PLAYING "WITH" THE PICK 3

[formula_2002]

One the outset let me say that no small effort went into this, but this is what I do :)

A sample of some of the stuff;
I looked at 430 pick 3's who's 1st , 2nd and 3rd leg odds were <10-1.
Everything was figured for a 18% win pool take, this inclued tracks where the take's were + or - 18%.
All pick 3 payoffs were based on 25% take.

37% of the payoffs exceeded the expected return, that is actual returns/ expected returns were >=1.

some of the most notable;
evd 8/19/2006 4th race
crc 8/11/2006 8th race
del 8/16/2006 6th race.

you can check the results free at
http://www.equibase.com/premium/eqbSummaryResultsDisplay.cfm?TRK=EVD&CY=USA&DATE=08/19/2006&STYLE=EQB

[formula_2002]

My pick 3 analysis program is completed, and if correct here are some stunning facts.

In my Aug data base, Bris Top Prime fig returned a win pool roi of .84, which is about expected, however, when the top Prime figure is singled in the pick three's the roi fell to .44.
The top Bris speed returned a win pool roi of .84 and a pick 3 roi of .25
The ML fav, returned a win pool roi of .77 and a pick 3 roi of .44
.
The prgram takes any single factor and compares the win pool roi with a pick 3 roi.

[ryesteve]

I looked at 430 pick 3's who's 1st , 2nd and 3rd leg odds were <10-1.
Why would you add this restriction? I can understand ignoring the oddball Pick 3 that comes back $5,000, but I don't think any analysis skewed towards chalky P3's will ever lead to anything that looks promising.

[formula_2002]

Why would you add this restriction? I can understand ignoring the oddball Pick 3 that comes back $5,000, but I don't think any analysis skewed towards chalky P3's will ever lead to anything that looks promising.

For now, I'm just throwing my hook into the water and seeing what comes back.

[ryesteve]

For now, I'm just throwing my hook into the water and seeing what comes back.
You gotta put at least a little bait on that hook... :D

[rrbauer]

My pick 3 analysis program is completed, and if correct here are some stunning facts.

In my Aug data base, Bris Top Prime fig returned a win pool roi of .84, which is about expected, however, when the top Prime figure is singled in the pick three's the roi fell to .44.
The top Bris speed returned a win pool roi of .84 and a pick 3 roi of .25
The ML fav, returned a win pool roi of .77 and a pick 3 roi of .44
.
The prgram takes any single factor and compares the win pool roi with a pick 3 roi.

Just so I understand your methodology.

You are using the Top whatever Fig horse in all three legs of a P3? And then you use the P3 payouts to measure the results of your P3 bets?

Assuming that is the case, how is it correct to compare the ROI from a single independent event to the ROI from three connected events where a dependency is established in each of the downstream legs from the preceding leg(s)?

[formula_2002]

Just so I understand your methodology.

You are using the Top whatever Fig horse in all three legs of a P3? And then you use the P3 payouts to measure the results of your P3 bets?

Assuming that is the case, how is it correct to compare the ROI from a single independent event to the ROI from three connected events where a dependency is established in each of the downstream legs from the preceding leg(s)?

RR..
I'm not trying to make "things happen", just trying to understand what "Is happening"

[formula_2002]

You gotta put at least a little bait on that hook... :D

I'm trying!! ;)

[The Judge]

Keep us posted.

[formula_2002]

One of my life’s racing axioms is, “If your picks don’t produce a profit in the win pool, they will not show a profit in any other pool”.

To give a bit of substance to the axiom, I have produced numerous computer runs on figures that return about a .79 to.84 win pool roi. (Most any published figure will be about .79 to .84)
When singled in a pick three, that same figure returns much less in the pick three pools (and much less than the pick three pool vig ) and if you want to combine your top fig and play it with your second figure, it returns even less.
It also seems that a profitable win pool figure becomes even more profitable in the pick three pool. But I don’t have many of those profitable figures to test. ;)

[Robert Fischer]

One of my life’s racing axioms is, “If your picks don’t produce a profit in the win pool, they will not show a profit in any other pool”.

To give a bit of substance to the axiom, I have produced numerous computer runs on figures that return about a .79 to.84 win pool roi. (Most any published figure will be about .79 to .84)
When singled in a pick three, that same figure returns much less in the pick three pools (and much less than the pick three pool vig ) and if you want to combine your top fig and play it with your second figure, it returns even less.
It also seems that a profitable win pool figure becomes even more profitable in the pick three pool. But I don’t have many of those profitable figures to test. ;)

Thats a great way to put it. I have to agree with your last point as well.

By their nature, Pick 3's will test any systems roi.
Interesting study.

[formula_2002]

Thats a great way to put it. I have to agree with your last point as well.

By their nature, Pick 3's will test any systems roi.
Interesting study.

I can now see my way clear to analyzing pick 3 returns based upon multiple selections in any of the legs..
For example the ticket could look like 3 x all x 1, or any combination thereof.
Of course, the program is not limited to pick 3’s, it could include any quanity of picks.
It’s a bit of work, but the methodology is very clear..

I have to think about some broader applications other than horse racing.