Anyone use Drone feed vs TPD?

[Brovashift]

Also, back testing in sports markets? Is that a good idea!?

Yes.

Where/how do you start back testing historical data? Is this where learning to model data comes in? I have dabbled with the idea but never really got anywhere with it.

I have got a python webscraper off GitHub that scrapes RP historical data and used that to build a desktop application to highlight lay sections on RP racecards based on my own analysis criteria. It was going well, but has been on the ToDo project pile for about 12 months now. I have thought about using this data (fresh scrape) to create a model, but I dont know what to model. I was interested in seeing data from around the 2min to post time pre off, but this data doesn't have pre off price data.

[JuiceyJones]

You get so much insight from TPD. Honestly, having seen drone pictures I think they are over rated.

It's quite easy to learn to race read, so it just becomes a process of who has the biggest balls, gets filled quickly and how far are you prepared to anticipate. It feels like a race to the bottom.

TPD gives you soooo much depth.

I am skewed by my own personal experience, but it's completely changed everything for me inplay.

I disagree with the drone take. My experience with drone feeds are that people are willing to smash the price up or down with no regard for value, as soon as they see the slightest signal for an entry from the horse. If you can combine race reading with roughly pricing up these scenarios in your head in a short time frame, there’s often value in being on the opposite side of the ladder to these players. I’ve found the general level of race reading quite poor from the speed merchants. Just my opinion.

Excatly that. I feel like drones are generally for people who don't want to take the time to put in the work learning how to race read. First horse that lookings squeezed or slightly boxed in gets price smashed. You can 100% catch a spiked up price on the back side when one of these guys smashes price out for a poor jump or when a horse is squeezed up up a hill of loses focus for a second. Goshen springs to mind when he was on the younger side. Could be an asshole of a horse to ride sometimes so price would get smashed up early. Catch a decent back price then ride it back down.

Also not sure why some people think race reading is easy. Took me forever to learn about the way horses can act, jockeys tendencies, different tracks, different tracks in different weather, different sized fields at different tracks, when draw bias actually counts, etc.

Drones are the ICT of in-play trading.

[LinusP]

Also, back testing in sports markets? Is that a good idea!?

Yes.

Where/how do you start back testing historical data? Is this where learning to model data comes in? I have dabbled with the idea but never really got anywhere with it.

I have got a python webscraper off GitHub that scrapes RP historical data and used that to build a desktop application to highlight lay sections on RP racecards based on my own analysis criteria. It was going well, but has been on the ToDo project pile for about 12 months now. I have thought about using this data (fresh scrape) to create a model, but I dont know what to model. I was interested in seeing data from around the 2min to post time pre off, but this data doesn't have pre off price data.

To be honest with you if you are not analysing / modelling data what are you doing?

[Brovashift]

Yes.

Where/how do you start back testing historical data? Is this where learning to model data comes in? I have dabbled with the idea but never really got anywhere with it.

I have got a python webscraper off GitHub that scrapes RP historical data and used that to build a desktop application to highlight lay sections on RP racecards based on my own analysis criteria. It was going well, but has been on the ToDo project pile for about 12 months now. I have thought about using this data (fresh scrape) to create a model, but I dont know what to model. I was interested in seeing data from around the 2min to post time pre off, but this data doesn't have pre off price data.

To be honest with you if you are not analysing / modelling data what are you doing?

Event driven markets like racing I primarily race read using BFP principles, and additional course knowledge. When the race selections are not the best I.e. prices a bit too competitive I sometimes pick lay selections by analysing the racecards, focusing on the higher priced 3rd or 4th favs on lower quality races, but I find this very convoluted even though profitable. Why I was building the winforms app.
I sometimes trade ATP500 & 1000 tennis tournaments, manually, but it's not my daily bread & butter.

Opinionated markets like pre off racing I simply trade order flow. I fell into sports trading while pursuing a career in Global Macro trading but covid sent me down this path, and order flow is something I've identified with from the start. Although not plain sailing lol, not until I started quantifying what I was doing. "Game changer"!

I'm always curious about what other people do though, can never stop learning in my eyes.

I've got an honours degree in computer science but the whole data modelling thing just doesn't appeal to me for some reason. Still interested though incase I'm missing something. What's your background, how did you start modelling?

I hope your question was not rhetorical

[Leb]

You get so much insight from TPD. Honestly, having seen drone pictures I think they are over rated.

It's quite easy to learn to race read, so it just becomes a process of who has the biggest balls, gets filled quickly and how far are you prepared to anticipate. It feels like a race to the bottom.

TPD gives you soooo much depth.

