TPD Services

[tico]

Hi Folks<
I have been looking at BAs in running facilities , what I want to ask is what subscription to TPD data services is adequate to do in running trading .The subscriptions vary from cheap(ish) right up to a price that would make it cheaper to buy and launch your own satellite
What package do other BA users use ?
Kind Regards
tico

[jamesedwards]

I'm on the 'High' package for £100 per month which covers all possible races (UK, US & UAE). If you just want to play UK then the 'Standard' at £60 should be fine most months. Anything below 'Standard' is reduced latency based on the info on their only subscription page, so not ideal for trading. https://www.tpd.zone/subscribe/

[millthuska]

I'm on the 'High' package for £100 per month which covers all possible races (UK, US & UAE). If you just want to play UK then the 'Standard' at £60 should be fine most months. Anything below 'Standard' is reduced latency based on the info on their only subscription page, so not ideal for trading. https://www.tpd.zone/subscribe/

It says 2000 races on that particular subscription. The amount of races in those countries combined over the course of a month would exceed 2000. Does it just allow you an unlimited amount of those races or would you be capped at 2000?

[jamesedwards]

I'm on the 'High' package for £100 per month which covers all possible races (UK, US & UAE). If you just want to play UK then the 'Standard' at £60 should be fine most months. Anything below 'Standard' is reduced latency based on the info on their only subscription page, so not ideal for trading. https://www.tpd.zone/subscribe/

It says 2000 races on that particular subscription. The amount of races in those countries combined over the course of a month would exceed 2000. Does it just allow you an unlimited amount of those races or would you be capped at 2000?

It would be capped, but it doesn't exceed 2000. Only about half of tracks are covered by TPD so about 1600 races across an average month.

[millthuska]

I'm on the 'High' package for £100 per month which covers all possible races (UK, US & UAE). If you just want to play UK then the 'Standard' at £60 should be fine most months. Anything below 'Standard' is reduced latency based on the info on their only subscription page, so not ideal for trading. https://www.tpd.zone/subscribe/

It says 2000 races on that particular subscription. The amount of races in those countries combined over the course of a month would exceed 2000. Does it just allow you an unlimited amount of those races or would you be capped at 2000?

It would be capped, but it doesn't exceed 2000. Only about half of tracks are covered by TPD so about 1600 races across an average month.

I see. Thanks.

[millthuska]

I'm on the 'High' package for £100 per month which covers all possible races (UK, US & UAE). If you just want to play UK then the 'Standard' at £60 should be fine most months. Anything below 'Standard' is reduced latency based on the info on their only subscription page, so not ideal for trading. https://www.tpd.zone/subscribe/

It says 2000 races on that particular subscription. The amount of races in those countries combined over the course of a month would exceed 2000. Does it just allow you an unlimited amount of those races or would you be capped at 2000?

It would be capped, but it doesn't exceed 2000. Only about half of tracks are covered by TPD so about 1600 races across an average month.

One question if you don't mind. How many mistakes are made these days by TPD? I've seen posts mentioning the wrong transponders were put on horses and so forth. I imagine it would be easy to set up conditions to test for mistakes like this and then put in some kind of rule that prevents any trading being done but I suppose your set up would have to be able to catch the error before bets went in. Am I being way off here? I just wouldn't be able to use it without having a failsafe option in there.

[jamesedwards]

It says 2000 races on that particular subscription. The amount of races in those countries combined over the course of a month would exceed 2000. Does it just allow you an unlimited amount of those races or would you be capped at 2000?

It would be capped, but it doesn't exceed 2000. Only about half of tracks are covered by TPD so about 1600 races across an average month.

One question if you don't mind. How many mistakes are made these days by TPD? I've seen posts mentioning the wrong transponders were put on horses and so forth. I imagine it would be easy to set up conditions to test for mistakes like this and then put in some kind of rule that prevents any trading being done but I suppose your set up would have to be able to catch the error before bets went in. Am I being way off here? I just wouldn't be able to use it without having a failsafe option in there.

Transponder errors was a major issue. Not that there were many, but the potential implication was huge; eg backing the wrong horse at 1.01 or laying the wrong horse at 1000.

But since the new warning flag system went in it's now easy to automatically halt any trading when a transponder swap error is flagged. Without digging into the figures, I would estimate perhaps a couple of swap errors flagged per year in the UK and a couple per month in the US. I've seen one or two swap errors without being flagged but they seem to be very rare.

z34.JPG

[millthuska]

It would be capped, but it doesn't exceed 2000. Only about half of tracks are covered by TPD so about 1600 races across an average month.

One question if you don't mind. How many mistakes are made these days by TPD? I've seen posts mentioning the wrong transponders were put on horses and so forth. I imagine it would be easy to set up conditions to test for mistakes like this and then put in some kind of rule that prevents any trading being done but I suppose your set up would have to be able to catch the error before bets went in. Am I being way off here? I just wouldn't be able to use it without having a failsafe option in there.

