Way back in the dark days of March, I listed betting the New York Yankees to win fewer than 95 games this year as one of the more solid bets baseball had to offer. In a follow-up post I pegged the Yanks " ... as more of a high-80s team with 92- 93-win upside, maybe a 30-35% chance of getting to 95 ... ". With two thirds of the year in the rearview mirror, New York's 69-43 record puts the Pinstripes squarely on a 100-win pace. Baseball Prospectus's (sadly proprietary, but still hugely useful) PECOTA projections have the Yankees winning 99 games. BP now ranks the Yanks as the best team in the American League.
Assuming I was right about New York's having only a 30-35% chance of winning 95 or more games, where did I go wrong?
Again we will turn to PECOTA, which not only spits about projections for each major-league player, but does so with a useful feature lacking in other projection systems: Error bars. PECOTA assigns different levels of performance to each player. A "20th percentile" projection means that a player had, as seen by PECOTA, an 80% chance of reaching this level. His 90th percentile projection suggested the player had only a 10% chance of hitting this level of performance. Here's the Yankees' current lineup, compared to their expectations.
1B - Mark Teixeira: .935 OPS, close to 75th percentile projection of .952. 4th among regular AL first basemen.
2B - Robinson Cano: .858 OPS, above 90th percentile of .856, 1st among regular AL second basemen
SS - Derek Jeter: .828 OPS, above 75th percentile of .811, 3rd among AL shortstops
3B - Alex Rodriguez: .871 OPS below weighted mean of .921, but still 3rd among AL third basemen
LF - Johnny Damon: .875 OPS, 2nd among AL left-fielders and his highest in nine years, close to 90th percentile of .890
CF1 - Melky Cabrera: .812 OPS, above 75th percentile of .786
CF2 - Brett Gardner: .748 OPS, near 75th percentile of .759
CF overall - Yankees are fifth in the the league in center-field OPS at .781
RF - Nick Swisher: .828 OPS, above 60th percentile of .816. Overall New York is fourth in the AL in right-field OPS at .853
C - Jorge Posada: .885 OPS, second among AL catchers to only Mauer the Great in Minnesota's 1.080, close to 90th percentile projection of .902
DH - Hideki Matsui: .858 OPS, third among DHs, right at 75th percentile of .855
Of the eleven players manning the ten lineup spots only Alex Rodriguez, an inner-circle Hall of Famer whose illustrious past pushes his projections to sometimes-unreasonable heights, is playing to the lower end of his expected performance range and even he is still the third-best hitter in the league at his position. *Eight* of the eleven are near, at, or above their 75th percentiles, meaning that PECOTA, the most sophisticated and accurate publicly-available projection system the game has to offer saw only a 25% chance of these players hitting at their current levels. Put another way, relative to position norms, New York's third-worst hitter is Teixeira, whose OPS ranks "only" fourth among AL first-sackers but who just happens to lead the league (in a tie) in home runs, is third in runs batted in, and is carrying a .382 on-base average to boot.
No comments:
Post a Comment