January 18, 2016
The Representativeness Heuristic and Baseball Positions
When filling out our fantasy baseball rosters, we do not simply select the 23 best players available to us; instead, we select the 23 best players that fit at the positions required by the league rules. This likely means that we will be selecting a player for each infield position, 3-5 outfielders, and eight-or-so pitchers. Depending on the league, we might be selecting a second catcher, an extra corner and middle infielder, and/or a specific number of starting pitchers and relief pitchers.
When our decisions change from picking the best to picking the best for a specific role, our decision making process can be negatively affected. At this point we welcome back our old friend, the representativeness heuristic. As written previously:
“The representativeness heuristic is the extent to which people judge probabilities “by the degree A is representative of B, that is, by the degree to which A resembles B.” Instead of predicting the likelihood that I am a cat owner based on the base rates of cat ownership in northeast America, cat ownership amongst age 20-30 males, etcetera; people will make their predictions based on how closely I resemble a stereotypical cat owner (whatever that looks like to them). Representativeness can be helpful when caution is beneficial and when being overly cautious has little or no cost. For example, a dog foaming at the mouth is representative of a dog carrying rabies; thus, if we assume a dog foaming at the mouth has rabies and, consequently, act more cautiously in the dog’s presence, then we are less likely to contract rabies. If the dog’s mouth was foaming for some normal reason or if it had really not been foaming at all, then our cautiousness only cost us a potentially nice encounter with a dog.”
While the above example explains the usefulness of using representativeness to make decisions (especially quick decisions where the payout is small and the potential negative outcome is severe), there is considerable downside to using representativeness when the odds are not as drastic. There is also more downside when the goal of the decision is changed from avoiding loss to getting the best odds.
You all see where this is going and you all see why this is being written during first-base week. When we think of first base, we think of 30+ home runs with a lot of RBI, a sufficient number of runs, and a batting average that could be anywhere from plus for the elite players to below average for the lesser options. Mostly, though, we think of the dingers. Consequently, we probably overrate players that are representative of this mold and underrate players that are not. I would guess that this is less the case in extreme cases—no one would be underrating a first baseman that stole 50 bases a season—while being more the case in moderate cases. An example to help make this case follows:
Eric Hosmer hit .297 with 18 home runs last season. Jose Abreu hit .290 with 30 home runs last season. In AL-only leagues, Hosmer was worth $27.62 and Abreu was worth $26.59. How is that possible? For starters, with stolen bases rarer and thus more valuable in AL rotisserie leagues last season, the value of the difference between Hosmer’s seven stolen bases and Abreu’s zero was the same as the difference between their home-run totals. Additionally, Hosmer bested Abreu by 10 runs scored while Abreu bested Hosmer by eight RBI. With runs scored being negligibly more valuable than RBI, the difference between the two ended up being mostly the batting average.
This is not to say that I will be taking Hosmer over Abreu this year fantasy drafts or auctions; nor will I be recommending anyone to do so. What this is to do, though, is to point out how molds can warp our valuation and decision making processes. Early ADP has Abreu being selected as the 22nd player overall and Hosmer as the 78th player overall. When doing draft or auction analysis, these are the kinds of gaps we should be looking for. Hosmer had a bad 2014 sandwiched between two good seasons, while Abreu has only had good seasons in his two years in the league, and that counts for something. Hosmer also just turned 26, while Abreu turns 29 in 11 days, and that also counts for something.
The point of all this is that valuation and decision-making through representativeness is the easier path to take—we are wired to do so and with so many decisions to make, it makes the overwhelming a little less so. The issue, though, is that in using this process, we will often be out-maneuvered by those doing a little more homework, by those doing more complete valuations and doing a better job of incorporating contextual factors. That said, let’s then strive to be the latter this offseason and let’s continue be conscious of the comparisons and molds we are using during the valuation process. Heck, if we do this, we probably have a shot at beating the representativeness heuristic more often than not.