The last couple of weeks, Matt Ferenchick has been using Out of the Park Baseball 21 to see what Eduardo Nunez would look like as a perfect baseball player. It turned out to be a remarkable success, with Nunez leading the league in batting average, on-base percentage, slugging percentage, runs, triples, RBIs, OPS, ERA, innings pitched, strikeouts, and WHIP, accumulating 40 WAR in the process.
The success of Eduardo Nunez made me wonder how many perfect players are needed to make a team successful no matter the quality of their supporting talent. Using OOTP 20, I decided to try to find out.
To begin, I edited all the ratings of Yankees players to be 0s across the board, then edited one player — Aaron Judge — to be perfect, just as in Matt’s articles. I then used the simulation module to set up a 1000-game series against the Boston Red Sox, to see how much one perfect player, playing shortstop, could offset the ineptitude of his teammates.
Needless to say, it was an epic disaster. The team lost every single game, scoring 273 runs while giving up 108,811; on average, they lost every game by over one hundred runs. Opposing hitters posted a .739 batting average, while Yankees batters hit 0.77. The defense, moreover, was so atrocious that even a perfect pitcher posted an ERA of 22.45). Clearly a second perfect player was needed — in this case, Gleyber Torres, who would play shortstop while Judge went behind the plate.
In all honesty, this broke the AI manager. The team still did not win a game, scoring only 316 runs and giving up 94,131. While this was an improvement, it’s not as much as I expected, so I looked through the stats and found a little quirk: the computer decided to have Gleyber Torres be the starting pitcher for 927 of the 1000 games, in which he threw only 881 innings. Even a perfect pitcher cannot handle that workload.
And so a third Mr. Perfect was added, Gary Sanchez. He stayed behind the plate, while Torres played short and Judge manned center (when they weren’t pitching of course). And while the team was still bad...hey, they won a game! Ten of them, in fact! They scored 932 runs, and only gave up 52,240, and while the majority of innings were not pitched by the perfect players, they for the first time pitched a noticeable amount of innings. Furthermore, the defense had improved well enough that they were able to post ERAs under 10.
Even so, ten wins in 1000 games is not what we’re looking for here, so I added another. Now, four spots in the rotation are filled by perfect players, as is catcher, second base, shortstop, and center field. This squad won 64 games, scoring 1796 runs and giving up 18,271. The defense had improved well enough that the perfect players posted ERA figures under two, although they only combined for about 500 starts (I’d guess the AI struggled to determine whether to prioritize the perfect players in the lineup or in the rotation).
At this point, I continued adding one more player until the team finally had a winning record, and that turned out to be 7 players, filling out the entire rotation (with two extra pitchers in the bullpen, although the computer only put them into the game as a pitcher when they had a day off in the lineup), as well as every position on the diamond except first base (the DH was also an imperfect player). This squad won 627 games, scoring 6469 runs and giving up 4854. With more than enough perfect players to fill out the rotation and still leave some of them in the lineup, the computer did not seem to struggle with trying to balance their rotation and lineup duties, employing them in a way similar to the way Shohei Ohtani has been when healthy.
Of course, given the reality that is injuries, chances are this team would struggle a bit more than this during a hypothetical season; however, considering the Red Sox were one of the league’s top teams (the test used the 2019 rosters, which were based on 2018 performance) and this squad competed well with them, it’s likely that this team would at least be competitive.