Opinions on TPD - is it worthwhile?

Integrate live racecourse data into your favourite Betfair trading interface from TPD.
Post Reply
User avatar
PDC
Posts: 2272
Joined: Sun Jul 24, 2016 5:52 pm

Dallas wrote:
Sat Dec 04, 2021 11:18 am
if it happened a thread would immediately pop up with quite a few unhappy people flagging it
Maybe anyone who sees it happening does exactly this, a kind of Betfair Down thread. Then people will be able to get an idea of how often it is occurring, if at all and the reason for it.
parrettt
Posts: 49
Joined: Sun Jul 28, 2013 11:31 pm

I think it is safe to say that you won't win anything if you follow their user guide.
They say that a low VF means a calm horse who will have conserved loads of energy and be in contention at the end. In fact, A high VF however can indicate a horse that was being held back ready to explode and win at the end. You end up looking at a lot of information with no real knowledge of whether it is useful. It is v expensive compared to bet angel I had high hopes for it but so far I look at the Bet Angel market overview more.
They say in one of their videos that the information you glean can tell you who the likely winner will be long before the final few furlongs. Sometimes you can and sometimes you cannot which does not really help because we can all do that.
If someone can harness their information usefully there may be hope. It has few UK races and the USA ones are not full of money so you are not going to get rich there!
SomeDude
Posts: 28
Joined: Fri Sep 24, 2021 5:32 am

I believe I got burned on this last night my time, day time UK. I'm still gathering data to support, what happened. On review of video, the horse TPD showed in first place was actually never leading and, in fact, fell before my automated bets were placed. This should never happen.
I'm fairly certain my automation is solid because I've been using it successfully for other TPD races where data matched events and video.
This was an anomaly, but it could have been catastrophic if I fully trusted the TPD feed and my stakes had been large. Thankfully, I'm still quite new to TPD and the stakes were relatively low. Also, since I'm just testing, I hadn't fully built in the error catch processes into my automation.. had I put max odds conditions on my back, the bets wouldn't have been placed.
Still, my experience showed me that errors on TPD's part definitely happen. Do not fully trust the data they provide. If you choose to use it, ensure you've got error correcting processes and logic in place.
User avatar
ilovepizza82
Posts: 490
Joined: Thu Nov 02, 2017 3:41 pm
Location: Sewers
Contact:

Cant upload the video here.
It says its too big tho its barely 6mb...lol.
sionascaig
Posts: 1053
Joined: Fri Nov 20, 2015 9:38 am

I'm sure TPD publish the tolerances of the system they deploy and defo read something about dropped / reconnected signals and how they are handled. Wouldn't any strategy have to factor this in to some degree (or not if you are happy with the variance)...
sniffer66
Posts: 1666
Joined: Thu May 02, 2019 8:37 am

ilovepizza82 wrote:
Sun Dec 05, 2021 6:51 am
Cant upload the video here.
It says its too big tho its barely 6mb...lol.
Pu it in the cloud somewhere and post a link
SomeDude
Posts: 28
Joined: Fri Sep 24, 2021 5:32 am

SomeDude wrote:
Sat Dec 04, 2021 9:34 pm
I believe I got burned on this last night my time, day time UK. I'm still gathering data to support, what happened. On review of video, the horse TPD showed in first place was actually never leading and, in fact, fell before my automated bets were placed. This should never happen.
I'm fairly certain my automation is solid because I've been using it successfully for other TPD races where data matched events and video.
This was an anomaly, but it could have been catastrophic if I fully trusted the TPD feed and my stakes had been large. Thankfully, I'm still quite new to TPD and the stakes were relatively low. Also, since I'm just testing, I hadn't fully built in the error catch processes into my automation.. had I put max odds conditions on my back, the bets wouldn't have been placed.
Still, my experience showed me that errors on TPD's part definitely happen. Do not fully trust the data they provide. If you choose to use it, ensure you've got error correcting processes and logic in place.
Hi mods... I'm not sure if this belongs here or needs to be moved to its own thread, but will post here for now. Apologies if incorrect!

Ok... so here's the video: https://www.attheraces.com/atrplayer-popup/VOD/1242941 of Chepstow R1 4 December 2021.

The automation I was testing for this is the "Back the Leader" Guardian automation bot at viewtopic.php?f=73&t=24687.

I had made minor alterations to staking, race percentage, and distance, but had not altered the ranking or selection criteria in any way. This automation had been working well prior to this race and has worked well since.

Bet Angel logs show market in play at 00:38:19 local time.

"The Galloping Bear" fell 5 minutes and 18 seconds into the race, or 00:43:37 local time.

