function than the to-date accumulated hits or averages over shorter-term windows, such as only the previous week's loading. So, it wouldn't surprise me if there are actually useful data out there that can be used to demonstrate that sustained peaks of loading are more detrimental to the arm. Of course, it may require Clinton-esque definitions of "detrimental", "sustained" and "peak" as the Rockies would be looking at terribly small sample sizes (what, 5? 8? 10? pitchers having reached those IP and years numbers?).

O’Dowd pointed out that a pitcher’s E.R.A. typically rises half a point every time he cycles through an opponent’s lineup during a game.

Isn't this the main reason for doing this? The optimal strategy for an NL team with crappy pitching would be to minimize:

a) The amount of at-bats that poor pitchers receive
b) The number of times poor pitchers go through a batting order.

“We have found that every starter who has pitched here for 185 to 200 innings for three consecutive years over the lifetime of this franchise has broken down with a significant injury,” he said.

THIS JUST IN: Pitchers get hurt a lot. Film at 11.

O’Dowd pointed out that a pitcher’s E.R.A. typically rises half a point every time he cycles through an opponent’s lineup during a game.

More teams need to realize this and act like they realize it.

O'Dowd and Vaux get it right.

How many starters have the Rockies HAD pitch 185-200 innings for three straight years? My guess was just Ubaldo Jimenez, Aaron Cook, Jeff Francis, Jason Jennings. The last two of those are right, Cook came close and Jimenez was traded in the middle of his third such year. Jennings's major injury made him start only 20 games in 2005, and then he came back to start 32 the next year.

Upon further inspection I'm pretty sure the entire list consists of Francis, Jennings, and Pedro Astacio.

THIS JUST IN: Pitchers get hurt a lot. Film at 11.

Sure, but that doesn't mean it's not worth occasionally trying something mildly crazy, in case you can make a difference. What the heck, right?

[7] For single seasons, Playing for the COL, (requiring At least 185 Innings Pitched), sorted by name and year

                                         
                  IP Year Age  G GS
Ubaldo Jimenez 198.2 2008  24 34 34
Ubaldo Jimenez 218.0 2009  25 33 33
Ubaldo Jimenez 221.2 2010  26 33 33

  Jeff Francis 183.2 2005  24 33 33
  Jeff Francis 199.0 2006  25 32 32
  Jeff Francis 215.1 2007  26 34 34

 Pedro Astacio 209.1 1998  29 35 34
 Pedro Astacio 232.0 1999  30 34 34
 Pedro Astacio 196.1 2000  31 32 32

If you expand the list to ERA qualifiers (min 162 IP), add the following pitchers:

                                       
                  IP Year Age  G GS
  Jason Hammel 176.2 2009  26 34 30 
  Jason Hammel 177.2 2010  27 30 30   
  Jason Hammel 170.1 2011  28 32 27 

Jason Jennings 185.1 2002  23 32 32  
Jason Jennings 181.1 2003  24 32 32  
Jason Jennings 201.0 2004  25 33 33 

    Aaron Cook 212.2 2006  27 32 32 
    Aaron Cook 166.0 2007  28 25 25 
    Aaron Cook 211.1 2008  29 32 32 

THIS JUST IN: Pitchers get hurt a lot. Film at 11.

But every one? I guess we would need to define "significant" and set a time frame (ie: guys who leave and get TJ surgery 8 season later shouldn't count), but still. I can't even remember the last time the Giants had a major injury in their rotation, for instance. I wonder if there is a measurable connection between injury rates and environmental factors like home park effects and weather. I wouldn't be at all surprised if playing in an extreme hitter's park and/or in hot weather produced more injuries.

I wouldn't be at all surprised if playing in an extreme hitter's park and/or in hot weather produced more injuries.

Hitters' park maybe, but Atlanta and St Louis are plenty hot and they've had durable pitchers in the past that lasted a long time.

