Do database players capture all extant wager value?

[lansdale]

As the question implies, I don't use a database, and wondered whether those who do have access to the complete range of whatever their handicapping indicates would be value bets across the full range of available tracks on any given day.

Any help appreciated.

lansdale

[ArlJim78]

i think that is a standard feature of most programs. I can speak only about jcapper which is what I use. I can download the cards from all around the country on any given day, process changes, calculate them according to my predefined models, and it will give me a summary of the plays I'm looking for. All in about 5 minutes.

Of course it takes some work to set up your criteria as to what you're looking for. But you do that in advance.

I don't know if that answers your question.

[CBedo]

I guess that depends on whether your truly understand where your value is coming from.

[46zilzal]

i think that is a standard feature of most programs.
Nope

[lansdale]

i think that is a standard feature of most programs. I can speak only about jcapper which is what I use. I can download the cards from all around the country on any given day, process changes, calculate them according to my predefined models, and it will give me a summary of the plays I'm looking for. All in about 5 minutes.

Of course it takes some work to set up your criteria as to what you're looking for. But you do that in advance.

I don't know if that answers your question.

This does answer my question. Thanks very much.

Cheers,

lansdale

[JustRalph]

As the question implies, I don't use a database, and wondered whether those who do have access to the complete range of whatever their handicapping indicates would be value bets across the full range of available tracks on any given day.

Any help appreciated.

lansdale

You said a mouthful..........but I think ArlJim78 answered it correctly. As long as you want to download it.......you can play em all........and in Jcapper they are pretty easy to see........depending on your criteria for a "value bet" the program has some automated version of that with automated highlighting etc.........

[lansdale]

I guess that depends on whether your truly understand where your value is coming from.

C Bedo,

That is assumed by the question, and of course, you're right.

Cheers,

lansdale

[lansdale]

You said a mouthful..........but I think ArlJim78 answered it correctly. As long as you want to download it.......you can play em all........and in Jcapper they are pretty easy to see........depending on your criteria for a "value bet" the program has some automated version of that with automated highlighting etc.........

JustRalph,

Yeah, that was a little convoluted. The question really is about the immediacy with which the effects of the handicapping research you've already done on a given day's cards - in the case of JCapper, I guess this would be the UDMs - are available. It does sound like they are available instantly, although I'm sure it takes some time to make choices. Thanks for your input.

Cheers,

lansdale

[CBedo]

For me currently, I have a model that runs every track (as long as there is data from TSN) every day. I have done enough research to believe (or fool myself) that the general principles of the model hold across all tracks. Then I (the computer, not me) makes some minor adjustments based on intricacies of specific tracks, but overall, if a track is running thoroughbreds, there's the potential to be bets for me.

I just looked at the records of the automation I'm running, and I have made bets at 25 different tracks in the last two weeks.

[GameTheory]

For me currently, I have a model that runs every track (as long as there is data from TSN) every day. I have done enough research to believe (or fool myself) that the general principles of the model hold across all tracks. Then I (the computer, not me) makes some minor adjustments based on intricacies of specific tracks, but overall, if a track is running thoroughbreds, there's the potential to be bets for me.

I just looked at the records of the automation I'm running, and I have made bets at 25 different tracks in the last two weeks.Do you have one model that covers all races/all tracks, or do you only play certain types, or different models for different X? My question is if you don't use one-model-fits-all, how do you stratify races? (i.e what is X?)

[CBedo]

Do you have one model that covers all races/all tracks, or do you only play certain types, or different models for different X? My question is if you don't use one-model-fits-all, how do you stratify races? (i.e what is X?)Interestingly, I have always handicapped tracks as different individual entities, but because of the "I need more data" mentality, I threw every track and distance together for the purposes of developing the simple model I'm running automated right now. Originally, I was throwing out certain distances, but as I've gathered more data, all distances seem to show roughly statistically the same results. There are a few tracks that have consistently given me poor results, and I have tossed those from the model (it's a pace based model, so it wouldn't surprise you which ones I'm tossing). There are a number of other minor adjustments that the model makes for track, distance, etc, but overall, it is looking at every race the same. Since the beginning of developing it, I've thought it was overly simplistic and couldn't work because of it, but so far, it has proven me wrong.

[Track Collector]

Interestingly, I have always handicapped tracks as different individual entities, but because of the "I need more data" mentality, I threw every track and distance together for the purposes of developing the simple model I'm running automated right now. Originally, I was throwing out certain distances, but as I've gathered more data, all distances seem to show roughly statistically the same results. There are a few tracks that have consistently given me poor results, and I have tossed those from the model (it's a pace based model, so it wouldn't surprise you which ones I'm tossing). There are a number of other minor adjustments that the model makes for track, distance, etc, but overall, it is looking at every race the same. Since the beginning of developing it, I've thought it was overly simplistic and couldn't work because of it, but so far, it has proven me wrong.

CBedo,

I am a database player and have pattern matches which would be basically equivalent to what has been referred to as UDM's. I do not handicap individual tracks for profitable patterns, as I am not comfortable with the sample sizes which can lead to higher risks that the pattern is not profitable as it appears in the sample. Experience with the methodology that I use has taught me to look for patterns that are profitable over a large number of tracks, so I start with ALL, and then eliminate a portion those (perhaps 20-25%) which are least profitable. Experience has also taught me that when I try to eliminate too many tracks to have only the highest ROI ones remaining, the remaining "super" set tracks often do not live up to their previous numbers, and some of the tracks one eliminates correspondingly show better results then in the past.

Past results still have quite a bit of risk when hoping to predict the future, otherwise, this game would be easy. ;)

[CBedo]

CBedo,

I am a database player and have pattern matches which would be basically equivalent to what has been referred to as UDM's. I do not handicap individual tracks for profitable patterns, as I am not comfortable with the sample sizes which can lead to higher risks that the pattern is not profitable as it appears in the sample. Experience with the methodology that I use has taught me to look for patterns that are profitable over a large number of tracks, so I start with ALL, and then eliminate a portion those (perhaps 20-25%) which are least profitable. Experience has also taught me that when I try to eliminate too many tracks to have only the highest ROI ones remaining, the remaining "super" set tracks often do not live up to their previous numbers, and some of the tracks one eliminates correspondingly show better results then in the past.

Past results still have quite a bit of risk when hoping to predict the future, otherwise, this game would be easy. ;)Makes perfect sense to me.