Question about late-money action

[markgoldie]

Not a straight-pool player, but a question came to me combining certain possibly correct assumptions, which are:

(1) Whale groups may attempt to maximize profits by playing into even somewhat lower-handle venues.

(2) Whale groups tend to put the money in late.

(3) Whale groups are using relatively good information on which they base their handicapping.

(4) At all but the highest-handle venues (and even some of them), the show pool will not be large enough to attract whale-betting interest.

If some or all of these assumptions are correct, then it might be reasonable to imagine that based on the strength of late money, a possible anomaly involving the value offered in the show pool versus the win (and possibly) place pools could arise.

My question is: Has anyone looked into this? And if so, are such show-pool anomalies already being exploited by non-whale bettors using last minute wagering-action information?

[Dave Schwartz]

"Last minute wagering information" is too late to use.

By the time you get it, the horses are 20 seconds into the race.

[Charli125]

I've been playing around with this exact type of thing. The basic premise was that if I could find an overlay in the Place or Show pool based on the Win pool odds, would it be profitable. My spreadsheet basically calculated the amount that should be bet on each horse in the Place and Show pools in order to have the same percentage of the pool as exists in the Win pool.

Dave mentions the biggest issue that arises in this type of exercise. When I would look at ROI using final pool totals, it was profitable(I don't have historic info since it overwrote each time but I am searching for a way to store it so I can keep working on this angle). When I looked at ROI at 1 MTP or anything prior, it wasn't profitable.

As I understand it, the computer teams make money by making their bets at the last possible moment, and thus they can be profitable. I can't make bets that late, so it didn't work for me.

[markgoldie]

"Last minute wagering information" is too late to use.

By the time you get it, the horses are 20 seconds into the race.
Hey Dave:

Haven't been following all the posts recently, but I'd imagine you've looked into reverse engineering the whale action so as at least to predict where it might fall. Any luck with that angle?

[Cholly]

For the sake of argument, I would contest the assertion that “whale groups tend to put money in late”. Certainly they are putting a large part of their wager in late, but my guess is they are constantly feeding money into the pool, guiding the odds to a desired spot, much as a tugboat guides a big boat into dock.

They don’t need to wait until the last minute to know the final odds. Surely they have a damn accurate picture of the final odds and also the exotic pay-offs well before the pools first flash on the tote. If their methodology is so good at predicting the performance of a horse, it should be even better at predicting how the public is going to bet a race. Aren’t people more consistent in action than horses?... especially if you are contrasting a performance of a single horse against the performance of a large group of humans.

A very large bettor who already knows (or has an opinion of) a horse’s chances and the public’s likely action will want to influence the latter to the extent possible. He wouldn’t want the horse’s odds to drift too high and therefore attract “value “bettors. But neither could he hammer all the money in at the first drop, because that would attract unwanted attention from those bettors who look for “steam”.

I’m not a “whale” and have never knowingly talked to one, so this is pure speculation on my part--but I question the supposition that all the very large bettors are putting all their money in at the last flash. It would seem that if they wanted to get a certain horse at a certain price, they would be staking out that position early, letting their money flow into the pool in direct proportion to the public’s rate of entry into that pool.

[Robert Goren]

The late money almost aways lands on a short priced favorite or so it seems to me. The late money at places like belmont is huge.

[46zilzal]

chartists used to be able to tell where the money was going but now with asynchronous pulses coming in from the 5 major betting hubs, it is hard to differentiate the source.

[podonne]

I've been playing around with this exact type of thing. The basic premise was that if I could find an overlay in the Place or Show pool based on the Win pool odds, would it be profitable. My spreadsheet basically calculated the amount that should be bet on each horse in the Place and Show pools in order to have the same percentage of the pool as exists in the Win pool.


Sorry to be that guy who talks about a study but can't attribute it, however, I did read a paper\book by some people who were doing this and they found it could be profitable, BUT only if you did the analysis after the fact (later on in time looking at the final pool totals. When they tried to do it just before the race it didn't work, because the late money throws off the pool %s.

Lesson: The place and show pools may be inefficeint enought to turn a profit shortly before a race begins, but by the time the horses cross the line, they are pretty damn efficient.

