The Death of Behavioral Economics

by habryka1 min read22nd Aug 202122 comments

136

Heuristics & BiasesReplication CrisisEconomicsWorld Modeling
Frontpage

Edit: I recommend reading Scott's response to this essay in addition to the essay itself.

I've been tracking the replication crisis and how it affects a bunch of behavioral economics for a while. I found reading this post useful for a particularly negative take. Some key quotes:

It sure does look alive... but it's a zombie—inside and out.

Why do I say this?

Two primary reasons:

  1. Core behavioral economics findings have been failing to replicate for several years, and *the* core finding of behavioral economics, loss aversion, is on ever more shaky ground.
  2. Its interventions are surprisingly weak in practice.

Because of these two things, I don't think that behavioral economics will be a respected and widely used field 10-15 years from now.

[...]

It turns out that loss aversion does exist, but only for large losses. This makes sense. We *should* be particularly wary of decisions that can wipe us out. That's not a so-called "cognitive bias". It's not irrational. In fact, it's completely sensical. If a decision can destroy you and/or your family, it's sane to be cautious.

"So when did we discover that loss aversion exists only for large losses?"

Well, actually, it looks like Kahneman and Tversky, winners of the Nobel Prize in Economics, knew about this unfortunate fact when they were developing Prospect Theory—their grand theory with loss aversion at its center. Unfortunately, the findings rebutting their view of loss aversion were carefully omitted from their papers, and other findings that went against their model were misrepresented so that they would instead support their pet theory. In short: any data that didn't fit Prospect Theory was dismissed or distorted.

I don't know what you'd call this behavior... but it's not science.

This shady behavior by the two titans of the field was brought to light in a paper published in 2018: "Acceptable Losses: The Debatable Origins of Loss Aversion".

I encourage you to read the paper. It's shocking. This line from the abstract sums things up pretty well: "...the early studies of utility functions have shown that while very large losses are overweighted, smaller losses are often not. In addition, the findings of some of these studies have been systematically misrepresented to reflect loss aversion, though they did not find it."

136

22 comments, sorted by Highlighting new comments since Today at 2:20 PM
New Comment

Even with Kaj's highlight comments, which are helpful, I don't feel educated enough in the economics area (one semester of high school Econ) to tell whether this is
1) vigorous academic debate or

2) damning evidence of academic fraud by Kahneman and Tversky.

Given how K&T are pretty central to parts of the Sequences, and that Judgment Under Uncertainty is at the top of my book pile--can someone with some expertise in this area give their take? I would donate $20 to Givewell for 500+ words that helped me to understand this situation.

Academic here, it's (1). Loss aversion is so popular that people think it underpins everything. Although loss aversion doesn't show up in every dataset, it does show up (https://doi.org/10.1002/jcpy.1156) - even the "second paper" shared by Kaj just says it appears sometimes. But does that mean it explains all these other findings? No! But some reviewers or authors think "isn't that just loss aversion?" and it seems authors take the easy route to publication (or just aren't well-read enough) instead of probing the psychological source of their findings more seriously. For example, loss aversion was the classic explanation for the endowment effect, but research in the last couple decades has generated results that loss aversion cannot really explain and that other theories readily explain, yet LA is sometimes still cited as the explanation the authors endorse.

I would also be interested in an explanation of how the replication crisis effects the sequences and willing to put in 10$ to givewell

How the heck do I update on this?

I don't feel like I have a graceful way to de-weight something when it turns out poorly in this fashion. I feel comfortable with unwinding an update I previously made, but in this case it amounts to just throwing out everything I have head-chunked as behavioral economics.

This feels wrong-ish, in the sense that it isn't as though all the research was a complete fiction; a more correct operation would be to adjust my priors in such a way as to capture what the research actually shows, rather than what I thought it showed.

Trouble is, this is even more work than making the initial updates, because the whole failure mode is an inability to have confidence in any existing distillation of the ideas. This means tackling the relevant studies one at a time, with only a few newer review or meta papers to help.

On the upside, it occurs to me that I integrated virtually none of the mentioned results well enough that it met the anticipated experiences standard; maybe that means I never really updated in the first place and this costs nothing to lose.

I note that I am confused. I am confused mostly because the claim "loss aversion exists only for large losses" seems to be completely disharmonious with my anecdotal experience, and I tend to view anecdotal experience as an often semi-reliable guide to the accuracy of social science. If the strong version of this claim is true, how would you explain the following facts?

