Tote as Benter's highest predictive value

[dnlgfnk]

If I understand correctly, Dave has stated that Benter's largest coefficient? for assigning predictive value in his model was public odds?

http://www.paceadvantage.com/forum/showpost.php?p=1004965&postcount=71

Doesn't this mean that the common perception of unsophisticated bettors in Hong Kong is overstated, and that rather Benter was working with slim advantages and capitalizing on huge wagering pools?

[gm10]

If I understand correctly, Dave has stated that Benter's largest coefficient? for assigning predictive value in his model was public odds?

http://www.paceadvantage.com/forum/showpost.php?p=1004965&postcount=71

Doesn't this mean that the common perception of unsophisticated bettors in Hong Kong is overstated, and that rather Benter was working with slim advantages and capitalizing on huge wagering pools?

Benter has never published his complete list of variables, and certainly not the weights that are assigned to them.

He has however described that there is a crucial bias in his odds. When they deviate a lot from the public, they are usually wrong. That's why he mixes his 'fundamental' probabilities with the public ones.

(Personally, while acknowledging a bias a la Benter's, I disagree with Dave. I completely ignore the public's odds and the morning line odds, on the grounds that mixing them in dilutes the value of my own odds-line.)

[fmolf]

Benter has never published his complete list of variables, and certainly not the weights that are assigned to them.

He has however described that there is a crucial bias in his odds. When they deviate a lot from the public, they are usually wrong. That's why he mixes his 'fundamental' probabilities with the public ones.

(Personally, while acknowledging a bias a la Benter's, I disagree with Dave. I completely ignore the public's odds and the morning line odds, on the grounds that mixing them in dilutes the value of my own odds-line.)
that in my book is the definition of an overlay.Finding and using an error in the public odds.This holds true in the stock market as well.The old buy low sell high axiom....Buy a horse you think should be 2/1...but cash a ticket at 4/1....sounds so eeeeasy!

[gm10]

that in my book is the definition of an overlay.Finding and using an error in the public odds.This holds true in the stock market as well.The old buy low sell high axiom....Buy a horse you think should be 2/1...but cash a ticket at 4/1....sounds so eeeeasy!


Yes agreed ... I meant that mixing in the public's odds with my own brings my own odds too close to the public's.

Instead, my view is that it is a matter of finding the sweet spot where your odds are reliable enough. In my case it's when they are within $3 dollar (to 1) to the public odds.

[Overlay]

Yes agreed ... I meant that mixing in the public's odds with my own brings my own odds too close to the public's.

Instead, my view is that is a matter of finding the sweet spot where your odds are reliable enough. In my case it's when they are within $3 dollar (to 1) to the public odds.
To me, it's more of a case-by-case situation. The public is usually remarkably accurate, but I wouldn't hesitate to back a horse that was at public odds that were more than $3 (to 1) higher than my personal odds, especially if i can identify a likely reason why the horse has been overlooked (often because one or more horse(s) in the race is/are being overbet for reasons that I can identify and discount, making other horses overlays by default). (I don't factor either the morning line or public odds into my personal fair odds.)

[dnlgfnk]

Regarding overlays, of course one must exploit public error to win.

My point was that the Hong Kong computer handicappers may not have been regularly crushing longshots, but rather had the capacity to wager massive sums on narrow advantages.

[Some_One]

I suggest you watch the HK feed every once in a while, if there is a longshot that the computer boys like, it won't be a longshot at the off as truckloads of money will come piling in late and thanks to the 15 sec cycle time on the HK tote, you can see it come in.

[fmolf]

we all tend to think of overlays as a wide disparity in ones own line and the publics...but the computer boys are betting a lot of favs they deem overlays say a 9/5 that their programs tell them should be 7/5 or 6/5...it also takes more money to move the odds because of the larger pool size.

[Overlay]

we all tend to think of overlays as a wide disparity in ones own line and the publics...but the computer boys are betting a lot of favs they deem overlays say a 9/5 that their programs tell them should be 7/5 or 6/5...it also takes more money to move the odds because of the larger pool size.
As you say, just because a horse is at low odds doesn't mean it can't offer value (and there's certainly a greater preponderance of winners at lower odds than at higher odds).