It may be from one of these:

"Part 5 discusses the place and show markets, and the existence of the most welldocumented of
the racetrack's inefficiencies. Ritter (1994)', Asch, Malkiel and Quandt (1984,86)' and Hausch, Ziemba
and Rubinstein (1981)' test various approaches for identifying inefficiencies. Hausch, Ziemba and
Rubinstein (1981)' study an optimal growth betting strategy to exploit these inefficiencies. This strategy
is extended in Hausch and Ziemba (19851,1990a',1990b'),"

[podonne]

Dave mentions the biggest issue that arises in this type of exercise. When I would look at ROI using final pool totals, it was profitable(I don't have historic info since it overwrote each time but I am searching for a way to store it so I can keep working on this angle). When I looked at ROI at 1 MTP or anything prior, it wasn't profitable.


But, you already knew that. Sorry. I should really read an entire post before I go looking on the internet...

[podonne]

Right after I wrote that, I found this. Should make interesting reading for anyone interested in this topic. Granted they found that the system could be profitable, but it was 30 years ago.

"Efficiency (http://www.powerswings.com/wp-documents/ZIEMBA_1980.pdf) of the market for racetrack betting - Ziemba 1980

http://www.powerswings.com/wp-documents/ZIEMBA_1980.pdf

[Stillriledup]

Whales, or someone, gets all the pools.

If you see a 4-5 shot with a small amount of show money with 1 mtp, someone always sticks a few grand on that horse to take away the value. I'm very aware of the show pool, i always watch it after i've put my exotic plays in, i search for some disparity in the P and S pools and almost always, someone adjusts that pool to the proper amounts.

[Charli125]

But, you already knew that. Sorry. I should really read an entire post before I go looking on the internet...

Great minds think alike and all that!

Thanks for the link, that is an interesting article. I find it an interesting puzzle no matter what the outcome.

[therussmeister]

Whales, or someone, gets all the pools.

If you see a 4-5 shot with a small amount of show money with 1 mtp, someone always sticks a few grand on that horse to take away the value. I'm very aware of the show pool, i always watch it after i've put my exotic plays in, i search for some disparity in the P and S pools and almost always, someone adjusts that pool to the proper amounts.

I use supertote which allows me to display the pools as a percentage, which makes it easy to see disparities between w-p-s pools. Stillriledup is right, any disparity at 1mtp is minimized (not completely eliminated) and it goes both ways, meaning when there is more money in the show pool than win pool, that too gets adjusted. This, of course, does not happen when there is a bridge jumper.

As far as late money goes, two years ago I did a study, recording win pool totals every minute starting at 5 mtp. I found typically three out of the top four favorites would get significant late money, and that "late money" isn't all that successful considering it is "choosing" three of the top favorites nearly every race. I was amazed how often the winner was the one of the top four that late money ignored.

[Dave Schwartz]

Haven't been following all the posts recently, but I'd imagine you've looked into reverse engineering the whale action so as at least to predict where it might fall. Any luck with that angle?

It is built into our software and is both highly predictive and advantageous to play.

That does not mean that you can wager on the favorites profitably but that the overbet horses win more races. No surprise there, but how do you profit from it?

I have been researching this for a very long time. Even though we know who is being over-bet, I could never make money from it. Now it appears to have become a viable approach.


That approach is part of HSH, version 7, so I'll need to keep it under my hat for awhile.

[markgoldie]

It is built into our software and is both highly predictive and advantageous to play.

That does not mean that you can wager on the favorites profitably but that the overbet horses win more races. No surprise there, but how do you profit from it?

I have been researching this for a very long time. Even though we know who is being over-bet, I could never make money from it. Now it appears to have become a viable approach.


That approach is part of HSH, version 7, so I'll need to keep it under my hat for awhile.
Thanks, Dave. So if we might combine two major themes of this thread, could we have a profitable approach? For example:

(1) The direction of whale money can be successfully predicted on a consistent basis (you said that), and

(2) Show-pool anomalies are routinely adjusted at the last second to come in line with the win-pool percentages (others have said that in this thread).

If we combined the two principles, what about early betting of the predicted whale target in the show pool? This might eliminate or greatly reduce the "adjustment" money that would normally be coming at the last second. We might assume that this adjustment money is at least nominally profitable (with rebates, of course), or else it wouldn't keep appearing. But with this strategy, it becomes "ours" and not "theirs."

Any at least theoretical validity in this line of thinking?