  • This video, in which a bunch of people on the street did not want to bet at favorable odds that a coin flip would land tails.
  • A large fraction of the population holds their money in low-interest savings accounts, rather than putting their money in a brokerage account where they can buy stocks.
  • People systematically refuse to throw away stuff that they own, even if they haven't used the items in years. But when asked whether they would pay a dollar to put one more useless item in their home, they will refuse.
  • People usually view petty theft as horrible when it's committed against them, even though they usually don't become extremely happy when you unexpectedly give them $50.
  • It's well known that people will panic-sell during a market correction, even though if the market rises by 10% people don't change their behavior by much, and even if they don't have much money in the stock market.
  • Anecdotally, I used to gamble small amounts of money and would feel really bad if I lost it, even if it was just $20. Lots of people don't like taking friendly epistemic bets for similar reasons.

My guess is that you could come up with ad-hoc explanations for all of these things without reference to loss aversion, but that doesn't seem very elegant to me. A proclivity for loss aversion present in 50+% of humans appears like the most natural, simple explanation.

[ETA: However, after reading the link from Kaj Sotala I'm starting to feel my mind being changed.]

There's also a second paper linked from that article which is quite interesting (some excerpts in child comments).

Here, we offer a review and discussion of the literature on loss aversion. Our main conclusion is that the weight of the evidence does not support a general tendency for losses to be more psychologically impactful than gains (i.e., loss aversion). Rather, our review suggests the need for a more contextualized perspective whereby losses sometimes loom larger than gains, sometimes losses and gains have similar psychological impact, and sometimes gains loom larger than losses.

Risky choice

Kahneman and Tversky (1979) propose that individuals will tend to demand a substantial premium over an expected value of zero to accept a bet with even odds of winning and losing the bet. In the words of Kahneman and Tversky (1979), “most people find symmetrical bets of the form (x, 0.50; !x, 0.50) distinctly unattractive.” In a typical demonstration, which we refer to as the risky bet premium paradigm, if individuals are offered a bet with a 50% chance of losing $5 and a 50% chance of winning X, on average, they demand that X be $10 or more in order to accept the bet. This finding is assumed to reflect the greater perceived psychological impact of a loss compared with a gain (Tversky & Kahneman, 1992).

Gal (2006) points out that the risky bet premium can be conceived as a special case of the status quo bias paradigm where not accepting the bet is the status quo (or inaction) option and accepting the bet is the change (or action) option. As a result, similar explanations to those that can explain the status quo bias and endowment effect can explain the risky bet premium. Therefore, it is unclear whether the risky bet premium reflects a general tendency of losses to loom larger than gains or reflects processes associated with a propensity to favor inaction over action.

In order to decouple losses and gains from inaction and action in the context of risky choice, Gal (2006) presented participants with a risky bet, where no difference in action or inaction existed with respect to accepting the bet and not accepting the bet. Gal found no evidence that losses loomed larger than gains. Specifically, in a hypothetical decision to allocate funds ($100) between a safe alternative that returned 3% for sure and a mixed even bet with an expected return of zero, nearly 80% of individuals allocated at least some funds to the even bet, that is, to a risky option with lower expected value than the safe option, and approximately 20% of individuals allocated all the funds to the even bet, an amount which matched the percentage of individuals allocating all their funds to the safe option.

Rather than evidence for loss aversion, if anything, the behavior documented by Gal (2006) appears, on net, to reflect gain seeking. Other researchers have similarly found that when given multiple investment options, individuals tend to choose risky investment options over safer investment options with higher expected value (Ben-Zion, Erev, Haruvy, & Shavit, 2010). Such findings appear difficult to reconcile with a general principle of loss aversion (see also Erev, Ert, & Roth, 2010; Sonsino, Erev, & Gilat, 2002 for results with similar implications) and provide evidence against both the strong and weak versions of loss aversion considered here.

