Combo Factors

[ilhcp]

I'm going to ask a question that yes I know in advance will not lead anyone to the win window.

I'll also preface the question by saying the only purpose of asking the question is to assist myself (and others) in creating a better betting line.

Say for example you have 3 independant handicapping factors that each are going to win 30% of the time: this could be 3 public selectors OR maybe a combination of a power number, speed figure and an overall software selection OR any combination of 3 things that independantly predict 30% winners.

For the GENERAL "note general" purpose of a betting line...will this 30% become 31%, 33%, 36%, stay at 30% or approach 39%?

I could go a step further and say what if the factors were 35% or MSW races etc? But really just looking for a simple answer base on your experience or better yet.....your database?

I have some general vibes based on my research but I always like to cross check it with someone elses experience?

[Larry Hamilton]

There is no way to know this:

For the GENERAL "note general" purpose of a betting line...will this 30% become 31%, 33%, 36%, stay at 30% or approach 39%?

you must test the combination as if it were a completely new and INDEPENDENT variable.

[Rick]

The "general" answer is that the win % will only go up a little and the ROI may get worse unless the factors are relatively independent of one another (a difficult task to accomplish). The best win % is achieved by using toteboard odds as one of the factors in combination with another good method. That would get you a more accurate line but the top rated horse would lose money.

As an example, when my own method agrees with the morning line favorite I get 39% winners but most of them would be underlays. Those that don't wind up being the post time favorite show a little profit but with a much lower win %. It seems that you can't have it both ways unless you play only a very small percentage of the races.

[Rick]

I should probably elaborate further. Combining my method and the morning line in EVERY race would get about 33% wins. Combining my method and actual odds would get about 36% wins. Dave Schwartz has done some research in this area and probably has better numbers than that.

[pmd62ndst]

In my experience, it's better to spend more effort trying to increase your ROI % then your Win %. Finding an angle that doesn't affect your Win % but increases your ROI % means you found something that the betting public doesn't value as much and should be guarded with your life!

Here's another way to look at it:

Let's say a particular method gives you the following:
30% Wins, -10% ROI.

Scenario A:
Adding Factor A will result in this:
35% Wins, -10% ROI.

Scenario B:
Replace Factor A with Factor B and you get this:
30% Wins, -3% ROI.

Scenario A happens all the time. Betting this factor will not help you. Scenario B happens rarely. Cherish this when you find it.

PMD

[Rick]

PMD,

I couldn't have said it any better. You're exactly right.

[ilhcp]

Thanks. But I just wanted this thread to stay on it's original track. That was "win percentage?" in an effort to calculate a line. Not ROI's. Or to not play against a legitimate legitamate favorite, etc. Forget the ROI's I know they're gonna suck.