New Research On Daily Double Bets Quantify Bias

feature

Recent research has quantified the qualitative analysis given by recent champion Lani Gonzalez in a recent #JeopardyLivePanel episode.

For the uninitiated, anchoring in economics is a cognitive bias that describes the common human tendency to rely too heavily on the first piece of information offered (the “anchor”) when making decisions. During decision making, anchoring occurs when individuals use an initial piece of information to make subsequent judgments.

As per the paper: This paper analyzes 12,596 wagering decisions of 6,064 contestants in the US game show Jeopardy!, focusing on the anchoring phenomenon in financial decision-making. We find that contestants anchor heavily on the initial dollar value of a clue in their wagering decision, even though there exists no rational reason to do so. More than half of all wagers occur within $500 of the initial dollar value, although the maximum possible wagering value averages $5,914. This anchoring phenomenon remains statistically significant on the one percent level, even after controlling for scores, clue category, time trends, and player-fixed effects.

You can find a PDF of Jay Walker and Michael Jetter’s recent paper “Anchoring in wagering decisions: Evidence from the field” here.

In the recent #JeopardyLivePanel episode, in talking about a Daily Double bet in the October 25, 2016 game Gonzalez said, “It was an $800 clue, I’ll just bet $800, because for some reason that seemed fair if I lost. I should have just bet $5, which I thought about later.” This certainly confirms that an anchoring bias exists, and may in fact give the cognitive reason for doing so (namely, that betting the clue value “seemed fair”).

I’ll be following up on this paper in the coming weeks here on The Jeopardy! Fan, as we might have our major reason why contestants routinely bet Final Jeopardy more aggressively than Daily Doubles.

Facebooktwittergoogle_plusredditpinterestlinkedinmail

1 Comment on "New Research On Daily Double Bets Quantify Bias"

  1. Finally read through the paper, and I’ve got two main issues:

    First, people generally still play the boards in order. The higher the clue value, the more money the player will have when they get to it. The paper mentioned that they control for score, but they didn’t appear to cover this specifically.

    Second, and this is the thing that drove me nuts on hearing the initial NPR story — there is *absolutely* a correlation between clue difficulty and clue value on Daily Doubles. A researcher can jump to the daily double stats for any reasonable length on j-archive and confirm this.

    For example, in seasons 21-32:
    Jeopardy: $400 DD clues were answered correctly at 73.25% overall, whereas $1000 DD clues were at 59.17%.
    Double Jeopardy: $800 DDs were 72.33%, and $2000 DDs were 56.92%

    From the paper: “the difficulty of the DD clue is not related to the initial dollar amount under which the clue is hidden. In fact, if a DD clue hidden under, say, $1,600 was more difficult than a DD clue hidden under $400, a rational response would be to wager more on clues with a smaller initial value, not less”

    I mean… maybe it would be, and is that even an important conclusion? The paper’s primary claim is that wagering should be done with complete disregard for clue value because there’s no link between clue value and clue difficulty. But when that’s objectively untrue, what next? It’s easy to say “well then the anchoring is even more apparent” but i don’t buy that as a theoretical knee-jerk rebuttal.

    Another note:

    “Note that the initial value of the question is made salient instantaneously before the wagering decision, as the contestant has to verbally state the initial dollar value of the clue”

    This is probably impractical because it’s not easily scrape-able data, but if there were a major difference between cases where the contestant said the clue value out loud and cases where they said “finish the category” or Alex pulled it up automatically as the last clue, that might be interesting.

    So, bottom line is that i think there is probably something here. The data shows that there’s a correlation between clue value and bet size. Whether this is due to some unconscious ‘anchoring’ phenomenon or something else seems to still be up in the air.

Leave a comment

Your email address will not be published.


*