FanPost

Spurious Correlations: Mets Edition Vol. 1

As many of you know, I'm a scientist by trade. This means I spend much of my time poring through piles of data in search of trends that reinforce a given hypothesis. This often involves detailed statistical analysis of the data to parse out which variables affect which other variables.

One common test to determine if two variables are related is the Pearson's Correlation Coefficient. This coefficient denotes the probability that when looking at two data sets, a change in one variable leads to a corresponding linear change in another variable. This value can range from 1 (meaning two variables change in a perfectly synchronized manner with respect to each other, to -1, meaning that an increase in one variable corresponds to a decrease in the other). A value of zero suggests that the variables change independently (or at least nonlinearly) with respect to each other. This value can then be used to determine the probability that changes in two variables are correlated.

I have applied the Pearson's Correlation Coefficient to a number of phenomena related to Mets players (and one bonus player) in an effort to determine the potential causes of performance variations from year to year and have come up with some very interesting relationships.

Bartolo Colon: The Hot Dog Hypothesis

Hypothesis: Bartolo Colon believes hot dogs to be sentient beings, and is thus greatly disturbed by their mass consumption.

Colonfinal_zps7f90f493_medium

*Information taken from Fangraphs and Wikipedia.com

The above plot shows Bartolo Colon's strikeout rate compared to the number of hot dogs Joey Chestnut successfully consumed at the Coney Island Hot Dog Eating contest. There is a modest but clear negative correlation (r=-0.627) between Colon's ability to strike out batters and Chestnut's wanton destruction of hot-dog kind. The knowledge of the tube steak genocide which has occurred clearly weighs heavily on kind Bartolo's shoulders, and it is demonstrated in his performance year after year. Should the contest ever be cancelled, one can only imagine what heights Bartolo's strikeout rate will raise to without being distracted by the horrors of Joey Chestnut.

Curtis Granderson: The Salmon Supposition

Hypothesis: Curtis Granderson is so salmon-crazed that fluctuations in its price affect his performance on field.

Grandersonfinal_zpsa6d40899_medium

*data adapted from 2014 Salmon Farming Industry Handbook and Fangraphs

Curtis Granderson's love of salmon has been well documented, and may have even played a role in his signing with the Mets. What the Mets didn't know is the dark obsession that underlies Granderson's love of salmon. Even small interruptions in Granderson's salmon supply can lead to mood fluctuations and increased aggressiveness. That is why it is unsurprising that increases in the cost of Atlantic Salmon lead to a direct increase in Granderson's aggressiveness at pitches in the strike zone. Knowing he needs to keep his salmon addiction a secret or risk facing league mandated penalties for substance abuse, Granderson unleashes his rage on baseballs, swinging with increased fervor as his supply of salmon becomes ever more expensive. So far an r value of .667 suggests that Granderson can keep his dark passenger in check enough to maintain some plate discipline, but it's only a matter of time before the salmonlust overtakes him.

Daniel Murphy: The Tremor Conjecture

Hypothesis: Throwing splitters to Daniel Murphy will eventually doom humanity

Murphycorr_zpsc85044ca_medium

*data taken from the US Geological Survey and Fangraphs

If you throw a split-fingered fastball to Daniel Murphy, you better hope he misses it. That's not simply because your job as a pitcher is to get outs, but because the lives of hundreds or even thousands of people rests on him not making contact with that pitch. The greater the Irish Hammer's success against splitters is in a given season, the greater the number of earthquakes experienced worldwide during that same year (r=0.761). Much as a butterfly flapping it's wings can create a hurricane across the sea, the spin on a splitter coming off of Murphy's bat and hitting the ground is just right to induce subtle tectonic shifts in the Earth. While these shifts are too small to feel at the local level, as they propagate outward these shifts lead to increased stress at distant fault lines triggering calamitous earthquakes all over the globe. It would probably be best for all parties concerned if pitchers stopped throwing this pitch to Murph, before some (other) mad scientist discovers that all you need to build a doomsday device is a pitching machine and Daniel Murphy.

Daisuke Matsuzaka: The Property Proposition

Hypothesis: Dice-K pitches so slowly, an increased strikeout rate provides fans enough time to start a family, settle down, and buy a house.

