Anyone know how the 'Percentage Projections' work?

[classhandicapper]

Originally Posted by ubercapper

I received this from TPD


Attachment 33758

Looks good.

I think this is the early preparation and testing for eventually taking bets during the race.

[David Siegel]

For context, I used to be the president of TrackMaster and currently work for Equibase on a limited basis advising and working with tracks regarding the potential for installing Equibase's Hybrid GPS system which consists of trackers on horses and wireless beams.

There are two reasons for the in-race win percentages (odds) being posted in Maryland. First, is the entertainment value. Many sports broadcasts show such things during the game, and while in-game wagers exist in certain jurisdictions, way less than 1% of the viewers use these win percentages to actually place a wager.



Second, the win percentages (converted to odds less a small vig) are used by bookmakers TODAY in the UK where people can bet on a a horse to win at any time during the race, so long as the bookmaker subscribes to the algorithm our technology partner (TPD) offers which intelligently generates these odds every second or so during the race.


So far, the algorithm is clearly working as if it was way "off", the bookmakers would be getting slammed and they would stop taking such wagers. This has not happened - quite the other way around - as the handle on in-race wagers has grown nicely.


We would hope that since in-race wagering could never be pari-mutually funded, this could be the first non-parimutual offer (outside NJ) and could help move jurisdictions to accept fixed odds wagering.


I rarely post here, but thought I could help. Happy to address other questions on this topic.

[Poindexter]

Originally Posted by David Siegel

For context, I used to be the president of TrackMaster and currently work for Equibase on a limited basis advising and working with tracks regarding the potential for installing Equibase's Hybrid GPS system which consists of trackers on horses and wireless beams.

There are two reasons for the in-race win percentages (odds) being posted in Maryland. First, is the entertainment value. Many sports broadcasts show such things during the game, and while in-game wagers exist in certain jurisdictions, way less than 1% of the viewers use these win percentages to actually place a wager.



Second, the win percentages (converted to odds less a small vig) are used by bookmakers TODAY in the UK where people can bet on a a horse to win at any time during the race, so long as the bookmaker subscribes to the algorithm our technology partner (TPD) offers which intelligently generates these odds every second or so during the race.


So far, the algorithm is clearly working as if it was way "off", the bookmakers would be getting slammed and they would stop taking such wagers. This has not happened - quite the other way around - as the handle on in-race wagers has grown nicely.


We would hope that since in-race wagering could never be pari-mutually funded, this could be the first non-parimutual offer (outside NJ) and could help move jurisdictions to accept fixed odds wagering.


I rarely post here, but thought I could help. Happy to address other questions on this topic.

How are delays handled? I can’t watch an nba game without it being significantly behind the live action. I check my sports book app not the tv for the most current score.I know there are delays in horse racing broadcasts. Does anybody actually have a completely live broadcast. Will we have that technology at some point if not. The concept is very interesting. Betting against people who may be 7 seconds ahead of you sounds like a recipe for disaster.

[Saratoga]

I agree with with whomever calls this useless.....

In Roberts post with the 2 pictures , you see the 2 in 4th position as the 2-1 favorite in picture 1..

That is exactly where he wants to be in his winning races but the AI gave him a

17% shot ....and then he gets a 54% when everyone knows hes winning???

Totally crap!!!

[denniswilliams]

Originally Posted by David Siegel

For context, I used to be the president of TrackMaster and currently work for Equibase on a limited basis advising and working with tracks regarding the potential for installing Equibase's Hybrid GPS system which consists of trackers on horses and wireless beams.

How accurate is this data? Why are there missing races?



Thanks

[ARAZI91]

It definitely does not originate from Betfair directly - as a 100% exchange player in the UK, in running betting is shown in the form of decimal odds on every race in the UK, so you could just convert those implied odds to percentages, although they change that fast you would be better off using software such as BetAngel or Betmover which now utilise sectional times within the in running process

BetAngel

BetMover


* There have been teething problems and in-accuracies galore though, you just need to look at some of the threads on the BetAngel forum - the sectional data is supplied by Total Performance Data /G-Max who i understand are involved in some US tracks but they have recently went back to the old "beam" technology - is that correct ??