Other researchers have found that when accepting a risky bet is not framed as the sole action option, but as one option in a choice between two action options, no evidence for loss aversion emerges (Erev, Ert, & Yechiam, 2008; Ert & Erev, 2013; Ert & Yechiam, 2010; Hochman & Yechiam, 2011; Koritzky & Yechiam, 2010; Yechiam & Ert, 2007). For example, Erev et al. (2008) offered participants a choice between either (a) receiving 0 points for sure or (b) receiving a bet that offered a 50% chance to win 1000 points and a 50% chance to lose 1000 points (points were to be converted to money at a known ratio). Erev et al. found that 48% of participants chose the safe option (i.e., receiving 0 point for sure) and 52% of participants chose the risky option. Consistent with this finding, a review of over 30 papers finds little evidence that losses loom larger than gains in the context of risky choice when a bet with even odds of gaining and losing is not framed as the action option (Yechiam & Hochman, 2013). We recently found additional support for this conclusion in two separate runs of an experiment conducted with participants from MTurk. In particular, we asked participants to imagine they faced a choice between either (a) receiving $0 with 100% chance or (b) receiving $15 with 50% chance or losing $15 with 50% chance. In both runs, participants exhibited a trend toward the choice of the risky option (Figure 2). Thus, we did not find evidence for participants to avoid loss any more than they pursued gain in risky choice.

The stakes of the outcomes in risky choice experiments that do not show evidence for loss aversion tend to be low to moderate (from less than $1 to as high as $100). Conversely, some experiments that involve higher stakes (e.g., several hundred dollars) have shown a tendency among individuals to choose the safer alternative. However, loss aversion is assumed to be independent of the level of the stakes involved (Kahneman & Tversky, 1979). In fact, that the effects attributed to loss aversion have been found with small stakes is cited as particularly strong evidence for loss aversion (Rabin, 2003; Rabin & Thaler, 2001). The reason scholars have focused on small stakes is because avoidance of large magnitude losses can be explained by ordinary risk aversion for changes in wealth/circumstances, which is entirely consistent with rational choice theory, whereas the same is not true of avoidance of low stakes losses that do not materially impact wealth/circumstances. For example, it is rational to perceive a greater impact from losing $1000 that is needed to pay the rent than from gaining $1000 when basic needs are already covered. Conversely, if neither losing nor gaining $5 materially changes one’s circumstances, it can be viewed as irrational to view its loss as more impactful than its gain. Thus, the finding that people often exhibit risk neutrality in choices among low-stakes mixed gambles is evidence against loss aversion.

Gain seeking (the opposite of loss aversion) in the stock market

Moreover, whereas real-world phenomena exist that appear consistent with loss aversion, as pointed out by Ert and Erev (2013), other phenomena occur that appear consistent with the opposite, namely gain seeking. For example, Barber and Odean (1999) identified the phenomenon of overtrading in the stock market, whereby investors trade more than would be justified by rationality assumptions. To the extent that maintaining the status quo is thought to represent loss aversion, this excess trading (i.e., changing of the status quo) could be interpreted to support gainseeking behavior. Further, individual investors exhibit insufficient diversification among assets (Barber & Odean, 2000). To the extent that diversification reduces risk, this behavior can also be interpreted as gain seeking.

Self-rated losses vs. gains

Arguably, perhaps the most straightforward test of loss aversion is to simply ask people to evaluate the impact of losing versus gaining the same object. However, when researchers have examined how people rate the impact of losing versus gaining the same amount of money, little support for loss aversion has emerged (Harinck, Van Dijk, Van Beest, & Mersmann, 2007; Liberman, Idson, & Higgins, 2005; Mellers, Schwartz, & Ritov, 1999; Mukherjee, Sahay, Pammi, & Srinivasan, 2017; Rozin & Royzman, 2001). For example, Rozin and Royzman (2001) write: “In its boldest form, losing $10 is worse than winning $10 is good. Although we are convinced of the general validity of loss aversion, and the prospect function that describes and predicts it, we confess that the phenomenon is only realizable in some frameworks. In particular, strict loss and gain of money does not reliably demonstrate loss aversion (unpublished data by the authors)” (Rozin & Royzman, 2001, p. 306). In fact, with low stakes, gains actually appear to loom larger than losses when using this paradigm (e.g., Harinck et al., 2007).

