Robotic Wagering Rears It's Ugly Head Again

[Kelso]

Originally Posted by cj

There are restrictions with some of the formats. For example, with both TwinSpires and PTC, you can only have the program attempt to place the bet one time. The condition is either met or it fails, and that is it.

If you write your own program, you can have it keep checking if the race went off yet, and if not, check the condition again. This is just one example.

Another would be you can only make the bet one size. With your own program, you could have the bet size vary according to the odds. There are many others, including some I've never thought of as well.

Understood ... and way beyond my pay grade. (Do you think I should retire my PC with the 5.25" floppy drive and get one of those newer ones that take the 3.5" discs?)

Thanks, CJ.

[cj]

Originally Posted by Kelso

Understood ... and way beyond my pay grade. (Do you think I should retire my PC with the 5.25" floppy drive and get one of those newer ones that take the 3.5" discs?)

Thanks, CJ.

I've never been one to upgrade just to have the latest. If it works, it works.

[JustRalph]

Originally Posted by cj

I've never been one to upgrade just to have the latest. If it works, it works.

Don't tell my first wife that.................

[green80]

are any of these robots made available for sale to the public?

[Robert Goren]

Originally Posted by green80

are any of these robots made available for sale to the public?

I figured this question would have been in first 10 post.

[GameTheory]

Originally Posted by Robert Goren

I figured this question would have been in first 10 post.

It is fairly easy to make one for personal use using the available on-line toteboards, but what the original poster was getting it I think is of another kind altogether. There was a dust-up a few years ago about a group making 1000s of bets at the very last second with the help of an ADW or track -- I forget, I think it involved Gulfstream? -- and possibly even having access to the trifecta matrix. They were eventually denied such access (or it was somehow hushed up) because there was such a backlash. There are threads about it archived somewhere here...

[classhandicapper]

Originally Posted by cj

I've never been one to upgrade just to have the latest. If it works, it works.

I spent 25 years in data processing and I never upgrade anything unless I HAVE TO. I live by the motto "If it isn't broke don't fix it".

[CBedo]

Originally Posted by classhandicapper

I spent 25 years in data processing and I never upgrade anything unless I HAVE TO. I live by the motto "If it isn't broke don't fix it".

You need to crank it up a notch then. My motto when it comes to technology, is that if you don't break it, you aren't pushing it hard enough!

With regards to automated betting, it's really not hard for anyone to do; the question is can you do it effectively. If you have something as simple as a list of rule based angles, why not program it to do it automatically so you don't miss an opportunity?

Some have tried to liken this to bots playing poker, as online poker houses try their best to keep these bots out, interestingly, more for the integrity of the game than for the real financial damage they do. I have consulted with one online shop, and results of their studies showed that players were more afraid of the bots than the reality said they should be. In fact, the bots are OK playing limit poker, but when you institute more "fuzzy" rules of playing no limit, the bots don't do very well at all. I think horse racing is at least as complicated (probably way more) than no limit hold'em, and noone should be too worried; in fact, I think we should welcome the liquidity.

If I look at my own experiences in automated handicapping/betting (and I'm sure there are many more experienced at this than me), I have yet to be able to get the computer to be able to handicap any one race as well as I can do it without the computer, but it can obviously handicap exponentially more races than I can by hand. The goal is handicap "only somewhat less optimally" and then make up for it with asset turnover. If other computer programs are similar, they could possibly alter what factors are weighed most in the betting pools, but it is doubtful that they will take the profitability out of the game for any good handicapper.

[Robert Goren]

Originally Posted by CBedo

You need to crank it up a notch then. My motto when it comes to technology, is that if you don't break it, you aren't pushing it hard enough!

With regards to automated betting, it's really not hard for anyone to do; the question is can you do it effectively. If you have something as simple as a list of rule based angles, why not program it to do it automatically so you don't miss an opportunity?

Some have tried to liken this to bots playing poker, as online poker houses try their best to keep these bots out, interestingly, more for the integrity of the game than for the real financial damage they do. I have consulted with one online shop, and results of their studies showed that players were more afraid of the bots than the reality said they should be. In fact, the bots are OK playing limit poker, but when you institute more "fuzzy" rules of playing no limit, the bots don't do very well at all. I think horse racing is at least as complicated (probably way more) than no limit hold'em, and noone should be too worried; in fact, I think we should welcome the liquidity.

