old pool data

[green80]

is there any way to see day old pool data? Such as how much was bet to win on a horse. I know twinspires shows this live but after the race is over I can't find pool data anywhere. Any suggestions?

[Track Collector]

is there any way to see day old pool data? Such as how much was bet to win on a horse. I know twinspires shows this live but after the race is over I can't find pool data anywhere. Any suggestions?

Equibase and Brisnet results charts can help.

They do not show the individual pools (win/place/show) but the entire total. I have found that the pool breakouts are approximately 65%/25%/10% respectively of the total pool. From there you can do the math (using the odds to $1 figure in the charts) to get a very close idea of how much was wagered on to win on a specific horse. Remember too that in your math calculations you will need to know the specific track takeout for the track in question.

In a more precise manner, I know there exists some special programs that grab various tote board info. Maybe someone who uses one of these programs can chime in.

Hope this helps.

[davew]

xpressbet has the results of all races that day until around midnight PST when they reset for next day

I can pull up results and closing pools (WPS) as well as probables for exactas, DD,P3,... for every race they carried that day

[TheGhostOfOscarB]

I can't imagine what use that information would be , unless you're looking to do some w/p/s analysis.

The amount bet on the winner is pretty much described by the closing odds, as has been pointed out.

[trying2win]

Try this link:

https://classic.ebetusa.com/getting-started/


They keep the pool data online for each race on every track they carry for awhile. Just choose a track, then a particular race you want the results for, then click on the 'DISPLAY' button...after that, look underneath their online toteboard, and click on the line that reads 'VIEW POOLS'.

What I don't know is how long they keep that data online after midnight, before they cancel the previous day's parimutuel pool totals.

T2W

[maddog42]

xpressbet has the results of all races that day until around midnight PST when they reset for next day

I can pull up results and closing pools (WPS) as well as probables for exactas, DD,P3,... for every race they carried that day

I too am interested in this Data for WPS study. I have been meaning to do a study of efficiency of exacta versus win pools.
thnaks

[green80]

some horses show unusual betting activity at the local track before the off-track money comes in. Looking at a horse at LAD the other day that had never finished better than 6th lifetime, always went off at long odds, moving up from 5000 maidens to 20k maiden, and bet down to 4/1 until the late $ came in. Ran a close 2nd. Would have won with a clean trip.

[davew]

one disadvantage with doing anything with closing data is that varying percentages of money shows up into the pool after you can make a bet

it seems the smaller the track with simultcasting, the higher percentage of money shows up after they start

I have seen some pools, where over 75% of the total appears after they are off


so big overlays, turn into underlays when all is said and done

if you want to do some exacta/straight pool analysis, you probably need to do it at 1-2 minutes before post - this would not be extremely difficult if you printed the screen of most ADWs odds board

[Pensacola Pete]

Premier Turf Club keeps that information until the end of the day.

[iceknight]

Equibase and Brisnet results charts can help.

They do not show the individual pools (win/place/show) but the entire total. I have found that the pool breakouts are approximately 65%/25%/10% respectively of the total pool. From there you can do the math (using the odds to $1 figure in the charts) to get a very close idea of how much was wagered on to win on a specific horse. Remember too that in your math calculations you will need to know the specific track takeout for the track in question.

In a more precise manner, I know there exists some special programs that grab various tote board info. Maybe someone who uses one of these programs can chime in.

Hope this helps.

Remember too that in your math calculations you will need to know the specific track takeout for the track in question.
Arent the pools already shown after subtracting the takeout?

[therussmeister]

No, pools shown are before takeout.

[Dr Jersey Jim]

If you spend enough time working with available pool data, you will find that the pool numbers for individual horses are pretty fuzzy and the math does not work out cleanly. I'm pretty sure that's why they are harder to find than pool totals.

<RANT>
A lot of the numbers on the tote are fuzzier than people expect. For example, it's 2011 but when you cash in a win ticket on a horse that is 5-1 on the tote, you might get back $12.40 because the odds are really 5.2 after breakage (maybe something like 5.2999999 before breakage) but the toteboard has not been updated to display odds in the form 5.2-1 properly even though it's the 21st century.

Don't get me started on how this effects understanding probabilities and betting strategies UGH!

I believe that the pari-mutuel tote system needs a major overhaul to bring it up to date but there are many with the power to make it happen that believe that it would be detrimental to the sport.

Being involved in that overhaul would be a 'dream come true' for me :)
<END OF RANT>

JJ

[therussmeister]

That's why I import Supertote into Excel and calculate the odds down to the penny. I allows me to see trends developing. Instead of the track tote board showing:

4 mtp - 5/1
3 mtp - 5/1
2 mtp - 5/1
1 mtp - 5/1

I see:

4 mtp 5.96/1
3 mtp 5.73/1
2 mtp 5.44/1
1 mtp 5.03/1

[Dr Jersey Jim]

When you calculate the final odds and knock off the breakage, does it match the payout? I often come out $0.20 higher than it pays. Occasionally more.

I use a technique that maps the distribution of probabilities based on the odds and the track take. I stopped using pool data because the final pools posted rarely added up to the final pool total (which I imagine comes from rounding parlay bets and such).