Transponder errors was a major issue. Not that there were many, but the potential implication was huge; eg backing the wrong horse at 1.01 or laying the wrong horse at 1000.

But since the new warning flag system went in it's now easy to automatically halt any trading when a transponder swap error is flagged. Without digging into the figures, I would estimate perhaps a couple of swap errors flagged per year in the UK and a couple per month in the US. I've seen one or two swap errors without being flagged but they seem to be very rare.


z34.JPG

Those rare occasions could be extremely damaging if not prepared for I imagine?

[Fugazi]

One question if you don't mind. How many mistakes are made these days by TPD? I've seen posts mentioning the wrong transponders were put on horses and so forth. I imagine it would be easy to set up conditions to test for mistakes like this and then put in some kind of rule that prevents any trading being done but I suppose your set up would have to be able to catch the error before bets went in. Am I being way off here? I just wouldn't be able to use it without having a failsafe option in there.

Transponder errors was a major issue. Not that there were many, but the potential implication was huge; eg backing the wrong horse at 1.01 or laying the wrong horse at 1000.

But since the new warning flag system went in it's now easy to automatically halt any trading when a transponder swap error is flagged. Without digging into the figures, I would estimate perhaps a couple of swap errors flagged per year in the UK and a couple per month in the US. I've seen one or two swap errors without being flagged but they seem to be very rare.


z34.JPG

Those rare occasions could be extremely damaging if not prepared for I imagine?

You could mitigate it by resetting your betfair balance daily.

So when you occasionally get hit its just part of your expenses rather than blowing years of work

[millthuska]

Transponder errors was a major issue. Not that there were many, but the potential implication was huge; eg backing the wrong horse at 1.01 or laying the wrong horse at 1000.

But since the new warning flag system went in it's now easy to automatically halt any trading when a transponder swap error is flagged. Without digging into the figures, I would estimate perhaps a couple of swap errors flagged per year in the UK and a couple per month in the US. I've seen one or two swap errors without being flagged but they seem to be very rare.


z34.JPG

Those rare occasions could be extremely damaging if not prepared for I imagine?

You could mitigate it by resetting your betfair balance daily.

So when you occasionally get hit its just part of your expenses rather than blowing years of work

A good idea. I'll definitely do that. To put things into context, I'm someone who has been around betfair for a time and profitable as a pure position taker but also profitable doing manual trading and have many many ideas when it comes to actual trading that can be done with BetAngel but haven't taken the plunge as of yet (will be trialing systems next week though) so take my posts as coming from an absolutely inexperienced beginner when it comes to trading. I think when it comes to automation, given that there absolutely will be races where there are TPD mistakes e.g transponder issue, I think (could be wrong here) that putting in some condition that somehow tests for the accuracy of TPD data before placing bets is an absolute must. Now this might be a complex thing to do or not a complex thing to do (I have programming experience so I can imagine how it would be done in a purely programming sense, depending on your system) but it would surprise me if experienced traders have not got such a failsafe system embedded into their ruleset? Surely it's something the market would teach you to do? We must have players in here who have lost money due to TPD mistakes.

[Fugazi]

I think Linus is probably the man to ask he will know

[LinusP]

You can do a few things

- monitor the warning flag (as mentioned)
- no runners with 0 or ‘stuck’ progress
- confirm you have an update for every active runner and the feed time is the same
- check latency between feed time and server time
- use the market

The last one is the most important, ie base your risk / stakes on what the market is telling you and don’t fight it.

[Fugazi]

You can do a few things

- monitor the warning flag (as mentioned)
- no runners with 0 or ‘stuck’ progress
- confirm you have an update for every active runner and the feed time is the same
- check latency between feed time and server time
- use the market

The last one is the most important, ie base your risk / stakes on what the market is telling you and don’t fight it.

Damn that was quick

[millthuska]

You can do a few things

- monitor the warning flag (as mentioned)
- no runners with 0 or ‘stuck’ progress
- confirm you have an update for every active runner and the feed time is the same
- check latency between feed time and server time
- use the market

The last one is the most important, ie base your risk / stakes on what the market is telling you and don’t fight it.

Thanks. Some things to think about. I was coming at it from the point of view of automation, so not actively managing the market in real time so the last point wouldn't apply but all others seem good. I am so inexperienced that every piece of advice will help.

[LinusP]

You can do a few things

- monitor the warning flag (as mentioned)
- no runners with 0 or ‘stuck’ progress
- confirm you have an update for every active runner and the feed time is the same
- check latency between feed time and server time
- use the market

The last one is the most important, ie base your risk / stakes on what the market is telling you and don’t fight it.

Thanks. Some things to think about. I was coming at it from the point of view of automation, so not actively managing the market in real time so the last point wouldn't apply but all others seem good. I am so inexperienced that every piece of advice will help.

I am 100% automated and the last point is the most important.