Whereas past work has focused on a comparison between losing versus gaining monetary amounts, we have recently examined how people react to losing nonmonetary objects (Gal & Rucker, 2017b). For example, how do people rate the impact of losing versus gaining a mug? For most everyday objects we examined (mugs, flashlights, notebooks), the positive impact anticipated from gaining the object was rated to be greater than the negative impact anticipated from losing the object. For example, using a scale ranging from !5 (“extremely negative”) to +5 (“extremely positive”) to describe their feelings, participants who rated their feelings about losing a mug said their feelings would be less affected (M = 1.38) than did participants who rated their feelings if they were to gain a mug (M = 2.71). Notably, for some objects, we found no statistical difference between the impact of gains versus losses (a watch, a mountain view, lakefront access), and for no object did we find losses were rated to be more impactful than gains.

McGraw, Larsen, Kahneman, and Schkade (2010) attempted to reconcile the inconsistency of such findings with loss aversion. Specifically, the authors proposed that losses and gains are evaluated on different subjective scales. Consequently, the comparison of the impact of a loss evaluated independently with the impact of a gain evaluated independently does not provide a fair relative comparison of the impact of losses versus gains. Instead, they argue for a fair comparison, the loss and gain of an object need to be evaluated jointly with respect to each other. To this end, McGraw et al. (2010) asked participants to evaluate the relative impact of losing versus gaining the same amount of money; for example, they asked participants which of losing or gaining $50 they thought would be more impactful. With this approach, McGraw et al. (2010) identified a pattern of results consistent with loss aversion: the majority of participants stated that the loss of money would be more impactful than its gain. [...]

... an important caveat is in order. Namely, the studies of McGraw et al. (2010) involved potentially significant amounts of money for the participants involved (i.e., $50 and $200 for undergraduates). As noted previously, when large amounts of money are involved, loss aversion is indistinguishable from risk aversion for changes in wealth, which is fully consistent with rational choice theory (cf. Rabin & Thaler, 2001). To put this in context, if losing $50 is more likely to impact one’s lifestyle and wellbeing than gaining $50 is likely to impact it, then it is perfectly rational that individuals would be more psychologically impacted by losing $50 than by gaining $50. However, it is assumed that the loss versus gain of small amounts of money do not differentially impact one’s objective wellbeing, and hence, it is considered irrational for losses to loom larger than gains when small amounts of money are involved (Rabin & Thaler, 2001).

Indeed, in a recent paper by Mukherjee et al. (2017), the authors replicated the procedure of McGraw et al. (2010) with low stakes. They observed that when stakes were low, gains were rated as having more psychological impact than losses. Conversely, when stakes were high, Mukherjee et al. (2017) found that participants tended to rate losses as more impactful than gains. Thus, consistent with the possibility of contextual factors affecting the relative impact of losses and gains, the findings of Mukherjee et al. suggest a moderator of when losses loom larger than gains. On the other hand, the definitiveness of this moderator must be tempered by potential concerns about the validity of the particular methodology used for testing the impact of losses versus gains and the fact that for high stakes it is difficult to distinguish risk aversion from differences in the psychological impact of losses and gains. Finally, in recent work (Gal & Rucker, 2017b), we also asked participants to rate the impact of gaining and losing various goods using McGraw et al.’s procedure. Although our results varied based on the nature of the good, we found no evidence for a predominance for losses to loom larger than gains. 

The endowment effect and loss aversion

The endowment effect is the phenomenon perhaps most often cited as evidence for loss aversion in the context of riskless choice (Kahneman et al., 1990; Thaler, 1980; Tversky & Kahneman, 1991). The endowment effect refers to the finding that owners of an object demand more to part with the object than nonowners are willing to pay to obtain it (Thaler, 1980). For example, in a classic study, Kahneman et al. (1990) found that individuals endowed with a mug demanded, on average, about $7 to part with it. In contrast, individuals not endowed with a mug were, on average, willing to pay only about $3 to obtain the same mug. The finding that individuals’ willingness to accept (WTA) is greater than their willingness to pay (WTP) appears robust across many different instantiations of the endowment paradigm (Kahneman et al., 1991). It is this central finding that is viewed as evidence for the general principle that losses exert a greater impact than gains.

Although taken as evidence for loss aversion, the endowment effect can be understood as a case of the status quo bias where maintaining the endowed option is the inaction (or status quo) alternative. As such, the endowment effect is subject to the same alternative explanations (e.g., inertia) to loss aversion as those described for the status quo bias. For example, the inertia explanation suggests that when individuals are indifferent between the endowed option and the nonendowed option, they will opt to maintain the endowed option due to lack of incentive to trade, not because the loss of the endowed option looms larger than the gain of the nonendowed option.

