Let me be clear, I am not a huge stats guy. I am also not a pitcher. I have not, since maybe in the Little League minors, or playing stickball with my friend Barry in high school, pitched a ball to a batter. I have pitched batting practice to little leaguers, and tons of softball, but that of course is a bigger ball, underhand, and slow. Clearly l cannot possibly know what is involved with preparing to throw, actually throwing, and trying to recuperate from throwing 100-plus pitches every five days.
That has not stopped many of my friends and I from discussing, frequently, how it has evolved, or perhaps devolved, to the point where the complete game has all but disappeared and pitchers’ lives are controlled by the “pitch count.” Of course, it is not that the concept and value of a complete game is not appreciated and aspired to; it is just that it can rarely be accomplished because of the obsessive preoccupation with limiting the number of times a pitcher is permitted to see the catcher’s sign and let it fly.
A few decades ago, it was simply taken as a given that pitchers would complete games or, at least go deep into them. Whereas teams now carry 12-13 pitchers on their active roster, back then my memory, if not the data, tells me that 10-11 was more the norm, certainly in the years of “four-man” rotations. I also made some assumptions that have been disproved. I thought that every team had four pitchers who, barring injury or ineffectiveness, or fewer starts because they were skipped as lower rotation guys because of off-days, would start 38-40 games, certainly so after the season expanded to 162 games in 1961 before the era of the five-man rotation. That is not correct; while a handful of pitchers each year may have made between 38 and 40 starts, occasionally up to 41, most regular, reliable hurlers took the ball for the first pitch 30-33 times. A quick look at the stats showed that teams averaged about 4 ½ starters in a season. Interestingly, 40-50 years ago, well within my time of following the game, many rotation guys also pitched 2-7 times in relief. Except for MadBum in the World Series last year, when will you see that again?
But the data does prove that starters do not regularly complete games or throw the number of innings they did decades ago. So, to get back to that initial question, how did the complete game seemingly vanish? Most front-line starters are starting maybe four times fewer than in the past. They have access to top-notch nutrition, training equipment and medical care. Shouldn’t they be at least as strong as they were in the days of yore, capable of pitching as much as their forerunners? If they don’t accrue as many innings because of fewer starts, shouldn’t they be strong enough to pitch deeper into the game?
The flip side, of course, is what has caused them to NOT be able to do this. What occurred to make the hundred-pitch limit almost a religious requirement to be followed? Did teams decide, by just a sense, or evidence, that pitching less frequently would result in fewer injuries (seemingly, it hasn’t)? Has the development of relief roles, set-up men, closers, one-out lefty specialists, forced this evolution, or did these roles evolve because starters stopped being able to go the distance? Why are so many pitchers having Tommy John surgery by the time they are 24-years-old, or sooner? Do I just not remember so many pitchers going down because they didn’t have a name for this procedure? What on earth is an oblique? I had never heard of it until Noah Lowery was diagnosed with this injury about a decade ago, and now everyone is getting one. Tommy John and obliques are becoming as trendy as wearing pants down to the feet and having caps on at a slight angle.
Is the focus on playing baseball, particularly pitching, causing this injury epidemic? Is it, as has been suggested, that throwing certain breaking pitches at too young an age can cause these problems? Is the pitch count limit simply now seen as a way to protect pitchers from overthrowing, even though 40 years ago they could throw far more pitches and innings? Are pitchers being babied now, or are they somehow more fragile? Has the economics, huge salaries commanded, become a factor?
Let’s take a look at a comparison of pitching accomplishments over the last six or so decades. This is not a thorough statistical analysis, just a rudimentary, first effort look at some basic pitching categories: games started, complete games, and innings pitched. I considered looking at shutouts, but of course this is already affected by the number of pitches permitted, so as complete games have decreased, so obviously have shutouts. Innings pitched is only somewhat of an indicator, since a pitcher, under current day philosophy, who averages 12 pitches an inning might be able to complete a game, where as a hurler who throws just three more per inning, will in this scenario last just 7 innings. So, again, there are various factors at play here.
The data presented below is taken from baseball-reference.com. I acknowledge that as my eyes went up and down through endless columns, I imagine I mis-counted. Still, I am well within the ball park, so to speak, with my stated figures within the usual acknowledged acceptable deviation rate of 2-3. I started with the 1950 season, and then looked at the stats every ten years thereafter, tossing in 2014 as the most recent indicator. To compare pitching statistics, I arbitrarily set standards of 34 games started, at least ten complete games, and innings pitched thresholds of 220 and 250 innings.
In 1950, twelve pitchers had at least 34 starts, with three topping out at 39. Overall, ten starters had 20 complete games. Twenty. And remember how few MLB teams there were, so how many fewer pitchers there were. Vern Bickford had 27 complete games, and his Boston Braves teammates, Warren Spahn and Johnny Sain, 25 each (1950 was obviously not the year of the famous phrase, Spahn and Sain and pray for rain). Many pitchers had more than 250 innings pitched, with Bickford leading the pack at 311.
In 1960, the last year of the 154-game season and still four-man rotations (even though as I noted above it really was not that simple an arithmetic problem), the numbers dropped. There was the same number and range of starts, but far fewer complete games (only 23 topped 10, with Lew Burdette, Spahn and Vern Law throwing 18) and fewer innings pitched. Larry Jackson led the majors with 282 innings, and only 14 threw more than 250. Burdette and Spahn completed 18 of their 32 and 33 starts, respectively. Keep in mind that all the pitching leaders also threw in relief, with Spahn stepping in as a reliever an amazing seven times.
