When Deciding How to Bet, Less Detailed Information May Be Better

[atlasaxis]

Key takeaway: “In everyday life, people often try to be specific to be accurate,” observe Suk and colleagues, but this new research suggests that specificity and accuracy don‘t necessarily go hand-in-hand. In weighing detailed information, we tend to give “greater weight to attributes that are more salient, justifiable, and easy to articulate,” say the researchers. As a result, we often lose sight of more general attributes that actually matter.

Read the rest here:

http://www.sciencedaily.com/releases/2013/05/130513123341.htm

[Overlay]

When confronted with the amount of detailed data available in the past performances, the key for me has been to separate the "significant few" from the "trivial many". I have found impact values particularly helpful in that regard by allowing the comparison of factors using a common scale.

[GaryG]

The Triple Crown is the best example of info overload. Best advice is to handicap the TC races the same way you handicap every other race. You probably know more than the talking heads dispensing the often trivial info anyway.

[jfdinneen]

atlasaxis,

You may wish to review this article from 1999 - Do You Really Need More Information? (http://www.au.af.mil/au/awc/awcgate/psych-intel/art8.html) .

John

[CincyHorseplayer]

If you have this level of trepidation you haven't arrived yet.I know because I'm just out of that phase.

[Magister Ludi]

Many thanks to Mssrs. Atlasaxis and Dinneen. Somewhat tangentially related to the subject is the Feynman Problem-Solving Algorithm, proposed by Dr. Richard Feynman: (1) write down the problem; (2) think very hard; (3) write down the answer.

You may find #1 to be the most difficult of the three steps.

[jfdinneen]

Magister,

Feynman is "primus inter pares".

However, in the narrow world of handicapping, it is always worth asking yourself Bill Benter's fundamental question of handicapping: what additional variable(s) (your edge) do you have that could explain a significant proportion of the variance in a horse's past performances, which is not already accounted for by the public odds (wisdom of crowds)?

John

[Fingal]

The Triple Crown is the best example of info overload. Best advice is to handicap the TC races the same way you handicap every other race. You probably know more than the talking heads dispensing the often trivial info anyway.

The perfect horse this year was the #9, OVERANALYZIZE.

Remember, you can't spell analyst without anal.

[Magister Ludi]

Magister,

Feynman is "primus inter pares".

However, in the narrow world of handicapping, it is always worth asking yourself Bill Benter's fundamental question of handicapping: what additional variable(s) (your edge) do you have that could explain a significant proportion of the variance in a horse's past performances, which is not already accounted for by the public odds (wisdom of crowds)?

John

Mr. Dinneen, I wholeheartedly agree. Mr. Benter has succinctly summarized the Fundamental Question Of Handicapping. Fortunately, it has been almost effortless for me to find such variables. I simply get out into left field at about 5,000 feet altitude and it all comes into sharp focus.

[pondman]

Magister,

Feynman is "primus inter pares".

However, in the narrow world of handicapping, it is always worth asking yourself Bill Benter's fundamental question of handicapping: what additional variable(s) (your edge) do you have that could explain a significant proportion of the variance in a horse's past performances, which is not already accounted for by the public odds (wisdom of crowds)?

John

Be careful that you don't assume that all high end players and whales are preparing precise, uniform representations of past performances, and then adding epsilon.

I would suggest another option of considering the size of the epsilon first. Make it large (60%.) And at that point find the simplest of the variables to make up the characteristic of a horse, which might be good enough to win. Note the word good enough. Often the simplest of the tip off variables are just that. They are simple, as the article suggests.

[atlasaxis]

atlasaxis,

You may wish to review this article from 1999 - Do You Really Need More Information? (http://www.au.af.mil/au/awc/awcgate/psych-intel/art8.html) .

John

John,

Saw that awhile back, good to read again, thank you!

Tom

[jfdinneen]

pondman,

I agree that it is irrelevant how you "divine" the variable(s) that define your edge so long as the crowd is either unaware of (best outcome) or overly discounts that information in the relevant markets.

With respect to your usage of the error term, epsilon, could you provide clarification for more general consumption.

John

[jfdinneen]

Tom,

As you are probably already aware, Kahneman, Slovic, And Tversky (1982) published the classic work in this field, "Judgment Under Uncertainty". In that context, you may wish to look at Kahneman's (Nobel in Economics, 2002) latest publication, Thinking Fast And Slow (http://www.amazon.com/Thinking-Fast-Slow-Daniel-Kahneman/dp/0374533555/ref=sr_1_1?ie=UTF8&qid=1368641894&sr=8-1&keywords=kahnemann).

John

[Valuist]

I love how they do a study on sports betting without taking into account a pointspread or moneyline. Without those factors, the study is pretty meaningless. Squares bet on teams; veteran bettors know they must get their price.

According to them, all one needs to do is get out a newspaper and just bet on the team with the better record; i.e. how the squares love to bet.

[Capper Al]

It goes both ways for me. Sometimes I'll complicate my capping and miss the obvious. Other times, i'll bet the obvious and look back after the race and find a hindden factor or angle. It's balance in the end, knowing when to go deep or not.