Past posting

[Gorrex]

Mid-Late 80s some of the syndicates developed systems to interface with the actual wagering terminals and input bets faster than any teller could. THAT was an actual huge leg up on what others could do.

Today.... You don't need anything special from a tote or an ADW to do what syndicates do. If you have the programming chops (and a bankroll) just open an account at every ADW out there. Write code to scrape their sites and push bets through the way their website does. There is nothing anyone can do to stop that as long as betting online exists.

[Tom]

Originally Posted by Gorrex

Mid-Late 80s some of the syndicates developed systems to interface with the actual wagering terminals and input bets faster than any teller could. THAT was an actual huge leg up on what others could do.

Today.... You don't need anything special from a tote or an ADW to do what syndicates do. If you have the programming chops (and a bankroll) just open an account at every ADW out there. Write code to scrape their sites and push bets through the way their website does. There is nothing anyone can do to stop that as long as betting online exists.

This is why it is not a fair betting system.
Not everyone has the time, money, now expertise to do this.
So anyone who does is automatically at a huge advantage.
Yet tracks and ADWs never mention this, then encourage you to sign up and bet now!

This is why my betting is 90% or more less than it used to be.
Why bother having anyone on site, just run races now, no fans, and let only hales bet through their access points. Don't pretend you are in business for everyone when 90% of your customers are nothing more than a bother to you.

It's like having a bingo game where some players get all the corners covered as free spaces, but most do not.

[PaceAdvantage]

Originally Posted by Tom

This is why it is not a fair betting system.
Not everyone has the time, money, now expertise to do this.
So anyone who does is automatically at a huge advantage.
Yet tracks and ADWs never mention this, then encourage you to sign up and bet now!

Of course it's fair.

How is it any more or less fair than someone who figures out some super-secret angle that nobody else knows of and makes gobs of money on it? Would you tell that person it's not fair he's smarter or more clever than the rest of us?

[PaceAdvantage]

Originally Posted by Gorrex

Mid-Late 80s some of the syndicates developed systems to interface with the actual wagering terminals and input bets faster than any teller could. THAT was an actual huge leg up on what others could do.

Today.... You don't need anything special from a tote or an ADW to do what syndicates do. If you have the programming chops (and a bankroll) just open an account at every ADW out there. Write code to scrape their sites and push bets through the way their website does. There is nothing anyone can do to stop that as long as betting online exists.

I was doing this 10-15 years ago....and I'm no great programmer....

[JerryBoyle]

Originally Posted by Tom

This is why it is not a fair betting system.
Not everyone has the time, money, now expertise to do this.
So anyone who does is automatically at a huge advantage.
Yet tracks and ADWs never mention this, then encourage you to sign up and bet now!

This is why my betting is 90% or more less than it used to be.
Why bother having anyone on site, just run races now, no fans, and let only hales bet through their access points. Don't pretend you are in business for everyone when 90% of your customers are nothing more than a bother to you.

It's like having a bingo game where some players get all the corners covered as free spaces, but most do not.

This is silly. It's absolutely fair. Otherwise, you're stating that being more skilled at something isn't fair.

It's not hard to write or pay someone to create a program that places wagers as close to the bell as you can. However, it's not trivial to make sure all the calculations are done prior to the race starting and STILL get all your bets down. I get shut out all the time. It's a hard problem. I guarantee that every single syndicate gets shut out occasionally.

The problem basically becomes:

1. I know my system takes X seconds to calculate all the things it needs to calculate to determine the bets it wants to place

2. I know that, given a certain point in the video feed, the race will likely go off within Y seconds

3. I make sure to press my "GO" button at some point where Y > X with some slight fudge factor built in (like PA said). For me, this is often when the horses are just starting to get loaded into the gate, with some increase in fudge factor for increase in number of runners.

However, for any number of reasons, X or Y can move slower/faster than I expect. Horses can be loaded in to the gate faster/slower than I expect. My program can take longer to calculate some of the inputs than I expect. I've had cases where a given horse or jockey has a lengthy racing history, so it takes longer to pull and work with the data than I'd expect. Perhaps the race has so many runners that there ends up being a signficant # of combinations in a given pool, and calculating the size of bets takes longer than I'd expect. It could even be that my estimated chance for each combination combined with expected payoffs makes the bet sizing algorithm take longer to calculate (think abt the case when there is more edge spread across more combos). It's even possible that for whatever reason, my computer itself is running slower than I expect.

