Finding Meaning in the Meaningless

By Miles Goodrich

We are taught about numbers from a young age. Probability is a concept built upon in school starting as early as preschool, and up all the way through college. Greenwood High School’s AP Statistics class, taught by Holly Bush, is one example. Learning about statistics is reinforced all throughout schooling. Why, though, is learning to recognize patterns such a useful trait? Making meaning from the meaningless seems abstract conceptually, but it is used in a large variety of different fields. “One example I use for my classes is for them to name a profession that doesn’t use statistics,” said Bush.  Being able to decipher and understand numbers is a very important trait. Holly Bush stated that probabilities can help us make predictions about things and plan accordingly, which is the result of interpreting statistics. Questioning statistics is important, as people can misuse them to mislead others, said Bush. Curiosity is a basis for most statistical analysis. Wanting to know why something happens is another core component of early learning development. Many problems in the world have been solved using statistics, but there are also optimizations that are less obvious to the public eye. Holly Bush said data can help us make predictions, but not absolute decisions. Seeking to perform operations in a “better” way is crucial to upholding society. Imperfect systems can lead to imperfect products. That is the job of a statistician. Fixing problems. Finding meaning in the meaningless.

One example of how statistics were used was in baseball in the early 2000s. The Oakland Athletics were financially a very poor team. In 2002, their team payroll was only 41 million dollars. The New York Yankees, for example, had 126 million dollars, more than triple that of the Athletics. The Athletics lost Jason Giambi, Jason Isringhausen, and Johnny Damon, the three cornerstone players on the team. With little money and little talent, Oakland manager Billy Beane was running out of options. With nowhere else to go, Billy Beane tried something previously unheard of. He disregarded his elder scouts’ opinions and advice on who to draft and trade for, and he decided to use statistics. Beane ended up firing the head scout, who refused to listen. Beane and his colleague, Paul DePodesta, a Harvard graduate with a degree in economics, transformed the Athletics program, bringing in “flawed” players whom their scouts had written off as nonfactors. At first, the team’s manager refused to play the new players signed by Beane, believing that they were wrong and that the old-fashioned way of picking players was better. Beane counteracted this by trading away any player that could be played other than his own. The Athletics went on to win 20 games in a row, and make the postseason, despite how bleak their season looked at the beginning of the year. The only way they were able to win was by using sabermetrics, a type of statistics used in baseball in order to analyze the game. Beane and Depodesta used these sabermetrics in order to craft a cheap and overlooked team that was truly full of talent and role-players. Instead of finding one expensive do-it-all player, they found two cheap do-one-thing players that, when combined, equaled the do-it-all players, but for a fraction of the cost. Using undervalued statistics that other teams saw no value in was another ideal, as these numbers strongly correlated with success in players. This started the moneyball revolution in baseball, with teams optimizing their rosters with previously undervalued players, all based on statistics.

During World War II, the U.S. Air Force was losing a significant amount of airplanes to the Nazi onslaught. Many airplanes did not return from battle, and the few that did came back with numerous patches of bullet holes across the planes. The majority of the bullet holes were located on the tips of the wings, the center of the plane where the wings connect with the body, and on the tail wing of the airplane. The army officers saw these planes, and immediately called for reinforcement of those areas. Abraham Wald, though, had a different idea. Wald, a Hungarian and American statistician and mathematician, saw this problem in a different way than the army officers. By analyzing the data set he was given, he concluded that it was actually best to not reinforce the spots with the bullet holes. Why, though? His explanation was that if the planes could safely return with damage to those spots, those spots must not be significant to the airplane’s structural integrity. He decided, then, to reinforce the spots without any damage, as that is most likely where the fallen planes were struck, and why they were unable to return to the US military bases. This is called survivorship bias, and is a historical example of how statistics were used in history to help find a better solution to a problem.