If I look at my own experiences in automated handicapping/betting (and I'm sure there are many more experienced at this than me), I have yet to be able to get the computer to be able to handicap any one race as well as I can do it without the computer, but it can obviously handicap exponentially more races than I can by hand. The goal is handicap "only somewhat less optimally" and then make up for it with asset turnover. If other computer programs are similar, they could possibly alter what factors are weighed most in the betting pools, but it is doubtful that they will take the profitability out of the game for any good handicapper.

It is not hard only if you are a programer. I take the answer is that no one has one for sale. Not that at this stage of my life I would be interested, but I think some posters might be.

[GameTheory]

Originally Posted by Robert Goren

It is not hard only if you are a programer. I take the answer is that no one has one for sale. Not that at this stage of my life I would be interested, but I think some posters might be.

You pretty much need a custom solution to tie up all the threads that you personally use or have access to. I do this kind of custom work for people sometimes, I'm working on an automated system for someone on this board right now. The first toteboard monitor I made (for someone else) was actually for Richard Bauer.

[CBedo]

Originally Posted by Robert Goren

It is not hard only if you are a programer. I take the answer is that no one has one for sale. Not that at this stage of my life I would be interested, but I think some posters might be.

It's always for sale; it's just also always a matter of price, lol.

You really don't have to be a programmer to move towards a solution somewhat. Take some of what the excel guys are doing here on the baord, and tie it into some of the past posts on using excel to get tote odds. Then for a crude automated system, just have it generate the needed csv/text file to upload to TS, PTC or other. It's not a perfect solution, but it's a beginning for someone who doesn't want to pay for it.

[Que]

The concerns about Robotic wagering is akin to the concerns about high-frequency trading, i.e. price manipulation, colocation, volatility, rogue trading, and other risks. It is estimated that 70% of all stock/option trading is now attributed to high-frequency trading. However, these systems are just as fallible as the humans that create them.

But if I were to create a robo-wagering system I would utilize the same lessons learned from today's high frequency traders where time is measured in milliseconds.

First, I would write separate interacting programs, i.e. thinkers, pricers, listeners, bettors, and managers. The thinkers take my directions (strategies) and convert it into instructions for the other components. The pricers would calculate in real-time the theoretical value of every possible wager. The listeners would take-in pari-mutuel data directly from the data centers and make it available to the other components. The bettors would interface via a native API directly with the data centers or ADWs for the purpose of submitting wagers. The managers would control the work of all the other components, principally the bettors, based on betting activity and other meaningful events (i.e. scratches, jockey changes, etc.) And all of these actions would be accomplished very, very quickly.

Next, to further reduce latency, I would colocate my computer servers in Sacramento, CA and Mount Laurel, NJ where wagers are processed from all the tracks, simulcast and off-track sites, and ADW wagering platforms. I would then, for a fee, negotiate to place my server on-site at each data center.

Finally, I would negotiate the highest rebate schedule possible--rates only available to the highest of the high-end players.

To maximize my profits I would then instruct my robo-bettor to begin placing as many wagers as possible (i.e. as theoretical values allow), in order to collect the rebate from the ADW.

Sound like a level playing field?

[Tom]

Hey Que!
Welcome back - been a while.

[CBedo]

Co-loing might help your bet submission times, but when noone has real-time pool information (old batch wagering submission technology), then it won't help you "get" better information than anyone else.

[mountainman]

Ho Hum. Any handicapper that can't see most tote corrections coming almost certainly lacks the acumen to beat the game to begin with. Moreover, since the majority of players just aren't disposed- or in a position- to bet with two horses left to load, an ability to forecast closing odds becomes crucial. It's a much overlooked aspect of handicapping. The whole world can't wager at zero to post.

Concerning robots: A seasoned human handicapper carries quite a computer between his own ears and can reprogram it on the fly. His prime advantage, though, lies in detecting and exploiting intangibles that the machine has no grasp of.