Biases, Conflicting or Otherwise

People have an absurd amount in common with one another. Just one characteristic that humans share is our proclivity to be biased. That is what this post is about. Bias. 

One reason bias is particularly tough to recognize and quell is that there are many types. Glancing through the list of cognitive biases that are featured even on Wikipedia really drives this point home. Given that there are many types of bias, so too are there many opportunities for those biases to conflict and distort one another even further.

In baseball, purging bias from player evaluation is tricky to say the least, but numbers help to either validate or discredit our more qualitative notions. A pitcher’s ERA is tracked to uncover bias that his win-loss record might obscure. A pitcher’s FIP is employed to uncover the bias generated by the poor defensive team that plays behind him. Furthermore, FIP+ elucidates the bias created in FIP, which ignores the ballparks he pitches in. And on and on and on.

Here we focus on two types of bias specifically. 

The first form of bias examined here is referred to as anchoring. Anchoring is a phenomenon that entails an overly weighted emphasis on occurrences that took place early on and distort our perception of events thereafter. If a player has a torrid April, the perception of him, especially for a more casual fan, might be that he is having a great season even after his performance has trailed off well into the summer. Anchoring is what might leave some of us under the impression that Todd Frazier is having a terrible season, despite him simply being injured and later getting off to a slow start.

Second, recency bias. An eponymous term, recency bias illustrates our emphasis on what has occurred most recently. In essence, recency bias begets the question, “What have you done for me lately?” In baseball, recency bias is big. Players who perform especially well in the year leading up to free agency get rewarded accordingly (or at least so they hope). Players who are playing particularly well in June and July opposed to April and May might generate greater trade leverage for their team should that team be a seller. 

These two phenomena are good examples of types of bias that conflict, compete, and impact our perception accordingly. In order to evaluate a couple instances wherein these biases may have snuck into our perception in 2019, four tables are featured below. Two are traditional leaderboards. Two are leaderboards of a more ignominious nature. 

This first table is very simply a leaderboard for wRC+ through April 30th, a solid catchall for players’ offensive performance. These players are good examples of ones who dominate the headlines early on in the season, and given that humans will be humans, might stick in our minds as having strong seasons for an undeservingly long time. 

This isn’t to say that some of the hype isn’t warranted. Cody Bellinger and Christian Yelich have been legitimately phenomenal all year. Mike Trout is doing Mike Trout things. 

But baseball is a game of attrition, and playing at the very highest level for a sustained period is incredibly tough. If the first hypothetical commercial for this article was brought to you by cognitive bias, the second commercial would be for regression to the mean. BABIP’s have been highlighted in order to illustrate those most likely candidates for regression at April’s end.

Below is an identical leaderboard, but ranging from May 1st onward. Its purpose is to highlight who has been able to sustain their incredibly strong starts. Those who made both lists are highlighted. 

Just five players are common among both leaderboards, three of which are the aforementioned MVP candidates. Pete Alonso illustrates his staying power though, remaining on both boards. Alex Bregman is the fifth player to make the cut both times. For the rest though, there was regression to varying degrees.

 On the flip side of things, there are those who stumbled mightily out of the gate. Here is a table with data through April 30th, but for those players off to the slowest starts.

As one might expect, the BABIP’s for these players pretty uniformly range from incredibly poor (or unlucky) to just average. Some items stick out though. For one, this is not the first time Ian Desmond has come up. He owns the 20th best wRC+ since May 1st (at the time of this writing), but that fact might be lost on many given his incredibly slow start, prior struggles with the Rockies, and continued defensive struggles. But don’t let Desmond’s bat be a victim of anchoring.

Like the second table featured, this fourth table includes dates from May 1st on. Here though, those players who have performed worst in terms of wRC+ are listed.

Here there are three names that are common between the first and fourth table. In other words, three players had incredibly strong offensive Aprils and have subsequently struggled considerably. Those players are Elvis Andrus, Adalberto Mondesi, and Alex Gordon. Unfortunately, Jose Ramirez continues his season-long (and beyond) slump, as he is listed on both of the slump leaderboards. 

It seems that within a season (and especially a season’s first half) anchoring plays a greater role mentally than recency bias. Mondesi, Andrus, and Gordon are players whose 2019 performance to date I associate as being quite positive, despite offensive struggles over a nearly two month sample size. 

Toward the end of any given season, I imagine recency bias takes on additional meaning. Pundits love to point to arbitrarily split second-half performance. The anchoring that garnered a lot of All-Star support for players off to hot starts has subsided. During playoff runs, “What have you done for me lately?” is a question increasingly asked. 

Mulling over these opposite leaning cognitive forces is a worthwhile exercise. So is appreciating the players who have shaken off slow starts and the ones finding their way after having come back down to earth following really robust ones. Or taking into account those players whose stats maybe aren’t cumulatively fantastic but solid despite having weathered particularly rough starts to their seasons. Really, it’s just about being conscious of what skews our own perspectives.