HANDICAPPING AND BEHAVIORIAL FINANCE

[Que]

I was interested in getting other people's thoughts regarding handicapping and how it relates to a new field that some psychologists are calling "Behavioral Finance." Behavioral Finance applies to how people make decisions, i.e. financial decisions, when the outcome of an event is uncertain. As handicappers, we all know that there are countless decisions that we have to consider before placing a bet, and there are an equal number of counterproductive behaviors that can lead us to make irrational or poor decisions.

Some of these irrational behaviors were recently described in a book, "A Mathematician Plays the Stock Market," by John Allen Poulous, and should apply to handicapping as well. My favorite counterproductive behavior, which I think has the most application to handicapping, is described as "Anchoring." Anchoring occurs when people become "attached" to a number, actually any number, we might see or hear. An example of this is described in the book, where an experimenter asks people to estimate the population of the Ukraine, the distance to Saturn, or some other question where the person is uncertain of the actual answer. Specifically, when the subjects where asked if the "population of the Ukraine was more or less than 200 million people?"--the subject's estimate would vary, but were generally a bit less than the figure given, i.e. averaging about 175 million. However, if the experimenter prefaces his question, "Is the population more or less than 5 million?"--the subject's response would still vary, but this time their responses were generally more than what was prefaced--averaging about 10 million. In these instances, the subjects estimate would usually move in right direction from whatever number was presented to them, but their estimates were nevertheless anchored to the prefaced figure.

Now this all makes perfect sense, since the average person doesn't know what the population of the Ukraine is, hence they will assume the experimenter is knowledgeable so they will stick close to the number presented. Now what's really interesting is when the experimenter obtains the preliminary number by some chance means, i.e. by spinning a wheel that has numbers around its periphery, e.g. 300 million, 200 million, 100 million, 50 million, 5 million, and so on. In this case, the experimenter would spin the wheel in front of the subjects, point out where it stopped, and then ask the subjects what was the population of the Ukraine was, and if the population was more or less than the number on wheel. Of course the subject’s responses still varied, but what’s really amazing is that the subject’s responses were still anchored to the figure displayed on wheel--even though the subject’s had to know that the wheel knows nothing about the Ukraine!

Now consider this same irrational behavior as it might pertain to handicapping, i.e. trying to determine the horse's actual probability of winning vice having everyone’s judgment anchored by a number, the morning line, etc. Moreover, based on this research, even if the morning line was created totally by chance (i.e. using a wheel of potential odds), the public’s estimated probabilities would still be anchored to this faulty information—no matter how inaccurate or random the initial estimates might be. Anyway, I found this concept rather fascinating...

[cratman]

Doc Sartin provided useful insights concerning behavioral finance with regard to the value of risk profiles, in connection with a project I worked on. I understand that his conclusion and analysis, base on his work with gamblers, was confirmed by perhaps the leading proponent in this field, Professor Thaler of the University of Chicago. This indicates that, at least for someone with the background and ability of Doc Sartin, lessons and experiance from each field can be used in the other.

[Dave Schwartz]

Que,

Book is so-ordered from Amazon.

Thanks.

Dave

[Robert Fischer]

I will at times walk a thin line between using a point of reference and full out "anchoring". In another thread I mentioned a probability for an underlayed favorite being between .20-.30 and probably in the low .20s...
Now theres a good possibility I should have guessed lower, but then theres that friendly 33% win probability for the "average" favorite...
Hey it is "scary" if we are handicapping an outcome that is off of the familar paths and different from our familar numbers ;).
The crowd (and the mass of data) is easier to predict than an individual or specific outcome - providing you know what to look for.
Behavorial Finance has its place in horse racing.

[PlanB]

Oh yes, Robert Fischer, behavorial finance is now. At work there was a memo about China & it read like a social psych letter. In another thread there was the Krugman NYT article & that included social forces. Freakanomics 102?

ps: I think I took your post the wrong way.

[traynor]

I was interested in getting other people's thoughts regarding handicapping and how it relates to a new field that some psychologists are calling "Behavioral Finance." Behavioral Finance applies to how people make decisions, i.e. financial decisions, when the outcome of an event is uncertain. As handicappers, we all know that there are countless decisions that we have to consider before placing a bet, and there are an equal number of counterproductive behaviors that can lead us to make irrational or poor decisions.