You can try to mitigate all of these situations as much as possible. Stratify Y by track/pool/number of runners/etc to get a better understanding of how long each race should take to load. Precalculate as much as possible to make sure you only calculate the things that could change right before a race. Try to get a sense for the edge you might have prior to the race running using something like morning line or your own estimate of what the final odds might be. However, it will still alwys be a guessing game, and teams will always get shut out from time to time.

For what it's worth, my program has attempted to place wagers past the close on 100s or 1000s of occasions, and 0 of those wagers have ever gotten in. My intentions aren't nefarious, it's just that I pressed the "GO" button too late. And in every single case, the ADW has blocked me from wagering, as it should.

[Gorrex]

You can do the same things on the stock market, betting exchanges, and even fixed odds.

This isn't unique to horse racing by any stretch.

[sharkey11]

ill tell you whats not fair dumb track mangment cancling cards in the middle of summer for weather excuses of no reasonable threats like DEL an PHA when are the horsemen gonna get away from the stupid casinos an run there game like they want . and to stay on topic any one rember the bell rinngers back in the late 70s horses would be 25 35 feet out of gate before totes would lock an bell would ring an have a guy at the 100 dollar window with a walkie talkie and guy outside with them also calling the leading off horse back to windows man now thats past posting

[ubercapper]

Originally Posted by Sea Hero

Just because the common guy can't get a bet down after the gate opens doesn't mean that nobody is able to do it. If you want to believe that this industry is pristine, then believe it. I don't. Wherever there are large amounts of money, there are unscrupulous people who want it.

If someone is able to do this, electronically, it will be flagged by the TRPB and the money will likely be taken from the bettor's account, as well as they may lose their access.

I'm sure someone will call me naive but I have confidence the TRPB, which is a part of the TRA, looks for this activity constantly, and, particularly using the Keeneland example, if asked can easily examine the data to be sure the timestamps on all wagers occurred before the stop betting button was pressed.

I am not addressing the perception based on the time delay from wagers coming in the last minute before the race and displaying while the race is running except to say some of that is related to graphics software delays updating from tote which are to be expected, the same reason horse numbers on the screen may not be correct compared real time (what we are seeing with our eyes) as horses pass each other at 30+ mph.

[metro]

Originally Posted by ubercapper

I'm sure someone will call me naive but I have confidence the TRPB, which is a part of the TRA, looks for this activity constantly, and, particularly using the Keeneland example, if asked can easily examine the data to be sure the timestamps on all wagers occurred before the stop betting button was pressed.

TRPB over sighting the TRA, is like Navarro over-sighting drug testing. They are always going to go with what the tracks want them to say.

Projecting transparency is different than actually providing it.

[Gorrex]

One other note..

Tote is always blamed for the odds shifting late on final. While much of it is related to the tote systems and remote systems taking to long to transmit, not all of it is.

Many times the displays (ADW, Toteboard, TV display) update at a very slow rate which causes the odds to show anywhere from 10 seconds to 2 minutes after the tote has actually done the update.

Tote isn't the only issue and many times not the biggest issue.

[comet52]

Originally Posted by JerryBoyle

This is silly. It's absolutely fair. Otherwise, you're stating that being more skilled at something isn't fair.

It's not hard to write or pay someone to create a program that places wagers as close to the bell as you can. However, it's not trivial to make sure all the calculations are done prior to the race starting and STILL get all your bets down. I get shut out all the time. It's a hard problem. I guarantee that every single syndicate gets shut out occasionally.

The problem basically becomes:

1. I know my system takes X seconds to calculate all the things it needs to calculate to determine the bets it wants to place

2. I know that, given a certain point in the video feed, the race will likely go off within Y seconds

3. I make sure to press my "GO" button at some point where Y > X with some slight fudge factor built in (like PA said). For me, this is often when the horses are just starting to get loaded into the gate, with some increase in fudge factor for increase in number of runners.

However, for any number of reasons, X or Y can move slower/faster than I expect. Horses can be loaded in to the gate faster/slower than I expect. My program can take longer to calculate some of the inputs than I expect. I've had cases where a given horse or jockey has a lengthy racing history, so it takes longer to pull and work with the data than I'd expect. Perhaps the race has so many runners that there ends up being a signficant # of combinations in a given pool, and calculating the size of bets takes longer than I'd expect. It could even be that my estimated chance for each combination combined with expected payoffs makes the bet sizing algorithm take longer to calculate (think abt the case when there is more edge spread across more combos). It's even possible that for whatever reason, my computer itself is running slower than I expect.

