Talking Handicapping with Dave Schwartz

[PICSIX]

Originally Posted by Capper Al

Hi Guys,

Just want to let you know that I'm talking handicapping on Dave's Facebook page at the link below:

https://www.facebook.com/groups/6057...7058819340430/

Last night Dave showed me how he wins with pace which was very informative and helpful. It's a new approach at dealing with pace. Hope to see some of you at Dave's Facebook page.

Is this New Pace or a newer New Pace?

[Exotic1]

Originally Posted by ReplayRandall

Why hasn't this thread been closed? Are you kidding me here? Redirecting PA'ers to someone's Facebook page so they can talk about handicapping?.....Just when I think I've seen everything on this board, we get this nonsense....

Agree. This is somewhat bizarre.

[Dave Schwartz]

Originally Posted by PICSIX

Is this New Pace or a newer New Pace?

While it is still early vs. late, we've developed a greater understanding of HOW to handicap races based upon the idea that the 1st call determines a lot about what happens.

We've learned that if the early leaders aren't around at the end, the race times will be very slow, which allows for horses that didn't figure to win.

Now, this may not sound earth-shattering but it really is a handicapping game changer. It causes the handicapper to look at a race from the two distinctly different viewpoints:

1. The race to the 1st call becomes a battle of its own, and those 2 or 3 horses will run around the track together to some degree.

2. The rest of the horses are in their own race, and the profile of what wins in such as race is far different.

For those who are familiar with the NewPace concept, you say, "Well, this is just NewPace all over again." And in many ways it is.

But the systemology has progressed. It has become more powerful and, at the same time, become simpler.

Of course, the catch is to predict this ahead of time, to whatever extent it is predictable. But even just knowing that this is how it works makes

The key change in the thinking process is to treat every race as a #1 race - SPEED WILL WIN - then ask the question who will win if it becomes a #2 race?


It will probably be a couple of months before I will have materials together to explain all this stuff. Stay tuned.


Best Regards,
Dave Schwartz

[DeltaLover]

Originally Posted by Dave Schwartz

We've learned that if the early leaders aren't around at the end, the race times will be very slow, which allows for horses that didn't figure to win.

What I find more interesting in your posting, is the use of the word learned .

If we can describe the “learning” process in a clear and objective way, then we can use it for any other possible hypotheses suggested by our handicapping experience and automate the discovery of potential market anomalies.

For example, in the quoted statement you are mentioning “very slow times” which opens the discussion of time comparisons. How do you arrive to the conclusion that a final time is very slow? Do you simply use final times, if not how you calculate your track variant / speed figures? Do you consider pace in your figure estimates or not? What are the starter profiles that “do not figure to win”?

More than this, the most important thing for the bettor is not to correctly predict what might happen given certain pace conditions but to what extend this event will be misestimated by the betting crowd. If you accept this premise, it is reasonable to shift your focus towards the understanding of how the crowd is forming the betting markets and realize to what extend your findings (similar to the quoted factor) are reflected to them.

A successful approach to betting must be multidimensional (meaning that it cannot rely on single or even a small set of factors like pace for example) and more than this it must be aware about the 101 factors used by novices and traditionalists.

It is exactly this type of factors that we go against when we try to make a profit and the most difficult challenge we are facing lies in the representation of the input data in a comprehensive and noise free format and also the description of the output which can take the form of classification, regression or clusterization based on whether we will follow supervised, unsupervised or reinforced learning process. Again, all these are related to the learning procedure that you are referring to, which consists the real problem we are trying to solve and I believe that this is where is should start our conversations from.

[Dave Schwartz]

Originally Posted by DeltaLover

What I find more interesting in your posting, is the use of the word learned .

If we can describe the “learning” process in a clear and objective way, then we can use it for any other possible hypotheses suggested by our handicapping experience and automate the discovery of potential market anomalies.

In the above paragraph, (to me) the most important word is actually not "learning," but rather "PROCESS."