I am skewed by my own personal experience, but it's completely changed everything for me inplay.

I disagree with the drone take. My experience with drone feeds are that people are willing to smash the price up or down with no regard for value, as soon as they see the slightest signal for an entry from the horse. If you can combine race reading with roughly pricing up these scenarios in your head in a short time frame, there’s often value in being on the opposite side of the ladder to these players. I’ve found the general level of race reading quite poor from the speed merchants. Just my opinion.

Excatly that. I feel like drones are generally for people who don't want to take the time to put in the work learning how to race read. First horse that lookings squeezed or slightly boxed in gets price smashed. You can 100% catch a spiked up price on the back side when one of these guys smashes price out for a poor jump or when a horse is squeezed up up a hill of loses focus for a second. Goshen springs to mind when he was on the younger side. Could be an asshole of a horse to ride sometimes so price would get smashed up early. Catch a decent back price then ride it back down.

Also not sure why some people think race reading is easy. Took me forever to learn about the way horses can act, jockeys tendencies, different tracks, different tracks in different weather, different sized fields at different tracks, when draw bias actually counts, etc.

Drones are the ICT of in-play trading.

But wouldn’t it still be most profitable to operate as you do now, but with the addition of having the fastest information (drones)?

[JuiceyJones]

I disagree with the drone take. My experience with drone feeds are that people are willing to smash the price up or down with no regard for value, as soon as they see the slightest signal for an entry from the horse. If you can combine race reading with roughly pricing up these scenarios in your head in a short time frame, there’s often value in being on the opposite side of the ladder to these players. I’ve found the general level of race reading quite poor from the speed merchants. Just my opinion.

Excatly that. I feel like drones are generally for people who don't want to take the time to put in the work learning how to race read. First horse that lookings squeezed or slightly boxed in gets price smashed. You can 100% catch a spiked up price on the back side when one of these guys smashes price out for a poor jump or when a horse is squeezed up up a hill of loses focus for a second. Goshen springs to mind when he was on the younger side. Could be an asshole of a horse to ride sometimes so price would get smashed up early. Catch a decent back price then ride it back down.

Also not sure why some people think race reading is easy. Took me forever to learn about the way horses can act, jockeys tendencies, different tracks, different tracks in different weather, different sized fields at different tracks, when draw bias actually counts, etc.

Drones are the ICT of in-play trading.

But wouldn’t it still be most profitable to operate as you do now, but with the addition of having the fastest information (drones)?

Have only watched drone feeds 4-5 times of live racing. Pictures were not great on any of them. You can see the obvious big mistakes but that was about it. RTV pics (IMO) better overall quality wise for seeing the more subtle moves. But sure. Those were just the feeds that I watched and in fairness some of the Irish races are terrible. Almost like they on have 1 camera at some courses.

I think TPD is good for pace but thats about it. For me atleast.

I think I just get salty about shit like this because it just seems like everyone these days wants to deposit $100 (Betfair - forex - crypto - whatever) and run it up into $1000 in a month in the quickest way possible without learning anything. Its not professional. Drones fall into that category for me. Completely off topic for this thead and just my $0.02s so my apologies.

[Brovashift]

The drone feed is great for in-play traders, and it's fantastic that someone is teaching how to trade using it.

Thought this was going to be a plug then for Carl Traders new course

No thanks!

I am new here, interested in trading in play as someone I know once mentioned he had a course by 'Carl' (not sure of second name) and said it seemed decent/genuine, although he hasn't studied it fully never mind tried the methods! I was considering looking for his course but saw this comment above and wondered if I am about to make a mistake. Wondered if you could elaborate on your comment, is it best avoided?

Back For Profit is better... and InPlayWithIvan, both compliment each other.

Carl's is OK, but BFP takes the top spot for me. Carl has got a new course out now showing in play with the drone, but he's dropped the price from about £300 to £95, he can't even give it away which should speak volumes.

[Fugazi]

Yes.

Where/how do you start back testing historical data? Is this where learning to model data comes in? I have dabbled with the idea but never really got anywhere with it.

I have got a python webscraper off GitHub that scrapes RP historical data and used that to build a desktop application to highlight lay sections on RP racecards based on my own analysis criteria. It was going well, but has been on the ToDo project pile for about 12 months now. I have thought about using this data (fresh scrape) to create a model, but I dont know what to model. I was interested in seeing data from around the 2min to post time pre off, but this data doesn't have pre off price data.

To be honest with you if you are not analysing / modelling data what are you doing?

Trial and error.

I got there in the end.

But I would be much more profitable if/when I get round to analysing properly.