Another explanation of the endowment effect, which similarly does not require loss aversion, comes from Weaver and Frederick (2012) and Isoni (2011) (see also Simonson & Drolet, 2004; Yechiam, Ashby, & Pachur, 2017). These authors provide a differential reference price account. They argue that buyers and sellers face fundamentally different decisions that lead them to focus on different reference prices when setting WTP and WTA amounts, respectively. For buyers, their own personal utility from the acquisition of the object is the most salient reference. In contrast, for sellers, the market value of the object is the most salient reference. As a consequence, if market prices tend to exceed personal valuations, owners will ask more for a product than a prospective buyer is willing to pay. For example, if both owners and nonowners value an object at $3, but the market price is $7, owners will demand $7 to part with it, whereas nonowners will only be willing to pay $3 to acquire it. This account, as with inertia, requires no differential sensitivity to losses to explain the endowment effect.

Other potential confounds exist in the endowment effect paradigm. For example, WTP and WTA are assessed on different scales. WTP is bounded by one’s ability to pay (i.e., budget constraints), whereas WTA is not.

In the possible alternative explanations to loss aversion for the endowment effect discussed so far, the valuation of an option when it is the endowed versus the nonendowed option does not differ. However, research also shows that individuals confronted with the decision of whether to give up an endowed option tend to focus more on positive features and less on negative features of the option than those faced with the decision of whether to acquire the option (Carmon & Ariely, 2000; Nayakankuppam & Mishra, 2005; see also Johnson, Haubl, & Keinan, 2007). This process could result in greater valuation for an option when it is endowed than when it is not endowed and, therefore, could be interpreted as a process that leads losses to loom larger gains in the context of the endowment effect. However, two caveats are in order. First, because loss and gain are confounded with inaction and action in the endowment paradigm, rather than reflect a tendency to elevate the option that might be lost, this process could just as well reflect a tendency to elevate the inaction alternative. Second, even if one accepts the idea that a tendency exists to elevate options that might be lost in the context of the endowment effect, it would not imply acceptance of loss aversion itself; that is, the acceptance of a general principle whereby losses loom larger than gains. In particular, to accept a general principle of loss aversion would, at the least, require evidence that losses loom larger than gains across different contexts, including in contexts where losses and gains are not confounded with inaction and action.

Status quo bias and loss aversion

The status quo bias, the name given for individuals’ propensity to prefer the status quo to an alternative option, has been attributed to loss aversion (Kahneman, Knetsch, & Thaler, 1991) and thus taken as evidence supportive of loss aversion. In particular, the loss aversion account suggests that the loss of the status quo option looms larger than the gain of an alternative (change) option. However, Ritov and Baron (1992) provided evidence that the status quo bias was not a propensity to remain at the status quo per se, but a propensity to favor inaction over action (i.e., omission over commission).

In particular, Ritov and Baron showed that when presented with a choice that involved the option to do nothing or to do something, people tended to choose to do nothing; this decision resulted in a tendency toward the choice of the status quo option when doing nothing maintained the status quo, but a tendency toward the choice of the change option when doing nothing resulted in a change from the status quo. Others have found that a propensity toward the status quo sometimes persists even when action is required to maintain the status quo (Schweitzer, 1994), though Ritov and Baron (1992) did not find this to be the case.

Regardless, acceptance of the idea that individuals tend to favor inaction over action (rather than to favor the status quo over change per se) does not preclude the loss aversion explanation for the status quo bias. Instead, this observation merely qualifies the loss aversion explanation: if loss aversion explains the status quo bias, then the reference point must be inaction (i.e., the default situation of doing nothing) rather than the status quo. In other words, it is not the loss of the status quo that looms larger than the gain of the alternative; rather what is to be lost by action looms larger than what is to be gained by action.

At the same time, a propensity toward inaction does not, by any means, require loss aversion. Gal (2006)’s inertia account states that when people are indifferent between options, they should favor inaction over action because doing something requires a psychological motive. Alternatively, a preference for inaction might occur because individuals will tend to favor options that reduce processing and transaction costs. Other explanations for a propensity toward inaction are that errors of commission tend to involve greater regret than errors of omission (Ritov & Baron, 1995) and that individuals might rely on an “if ain’t broke, don’t fix it” heuristic (alluded to by Baron & Ritov, 1994).