I wouldn't be at all surprised if playing in an extreme hitter's park and/or in hot weather produced more injuries.

I would think cold weather would be more likely to cause pitching injuries than hot. Stiff muscles, harder to grip the ball, worse circulation, more throwing to get warm.

Playing in a hitter's park drives up pitch counts, and especially pitch counts per inning, so I too would not be surprised if that resulted in more injuries. If someone's going to try the 13 Relief Pitchers method of running a pitching staff, the Rockies would be the team for it, probably.

I'm not faulting O'Dowd for thinking that way. But this is really about maximizing performance more than minimizing injuries. For all the effort I'm still yet to see any solid evidence that pitchers in 2012 are getting hurt at a significantly different rate from pitchers in 1935. We have always talked a lot about keeping pitchers healthy, but I think the driving force behind the long-term historical trend toward spreading innings out among more and more pitchers is about maximizing performance.

That's why if you're the Rockies and you play in a hitter's park and you have no particularly good pitchers, it might make sense to run the 13 Relief Pitchers staff. Not so much if you're the Brewers and you have a couple of very good pitchers. Greinke or Gallardo the third time through the lineup is still better than most relievers the first time through. You don't really want to take more innings away from those guys than you have to; but it's hard to run the 13 Relief Pitchers staff only half the time, because it will rely on regular and well timed workloads for everyone.

Since 1993 here are the SP for the Mets to have 3 consecutive years of 185 IP:

Al Leiter
Bobby Jones
Tom Glavine

Mike Pelfrey did it three times in four years (184.1 the other year) and he got hurt this year.

I can't even remember the last time the Giants had a major injury in their rotation, for instance.

Noah Lowry, you're a forgotten man.

For all the effort I'm still yet to see any solid evidence that pitchers in 2012 are getting hurt at a significantly different rate from pitchers in 1935. We have always talked a lot about keeping pitchers healthy, but I think the driving force behind the long-term historical trend toward spreading innings out among more and more pitchers is about maximizing performance.

Agreed that modern protocols haven't resulted in a detectable reduction in injury rates. (Modern diagnostics and medicine have had a tremendous effect on injury recovery.) But I think it's fairly clear that the moves to the fixed 5-man rotation in the 1970s/80s, and particularly the imposition of pitch count routines and the reduction in workloads of relief aces in the 1980s/90s, were motivated by a quest for injury reduction, but have remained orthodox because they do have a posititve impact on staff effectiveness.

Since 1993 here are the SP for the Phillies to have 3 consecutive years of 185 IP:

Cole Hamels

If you lower it to 180 IP, Curt Schilling too.

Steve: I still think the five man rotation is itself a losing concept for any team with two good pitchers. Taking eight starts each (around 50-55 innings) away from those two good pitchers and giving all of those starts/innings to an assorted cast of Quad-A guys (which is what comprises the fifth starter role on nearly all teams) is yielding a ton of value, and I very much doubt they're getting that value back in the form of better performance from the 1 through 4 starters.

If it's not keeping the pitchers healthier--and I don't think there's any evidence it is--it's strictly a performance question. I think teams could do a lot more to monitor fatigue factors in game (including, but not limited to, pitch count per inning, elapsed time, weather and high-stress situations) and direct more innings to their best pitchers. I wrote an article about this years ago, it's probably been five years now, that I still mostly stand by.

More teams need to realize this and act like they realize it.

What do you want them to do? The same thing the Rockies are doing?

... the reduction in workloads ... in the 1980s/90s ..., but have remained orthodox because they do have a posititve impact on staff effectiveness.

There have been a number of occasions in baseball history in which reducing individual workloads and spreading the pitching effort out has accompanied improved staff effectiveness. Note the success of the Cubs and Pirates in the 19-oughts. Those teams spread the pitching load to more pitchers then most of their rivals at the time, and were very good at it. (Good defense helped.) And from the 1930's, there's the Ted Lyons old pitcher plan: let the old guy start once a week or so. Lyons is the headline, but there were a number of other older pitchers turning in effective 140 inning seasons as starters.

