Just lost money on TPD
Number 2 won
TPD delay or wrong
TPD use two in each saddlecoth to make them as accurate as possible, but if you are using them to judge small distances, you are going to have to build tolerances into them. It's just not realistic to bet or trade on those tolerances, it sort of misses the point of all that glorious data that they pump out.
- wearthefoxhat
- Posts: 3205
- Joined: Sun Feb 18, 2018 9:55 am
Just watched a re-run of that race.
Jens boy kicked 2 lengths clear off the final bend (1 to jump), as they approached the last hurdle, the actual winner jumped to the front before the last hurdle and was never headed on the short run in, albeit a close finish in the end.
Would be interesting to see what the TPD showed on that one before the final hurdle was jumped.
Just guessing, it probably flashed up a clear advantage around the bend, so those that acted on that information would have quickly seen it change within seconds.
My bot was kicking in when Jens Boy was leading with over 8m.
If it now was 8m? Could have been 3m or less.
I have also seen delays in TPD data several times. And I am not the first one.
It misses the point to offer this service if it is not more correct than this. It is just a clever way to make money until people realise that it is not trustworthy.
If it now was 8m? Could have been 3m or less.
I have also seen delays in TPD data several times. And I am not the first one.
It misses the point to offer this service if it is not more correct than this. It is just a clever way to make money until people realise that it is not trustworthy.
Bet Angel recieves no funding at all from TPD, we have just integrated for people to use. There is no incentive for us to promote it's use.
I realise it's painful to take a loss, but you have to be realistic when using data and expect odd things to happen now and again. You have to trap those errors and work within tolerances. Just the same as if you automate something in a market. Strange and odd things will happen. You need to account for that.
I realise it's painful to take a loss, but you have to be realistic when using data and expect odd things to happen now and again. You have to trap those errors and work within tolerances. Just the same as if you automate something in a market. Strange and odd things will happen. You need to account for that.
- ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
- ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
Or accept it. Automation is about the 90% in the middle and trying to trap edge case can sometimes impact on that. Automation is dumb and occasionally does dumb things, that's the price you pay.
(I'm not doubting you can reduce the edge cases, it's just not especially easy)
I put this thread up to warn people who want to try TPD to be careful. I have been working with electronics for 40 years so I know what can be expected. When I write that you should be able to expect more from TPD, I do that from a very realistic stand point. I have lost money before so that do not become me very much. I usually make my money from watching racing on video. This I will continue with. For all you who want to have more data I suggest that you buy the services of TPD and try it out. I will not. If anyone wants to live in a dream world where you think it is possible to guard yourselves against poor GPS quality, well that is up to you. Everyone is entitled to their own opinion.
- jamesedwards
- Posts: 2234
- Joined: Wed Nov 21, 2018 6:16 pm
The initial mistake that I made was assuming TPD is about smashing 1.01 on the winner a second ahead of everyone else. It is far more than that. The plethora of data is far bigger an edge than the time advantage over live stream.