For example, in the quoted statement you are mentioning “very slow times” which opens the discussion of time comparisons. How do you arrive to the conclusion that a final time is very slow? Do you simply use final times, if not how you calculate your track variant / speed figures? Do you consider pace in your figure estimates or not? What are the starter profiles that “do not figure to win”?

a final time is very slow when it is "substantially" slower than would be expected using what ever method you choose. I use speed ratings.

For example, if a particular race is projected to run in "about" 95, it is not uncommon to see such a race run 10 points slower if the race runs late.

Perhaps a better way to say it would be, "If the early runners fail, speed ratings will not matter."

More than this, the most important thing for the bettor is not to correctly predict what might happen given certain pace conditions but to what extend this event will be misestimated by the betting crowd. If you accept this premise, it is reasonable to shift your focus towards the understanding of how the crowd is forming the betting markets and realize to what extend your findings (similar to the quoted factor) are reflected to them.

A successful approach to betting must be multidimensional (meaning that it cannot rely on single or even a small set of factors like pace for example) and more than this it must be aware about the 101 factors used by novices and traditionalists.

It is exactly this type of factors that we go against when we try to make a profit and the most difficult challenge we are facing lies in the representation of the input data in a comprehensive and noise free format and also the description of the output which can take the form of classification, regression or clusterization based on whether we will follow supervised, unsupervised or reinforced learning process. Again, all these are related to the learning procedure that you are referring to, which consists the real problem we are trying to solve and I believe that this is where is should start our conversations from.

IMHO, this is way too deep and analysis for most players. Neither is it necessary. At least not directly.

Of course, if it is working for you, keep doing it.

Earlier in this thread I made the statement that I had a, "Change of heart." There are several components to that change. One of them is to move away from the BIG DATA approach.

Sure, I have a lot of data. Typically I keep my database at around eight years or so, (480,000 races).

However, if I take a "Big Data Analysis" approach, I am in direct competition with The Whales. (I realize you believe they are completely mythical, much like unicorns. Your beliefs are different than mine.)

My belief is that if I try to compete in the same arena with players that are clearly superior to me I am destined to fail. If for no other reason than they have massive resources and I don't.

This is why, in my personal approach, I have chosen to use small, narrow samples. Even in the large project I have just completed (known as "Melville"), we still use a different approach. (This is not a topic for this thread.)

Since our personal approaches to the game are so vastly different, I doubt that we will ever be on the same page.


Regards,
Dave Schwartz

[PaceAdvantage]

Originally Posted by ReplayRandall

Why hasn't this thread been closed? Are you kidding me here? Redirecting PA'ers to someone's Facebook page so they can talk about handicapping?.....Just when I think I've seen everything on this board, we get this nonsense....

Dave is an authorized advertiser here...you do realize this, right?

[DeltaLover]

Originally Posted by Dave Schwartz

For example, if a particular race is projected to run in "about" 95, it is not uncommon to see such a race run 10 points slower if the race runs late.

The question is how do you validate the 95 figure?

Originally Posted by Dave Schwartz

IMHO, this is way too deep and analysis for most players. Neither is it necessary. At least not directly.

I think that a deep analysis is required if you want to maintain valid expectation to beat the crowd.

Originally Posted by Dave Schwartz

Typically I keep my database at around eight years or so, (480,000 races).

However, if I take a "Big Data Analysis" approach, I am in direct competition with The Whales. (I realize you believe they are completely mythical, much like unicorns. Your beliefs are different than mine.)

400K races do not qualify as Big Data; you can easily load all of them in memory and treat them as a closed universe. The reason you need as much data as possible is because part of the research consists of understanding how the crowd is changing preferences both though time and based on locality as well. I believe that the objective is not to discover a global set of concrete rules but to identify the meta-handicapping process of the crowd which conceptually will always remain the same.

[ReplayRandall]

Originally Posted by PaceAdvantage

Dave is an authorized advertiser here...you do realize this, right?

My problem is clearly with Capper Al having anything to do with this thread, let alone being the OP......You do realize this, too?

[PaceAdvantage]

Originally Posted by ReplayRandall

My problem is clearly with Capper Al having anything to do with this thread, let alone being the OP......You do realize this, too?