You can try to mitigate all of these situations as much as possible. Stratify Y by track/pool/number of runners/etc to get a better understanding of how long each race should take to load. Precalculate as much as possible to make sure you only calculate the things that could change right before a race. Try to get a sense for the edge you might have prior to the race running using something like morning line or your own estimate of what the final odds might be. However, it will still alwys be a guessing game, and teams will always get shut out from time to time.

For what it's worth, my program has attempted to place wagers past the close on 100s or 1000s of occasions, and 0 of those wagers have ever gotten in. My intentions aren't nefarious, it's just that I pressed the "GO" button too late. And in every single case, the ADW has blocked me from wagering, as it should.

I am curious, why is there such an emphasis on getting a bet in at the last second. What is it about 1 second prior to post that is better in some way or more profitable than say, 5 or 10 or 30 seconds, or two minutes? Or half an hour if you feel you have the right horse? Not that feel enters into some computation but still...

I am not a computer whiz, a whale, a syndicate guy, a math guy, I have my own way of capping horses which is probably very different from anyone else's. So this is all sort of greek to me. Hope you can enlighten me.

[Gorrex]

The same reason(s) that most people go up to the teller at the last moment.

Betting as late as possible gives you the best idea of the payouts. If the payout isn't high enough you don't make the bet regardless of if you think its a winner.

[comet52]

Originally Posted by Gorrex

The same reason(s) that most people go up to the teller at the last moment.

Betting as late as possible gives you the best idea of the payouts. If the payout isn't high enough you don't make the bet regardless of if you think its a winner.

But the payouts change quite often during the race - something that people complain about around here often. I had an 8-1 shot going into the gate the other night in Australia and was excited to see it win, but the odds were 2-1 at close. That's kind of extreme but I've experienced similar many times. I've also had horses where the odds go up during the race and the payout is better than expected. I get that closest to post time is probably most accurate but it's not perfect. Would I make that 8-1 bet if it was 2-1 at post? Yes. The other night i had a 3-1 drop to 3-5 as it crossed the line, I would not have bet it. But it did win so I got the whopping $3.10 payout regardless.

[metro]

Originally Posted by comet52

I am curious, why is there such an emphasis on getting a bet in at the last second. What is it about 1 second prior to post that is better in some way or more profitable than say, 5 or 10 or 30 seconds, or two minutes? Or half an hour if you feel you have the right horse? Not that feel enters into some computation but still...

I am not a computer whiz, a whale, a syndicate guy, a math guy, I have my own way of capping horses which is probably very different from anyone else's. So this is all sort of greek to me. Hope you can enlighten me.

In general most of the betting public gravitates to horses that are getting played since favorites and lower odds horses win more frequently than the double digit ones.

When the guys in the know "know" a horse should be 5-1 instead of 10-1, or 2-1 instead of 4-1, they aren't going to bet their money with 5 minutes to post and let the rest of us common folk in on their intel, they are going to bet it at the last possible second, or, imo, if you are part of the syndicate or whale crowd, have the host track bet it for them when the bell rings as long as their bets are queued and ready to be placed.

The 10-1 horse still winds up at 5-1 but more likely it would have been 7-2 or 4-1 if the general public had time to adjust to what the tote board was saying in those last few minutes to post.

[JerryBoyle]

Originally Posted by comet52

I am curious, why is there such an emphasis on getting a bet in at the last second. What is it about 1 second prior to post that is better in some way or more profitable than say, 5 or 10 or 30 seconds, or two minutes? Or half an hour if you feel you have the right horse? Not that feel enters into some computation but still...

I am not a computer whiz, a whale, a syndicate guy, a math guy, I have my own way of capping horses which is probably very different from anyone else's. So this is all sort of greek to me. Hope you can enlighten me.

Pretty much what Gorrex said. The variations in final price can be fairly large, as everyone has said in this thread, so waiting as long as possible gives me the most insight into what I'll actually get paid. Remember that the handicapping equation is 2 parts: what do I think the chance of a given combination winning is, and what will I get paid if it wins? You shouldn't bet without both parts. Waiting as long as possible gives me more information about the second part.

It's also important to remember that probably all teams use some type of model that blends their prediction with the public's prediction to create a sort of ensemble prediction. So waiting as long as possible actually benefits my final prediction, as the part that's contributed to by the public gets "better" the longer I wait.

A third, but less important benefit, is that waiting as long as possible makes it less likely that anyone benefits from seeing your action enter the pool. Although I've heard rumors of teams getting bets down earlier so they can push other teams off their action. This isn't something I've tried, because it's hard to test that it actually works without spending the money to bet.