[LinusP]

Where/how do you start back testing historical data? Is this where learning to model data comes in? I have dabbled with the idea but never really got anywhere with it.

I have got a python webscraper off GitHub that scrapes RP historical data and used that to build a desktop application to highlight lay sections on RP racecards based on my own analysis criteria. It was going well, but has been on the ToDo project pile for about 12 months now. I have thought about using this data (fresh scrape) to create a model, but I dont know what to model. I was interested in seeing data from around the 2min to post time pre off, but this data doesn't have pre off price data.

To be honest with you if you are not analysing / modelling data what are you doing?

Trial and error.

I got there in the end.

But I would be much more profitable if/when I get round to analysing properly.

I have been working with this data for over 6 years almost every day, no where near the end.

[tico]

Interesting discussion . However would not Drone Feeds and TPD be factored into the market very quickly resulting in a reaction from autobots and humans alike. I might be a bit old fashioned but thorough form study is still IMHO the best way to suss out a race ,and if you can get some reliable speed figures and pace analysis you have a good edge to start.
Kind Regards
Tico
PS how long will it be before race tracks ban the use of drones because they are distracting the stewards whilst they are napping

[DerbyTrader]

The drone feed is great for in-play traders, and it's fantastic that someone is teaching how to trade using it. It's obvious to most traders that you're not the only one using drones.

The drone feed provides a much better angle on RTV tracks and is essential for SKY tracks, given the huge delay there.

Euler promotes TPD because he's a coder, so that makes sense. However, TPD has errors and is a bit slower than the drone feed, even the reshared version, and you can't see much about the horses.

It does work well if you have more advanced code, but then again, many traders use the expensive 20ms GPS feed, which puts you at a disadvantage compared to them.

The census among drone community is that the drone is behind a massive edge, particularly at Sky venues. It's a much fairer playing field on RTV fixtures.

There's often a warning message telling people not to trade races when another drone has been spotted at the venue.

[Scotty66]

The drone feed is great for in-play traders, and it's fantastic that someone is teaching how to trade using it. It's obvious to most traders that you're not the only one using drones.

The drone feed provides a much better angle on RTV tracks and is essential for SKY tracks, given the huge delay there.

Euler promotes TPD because he's a coder, so that makes sense. However, TPD has errors and is a bit slower than the drone feed, even the reshared version, and you can't see much about the horses.

It does work well if you have more advanced code, but then again, many traders use the expensive 20ms GPS feed, which puts you at a disadvantage compared to them.

The census among drone community is that the drone is behind a massive edge, particularly at Sky venues. It's a much fairer playing field on RTV fixtures.

There's often a warning message telling people not to trade races when another drone has been spotted at the venue.

I've seen that message a lot when I've used the drone but I don't understand it. I just thought it was another group using a different drone service. Why should you not trade via a drone when there is another one in the sky?

[ShaunWhite]

The advantage with TPD, as with all automated tech, is being able to trade all of the horses simultaneously, every race everywhere every day all day, rather than the relative handful any human can trade effectively in a year with their one set of eyeballs.

It might not return the same peak RoI on any given race but it benefits from much larger scale. I can't see many parallels with drones apart from the market type it's being deployed in, it's a different take on the game entirely.

You certainly don't need to be a 'coder" to use it these days, just be able to set up some fairly simple software. We're past the days when all we got was 2d coordinates and had to do the course mapping ourselves, and pay £20 a race for the pleasure of doing it.

[Euler]

Drone footage feels like a bit of a cult thing to me currently. Everybody feels if only they had a feed they could make it work, but the reality is very different.

It reminds me of when there were in-play trading rooms up and done the country. Of course, there are none now.

How many people are seeing exactly the same feed, which is probably a feed of a feed of a feed. What's to stop the upstream feed of a feed delaying it by 0.5s, just so it's more useful to them?

I've seen and used drone footage and yes, if you can race read, you may be able use it, but so can anybody else.

[Scotty66]

It's not just about the speed. Compared to the racing TV tracks there's hardly any difference, but the difference is the camera angle the drone gives you. You can see the entire field and that alone gives you an edge.

TV camera angles are shocking. I have the drone footage and the TV feed both open at the same time and there's often a big difference in what you're seeing. Head to head on the TV for example when it isn't actually head to head.

I've tried tpd but I just didn't feel comfortable with it. Personally I wouldn't like to be using big stakes relying on something telling me what happening which could be slightly wrong, when I can use my own eyes and see exactly what is happening.

Also you can't trade with TPD every race as it's only certain tracks.

But that's just my own experience with it. I'm sure there's definitely an edge to be found if you look deep enough.