In a recent post, Baseball Prospectus’ Jay Jaffe shed some light on offensive contributions across the defensive spectrum, which, for those unfamiliar with the concept, is the sequential ordering of each position based on how difficult it is to play (from easiest to hardest: DH, 1B, LF, RF, 3B, CF, 2B, SS, C). In addition, the spectrum also provides a frame of reference for expected offensive production from each position. According to the logic, because 1B is easier to play than most other positions, for example, those who fill the role need to provide more offense to enhance their value. Using a BP construct called True Average, Jaffe concluded that offensive levels have mostly followed the spectrum’s roadmap, even though there have been a few diversions along the way. However, the analysis didn’t make a connection between spectrum-based offensive production and winning games, so I figured I’d use the Yankees as a quick case study.
Yankees Positional OPS+, 1961-2011 (click to enlarge)
Note: OPS+ is a relative measure of OPS against the park-adjusted league average. An OPS+ of 120 is 20% better than the league average.Source: baseball-reference.com
Using OPS+ instead of Total Average to measure offensive performance, the chart above plots the production the Yankees received from each position since 1961 (the start of the expansion era). Whereas Jaffe’s chart had some discernible patterns, the graph above looks as if it was created by a child with a box of crayons. That’s because over the last 51 years, the Yankees have experienced significant peaks and valleys with regard to the relative offensive values of their position players. However, based on the percentage of seasons with above league average production, the Yankees strength at key positions, particularly catcher and center field (and more recently short stop and second base), becomes more evident, which, considering the team’s overall success, lends credence to the importance of building a ball club up the middle.
OPS+ > 100 Seasons by Position, 1961-2011
Note: OPS+ is a relative measure of OPS against the park-adjusted league average. An OPS+ of 120 is 20% better than the league average.
Source: baseball-reference.com
Since 1961, the Yankees’ winning percentage has correlated most closely to the team’s relative production from short stop. Although Derek Jeter is mostly responsible for that relationship, several prior successful periods in franchise history were also punctuated by an above average offensive short stop. Production from right field, which has been manned by Reggie Jackson, Dave Winfield, and Paul O’Neill, among others, has also exhibited a relatively strong correlation to winning. Grouped together behind those positions have been third base, catcher, second base, and center field, which have shown a similar link to winning ballgames. Finally, the three positions with the lowest correlation to winning are the ones at the easy-end of the defensive spectrum. In fact, left field has shown a negative correlation, which can’t make Brett Gardner feel very good.
Correlation Between Yankees’ Winning Percentage and Positional OPS+, Sine 1961
Note: OPS+ is a relative measure of OPS against the park-adjusted league average. An OPS+ of 120 is 20% better than the league average.
Source: baseball-reference.com
There’s more than one way to build a championship team, but collecting good hitters at premium defensive positions is a blue print that has withstood the test of time. Since 1996, the Yankees have used that approach to great effect, and it wouldn’t be surprising if their ability to continue as a perennial contender rests on the organization’s ability to continue to acquire or cultivate top players at key positions on the spectrum. That’s all the more reason to appreciate what players like Jeter and Jorge Posada have done, enjoy the excellence of Robinson Cano and Curtis Granderson, and, perhaps, look forward to the arrival of potential future stars like Gary Sanchez and Mason Williams.
interesting article but does the fact that oneill and the rest played right field instead of left really that significant? also how much of that negative correlation attributable to just the late 90s- chad curtis, ricky ledee, etc while the yankees were winning- for which i guess the lesson would be that great teams can afford one bad player, i dont think the fact that it was a lf really matters.
thanks
It’s not that significant. The difference between RF and LF isn’t great, and at Yankee Stadium, you probably need a better left fielder anyway. It wasn’t my intention to imply significance. Rather the negative correlation was of anecdotal interest, probably because, as you pointed out, it was one of the rare spots on the field for which the Yankees didn’t have an All Star.