The question of whether or not pitchers pitch to the score is one that never seems to go away. In recent years it has frequently cropped up in discussions and debates regarding the Hall of Fame credentials of Jack Morris who many believe pitched to the score to such an extent that his ERA understates how good a pitcher he really was. The subject has recently been fodder for numerous tweets by ESPN writer Buster Olney.
In general, proponents of both sides of the argument make the mistake of seeing the question in more or less absolute terms. It is less dramatic, but more accurate to conclude that pitchers pitch to the score, but only sometimes. Many pitchers only have limited opportunity to pitch to the score. Moreover, it should not be overlooked that the score influences what managers do far more than what pitchers do.
There are specific situations where the score will change what pitchers do. With a big lead, pitchers may be less afraid to challenge hitters, a strategy which often results in more strikeouts and hits, but fewer walks. Similarly, in close games pitchers may be reluctant to throw the ball down the middle of the plate, a strategy which would result in fewer home runs, but possibly more walks. Even if we accept this at face value and assume that all pitchers have a sufficient breadth of approaches and sufficient command to implement these strategies, the impact of these approaches over the course of a season or even a career is probably somewhat limited.
Pitching to the score is rarely applicable to anybody but the starting pitcher. Closers, setup men, and LOOGYS, for example, almost always appear in close games, so they have very little opportunity to pitch when their team is leading or behind by a lot of runs. The relievers who pitch in games that are not close are usually pitchers with something to prove so are generally trying to simply pitch as well as possible to make an impression on their manager. There are some exceptions, but for the most part, the question of whether or not pitchers pitch to the score applies to starters.
In the first three innings of a game, pitching to the score is rarely possible or a good idea. In most of these innings the score is either tied at zero or sufficiently close that the starter just wants to allow as few runs as possible. With a 6-0 lead in the third inning, it is probably true that starters are more likely to challenge hitters, but this is not true of a 1-0 or 2-0 score. Of course, few starting pitchers last long in games when they give up a lot of runs, so they rarely are around to pitch a specific way when they are down by many runs.
After the seventh inning, few starters are still in the game. There are of course exceptions. Pitchers like Roy Halladay, Cliff Lee or Tim Lincecum who frequently pitch late into games may adjust their pitching approach based on the score, but this is a small group of pitchers who are only changing how they pitch in a small percentage of innings. More ordinary starters such as Ivan Nova or Chad Billingsley rarely see the 8th or 9th inning.
There is another reason why the notion of pitching to the score bears more scrutiny, despite players frequently claiming that they do this. It is possible that pitchers say they pitch to the score simply because they recognize this is the answer they are expected to give and that it is the answer that will cause the least controversy. However, most big leaguers are extremely competitive on both a team and individual level. It is certainly possible that in games when the score is not close, a pitcher might be more interested in demonstrating his ability rather than simply very marginally improving his team’s chances of winning.
This is probably not true in a pennant race or during the post-season, but there are a lot of games played by teams that are not in a pennant race and not going to the post-season. On these teams, pitchers are competing for spots on the team and in the big leagues. Most people making decisions about who to keep on a team and how much to pay players are probably looking at things like strikeout to walk ratio or even ERA rather than things like holds or whether a pitcher goes 4-5 or 6-4 over 50 relief appearances. Starting pitchers are different, but they are no longer evaluated so heavily on wins either. It is almost certain, for example, that most baseball people recognize that Matt Cain is a top pitcher despite his poor win-loss record and that this record is due to Cain’s lack of run support rather than some mysterious inability to pitch to the score.
While it would be a wise career strategy for pitchers, particularly lesser known and less established pitchers, to simply pitch their best regardless of the score in order to make their numbers better, it would be a career threatening mistake to say this publicly. It is probably wrong to say that pitchers never pitch to the score, but the opportunities to do that are few and the incentives are not obvious for many pitchers.