on January 16, 2012
I am deeply skeptical of psychological studies that purport to extrapolate from the general to the particular as to human behavior. I think that approach runs afoul of what Dr. Kahneman criticizes, substituting an easy answer for the hard question. The whole analysis of prospect theory and loss aversion seems to me flawed in that respect. We often resort to group averages to cover up our lack of understanding as to the physical reasons an individual brain operates the way it does. Witness the alcoholic or the gambling addict, who pay no attention to loss aversion.
When I was in college and took sociology and psychology courses, the joke on campus was that both disciplines could be described by the phrase: "The science of what everybody knows." Dr. Kahneman likes to say that the results of the experiments he describes are "surprises," but I really don't think so. From my perspective, using my, as Dr. Kahneman describes, System 1, I found most of the results of the studies he described as unsurprising and a match for my intuitive (counter-intuitive). As another reviewer commented, Dr. Kahneman's studies are flawed in a few ways: (a) they extrapolate from general averages to a particular; (b) he defines intuitive based on those averages, rather than on individual brain chemistry and how a particular individual may think; (c) he designs experiments on loss aversion that guarantee a certain result, because the students in the experiment do not get the benefit of the law of large numbers. Having presented these criticisms, however, i think his book makes a valuable contribution to people who read it and want a better understanding of how most people think.
The problem I have with Dr. Kahneman's approach is that it is a classic example of that in which he finds fault. Substituting the easier, irrelevant question for the harder question. Psychology experiments such as he cites are substitutes for an understanding of the physical brain functions that cause an individual to behave in a certain way. Neither gambling addicts nor alcoholics nor serial murderers exhibit any loss aversion whatsoever.
So, to cite one particular example, here is what I wrote the authors of a study Dr. Kahneman cites in support of loss aversion and his prospect theory.
Dear Dr. Schweitzer and Dr. Pope:
I read with interest your study after reading Thinking Fast and Thinking Slow.
I have some questions/observations that I would very much appreciate your comments.
By way of background, I have been a trial lawyer for 40 years and have a substantial interest in human behavior and loss aversion and related concepts of risk.
After reading your article I remain a bit skeptical about the connection between your results and loss aversion, at least insofar as the latter is viewed as non-rational. I should admit I have a bias against psychological or any kind of group behavior that attempts to establish a causative link with individual behavior, but having admitted my bias, I have the following to say:
1. Did your study examine the statistical comparison in the last nine holes of a tournament by a tournament leader who is ahead by four strokes or more, or any other number in the lead. Since, for example, the odds of the second place player proceeding to four under par for the last nine holes seems low, might it not be rational for the leader to avoid bogeys, regardless of whether putting or conservative play in general is concerned.
2. Some golfers (perhaps many) are leads concerned about winning than ending up in the money. The tour is replete with examples of top annual money winners having won few if any tournaments. For those golfers who are not intensely interested in winning, their putting may be differently motivated.
3. I'd be interested if you looked at the following correlations:
a. Percentage record in your study by individual golfer versus wins and where they finished on the money list;
b. The individual golfer's putting performance as related to drives in fairway, or greens in regulation;
c. None of the statistics currently promoted by the PGA, e.g. drives in fairway, sand saves, putting, greens in regulation, etc. seem to correlate well with winnings.
4. I noticed that some of the golfers with the lowest percentage difference in your study were relatively unsuccessful in winning tournaments, e.g. Sergio Garcia, Justin Leonard, John Daly, where Tiger Woods with a fairly high or average percentage was the top winner by far until a couple of years ago.
5. As in football turnover ration is there a correlation between the putting percentages you found and winning. In football, there is a high cost to pay for turnovers (in field position, lost opportunity to score, for example), which may be the reason coaches stress the importance of avoiding risk by engaging in less aggressive play (carrying the ball with one hand enhances the runner's ability to avoid tackles or)
the longer the pass, perhaps the greater chance for an interception or conversely, the greater chance for a big gain).
6. A golfer, as a young player, often participates mostly in match play. This may "train" the player to think in terms of avoiding bogeys at all cost.
7. Did you analyze the casual player, who is not playing competitively. He may be more aggressive in trying for birdies. Your acknowledgment that the percentage difference declines in the last two rounds seems to indicate what I would call the "match play" mindset, because birdies may become more important during the last two rounds as having a larger psychological effect on a competitor who sees the birdie on the scoreboard from a close competitor.
8. Downhill putts are generally considered more difficult than uphill putts; did you analyze those statistics. Your division of the green in segments may be to broad.
9. Golfers often say that putting past the hole offers a better chance of making the next putt, because they immediately see the correct line. Does your study contradict that approach?
10. Is the prospect theory of loss aversion undermined, at least a little, by the relative lack of success that appears to come to an aggressive player, e.g. Phil Mickleson's early relative lack of success, until he played more conservatively. Is there a relationship between the high percentage putt players and conservative play in general? (But see John Daly-aggressive- vs. Justin Leonard-conservative) Or, does aggressive play from tee to green correlate with conservative play on the green?
11. Is there a relationship to examine between the leader after two rounds and the ultimate winner?
I have other similar observations, but I suppose the ones above are all based on wondering whether you should take into account each golfer's overall play before deciding whether a particular birdie vs. par putting percentage is consistent with loss aversion theory in general?
Thank you in advance for reading this email. I certainly found Dr. Kahneman's book and studies and your own study as thought provoking.