Back $4 on 1000 odds taken

[Dag]

Yesterday I placed $0.20 on the outsider. My back of $4 were accepted on his back quote 1000. I lost it of course. Why that happened?

Horces racing - 5f distance.

[GaryRussell]

Below minimum stake bets are not officially accepted by Betfair so the software has to "cheat" to place the bet. The first part of the process is to place $4 at 1000 with the intention of not getting matched so it can be updated. If the odds are already 1000 then it will get matched.

[Dag]

Below minimum stake bets are not officially accepted by Betfair so the software has to "cheat" to place the bet. The first part of the process is to place $4 at 1000 with the intention of not getting matched so it can be updated. If the odds are already 1000 then it will get matched.

Thanks Mark.
I know about "cheating" but seems that I've cathed the moment 2 or 3 seconds before race finish so horse odds jumped from 20 on 1000 (as loser). Maybe this issue is something to think about to protect future?:) If odds are >99 then cancel bet. Whith this "fisculture" our betting time is 3 seconds (each stake that is bellow $4) not 1 (for horses)?

[GaryRussell]

We will consider making something like your suggestion an optional feature in the future.

[Dag]

We will consider making something like your suggestion an optional feature in the future.

And again:
31558399385 The Tullow Tank L $-8.33 120.00 RESULT_LOST

I didn't place bet like this. 2 lays accepted. Maybe some smart as^%$#s are waiting machines that "cheating" betfair?

[Captain Sensible]

You need to ensure you put as many safety conditions within your code as possible to trap any errors. With in running markets odds change far too quickly to expect Gruss to trap them for you, you may well place a bet at 2's but by the time that bet gets thru the in running delays, imposed by Betfair, that price could now be 1.01 or 1000. Once submitted there's no way Gruss can override that.

[Dag]

You need to ensure you put as many safety conditions within your code as possible to trap any errors. With in running markets odds change far too quickly to expect Gruss to trap them for you, you may well place a bet at 2's but by the time that bet gets thru the in running delays, imposed by Betfair, that price could now be 1.01 or 1000. Once submitted there's no way Gruss can override that.

Thanks Captain.

I thought about this issue and you are right! Everything is on the relation between me <--> market. I know that that bets were placed during the In Play and the speed did all but I still can't understand how bets were taken? Was it just random case (unintentionally; just happens) or some waited for players / bots like me (means intentionally)?

In two words: that program cheating (back on 1000 or lay on 1.01) - is it safe? That's all.

[Captain Sensible]

I know that that bets were placed during the In Play and the speed did all but I still can't understand how bets were taken?

Not too sure what you mean by that? As soon as a horse looks like it has no chance of winning plenty of players on course and off will lay at 1000 so any backs bets will be taken.

The reason you're more likely to encounter it at high odds is because you're placing bets under the minimum amounts so your bet is left up for 3 times longer as the program needs to do three steps and therefore 3 inrunning delays to get around the minimum limit. 3 seconds on horse racing is a long time when you're competing against people on course.

In two words: that program cheating (back on 1000 or lay on 1.01) - is it safe? That's all.

I think Gary may have been better using the word 'loophole' rather than 'cheat' in his explanation. The program is not doing any cheating at all just using a method or getting around the minimum limits, which Betfair tolerate if not abused. If we wanted to be pedantic it's actually the user requesting and submitting bets under minimum limits that is 'cheating'.

As for being 'safe', again I'm not too sure what you mean. The loophole is well known and most people are well aware of how it works to place bets and it'll only ever begin with a minimum stake bet. At 1.01 the worst you could lose is a few cents, at 1000 you could lose up to almost 2 times the minimum stake if odds went quickly to 1000 in running. Only way to avoid the 3 times delay is to ensure your bets are above the minimum limits and even then a losing bet will still be a losing bet regardless of what odds it's matched at.

[Captain Sensible]

Basically it's just one of the drawbacks of placing bets under the minimum limits that occasionally things will move too quick for your opening minimum bet to bet cancelled. If you get caught out too often that it's affecting your winnings you'd probably need to reassess your strategy or stakes as there's no way to get round it. Betdaq have lower limits but in running is dead at the moment but who knows if that will change in the future

[Dag]

Thanks again! Everything clear now.

Re Strategy. My strategy is not to play 1 or 2 cents per horse so this (potentially) dangerous situation of delayed bet time x 3 will have no any influence when I start to play above min. bet amount.

Each knowledge costs money. Those cases (re: this thread) were not too expensive I just saw what can happened. Good experience for the future.

Agree about “cheating”. Should be very conditionally used: I am “cheating” but not the program (which makes me available so). On mine risk of course.

And in one previous post, you wrote nice: check twice everything on your side (re: Excel). I added some limitation.

[Captain Sensible]

Glad to have been some help, yep it's always best to use small stakes when trying out new strategies/bots, much wiser to find any flaws with small stakes

[Dag]

Captain, do u maybe know where I can find some good data base about actual racing (or winners) times? I have some XML from Geekstoy but there are many tough mistakes there... or I have to build it by myself?

[Captain Sensible]

Afraid not, the Racing Post does the standard times for each course but I think you have to pay for access these days. I'd have thought standard times wouldn't be much use though as to do things accurately you need going stats too and that data doesn't get sent via the API and can also change throughout the day plus be different on chase or hurdle courses. Unless you're using reliable data sometimes you're just better omitting it altogether.

Might be worth searching the net for some horse racing forums and asking if anyone will share the data on those.

[Dag]

Irishracing.com publish winners time (occasionally seems to me:)
http://www.irishracing.com/results?meet ... d=20131113
That times could be collected… within month or so: DB will be nice and big.

I am grabbing the distance from base race line, placing that in the proper cell and automatically, another cell starting the counter (down) for that race distance when In Play start. Can be useful probably…

I can post here geekstoy distance times if you think that it can be interesting. Or maybe in separate thread?

[Captain Sensible]

Thanks but I doubt the geek would be happy if someone posted his data on a 'rival' API vendors forum

And to be honest I can't imagine standard data for distances without going stats would be of any real value other than a very rough idea of what stage the race was at. The irish racing page would probably be more useful to any speed/form based methods as that gives times for all runners and should be relatively simple for a scraper to gather the data.

For botting I've always found the simplest approach make me the most money.