This doesn't affect the game much as trends and distribution are pretty accurately preserved but it does not allow me to measure accurate effects of breakage and I like my math to come out correctly :)

I know there's nothing sneaky going on but I think the numbers you see are not as accurate as the numbers being used to calculate the actual payout. If that is in fact the case, I would like access to more accurate numbers.

JJ

[therussmeister]

My spreadsheet gives the proper payout very close to 100% of the time. The few times it is off, I think, is due to net pool pricing.

With net pool pricing there are, I assume, two different takeout rates depending on where the bet was made. One rate for US bettors, and a different rate for Canadians. In order to properly calculate exact odds, you would have to know how much money was bet on each horse at each rate. Information not available to me.

I believe calculating odds based on US takeout rate gives you the wrong answer only when Canadian bettors have a significantly different opinion, such as if US bettors wagered 10% of the win pool on "Horse A", but Canadians wagered 20% on "Horse A".

[Dr Jersey Jim]

I'll have to give pool figures another look then.

My error comes from adjusting everything evenly to maintain the distribution.

It's kind of like assuming everyone in a row moves forward two steps when in reality some move forward one step and some move forward three steps. The average is still two steps but my numbers are off for individuals.

To get back to the topic of this thread, I too would like a source for old pool data thanks to therussmeister.

JJ

[Dr Jersey Jim]

I've tried using pool data from three different places including my ADW and I don't even come close to calculating correct payouts.

Therussmeister, where do you get your pool data from?

Thanks
JJ

[GameTheory]

I've always used the Racing Channel/Phonebet toteboard for recording data (I have a program that does that) -- it is the easiest to deal with and the most stable. It is true that the final pools often don't add up the amount that is shown in the Equibase chart for the WPS pool, but nevertheless the displayed final odds always seem to be correct. I usually don't calculate exact odds from the dollar amounts, but I'll do a couple of tests. A quick look at the final race from Belmont yesterday, and the pool totals I've got on the toteboard add up to $218,702, but the chart says $260,557, quite a bit more.

[GameTheory]

Yep, looking at a few races and using the raw dollar amounts I was able to calculate the exact final odds as shown in the chart. Not sure though where that extra money comes from shown in the pool total on the chart.

[Robert Goren]

This might be area for someone to pursue. There does seem to at least some market for this data. If I was in better health, I would look into providing this service for a fee. I don't know how big the market is or if it can be done profitably, but it might be an area to look at if you in the business of marketing horse race data.

[GameTheory]

This might be area for someone to pursue. There does seem to at least some market for this data. If I was in better health, I would look into providing this service for a fee. I don't know how big the market is or if it can be done profitably, but it might be an area to look at if you in the business of marketing horse race data.You mean the old data? Probably not good for much unless you have a program to also record and do something with the real-time data. (The real-time feed you couldn't provide legally to a third-party -- the archived data probably no one would make a fuss.) I've sold my program to a few people, and then they can record it themselves providing they have an account to the toteboard. There is also the ATR program with all its fancy graphs, etc.

[Robert Goren]

I mean old data. I am not sure there is large market for it, but someone is always asking where to get old pool data. Live pool data is readily available online. I am not willing pay very much for it and probably nothing right now, but old show pool holds some interest to me. Not just the total show pool, but how much was bet on each horse. If my health gets better, I would mind sorting through a bunch of that kind of data.

[therussmeister]

Therussmeister, where do you get your pool data from?

Thanks
JJ
I get it from BRIS supertote, real time, not old pool data. Make sure you have the right takeout rates, there are some sources that don't always have up-to-date information, (including DRF).

[GameTheory]

I get it from BRIS supertote, real time, not old pool data. Make sure you have the right takeout rates, there are some sources that don't always have up-to-date information, (including DRF).If you've got the charts with final odds, it is fairly easy to reverse-engineer the takeout rates given enough sample races from a particular track.

[BIG49010]

Had a friend who went to the Illinois racing board to try and get the data from the Chicago tracks, he was told sorry we can't make that data available at any price! I really never could figure what he thought he would learn from it.

[vikingrob]

Had a friend who went to the Illinois racing board to try and get the data from the Chicago tracks, he was told sorry we can't make that data available at any price! I really never could figure what he thought he would learn from it.

I would consider making this the subject of a request under open records laws.

[Ted Craven]

is there any way to see day old pool data? Such as how much was bet to win on a horse. I know twinspires shows this live but after the race is over I can't find pool data anywhere. Any suggestions?
AmWest has at least 60+ days of historical pool data for detail pool queries.

For example, here's BEL 9/28 race 9. $2,063 was bet in the Place pool on horse #8.

Also, PRX 8/06 race 1. $349 bet in the Show pool on horse #2

All pools are available.

FWIW, if anyone wants a referral to AmWest ADW (US based) along with whale-size rebates, please feel free to PM me.

cheers,

Ted

http://sartinmethodology.com/images/pa/BEL0928-9.png

http://sartinmethodology.com/images/pa/prx0806-1.png

[Dr Jersey Jim]

Those look much more accurate than what I was finding.
Thanks
JJ