Quiz: Chapter 18

Instructions

Select the best answer for each question. Some questions may have multiple correct answers; select all that apply where indicated.

Question 1

The favorite-longshot bias refers to the tendency for:

A) Favorites to be overpriced and longshots to be underpriced B) Favorites to be underpriced and longshots to be overpriced C) All contracts to be overpriced relative to their true probability D) All contracts to be underpriced relative to their true probability

Answer: B Explanation: The FLB is one of the most robust findings in betting and prediction markets. Longshots (low-probability events) are systematically overpriced relative to their true probability, while favorites (high-probability events) are systematically underpriced. This means buying favorites and selling longshots has positive expected value on average.

Question 2

In the prospect theory weighting function $w(p) = \frac{p^\gamma}{(p^\gamma + (1-p)^\gamma)^{1/\gamma}}$, the favorite-longshot bias pattern is produced when:

A) $\gamma > 1$ B) $\gamma = 1$ C) $\gamma < 1$ D) $\gamma = 0$

Answer: C Explanation: When $\gamma < 1$ (empirically estimated around 0.65), the weighting function overweights small probabilities (making longshots appear more likely than they are) and underweights large probabilities (making favorites appear less likely than they are). This produces the characteristic S-shaped distortion that underlies the FLB. When $\gamma = 1$, the weighting function is linear (no distortion).

Question 3

A trader observes that a prediction market price has barely moved after the release of a major economic report. The report strongly suggests the probability should shift by 10 percentage points. Which bias best explains the market's under-reaction?

A) Herding B) Anchoring C) Availability heuristic D) Overconfidence

Answer: B Explanation: Anchoring causes traders to stick too close to the pre-existing price (the anchor) and adjust insufficiently to new information. The market's failure to move the full 10 points suggested by the economic report is a classic example of anchoring and insufficient adjustment. The pre-report price acts as an anchor, and traders move only partway to the information-justified price.

Question 4

Which of the following is NOT one of the three faces of overconfidence described in behavioral finance?

A) Overestimation B) Overplacement C) Overprecision D) Overreaction

Answer: D Explanation: The three faces of overconfidence are overestimation (overestimating one's own ability or accuracy), overplacement (believing one is better than average), and overprecision (having confidence intervals that are too narrow). Overreaction is a separate phenomenon related to how markets respond to new information, not a form of overconfidence.

Question 5

In an information cascade model, a cascade forms when:

A) All traders receive the same signal B) The public information from previous traders' actions overwhelms each individual's private signal C) Traders collude to move the price D) A single large trader dominates the market

Answer: B Explanation: An information cascade occurs when the accumulated public information (inferred from observing previous traders' actions) is so strong that it becomes rational for each subsequent trader to ignore their own private signal and follow the crowd. This can happen even when the crowd is wrong, because each individual trader rationally concludes that the collective evidence outweighs their personal signal.

Question 6

A prediction market contract rises from $0.40 to $0.65 over five days with sharply increasing volume and no identifiable news events. This is followed by a decline to $0.52 over the next three days. This pattern is most consistent with:

A) Efficient price discovery B) Herd behavior followed by mean reversion C) A rational information cascade D) The favorite-longshot bias

Answer: B Explanation: The sharp price increase with rising volume and no news, followed by a reversal, is the classic signature of herd behavior. Traders pile in because they see others buying (not because of new information), creating a self-reinforcing price spike. When the herding impulse exhausts itself, prices revert toward fundamentals. The absence of news events and the subsequent reversal distinguish this from efficient price discovery.

Question 7

The availability heuristic predicts that prediction market contracts on which type of event will be most overpriced?

A) Mundane, frequently occurring events B) Dramatic, emotionally vivid, and heavily covered events C) Events with high base rates D) Events that have never occurred before

Answer: B Explanation: The availability heuristic causes people to overestimate the probability of events that are easily recalled — those that are vivid, emotional, recent, or extensively covered in the media. This leads to systematic overpricing of contracts related to dramatic events like pandemics, terrorist attacks, or financial crises, because the ease of recalling such events makes them feel more probable than they actually are.

Question 8

In prospect theory, loss aversion is characterized by a loss-to-gain sensitivity ratio ($\lambda$) of approximately:

A) 0.5 B) 1.0 C) 2.25 D) 5.0

Answer: C Explanation: The standard empirical estimate for the loss aversion parameter $\lambda$ is approximately 2.25, meaning that the psychological pain of losing $X is about 2.25 times as intense as the pleasure of gaining $X. This asymmetry has profound implications for trading behavior, including the disposition effect and risk-seeking behavior in the loss domain.

Question 9

The disposition effect predicts that traders will:

A) Sell winners too early and hold losers too long B) Sell losers too early and hold winners too long C) Sell both winners and losers too early D) Hold both winners and losers too long

Answer: A Explanation: The disposition effect, predicted by prospect theory, is the tendency to sell winning positions too early (because the value function is concave in the gain domain, making traders risk-averse for gains) and hold losing positions too long (because the value function is convex in the loss domain, making traders risk-seeking for losses and hoping for a recovery). This is one of the most well-documented patterns in trading behavior.

