How are traders using TPD Data.

Integrate live racecourse data into your favourite Betfair trading interface from TPD.
User avatar
xtrader16
Posts: 308
Joined: Sun Feb 26, 2017 1:00 pm

Euler wrote:
Tue Sep 13, 2022 3:37 pm
I recorded a ton of races next to the TPD data to tease out characteristics of why each horse won, why and where it was obvious.

That felt a really sensible way to approach the problem.

I can read a race visually, but I've never been able to put numbers on it. TPD solved that problem for me.
I think it would take a few hundred races before you could start putting real sums of cash into an Automated system. I was looking a race yesterday where the Horse that won never had the lowest split in any part of the race. So if you were going to automate a strategy based on times/splits its not going to be as clear a back the fastest horse in the last 2 furlongs out. The Horse that won was top of this list.

I am more inclined to read the VP VE CE numbers and read the race for manual trading with clearly defined offsets. Only 32 people have downloaded the pre-set Guardian rules set up by Dallas which mean maybe less than 20 people on Bet Angel are using TPD Data to trade with. I'm not seeing anybody put forward a positive strategy based on the data. My guess is it will die a slow death and people will lack trust in it mainly because they don't understand it. I am willing to share anything I learn over the next month or so if anybody else wants to work together let me know or it will just drift off.

TPD Splits.JPG
User avatar
jamesedwards
Posts: 1194
Joined: Wed Nov 21, 2018 6:16 pm

xtrader16 wrote:
Wed Sep 14, 2022 1:12 pm
Euler wrote:
Tue Sep 13, 2022 3:37 pm
I recorded a ton of races next to the TPD data to tease out characteristics of why each horse won, why and where it was obvious.

That felt a really sensible way to approach the problem.

I can read a race visually, but I've never been able to put numbers on it. TPD solved that problem for me.
I think it would take a few hundred races before you could start putting real sums of cash into an Automated system. I was looking a race yesterday where the Horse that won never had the lowest split in any part of the race. So if you were going to automate a strategy based on times/splits its not going to be as clear a back the fastest horse in the last 2 furlongs out. The Horse that won was top of this list.

I am more inclined to read the VP VE CE numbers and read the race for manual trading with clearly defined offsets. Only 32 people have downloaded the pre-set Guardian rules set up by Dallas which mean maybe less than 20 people on Bet Angel are using TPD Data to trade with. I'm not seeing anybody put forward a positive strategy based on the data. My guess is it will die a slow death and people will lack trust in it mainly because they don't understand it. I am willing to share anything I learn over the next month or so if anybody else wants to work together let me know or it will just drift off.


TPD Splits.JPG
Dallas has kindly provided a dozen or so templates, some of which have been downloaded hundreds of times. There will be many who have a deep knowledge of Guardian and don't need templates to utilise TPD within their rule files.

It's unlikely people will share their positive strategies because they will be sharing their edge, but there has been plenty of good advice already posted on this thread.

My advice is to play and play and play with the data and tease out the indicators that flag winning trades. I spent a full-time week learning TPD inside out last winter and incorporating it into my existing rules, and 9 months later I'm into the version 100+ and still continually tweaking.
User avatar
ShaunWhite
Posts: 8108
Joined: Sat Sep 03, 2016 3:42 am

xtrader16 wrote:
Wed Sep 14, 2022 1:12 pm
I think it would take a few hundred races before you could start putting real sums of cash into an Automated system. I was looking a race yesterday where the Horse that won never had the lowest split in any part of the race. So if you were going to automate a strategy based on times/splits its not going to be as clear a back the fastest horse in the last 2 furlongs out. The Horse that won was top of this list.

My guess is it will die a slow death.
People develop automated systems in many ways, and if you collect price and TPD data then you can backtest over thousands of races before using real money. You can do so relatively quickly and then scale up quite fast. You don't have to test an idea with live markets and real money.

Your observation about "a" race, as I said before every edge includes good picks and some terrible ones, only averages matter.