Some of these irrational behaviors were recently described in a book, "A Mathematician Plays the Stock Market," by John Allen Poulous, and should apply to handicapping as well. My favorite counterproductive behavior, which I think has the most application to handicapping, is described as "Anchoring." Anchoring occurs when people become "attached" to a number, actually any number, we might see or hear. An example of this is described in the book, where an experimenter asks people to estimate the population of the Ukraine, the distance to Saturn, or some other question where the person is uncertain of the actual answer. Specifically, when the subjects where asked if the "population of the Ukraine was more or less than 200 million people?"--the subject's estimate would vary, but were generally a bit less than the figure given, i.e. averaging about 175 million. However, if the experimenter prefaces his question, "Is the population more or less than 5 million?"--the subject's response would still vary, but this time their responses were generally more than what was prefaced--averaging about 10 million. In these instances, the subjects estimate would usually move in right direction from whatever number was presented to them, but their estimates were nevertheless anchored to the prefaced figure.

Now this all makes perfect sense, since the average person doesn't know what the population of the Ukraine is, hence they will assume the experimenter is knowledgeable so they will stick close to the number presented. Now what's really interesting is when the experimenter obtains the preliminary number by some chance means, i.e. by spinning a wheel that has numbers around its periphery, e.g. 300 million, 200 million, 100 million, 50 million, 5 million, and so on. In this case, the experimenter would spin the wheel in front of the subjects, point out where it stopped, and then ask the subjects what was the population of the Ukraine was, and if the population was more or less than the number on wheel. Of course the subject’s responses still varied, but what’s really amazing is that the subject’s responses were still anchored to the figure displayed on wheel--even though the subject’s had to know that the wheel knows nothing about the Ukraine!

Now consider this same irrational behavior as it might pertain to handicapping, i.e. trying to determine the horse's actual probability of winning vice having everyone’s judgment anchored by a number, the morning line, etc. Moreover, based on this research, even if the morning line was created totally by chance (i.e. using a wheel of potential odds), the public’s estimated probabilities would still be anchored to this faulty information—no matter how inaccurate or random the initial estimates might be. Anyway, I found this concept rather fascinating...

Anchoring is similar to two other psychological functions (similarly used in marketing and sales) called "priming" and "sleeper effect." Anchoring is closest to priming when used intentionally; for example, the subject is provided a number or reference point that may be unrelated, but apparently meaningful. The tendency is to fasten on the "solid fact" as if it applies to the subject at hand. Russo and Schoemaker refer to it as "strategic anchoring," and it is a commonly used strategy in negotiations.

Classic example from utility company, "You may have heard the rumor that rates are going to increase by 40%." Generic blah blah follows, tagged with "Fortunately, the rates will only be raised by 15%." Similarly, "Looks like a Ferrari, doesn't it?" General blah blah follows. When the price comes up, the subject has anchored the price of a Ferrari as the frame of reference.

Consider the strategy of realtors to show very expensive, very cheap, and intended to sell in that sequence. There is a reason for it. People tend to be "cognitive misers," and use anchoring as a heuristic shortcut in decision making.

The "sleeper effect" is like anchoring, but takes place a longer period of time before the event in question.

Applied to handcapping, consider the effect of seeing Beyers or Average Pace of one entry higher than the others; anchoring works in conjunction with cognitive dissonance to make the subject believe both that the "anchored" entry is superior, and that conflicting (but equally relevant) evidence is not as important.
Good Luck

[1st time lasix]

It is possible that number " anchoring" you describe is more likely to occur with a big Beyer than the morning line? Say if a recent 2nd place finisher ran a 90.....then the vast majority of recreational handicappers with a working knowledge of speed figures will throwout all the horses that haven't run a 85 or so. Of course that 90 may be a out of line or a potential "bounce" race that isn't going to be repeated today. A longer shot may be a horse than is about to improve....say to a 85-90 that puts him right in the exotic mix at an overlay price. Another extention of that theory may be recreational handicappers who let the tote board influence their handicapping. Say an entry is an early 8/5.....some players may just assume that horse will run 1st or 2nd...even though the pace scenario, distance or surface may not suit. I don't know,,,,,,The more I play this game .......the more I think it comes down to only playing races where the chalk favorite {or perhaps the second popular choice} of the wagering public is truly vulnerable. If you don't think that is the case....you must pass. Wait for another race or where one of your "watch" horses is entered. In those cases where you expect "vulnerability" because of sound handicapping fundamentals....you must seek the overlays, play within your money management bankroll and structure your tickets to the correct pools accordingly based on your experienced opinion. I hope everyone has a very Merry Christmas....and gives thanks to the brave young men and women in uniform for our Country all over the world. :ThmbUp:

[Dave Schwartz]

Wow! I never heard of this "anchoring" before, yet it is so obviously a part of all of our lives.