[gm10]

To me, it's more of a case-by-case situation. The public is usually remarkably accurate, but I wouldn't hesitate to back a horse that was at public odds that were more than $3 (to 1) higher than my personal odds, especially if i can identify a likely reason why the horse has been overlooked (often because one or more horse(s) in the race is/are being overbet for reasons that I can identify and discount, making other horses overlays by default). (I don't factor either the morning line or public odds into my personal fair odds.)

Again agreed.
I was just referring to the bias that made Benter choose to add in the public's odds. I acknowledge that I have such a bias too, but mainly when my odds are about $3 or more shorter.

Of course this sweet spot can be measured in terms of difference in probabilities rather than odds as well (and should probably should be).

[The Judge]

not only can the public tell the difference between two low odds horses say a horse that should be even money (1/1) versus a horse that should be say 2/1 we might be able to do that at a decent level, but it can tell the difference between high odds horses say a horse that should be 20-1 versus a horse that should be 25/1.

Now how do you do that? They are indistinguishable in the form yet the public can separate them in the long run. The horse that is 20-1 will win more then the horse that is 25-1.

At least that is what I have been told by guys will data bases the have every race that has ever been run in North America going back to the last umpteen years.

[lansdale]

sideIf I understand correctly, Dave has stated that Benter's largest coefficient? for assigning predictive value in his model was public odds?

http://www.paceadvantage.com/forum/showpost.php?p=1004965&postcount=71

Doesn't this mean that the common perception of unsophisticated bettors in Hong Kong is overstated, and that rather Benter was working with slim advantages and capitalizing on huge wagering pools?

dnlgfnk,

From Benter's perspective, Hong Kong's bettors were relatively unsophisticated, but he conjectured that the effectiveness of his tote-smoothing was a reflection of 'inside information' in the pools, information to which he couldn't possibly have access. As many have also pointed out, this doesn't mean that the individuals who were betting on the basis of this information could necessarily profit as individuals, but that as a distillation of the 'wisdom of crowds' it offered wager-value that Benter couldn't replicate. To those who think that Benter wasn't still uncovering huge overlays without this technique, the charts in his well-known article say otherwise. However, it's worth remembering that he was seeking to extract every last ounce of value from the HK pools with his 'strong information' approach - which is why he was finally banned.:-)

A good person to respond to this question is 'arkansasman' who has developed a profitable logit model. For me, one of his most intriguing comments is that most of his overlays are in the low-odds region.

Cheers,

lansdale

[gm10]

side

...
A good person to respond to this question is 'arkansasman' who has developed a profitable logit model. For me, one of his most intriguing comments is that most of his overlays are in the low-odds region.

Cheers,

lansdale

Intuitively, I would say that makes sense. A typical logit models seems to be at its most accurate when it is relatively close to the public's perception.

[TrifectaMike]

Interesting stuff...I'd like to hear Ark'mans comment on this subject.

Mike (Dr Beav)

[thaskalos]

To me, it's more of a case-by-case situation. The public is usually remarkably accurate, but I wouldn't hesitate to back a horse that was at public odds that were more than $3 (to 1) higher than my personal odds, especially if i can identify a likely reason why the horse has been overlooked (often because one or more horse(s) in the race is/are being overbet for reasons that I can identify and discount, making other horses overlays by default). (I don't factor either the morning line or public odds into my personal fair odds.)
Overlay...your post begs the question:

What if we cannot readily identify the reason for our horse being ignored on the tote board...what do we do then?

Suppose that a horse with questionable credentials receives an avalanche of surprising support, and stands at 9/5 - for reasons that we cannot "identify and discount"...while our "legitimate" 5/2 shot is totally ignored at 8/1.

We quickly look to make sure that we haven't overlooked anything in our handicapping...and we determine that we have made no noticeable mistake.

Is our 8/1 shot a legitimate overlay...or should we be concerned because the horse is "ice" on the board?

[Overlay]

What if we cannot readily identify the reason for our horse being ignored on the tote board...what do we do then?...Is our 8/1 shot a legitimate overlay...or should we be concerned because the horse is "ice" on the board?
Considering the fundamental, multi-faceted nature of the factors upon which my probabilities are based; the lack of value being offered by the 9-5 shot; and the degree of overlay represented by the 8-1 odds, I'd continue to back the horse at those higher odds. (I might also look at exotic prices involving my horse and the horse at unexpectedly low odds for a value comparison.)