To illustrate that loss aversion is not required to explain the status quo bias, Gal (2006) asked participants if they would trade one good (a quarter minted in Denver) for an essentially identical good (a quarter minted in Philadelphia). Kahneman (2011) has noted that loss aversion does not come into play when individuals exchange essentially identical goods (e.g., when trading one $5 bill for five $1 bills) because people do not code such exchanges in terms of losses and gains. Nonetheless, Gal (2006) found that more than 85% chose to retain their original quarter. We recently replicated this result by asking 149 MTurk participants whether they would prefer to trade a $20 bill they were slated to receive for another $20 bill (i.e., the change option) or to stick with the original $20 bill they were slated to receive (the status quo option). In one version, participants were only able to choose between these two options, whereas in another version, participants were able to indicate that they were indifferent between the options. Although, according to Kahneman (2011), loss aversion should not come into play in this context because the exchange would not be coded in terms of losses and gains, we observed a clear tendency of participants to indicate a preference for the status quo option (see Figure 1). Thus, again, the presence of a status quo bias should not be viewed as evidence of loss aversion.

In sum, the mere presence of a status quo bias (or inaction bias) does not provide insight into whether losses loom larger than gains. The status quo bias might be caused by the loss of the status quo looming larger than the gain of an alternative, but it might equally be caused by any of a number of other factors that lead toward a propensity toward inaction (and/or a propensity toward the status quo). As such, the presence of a status quo bias, in and of itself, cannot be taken as tantamount to evidence for loss aversion.

So the one that still stands is confirmation bias?

It turns out that loss aversion does exist, but only for large losses. This makes sense. We *should* be particularly wary of decisions that can wipe us out. That's not a so-called "cognitive bias". It's not irrational. In fact, it's completely sensical. If a decision can destroy you and/or your family, it's sane to be cautious.

 

It sounds like much of loss aversion is just an intuitive use of the Kelly Criterion?

It's hard to believe that scientists would deliberately manipulate their findings. The risk of getting caught and discredited is just too high – oh wait.

Link to pre-publication (but presumably near-identical) version of the 2018 paper: https://ie.technion.ac.il/~yeldad/Y2018.pdf.

Scott Alexander has written an in-depth article about Hreha's article:

The article itself mostly just urges behavioral economists to do better, which is always good advice for everyone. But as usual, it’s the inflammatory title that’s gone viral. I think a strong interpretation of behavioral economics as dead or debunked is unjustified.

See also Alex Imas's and Chris Blattman's criticisms of Hreha (on Twitter).

Here is a little detail I learned in behavioral finance class: you don’t need behavioral finance/econ to discover loss aversion. All you need is a rational utility maximizing agent in a standard neoclassical framework who has a concave utility function (such as LOG which is commonly assumed to model diminishing marginal utility). From this you see that the rational agent has more to loose from a one unit negative change than a one unit positive change i.e. loss aversion.

I'm not sure it's any more dead than other fields of social science. Which, maybe they're all actually zombies, but that sounds excessively strong. For example, take the effect sizes of nudges. I believe that the effect of "opt out" policies for organ donation have absolutely massive effects (see https://sparq.stanford.edu/solutions/opt-out-policies-increase-organ-donation ). So is the problem that the field is dead, or that it's just sick with the same diseases as psychology and better work needs to be done to separate wheat from chaff? Forgetting hypotheses that turn out not to hold up, doing more replications, etc. For example, I believe hindsight bias has held up as being real, having significant effects, and being difficult to overcome. 

Does this suggest they don't hold to loss aversion in any sense? I'm taking the claim, selective analysis/data presentation, at face value here. If true that seems like it would suggest a very significant loss to current and future status as well as position and potential future positions.

I'd like to recommend a book named Radical uncertainty, it does a great job at criticizing Behavioural economics (among other things) and how we should get many of their results with a pinch of salt. I think this community specifically can benefit greatly from it

I would be interested to read a review of it on LessWrong. (I have also not read the book, and do not have the book either.) The only review I found that was not just a summary of the book described the authors' recommendations as "Their alternative to probability models seems to be, roughly, experienced judgment informed by credible and consistent “narratives” in a collaborative process." That sounds to me like dressing up the non-apple of "not using probability models" as a banana.

(surprised) No way!! I bought that book three months ago, at the recommendation of no one. I haven't read it yet, but it's good to see that I have made a good investment on my own judgment.