I definitiely see it in our behavior with regards to the tote board. ("I make this horse 4/1 and... Wow! He's 18-1 on the board. I must have this wrong!)

Dave

[robert99]

Politicians also use "the staus quo is not an option" anchor strategy when they want to convince people to change something that actually works, to some other thing that will never work. People then suggest lesser evil alternatives, but do not often argue for retaining the status quo which they actually believe in. The politician has set the rules, and people tend to obey the rules within that given framework, rather than writing their own - the politician must have researched this better than me syndrome. Nazi Germany was the most tragic example.

Anchoring theory makes the "efficient market" theory in racetrack betting seem even more of a bizarre concept. What is going to happen to the all the thousands of "me too" economics papers on the EM subject?

[Overlay]

I think that maximizing the objectivity of the handicapping process to the greatest extent possible is a way to avoid this. If you're working with characteristics with known probabilities, and especially if you have confidence in their reliability based on your own experience with them, it makes it easier to filter out contrary opinion and other subjective "noise" (including the morning line).

[singunner]

Please don't obfuscate interesting points by invoking Godwin's Law...

[traynor]

The real problem with anchoring is that people want to maintain a self-image of insightful thoughtfulness regarding decision-making. The notion of anchoring (and a variety of related topics) is "uncomfortable," because it threatens the cherished self-image of logical, rational decisions. Unfortunately, that increases the effect, as anyone in sales or marketing is well aware.

While it may be mildly amusing to realize that you have been manipulated into (in Robert Cialdini's famous example) buying Girl Scout cookies, it is not as funny to realize that you just lost a major wager because of your "belief" that Horse A was "better" than Horse B, all based on trivial evidence (close finish last, big Beyer, etc.)

If there is any topic that is of value to a handicapper, it is understanding his or her own process of decision-making.
Good Luck

[thoroughbred]

Que,

Reading your message about "anchoring" reminded me of another, but related, experiment that was performed and reported in Scientific American a number of years ago.

A fairly large number of people were assembled in a room that had a projection screen in front for all to clearly see. Then various well known objects were projected on to the screen. But the way it was done, was that when an object first appeared it was completely out-of-focus, i.e., difficult to determine what it was. Then slowly, the focus was changed to make the object become, slowly, clearer and clearer.

Each person had to press a button as soon as they thought they could identify what the object was. The time was noted, and the person had to write down what he thought he was seeing. They then were allowed to change there minds when the object became even more clear.

Now here is the interesting result. Those who made their decision the fastest, saying, for example that the object was "fire hydrant", kept to their decision even long after it was very clear that the object was a "tree."

It was an example of how humans, once we make our minds up, find it difficult to change our minds even in the face of obvious evidence.

I'm sure we handicappers suffer from this malady. We may make our minds up about a possible winner early in the handicapping process, and find it difficult to let go of the conclusion in the face of subsequent facts about additional horses.

[PlanB]

We may make our minds up about a possible winner early in the handicapping process, and find it difficult to let go of the conclusion in the face of subsequent facts about additional horses.

Amen brother. How often I've done that.

[robert99]

Que,

Reading your message about "anchoring" reminded me of another, but related, experiment that was performed and reported in Scientific American a number of years ago.

A fairly large number of people were assembled in a room that had a projection screen in front for all to clearly see. Then various well known objects were projected on to the screen. But the way it was done, was that when an object first appeared it was completely out-of-focus, i.e., difficult to determine what it was. Then slowly, the focus was changed to make the object become, slowly, clearer and clearer.

Each person had to press a button as soon as they thought they could identify what the object was. The time was noted, and the person had to write down what he thought he was seeing. They then were allowed to change there minds when the object became even more clear.

Now here is the interesting result. Those who made their decision the fastest, saying, for example that the object was "fire hydrant", kept to their decision even long after it was very clear that the object was a "tree."