Dicekfinal_zpsa148e6e9_medium

*data taken from federalreserve.gov and Fangraphs

Daisuke Matsuzaka is infamous for taking an excruciatingly long break between each pitch. Whether he is using this time to prepare for his next offering, taking a moment of quiet reflection on existence itself, or just trying to remember what it was he was supposed to be doing right then, these pauses can drag games on incessantly. Therefore in seasons where Matsuzaka induces more swinging strikes it's only natural that fans would have a lot of additional time on their hands to attend to daily tasks. This increased productivity allows fans to finish up mundane tasks around the house between pitches, leaving more free time for professional activities during non-game hours. This leads to improved productivity at work, increasing the chances of raises and promotions. This increased socioeconomic standing eventually culminates in the acquisition of a family and house, the final phase of the proverbial American Dream. Noticing this trend, the Fed now sets it's 30-year mortgage rates directly in relation to Dice-K's swinging strike rate in a given season (r=0.891), keeping the rates high in years Matsuzaka induces the most swinging strikes in order to cash in on the massive influx of mortgage requests. So if you're looking to buy a house at a good mortgage rate, you better hope Dice-K is having a down year.

David Wright: The Carnage Correlary

Hypothesis: The very existence of David Wright sends Philadelphia fans into an insatiable bloodlust.

Wrightfixedfinal_zps8cd33b0a_medium

*data from Phillypolice.com and Fangraphs

There are two things that are abundantly clear to a Mets fan. 1) David Wright is the embodiment of all that is good in the world and 2) Philadelphia is the embodiment of all that is evil. I present here the greatest proof that Philadelphia is averse to all things good. The number of times David Wright appeared in a game in a season directly correlates with the number of murders committed in Philly in the same year. This correlation gives a convincing r value of 0.668, showing a significant correlation between the two variables. Just being reminded of the existence of David Wright (and thus goodness in the world), sends the city into an uncontrollable cheesesteak-fueled psychosis. The amount of calories in a single cheesesteak is enough to fuel a murderous rampage for days at a time, meaning once the metamorphosis into mindless killing machine begins, it can't be stopped until it has run its course. The best case scenario once the madness overtakes them is to hope they have a sufficient supply of Yuengling to drink themselves back into a stupefied state, suppressing their destructive urges for another day. So please Mets fans, as long as Wright is healthy, stay out of Philly for your own safety!

Bonus Content: The Pence Proof

Hypothesis: Hunter Pence is an alien, you know it in your heart to be true

Pencecorr2_zps85f6b0b0_medium

*data from CTVnews and Fangraphs

While SETI sends signals into space looking for intelligent life, I've found stronger evidence right here in front of us. It has long been postulated that Hunter Pence must be some sort of alien sent to observe humanity while attempting to mimic us. While his disguise is less than ideal, it has been sufficiently effective to cast doubt on his extraterrestrial origins, however I have uncovered an irrefutable piece of evidence that Hunter Pence is in league with aliens. The number of strikeouts Pence has in a season strongly and directly correlates (r=0.849) with the number of UFO sightings reported in Canada in a given year. Clearly as Pence struggles to make contact, the other aliens decide to take matters into their own hands and make contact of a different sort. So the next time you're abducted by aliens and probed in a very uncomfortable place, you can blame the Golden Sombrero Hunter Pence picked up the previous night. Sombrero's even look like UFO's... coincidence? I think not!

Conclusion

There are many undiscovered factors that influence baseball performance, and some factors of baseball performance which seem to affect the world at large as well. As the Sabermetric movement grows, more of these relationships are being discovered all the time. By utilizing the massive amounts of data available today, we can find correlations never before dreamed of between baseball and seemingly unrelated other topics, ranging from global economics to quantum physics. Stay tuned for future installments, who knows what I'll find next!

*Note: All correlations shown had sufficient degrees of freedom and r values to yield a p-value of 0.10 or lower meaning there is a 90% chance or greater that the null hypothesis (the variables are unrelated) can be discarded.

This FanPost was contributed by a member of the community and was not subject to any vetting or approval process.