So now what's bothering me is the one period in which the trends were running most starkly in the other direction: the late 60's and early 70's. We at the Hall of Merit have been working our way through that time in our "Most Meritorious Player" project. 300 inning seasons were commonplace, and pitchers' seasons of great value were commonplace, with the great value coming at least partly from the great bulk: Bob Gibson, Gaylord Perry, Steve Carlton, Tom Seaver, Wilbur Wood, and so on. And for the most part, these pitchers - at least the successful ones - didn't really break down under the load. Sure, Vida Blue never came close to matching his great early season, but even he had a long career. And at the same time, the top relief pitchers were carrying very heavy loads. Right now, we're on 1974, which features Mike Marshall's 200+ inning season. That's an outlier in usage, but there have been a lot of great 110-130 inning seasons from relievers. (The most valuable such season that I've seen so far: John Hiller's 1973.)

So here's the question: did that early-70's strategy of leaning so heavily on the #1 starter and the relief ace pay off? Or was it counterproductive in terms of whole-staff productivity? Perry in 1974 had an RA+ of 150 in 322 innings. If he'd come out of every game after 110 pitches, what would his RA+ have been? What would Marshall/Lyle/Hiller/Forster have done in a modern closer's role, and how would the pitchers you've never heard of performed in the suddenly vacated 6th/7th/8th inning roles?

Steve: I still think the five man rotation is itself a losing concept for any team with two good pitchers.

I don't disagree at all, and have written much to this effect. However, in much of the response I've received, I have been persuaded by the arguments about practicality as opposed to theory.

Which is to say: it's the rare team that KNOWS it has, say, two starters who are much better than their others, or more to the point, two starters who will prove to be much better than their other starters over the course of the full season to come. Practically speaking, the certainty around which pitchers will be better than the others each year is pretty weak. In retrospect it may be obvious, but it isn't nearly so clear before the fact.

Therefore, just giving all your regular starters the same number of turns and managing their workload within each game is more workable.

Anyway, that's the practical argument for the 5-man rotation, and while it's far less elegant than a nuanced solution, it tends to work out decently well for most teams in most seasons.

And the other practicality is that no pitcher is developed to make more than 32-35 starts in a season any more, anyway. To impose a 4-man (or at least 4-day) would mean increasing the workload of your most precious pitching assets in mid-career, and the downside of that is quite obvious ... nobody wants to be the manager who "shredded" Strasburg's (or whoever's) arm, as everyone would wildly scream. Public opinion has a force in this.

I think the best way to go about it would be to develop your best organizational starters to handle starts on 3 days rest coming up through the minors. But nobody does that. (Maybe the Rockies will start doing it?)

Cubs:
Steve Trachsel (1995-99)
Jon Lieber (1999-2001)
Carlos Zambrano (2003-2008)
Greg Maddux (2004-2006; in 2006, only 136 IP as a Cub, but another 73 with LAD)
Ryan Dempster (2008-2011)

Plus a couple of near-misses (most notably, Matt Clement throwing 181 IP after two years of 185+). Amazing to realize that 2002 was the only year since the 1994 strike when the Cubs were not in the midst of at least on such streak. Dempster might not make it this year because of his DL time; Garza is on pace for a second consecutive year as a Cub, and his 4th consecutive year (with the year before that being 184.2 IP). When I realize how uncommon such sustained streaks actually are, I suddenly realize I don't want the Cubs to keep and sign Garza; he would seem to be a virtual certainty to miss real time at some point over the next several years.

What do you want them to do? The same thing the Rockies are doing?

Pinch hit earlier.

Greg Maddux threw 194+ IP in 20 consecutive years, including both strike-shortened years (1994 and 1995). He was the only pitcher over that 185 IP threshold in 1994.