It was an example of how humans, once we make our minds up, find it difficult to change our minds even in the face of obvious evidence.

I'm sure we handicappers suffer from this malady. We may make our minds up about a possible winner early in the handicapping process, and find it difficult to let go of the conclusion in the face of subsequent facts about additional horses.

Seems to explain the lynch mob mentality quite well.

[DaveP]

.. as interesting as the subject is, I dont think its relevant to handicapping. I believe this phenomena is taken care of by the longshot bias and market correction factor.

Sorry to disagree.

[traynor]

Que,

Reading your message about "anchoring" reminded me of another, but related, experiment that was performed and reported in Scientific American a number of years ago.

A fairly large number of people were assembled in a room that had a projection screen in front for all to clearly see. Then various well known objects were projected on to the screen. But the way it was done, was that when an object first appeared it was completely out-of-focus, i.e., difficult to determine what it was. Then slowly, the focus was changed to make the object become, slowly, clearer and clearer.

Each person had to press a button as soon as they thought they could identify what the object was. The time was noted, and the person had to write down what he thought he was seeing. They then were allowed to change there minds when the object became even more clear.

Now here is the interesting result. Those who made their decision the fastest, saying, for example that the object was "fire hydrant", kept to their decision even long after it was very clear that the object was a "tree."

It was an example of how humans, once we make our minds up, find it difficult to change our minds even in the face of obvious evidence.

I'm sure we handicappers suffer from this malady. We may make our minds up about a possible winner early in the handicapping process, and find it difficult to let go of the conclusion in the face of subsequent facts about additional horses.

Very famous experiment by psychologist Asche, on the length of lines. The interesting part is that it is tied to group opinions (comparable to morning lines, rankings, expert opinion), and the difficulty of forming an "independent" opinion that differs from the group.

Another interesting topic for handicappers is the effect of consistency and commitment in combination. Either is a strong motivator alone, but they are expecially compelling in combination. When a handicapper has made his or her mind up, most would rather lose than admit the original decision was flawed; that is consistency. When that opinion is stated to someone else (or a wager made), the handicapper becomes locked into the decision; that is commitment.

In a job interview, a typical ploy is to ask the interviewee, "What are you going to do for us?" Verbalization of the intent to work like a maniac, 70 hours a week or more, thinking only of profit for the organization mandates that the interviewee, if hired, is internally compelled to perform consistent with the declaration of intent.

The same effect makes people bet on the wrong horses, because they are unwilling or unable to modify their decisions. The trick is to avoid deciding until later in the game. Once you pick Horse A over Horse B, there is a tendency to only process information that supports the decision, and to ignore information that conflicts with the decision.

The phenomena was explored at length by Festinger, under the label "cognitive dissonance."

[andicap]

.. as interesting as the subject is, I dont think its relevant to handicapping. I believe this phenomena is taken care of by the longshot bias and market correction factor.

Sorry to disagree.

Im not so sure. I was once at Belmont with my cousin and I loved a 27-1 shot and told him about it. He didn't bet it, reasoning that if the horse was any good it would have been bet more. It did win.

[valueguy]

Interesting concept.
I wonder if i could tighten up my odds line by anchoring my starting point to
the number of horses in the race( eg 10 horses every horse starts at 9/1 )
then work up from there .At the moment i scan for the highest speed or pace fig and work from there.
Its worth a try:).

[traynor]

Interesting concept.
I wonder if i could tighten up my odds line by anchoring my starting point to
the number of horses in the race( eg 10 horses every horse starts at 9/1 )
then work up from there .At the moment i scan for the highest speed or pace fig and work from there.
Its worth a try:).

Your comment is a perfect example of how anchoring functions; if you fasten on a particular number/rating/figure, the subsequent processing is less objective. That is, the tendency is more to support the initial (even very tentative) evaluation than to disconfirm it. The entire view of anchoring as a "problem" is that it tends to diminish objectivity; once the anchor is activated, it functions to decrease objectivity, which in turn diminishes the ability to make quality decisions.

Again, this is not my individual opinion; it is a simplified explanation of the view held by the majority of experts and scholars in the fields of decision-making and decision theory.

There is, of course, the flip side. If what you are doing is working (meaning "very profitable"), don't abandon it because it doesn't match theory. Milk it dry and THEN look for a newer, better way. ;)
Good Luck