Bartolo Colón won the 2005 AL Cy Young with 21 "wins," despite being 8th in the AL in ERA.
Many baseball fans place a kind of special significance on a starting pitcher "winning" 20 games in a season. Twenty game-winners are often by default considered serious candidates for that year's Cy Young Award. Even greater significance is given to pitchers who "win" 300 games over the course of their career. Every 300-game winner who has been retired at least five years has been inducted into the Hall of Fame.
However, the reality is that "wins" as an individual statistic are often a misleading measure of a starting pitcher's performance and value. Pitchers who play for teams with good offenses (that can give them a lead) and good bullpens (that can hold on to leads) have a significant advantage in accumulating wins over pitchers on mediocre teams.
When it comes down to it, there are only two statistics that are relevant for starting pitchers. First, earned run average (ERA), which measures the average number of runs a pitcher allows over nine innings. This is ultimately a pitcher's job: to prevent runs from scoring. It doesn't matter if he loads the bases every inning. If runs don't score, it ultimately doesn't matter. Second, innings pitched (IP), are a good measure of a starter's efficiency and stamina. Pitchers that can pitch 220+ innings over the course of the season take pressure of their bullpen, giving them an opportunity to rest.
However, in many cases wins and ERA go hand in hand. Many pitchers who win 20 games in a season have pretty good ERAs. However, there are also quite a few examples of pitchers with a lot of wins and an unremarkable ERA, perhaps because they play on a team with a good offense (Yankees, Red Sox).
Which pitching statistic is most meaningful?
A classic example of this was last year Phil Hughes won 18 games supported by the highest scoring lineup in the MLB. However, his ERA was a pedestrian 4.19. Felix Hernandez, on the other hand, only won 13 games yet led the Majors with a 2.27 ERA. Fortunately, the sportswriters who voted on the 2010 Cy Young overlooked King Felix's low win total and still gave him the award.
However, there are many examples when the opposite happened. In 2005, Cy Young winner Bartolo Colón "won" 21 games for the Angels while posting a 3.48 ERA. While a 3.48 ERA is pretty good, it was eighth in the American League that year. He was abused by the Yankees to the tune of four earned runs in Game 1 of the ALDS. Mark Buerhle, on the other hand, had a lower ERA and pitched more innings than Colón that year, leading the White Sox to the best record in the AL and eventually a World Series championship. It was obvious that Colón was not the best pitcher in the American League that year, but won the AL Cy Young by virtue of being the only pitcher with 20 "wins."
I realize that the 300 Win Club is populated by pitchers such as Randy Johnson, Greg Maddux and Nolan Ryan. I believe these three are all-time greats—not because they "won" 300 games, but because they maintained ERAs of 3.29, 3.16 and 3.19, respectively, over an entire career. That's a lot more impressive and meaningful than 300 "wins."