Question 10

Base rate neglect, a consequence of the representativeness heuristic, would most likely cause which mispricing?

A) Overpricing a contract on a novel pandemic when the base rate of pandemics is very low B) Underpricing a contract on a mundane economic indicator C) Correctly pricing a contract when the narrative matches the base rate D) Overpricing a contract on an event with a high base rate

Answer: A Explanation: Base rate neglect occurs when people judge probabilities based on how well a situation matches a stereotype or narrative (representativeness) rather than considering the underlying base rate. When a situation "looks like" a pandemic (matches the representative features), people overestimate its probability because they neglect the very low base rate of actual pandemics. This is especially pronounced when the narrative is vivid and emotionally engaging.

Question 11

Which debiasing technique involves imagining that your trade has already failed and working backward to explain why?

A) Red teaming B) Pre-mortem analysis C) Calibration training D) Decision journaling

Answer: B Explanation: Pre-mortem analysis, developed by Gary Klein, involves imagining a future in which your decision has failed and then explaining why. This leverages the brain's narrative-construction abilities for constructive skepticism rather than post-hoc rationalization. By imagining failure scenarios in advance, traders can identify risks they might otherwise overlook due to overconfidence or confirmation bias.

Question 12

A trader notices that their 90% confidence intervals contain the true value only 55% of the time. This is an example of:

A) Overestimation B) Overplacement C) Overprecision D) Underconfidence

Answer: C Explanation: Overprecision is the form of overconfidence in which confidence intervals are too narrow. If 90% confidence intervals contain the truth only 55% of the time, the trader's intervals are far too tight — they believe they know the answer with more precision than they actually do. This is one of the most common and costly forms of overconfidence in prediction market trading.

Question 13

The conjunction fallacy, related to the representativeness heuristic, can cause which problem in prediction markets?

A) The sum of probabilities for mutually exclusive outcomes may exceed 100% B) All contracts are uniformly overpriced C) Prices converge to 50% regardless of true probability D) Volume drops to zero

Answer: A Explanation: The conjunction fallacy causes people to rate specific, detailed scenarios as more probable than their component events. In multi-outcome prediction markets, this means that the individual probabilities assigned to specific scenarios may each be too high, causing the sum across mutually exclusive outcomes to exceed 100%. This creates a direct arbitrage opportunity for traders who recognize the violation of basic probability rules.

Question 14

Round-number anchoring in prediction markets refers to the tendency for:

A) Prices to cluster at multiples of 5 or 10 percentage points B) Traders to place orders in round lot sizes C) Markets to open at round-number prices D) Trading volume to peak at round-number times

Answer: A Explanation: Round-number anchoring is the tendency for prediction market prices to cluster near psychologically salient round numbers (10%, 25%, 50%, 75%, 90%). Traders find it difficult to move prices away from these focal points because the round numbers serve as cognitive anchors. A contract that should trade at 47% may get "stuck" at 50% due to this effect.

Question 15

Which of the following correctly describes the relationship between recency bias and Bayesian updating?

A) Recency bias is a form of Bayesian updating with a strong prior B) Recency bias gives too much weight to the most recent evidence relative to Bayesian updating C) Recency bias gives too little weight to the most recent evidence relative to Bayesian updating D) Recency bias and Bayesian updating produce identical results

Answer: B Explanation: A Bayesian updater weights each piece of evidence by its informativeness (likelihood ratio), regardless of when it was received. Recency bias causes the most recent piece of evidence to receive disproportionate weight, overriding earlier evidence that may be equally or more informative. This leads to overreaction to the latest data point and underweighting of the accumulated evidence base.

Question 16

The narrative fallacy creates which characteristic price pattern?

A) Random walk with no predictable patterns B) Gradual drift as the narrative builds, followed by sharp reversal when it breaks C) Steady increase to the resolution date D) Immediate jump to the correct price

Answer: B Explanation: The narrative fallacy drives a characteristic two-phase price pattern. First, as a compelling narrative takes hold among traders, prices drift gradually in the direction of the narrative (because more traders adopt the narrative over time). Second, when the narrative is contradicted by evidence (or simply fails to materialize), prices correct sharply as traders simultaneously abandon the narrative. This creates predictable contrarian opportunities at the point of narrative reversal.

Question 17

In the context of prediction markets, an echo chamber primarily amplifies which bias?

A) The favorite-longshot bias B) Anchoring C) Confirmation bias D) The disposition effect

Answer: C Explanation: Echo chambers — online communities where a dominant view crowds out dissent — are a social amplifier of confirmation bias. Members of the echo chamber selectively share confirming evidence, dismiss contradictory information, and become increasingly confident in the group's shared view. This collective confirmation bias can drive market prices away from fundamentals as community members all trade in the same direction.