[denniswilliams]

Originally Posted by ARAZI91

* There have been teething problems and in-accuracies galore though, you just need to look at some of the threads on the BetAngel forum - the sectional data is supplied by Total Performance Data /G-Max who i understand are involved in some US tracks but they have recently went back to the old "beam" technology - is that correct ??

You can scrape the GPS data on Equibase. Yet, their 'other' results charts remain in encrypted PDF. I asked earlier but it's fairly obvious.

[ARAZI91]

Most of the big in-running punters in the UK & IRE use "drones" now, basically controlling them near the racecourse. Years ago it was positional spotters at various points of the track and the use of live mobile phones feeding back info to a HQ.
Here's an article on it https://www.wired.co.uk/article/horse-racing-drone

The problem is that liqudity has dropped by around 70% from Betfair's Glory days back in the late 2000's.
You could also hire out rooms where apparently the video feed from the track is a lot faster than commercial racing TV which will always be around 4-6 seconds behind the live event depending on the racing channel.

[David Siegel]

In answer to your question and some of the comments earlier, no video is really live as it takes time for the video to be transmitted to the viewing platform. At a live track, it is as close to real time (live) as possible. However, when you get into all of the satellite distribution, it can take a few seconds.


That said, the data used in the algorithm which produces odds (win percentages) takes around 0.5 seconds. So our feeds are generally faster than the actual video.


Hope this helps (a little).

[ubercapper]

Originally Posted by denniswilliams

How accurate is this data? Why are there missing races?



Thanks

Please clarify if you are talking about the GPS charts missing for certain races, or the in-running percentages. I can address any questions about missing charts.

[ubercapper]

Originally Posted by ARAZI91

It definitely does not originate from Betfair directly - as a 100% exchange player in the UK, in running betting is shown in the form of decimal odds on every race in the UK, so you could just convert those implied odds to percentages, although they change that fast you would be better off using software such as BetAngel or Betmover which now utilise sectional times within the in running process

BetAngel

BetMover


* There have been teething problems and in-accuracies galore though, you just need to look at some of the threads on the BetAngel forum - the sectional data is supplied by Total Performance Data /G-Max who i understand are involved in some US tracks but they have recently went back to the old "beam" technology - is that correct ??

Nearly every North American track which uses GMAX for timing is using a combination of beams and GPS. This is known as "hybrid beam" technology where there is a beam at the start and finish, but the beams are wireless so do not require miles of cables and at times significant maintenance. Horses still carry two GPS trackers for everything else, including graphics and fractional times.

[denniswilliams]

Originally Posted by ubercapper

Please clarify if you are talking about the GPS charts missing for certain races, or the in-running percentages. I can address any questions about missing charts.

If you have a minute, look at the fractions for the HOLY BULL,


The race splits are consistent in the GPS and PDF charts.



However, 4F & 6F fractions for the leader/winner do not match with the race splits:


50.58 / 50.53

1:14.34 / 1:14.25

[ubercapper]

The GPS charts default sort is the winner. Each column is sortable separately. If you click on the 4 1/2F arrow on the green (leader's bar) and sort in order by time, you will see the leader's time at that exact segment is 50.53 which is the same as the time as the time in the header row, which matches the official chart data (below the GPS portion of the chart) as well. The same is true at the 2 1/2F mark with 1:14.25 displayed.

Holy Bull GPS chart.png

[denniswilliams]

Originally Posted by ubercapper

The GPS charts default sort is the winner. Each column is sortable separately. If you click on the 4 1/2F arrow on the green (leader's bar) and sort in order by time, you will see the leader's time at that exact segment is 50.53 which is the same as the time as the time in the header row, which matches the official chart data (below the GPS portion of the chart) as well. The same is true at the 2 1/2F mark with 1:14.25 displayed.

Attachment 33840

Sorry. Didn't look close enough. PDF chart has Hades leading all but the 3/4 call.

GPS has him 2nd at the 3/4 call and 3rd at the 1/2. I didn't pay attention and assumed he had led throughout.