The automation is supposed to trigger when a gap opens between the leader and the horse in second place and it works well almost all of the time.

The automation also has a condition that the race information has been updated from TPD within the last 0.8 seconds. A comms interruption should mean no bet is placed.

Automation first triggered at 00:43:43... 6 full seconds after "The Galloping Bear" fell... because a gap had indeed opened. The bot placed 30 back bets over the next 8 seconds on the horse in first place at 1.01. Each bet was fully matched at the time of placing at 1000...because the automation thought "The Galloping Bear" was leading.

"The Galloping Bear" never led in this race, though it pulled within a nose momentarily just prior to the fall.

As I see it, there are two possibilities here:
1) The TPD feed was wrong; either transponder attached to wrong horse, or a mix-up in transmission.
2) The Bet Angel automation mixed up horse order for some reason, i.e., a bug.

Am I missing something? Is there another explanation as to what happened? Is there a way to rule out one or the other, e.g., is there a full log of all TPD data accessible somewhere?

Thanks all!
StellaBot
Posts: 818
Joined: Thu Jan 26, 2017 11:52 am

Thanks for feedback

I apologise for negative posts I have submitted

In response I will only do positive feedback
based on evidence
I will present the evidence and outcomes
;)
Last edited by StellaBot on Mon Dec 06, 2021 12:37 am, edited 1 time in total.
StellaBot
Posts: 818
Joined: Thu Jan 26, 2017 11:52 am

May I add

The whole theory about times and distance is Clunky with TPD at mo
but a step in right direction
in the same vain in my opinion
I think time based systems are alwyas a step forward
Phil Bull , Nick Mordin etc
In my opinion there are other factors to consider
But Im not a pro ;)

I will also add on this thread
I apologise for negative threads I have posted on here

I have been passionate for many years about speed, time, going distance ,rail movements etc etc in racing
jamesg46
Posts: 3769
Joined: Sat Jul 30, 2016 1:05 pm

StellaBot wrote:
Mon Dec 06, 2021 12:35 am
May I add

The whole theory about times and distance is Clunky with TPD at mo
but a step in right direction
in the same vain in my opinion
I think time based systems are alwyas a step forward
Phil Bull , Nick Mordin etc
In my opinion there are other factors to consider
But Im not a pro ;) ;)

I will also add on this thread
I apologise for negative threads I have posted on here.

I have been passionate for many years about speed, time, acceleration etc etc in racing
Nobody has a problem with a positive or negative comment (over the long term anyway, trust me I’ve pushed the boundaries), the issue is that you submit 400 comments that could be condensed into one or two.
StellaBot
Posts: 818
Joined: Thu Jan 26, 2017 11:52 am

Read 100Os of comments
Pick 1 that makes sense
or move on ;)
jamesg46
Posts: 3769
Joined: Sat Jul 30, 2016 1:05 pm

StellaBot wrote:
Mon Dec 06, 2021 12:55 am
Read 100Os of comments
Pick 1 that makes sense
or move on ;)
Comments usually come in the form of a question or reply, In that very moment I’m sure the people asking the questions have genuine curiosity, which makes sense. The people answering have a knowledge worth sharing, that may not make sense to everyone but may help someone… either way the majority of those comments and replies take shape in 1 submitted question or reply & then possibly mutate into a conversation.

From here I’ll move on. Anymore and it’ll destroy the thread because I’m pulling out all the stops to be friendly.
User avatar
decomez6
Posts: 651
Joined: Mon Oct 07, 2019 5:26 pm

SomeDude wrote:
Sat Dec 04, 2021 9:34 pm
On review of video, the horse TPD showed in first place was actually never leading and, in fact, fell before my automated bets were placed. This should never happen.
custom ranking is more accurate when done using the current prices (refreshed at every instance). in my opinion , Bf Algo are the most accurate representation of IP% chances of the horse winning the race .
i think its safer to include an additional price custom ranking rule . this should work to complement the TPD custom ranking. BOTH must be in the affirmative before a bet is placed.
User avatar
megarain
Posts: 2041
Joined: Thu May 16, 2013 1:26 pm
Contact:

There is a telegram link where users discuss the product.

https://t.me/BetMover

The nature of these things is that you see more negative than positive, so maybe take with a pinch of salt.
User avatar
Shaung89
Posts: 181
Joined: Wed Aug 08, 2018 5:39 pm

megarain wrote:
Mon Dec 06, 2021 4:36 am
There is a telegram link where users discuss the product.

https://t.me/BetMover

The nature of these things is that you see more negative than positive, so maybe take with a pinch of salt.
If you scroll back through the chat when joining you will soon realise that these GPS failings are not a 'one off' also.
Post Reply

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