Question 18

A trader who doubles their position size after a losing streak, trying to "get back to even," is exhibiting behavior predicted by:

A) The anchoring effect B) The availability heuristic C) Prospect theory (risk-seeking in the loss domain) D) The representativeness heuristic

Answer: C Explanation: Prospect theory predicts that people become risk-seeking in the loss domain (when facing losses). The value function is convex for losses, meaning that the marginal pain of additional losses decreases, making risky bets more attractive. A trader who is losing money may take larger, riskier bets in an attempt to recover their losses — a behavior often called "doubling down" or "revenge trading." This is one of the most dangerous consequences of loss aversion.

Question 19

The endowment effect in prediction markets means that:

A) Traders who own a position value it more highly than traders who do not B) All positions increase in value over time C) Traders prefer contracts with larger payoffs D) Market makers set prices too high

Answer: A Explanation: The endowment effect is the tendency to value something more simply because you own it. In prediction markets, a trader who holds YES shares may believe the fair value is $0.60, but if they did not hold the position, they might estimate fair value at only $0.52. This creates a "zone of inaction" where the trader neither buys more nor sells, reducing market efficiency and slowing price discovery.

Question 20

When combining multiple bias signals, the text recommends which approach for typical prediction market datasets?

A) Deep neural networks B) Random forests C) Simple weighted sum D) Unweighted average

Answer: C Explanation: Given the relatively small sample sizes typical of prediction market data, the text recommends a simple weighted sum of bias signals rather than complex machine learning models. While logistic regression or other ML approaches can learn optimal signal combinations, they are prone to overfitting on small datasets. A simple weighted sum, with weights calibrated from historical data, provides a more robust approach that generalizes better to new markets.

Question 21

Which of the following is the most effective debiasing technique according to the research discussed in this chapter?

A) Simply being aware of biases B) Having high intelligence C) Structured decision-making processes (checklists, pre-mortems) D) Trading experience alone

Answer: C Explanation: Research consistently shows that simply knowing about biases (the "bias blind spot") has surprisingly little effect on reducing them. Intelligence and experience alone are also insufficient — smart, experienced traders still exhibit systematic biases. The most effective approach is implementing structured decision-making processes that force different behavior regardless of how the trader feels. Checklists, pre-mortems, red teaming, and decision journals create external accountability and interrupt the automatic cognitive processes that produce biases.

Question 22

The adjustment coefficient $\alpha$ in the anchoring model $\hat{p} = A + \alpha(p^* - A)$ is typically estimated in prediction markets as:

A) $\alpha \approx 0.1$ (almost no adjustment from anchor) B) $\alpha \approx 0.4 - 0.7$ (partial adjustment) C) $\alpha \approx 0.9 - 1.0$ (nearly complete adjustment) D) $\alpha > 1.0$ (overadjustment from anchor)

Answer: B Explanation: Empirical estimates suggest that prediction market traders adjust approximately 40-70% of the way from the anchor to the true probability. This means 30-60% of the mispricing (the gap between the anchor and the true probability) remains in the market price. This partial adjustment is the core mechanism by which anchoring creates exploitable mispricing.

Question 23

A prediction market where herding is prevalent would show which autocorrelation pattern?

A) Zero autocorrelation at all lags B) Positive autocorrelation at short lags, negative autocorrelation at longer lags C) Negative autocorrelation at all lags D) Positive autocorrelation at all lags

Answer: B Explanation: Herding creates positive short-term autocorrelation (momentum) as traders pile in, pushing prices further in the same direction. This is followed by negative medium-term autocorrelation (reversal) as prices correct back toward fundamentals when the herding impulse exhausts itself. This pattern — short-term momentum followed by medium-term reversal — is the empirical fingerprint of herding behavior.

Question 24

Select ALL biases that would contribute to overpricing of a contract on "Will there be a major cyberattack on U.S. infrastructure this year?" (Select all that apply)

A) Availability heuristic (recent media coverage of cyber threats) B) Representativeness heuristic (the scenario "looks like" something that would happen) C) Favorite-longshot bias (low-probability event) D) Recency bias (a recent minor cyberattack) E) Anchoring to previous year's contract price

Answer: A, B, C, D, E Explanation: All five biases could contribute to overpricing this contract. The availability heuristic makes the event seem more likely due to vivid media coverage. The representativeness heuristic makes the specific scenario feel probable because it matches our mental model of "cyber threat." The FLB causes overpricing of low-probability events in general. Recency bias overweights a recent minor incident. And anchoring might cause the market to stick near the previous year's price regardless of whether this year's risk profile is different. This is an example of how multiple biases can compound to create significant mispricing.

Question 25

A decision journal is recommended as a debiasing tool primarily because:

A) It provides legal documentation of trading decisions B) It forces rigorous reasoning before trades and enables pattern recognition in errors afterward C) It is required by prediction market platforms D) It increases trading speed by recording templates

Answer: B Explanation: A decision journal serves two primary purposes: (1) Writing down reasoning before a trade forces the trader to articulate specific reasons rather than relying on vague intuitions, which improves the quality of decision-making in real time. (2) Reviewing past decisions after resolution enables the trader to identify systematic patterns in their errors — e.g., consistent overconfidence in certain types of events, or larger positions when emotionally agitated. Over time, this creates a feedback loop that helps the trader recognize and correct their specific biases.