I don't think it shows any sign of dying, people were making good money with it even before BA included it and I know several people doing well using BA. In-running volumes are up too which all bodes well.
arbitrage16
Posts: 348
Joined: Tue Feb 14, 2017 7:27 pm

ShaunWhite wrote:
Wed Sep 14, 2022 7:04 pm
In-running volumes are up too which all bodes well.
I always hear conflicting reports about this Shaun, but great if this is true - where are you seeing this info and over what period please?
User avatar
ShaunWhite
Posts: 8108
Joined: Sat Sep 03, 2016 3:42 am

arbitrage16 wrote:
Mon Sep 19, 2022 7:24 pm
ShaunWhite wrote:
Wed Sep 14, 2022 7:04 pm
In-running volumes are up too which all bodes well.
I always hear conflicting reports about this Shaun, but great if this is true - where are you seeing this info and over what period please?
I'm glad you challenged me on that, it made me look at the data again that's incorrect. (the source is the api stream I've been storing since 2019-01)

I haven't updated these for the last few months but .....(1) total IP vol is accumulating at approx the same rate as usual. (2)The proportion of IP vol vs total vol is up slightly but hardly significant as that's on (3) an overall market decline. I'll get on top of my summer stats soon and let you know if there's a marked change. My monthly 'state of the markets' reports are another 90% completed project but they fall into the 'interesting' rather than 'essential' pile so keep getting overlooked. Seriously though always happy to be challenged and to be wrong, making assumptions is my pet hate and it looks like I made one based on hearsay.

(1)Cummulative vol
Screenshot_6.jpg
(2)Per race proportion
Screenshot_5.jpg
(3)Overall avg vol
Screenshot_7.jpg
Screenshot_7.jpg (24 KiB) Viewed 333 times
Trader Pat
Posts: 4324
Joined: Tue Oct 25, 2016 12:50 pm

Thats really interesting, thanks for posting Shaun.

I wonder if IP is up due in part to TPD?
User avatar
ShaunWhite
Posts: 8108
Joined: Sat Sep 03, 2016 3:42 am

Trader Pat wrote:
Mon Sep 19, 2022 9:18 pm
Thats really interesting, thanks for posting Shaun.

I wonder if IP is up due in part to TPD?
No probs. its tough keeping on top of the info so a reason to do it is welcome...... for a serial procrastinater.

Dunno, it's been around for a while. I don't know how much new money the BA TPD version has introduced. I don't have the stats to hand but the number of transactions is well up, and the avg size is down. That to me points to a growth in automation so a firm "maybe" :)
User avatar
jamesedwards
Posts: 1194
Joined: Wed Nov 21, 2018 6:16 pm

ShaunWhite wrote:
Tue Sep 20, 2022 12:14 am
Trader Pat wrote:
Mon Sep 19, 2022 9:18 pm
Thats really interesting, thanks for posting Shaun.

I wonder if IP is up due in part to TPD?
No probs. its tough keeping on top of the info so a reason to do it is welcome...... for a serial procrastinater.

Dunno, it's been around for a while. I don't know how much new money the BA TPD version has introduced. I don't have the stats to hand but the number of transactions is well up, and the avg size is down. That to me points to a growth in automation so a firm "maybe" :)
Don't forget the minimum bet decreased from £2 to £1 over this time. Did the change correspond with drop in avg size?
Trader Pat
Posts: 4324
Joined: Tue Oct 25, 2016 12:50 pm

Its a shame that pre off seems to be going in the opposite direction. I don't have the figures but even just looking at videos I recorded from 2016 and 2017 volume and liquidity have fallen off a cliff in a lot of cases.
User avatar
jimibt
Posts: 3302
Joined: Mon Nov 30, 2015 6:42 pm
Location: South West Scotland

Trader Pat wrote:
Tue Sep 20, 2022 9:26 am
Its a shame that pre off seems to be going in the opposite direction. I don't have the figures but even just looking at videos I recorded from 2016 and 2017 volume and liquidity have fallen off a cliff in a lot of cases.
i'm not saying that folk have switched out to crypto/options/futures etc, but given the skills required to master pre-race trading, then i can see a huge synergy in working against these emerging markets. i'm centered now on futures in the crypto markets which is in many ways akin to backing and laying on pre-race. add in a SMALL dash of leverage and your're in the arena for making decent money on small moves.

