Interesting Statistic Developing

[HHRA Basic-Pro]

NOTE: This doesn't seem to want to format when posted like it does when edited. Sorry.

I've embarked on some research. I'm currently downloading the XML formatted results from Equibase for one month. I have a Java program that parses the XML and sucks the results into a MySql database.

I also suck the parsed CSV files from BRIS (DRF Data Files (Single)) into MySql.

The last leg of the triad is to suck the computational output of my own handicapping program into MySql.

The idea behind all this is to bang all three against one another to improve the computational algorithm of my program.

I now have 393 races in the database (about 1/3 month) and I decided, as a curiosity mostly, to see how many times the betting favorite wins... drum roll please...





All Races: 393




Favorite Winners: 126
Favorite payoff when favorite wins: 608.8999999999997

So, if I bet $2.00 on the favorite, I would bet $786 for a return of $608 and change (-$178). But here's the interesting part. This same general ratio held true for all races except the 2nd race on a card when that race was not a Maiden race.





2nd Races (non-maiden): 29




Favorite Winners: 14
Favorite payoff when favorite wins: 66.10000000000001

So, if I bet $2.00 on the favorite in a non-Maiden race #2, I would bet $58 for a return of $66 and change (+$8).





3rd Races (non-maiden): 34




Favorite Winners: 14
Favorite payoff when favorite wins: 69.7

So, if I bet $2.00 on the favorite in a non-Maiden race #3, I would bet $68 for a return of $69 and change (+$1).

ALL other races 3 through whatever, whether maiden or non-maiden, produced losing results.

It may very well be that over the entire month (I still have about 2 1/2 weeks to go) that it all evens out... but I also have a theory.

Race 1: 171 Maiden, 181 Non-Maiden
Race 2: 135 Maiden, 201 Non-Maiden
Race 3: 81 Maiden, 263 Non-Maiden

I must say I thought there would be a ratio similar to the Race 3 statistics above for Race 2... but no matter. With many tracks, especially on the West coast, the 1st race often a Maiden. I think I'm safe in saying maiden races are less predictable than non-maidens... but I'm sure that's debatable.

In a hustle, the sucker is allowed to win early only to be set up for a late smack down. The question I will be trying to answer over the course of the month is, Is the 2nd/3rd race set up to single out a favorite for betters to have an early "good feeling" about the day only to have their hopes dashed later on.

Let's look at race 8, as most cards have at least 8 races

8th Races (non-maiden): 33
Favorite Winners: 9
Favorite payoff when favorite wins: 43.2

So, if I bet $2.00 on the favorite in a non-Maiden race #3, I would bet $66 for a return of $43 and change (-$23).

Your comments are invited.

[MiJan]

Personally, I wouldn't think that the secretary would set up the races to snooker the betting public. After all, the take is the take no matter the finish.



In a hustle, the sucker is allowed to win early only to be set up for a late smack down. The question I will be trying to answer over the course of the month is, Is the 2nd/3rd race set up to single out a favorite for betters to have an early "good feeling" about the day only to have their hopes dashed later on.

[ryesteve]

It may very well be that over the entire month (I still have about 2 1/2 weeks to go) that it all evens out... but I also have a theory.

I think your first theory is the correct one... it will all even out. Perhaps not within the next week and a half, but over any reasonable length of time, I can pretty much guarantee that it will.

[Dave Schwartz]

Not sure what you are looking for as I have not followed this thread, but let me save you the trouble.

Here are all the 1st and 2nd race favorites, non-maidens from 2007. I have left out the very minor tracks (i.e. Chipewaw downs, etc.)


1-rPubCh
-------------------
WIN BETS
Field1 Field2 Starts Pays Pct $Net IV PIV
--------------------------------------------------------
1st race 2,982 1,059 35.5 $1.59 2.36 0.95
2nd race 3,407 1,189 34.9 $1.62 2.50 0.98



Regards,
Dave Schwartz

[HHRA Basic-Pro]

Those figures mirror mine pretty much except for the race 2 and 3 non-Maiden. I'm sure you're right and it will flatten.

[stuball]

363 races does not make a database....small sample...
turf or dirt or poly?

right track but road needs to a lot longer.

JMHO

Stuball