By 1970 there was a 162-game season, so it would be reasonable to think that these various stats would rise. Rise they did, as pitchers demonstrated that they were capable of more starts, more complete games, and more innings pitched. Forty-one starters pitched at least 34 games. Six threw the first pitch 40-41 times. Curiously, of this group, Dobson and Hunter pitched for Oakland, with Cuellar and McNally throwing for Baltimore, with Palmer right behind at 39. That would seem to indicate something of organizational philosophy, or just reflect scouting and acquisition of some damned good starting pitchers. Twenty-eight pitchers completed at least 10 games, but only three–Ferguson Jenkins (24), Bob Gibson (23) and Mike Cuellar (21) exceeded 20. The most startling stat, though, is the number of starters who threw not just 220 innings or more, but exceeded 250. Twenty-eight starting pitchers topped this level, with four topping 300 innings, led by Garylord Perry with 328. Remember, this is in the era of nominally, if not in fact, four-man rotations, so the spike in these numbers would make sense.
1980 is the first year examined here when teams were using five-man rotations, with teams having started to migrate in that direction in the mid-1970s. The number of pitchers starting 34 games or more dropped to 24, with only four starting 38 games. There was a small decrease in the number of pitchers completing at least 10 games, down to 23, with the three leaders, Lankford (28), Norris (24) and Keough (20) interestingly all pitching for the Oakland A’s. Thirty-five pitchers tossed at least 220 innings, and 16 topped 250, with Steve Carlton having the only 300-plus inning season. This was a large drop from the previous decade. Fewer innings would naturally result from fewer starts because of five-man rotations, but with more rest shouldn’t there have been more complete games?
Up to this point, almost each decade I looked at reflected a change in factors. While 1950 and 1960 were similar, 1970 now had the 162-game season (more starts). Then, in 1980, there was now a five-man rotation (fewer starts). So, variations, could be expected and explained. What is hard to fathom, and is the basic point of this article, is what happens from 1980 on, because none of these specific variables, length of season or size of the rotation, changes. Except of course the weird experiment of a few years ago by Colorado, limiting every pitcher to 75 pitches.
From this point on, a downward trend in these stats continued and then more or less flattened out. In 1990 only 13 pitchers started at least 34 games, and only 3 completed more than 10 games, led by Ramon Martinez with 12. Twenty pitchers exceeded 220 innings pitched, but only one, Dave Stewart, with 267, exceeded 250. In 2000, even fewer pitchers exceeded 220 with again only one, Jon Lieber, over 250 (251). Most significantly, no pitcher threw as many as ten complete games; David Wells had 9. Same results in 2010, with the number exceeding 220, and Roy Halladay leading with 250 innings pitched and 9 complete games. By 2014, only eight pitchers completed 220 or more innings, led by David Price with 248. Clayton Kershaw led the league with 6 complete games.
So, what does this all tell us? The point of this article is the demise of the complete game as a significant statistic, caused in large part by the limiting pitch count. An equally interesting question is why has this occurred? Innings and complete games are tabulated at the end of the season, but the major factor in these totals is that which is just counted on a game-by-game basis, the pitch count. Of course there is a direct connection between pitch count, innings pitched and complete games, but there are subtle caveats and factors which can go undetected and unmeasured. Too many walks, or even foul balls, will increase pitch totals and affect how deep pitchers can go in a game, in turn limiting innings pitched and complete games (ironically large numbers of strikeouts tend to raise pitch totals).
Poor run support might result in pitchers coming out early, even with a low pitch count, as managers will send up a pinch hitter to try to manufacture runs, also reducing the number of innings pitched. Of course, this is the case primarily in the National League. Perhaps young pitchers are stopped from fully taxing their arms too early in their careers, especially if they are pitching in the majors for the first time. So, they may have their turn skipped (Ask Matt Harvey about this) in addition to the pitch count being scrutinized during the game.
If you look at the box scores, you can see examples flourishing of pitchers who have pitched well, given up few runs, thrown less than 100 pitches, and are taken out. So then the pitching threshold becomes even less than 100 when there does not seem to be a reason for it. In the past few weeks alone, Colby Lewis (97), Ruby De La Rosa (93), Jerard Eickhoff (75), Charlie Morton (92), Tyson Ross (91), Adam Morgan (92), J.A.Happ (83), Joe Kelly (100) , Nathan Karns (84), Patrick Corbin (91), Danny Duffy (94), James Shields (95), Jacob deGrom (100), and Michael Wacha (93) were among the pitchers who left the game with 100 pitches or less, a lead, sometimes quite comfortable, and no apparent indications (what you can read from a box score), of starting to “lose it”.
Gone are the days of a Sandy Koufax having 27 complete games out of 41 starts, or Juan Marichal (30-38, with far more complete games than wins). In the 70s, Tom Seaver once completed 30 of 41, and Jim Palmer 25-38. By the 80s, Fernando Valenzuela completed 20 games once, Dwight Gooden at best finished 16, and Roger Clemens 18. Into the 90s, Randy Johnson completed at least ten games just three times, Greg Maddux twice, and Pedro Martinez only once. John Smoltz never completed ten games in a season. And the recent leaders in this category, Justin Verlander, King Felix Hernandex, Kershaw each completed just six. It took six seasons for Max Scherzer to complete his first game.
So, I repeat what does this tell us? The data speaks for itself, although the reasons are as cloistered as, well, you fill in the analogy. You will have to ask baseball GMs, managers, pitching coaches, trainers and the pitchers themselves. I wish I had a dollar for every time I watched a pitcher roll his eyes with a look of disgust on his face when pulled from the game with a lead in the 6th inning after throwing 96 pitches.