like pre-race, it's all about having a decent R:R, rather than being right all the time.
arbitrage16
Posts: 348
Joined: Tue Feb 14, 2017 7:27 pm

jimibt wrote:
Tue Sep 20, 2022 9:51 am
Trader Pat wrote:
Tue Sep 20, 2022 9:26 am
Its a shame that pre off seems to be going in the opposite direction. I don't have the figures but even just looking at videos I recorded from 2016 and 2017 volume and liquidity have fallen off a cliff in a lot of cases.
i'm not saying that folk have switched out to crypto/options/futures etc, but given the skills required to master pre-race trading, then i can see a huge synergy in working against these emerging markets. i'm centered now on futures in the crypto markets which is in many ways akin to backing and laying on pre-race. add in a SMALL dash of leverage and your're in the arena for making decent money on small moves.

like pre-race, it's all about having a decent R:R, rather than being right all the time.
interesting, and this kind of tallies with my other thread about finding a second market to trade. Anywhere you'd suggest to get started with learning to trade crypto?
User avatar
xtrader16
Posts: 308
Joined: Sun Feb 26, 2017 1:00 pm

I've scrapped my TPD Data subscription. I've work on my discipline for a few years to avoid In Play gambling and in a matter of days TPD pulls me straight back in. Losses accumulating fast. The data just isn't accurate enough. I've been using VE - VP -CE on flat racing but it seems pretty inconsistent. I would have thought I could have downloaded the data on an excel file after each race or get access to historical data but it seems not. I'm sure if you have enough time on your hands you could find a profitable strategy but I'd rather stay away from In Play Horses - TPD is not for me unless it gives me an excel file of data in one second increments of every race.
User avatar
Dallas
Posts: 19603
Joined: Sun Aug 09, 2015 10:57 pm
Location: Working From Home

xtrader16 wrote:
Sun Sep 25, 2022 2:11 pm
I've scrapped my TPD Data subscription. I've work on my discipline for a few years to avoid In Play gambling and in a matter of days TPD pulls me straight back in. Losses accumulating fast. The data just isn't accurate enough. I've been using VE - VP -CE on flat racing but it seems pretty inconsistent. I would have thought I could have downloaded the data on an excel file after each race or get access to historical data but it seems not. I'm sure if you have enough time on your hands you could find a profitable strategy but I'd rather stay away from In Play Horses - TPD is not for me unless it gives me an excel file of data in one second increments of every race.
It can do it in 200ms increments (depending on your subscription)

You just put the information into Stored Values then export them at the end of each day into Excel

Here an example here which stored all data, there's a few more been added since that file was created so anyone using this will need to add the most recent types of data to SV's
viewtopic.php?f=73&t=24701
User avatar
ShaunWhite
Posts: 8108
Joined: Sat Sep 03, 2016 3:42 am

xtrader16 wrote:
Sun Sep 25, 2022 2:11 pm
Losses accumulating fast. The data just isn't accurate enough.
Can you use tiny stakes until you've sussed it? Or is it tempting you to let your full pre-race stake run? PP and IP are so unrelated they might as well be different sports so trade one, accept your result, then trade the other. If necessary set up an autoclose bot to remove the temptation (that should be an account level setting Imo or a BA setting it's so important) .

And data accuracy, if there's a bias then try to pin it down and exploit it, if there isn't a bias then it's annoying but long term it will be a wash.

I feel your pain though, getting on top of a new info stream on top of the price data stream takes some doing but ultimately all info is good info, so treat it as a whole new seperate project rather than lumping it into your existing methods.
arbitrage16
Posts: 348
Joined: Tue Feb 14, 2017 7:27 pm

ShaunWhite wrote:
Sun Sep 25, 2022 5:20 pm
xtrader16 wrote:
Sun Sep 25, 2022 2:11 pm
Losses accumulating fast. The data just isn't accurate enough.
PP and IP are so unrelated they might as well be different sports so trade one, accept your result, then trade the other.
This is good advice and I think is a useful principle, however in practice, at least for me, it falls apart. The result garnered in PR does a funny thing to that delicate balance we call trader mindset, some psychic weight is adjusted somehow and irrespective of whether the PR race is green or red, there is a tendency to adjust that weight IR, with predictable results.

For me, my advice is for each race, when learning, do one or the other. Would be interested to hear if established traders have no issue switching from one to the other.
Post Reply

Return to “Bet Angel - Total Performance Data - FAQ's”