Capper Al isn't banned from the horse racing side...so he's free to post...and I have it on good authority that it wasn't Dave who put him up to posting here...I don't really see the big deal in the end...

If I did, I would have deleted the thread.

[ReplayRandall]

Originally Posted by PaceAdvantage

Capper Al isn't banned from the horse racing side...so he's free to post...and I have it on good authority that it wasn't Dave who put him up to posting here...I don't really see the big deal in the end...

If I did, I would have deleted the thread.

If Capper Al was allowed all along to talk handicapping on this site, where has he been for the past year? Now, after all this time, his welcome back message to this board, is to invite us to a foreign site to "talk handicapping"? Isn't this what we do here at PA on a daily basis?.....IMO, this thread should have been deleted from the original post, but it's your site, your call.

[PaceAdvantage]

Originally Posted by ReplayRandall

If Capper Al was allowed all along to talk handicapping on this site, where has he been for the past year?

You would have to ask him that. I guess it has something to do with politics on off topic or something like that.

[Dave Schwartz]

Okay... let me put an end to THIS nonsense.

First, I have a small group of guys - 3 to be exact - that I have enlisted as potential teachers of the systems & methods that I am going to put forth in the coming months.

I invited them to the first of many private workshops to explain what and how we are doing this. Basically, I am giving them private coaching in return for being my support team. Al is not one of those 3.

However, Al has continually demonstrated an eagerness to help people and share ideas. Often his ideas are not all that well-received. Despite this, he has continued sharing his ideas and maintained a positive attitude.

Within our Facebook group he makes a couple of posts a week, often really good topics. Almost always about handicapping. (I usually delete posts that are about horse racing but not handicapping.)

Because Al is a positive contributor, building HIS OWN stuff, I invited him to attend this private workshop thinking that it might benefit his handicapping endeavors.

(Please resist the urge to play that same old song about how nobody gives away anything of value. Frankly, I do it all the time. It is what has kept me in business for 27 years.)

So, when Al heard me speak about sharing stuff in the group down the road, he missed the key elements: WHEN, WHERE AND WHO and just jumped out to do what he saw as positive.

When I saw his post here, I supported his effort because... well, he deserved the support.


Finally, I rarely thump my stuff anywhere. When I have a new product, I generally start one thread, put up the announcement, and move on. Pretty much the same thing in the Facebook group.

Of course, I study marketing all the time. But in the handicapping world we've all seen so much of that over-hyped "Going-to-get-rich-for-$47" crap. My idea of marketing is to drive people to our site and NOTIFY about new stuff. I NEVER hype get-rich-stuff because I like sleeping at night.


Regards to All,
Dave Schwartz

[Light]

Originally Posted by Dave Schwartz

handicap races based upon the idea that the 1st call determines a lot about what happens.

Now, this may not sound earth-shattering but it really is a handicapping game changer.

This stuff is ancient isn't it? I picked the winner of the Preakness based on CC having the best E1 call in the field. Not that difficult to see. Many,many handicappers check for a lone speed horse as part of their handicapping regiment every day. Granted CC came off the pace but his E1 call told me that if he doesn't get the lead, he would be the first to pounce.

Not trying to knock you but what is unique about paying attention to the E1 call using your method?

[Dave Schwartz]

Originally Posted by Light

This stuff is ancient isn't it? I picked the winner of the Preakness based on CC having the best E1 call in the field. Not that difficult to see. Many,many handicappers check for a lone speed horse as part of their handicapping regiment every day. Granted CC came off the pace but his E1 call told me that if he doesn't get the lead, he would be the first to pounce.

Not trying to knock you but what is unique about paying attention to the E1 call using your method?

You will have to wait to find out.

[Clocker]

Originally Posted by Dave Schwartz

We've learned that if the early leaders aren't around at the end, the race times will be very slow, which allows for horses that didn't figure to win.

Now, this may not sound earth-shattering but it really is a handicapping game changer. It causes the handicapper to look at a race from the two distinctly different viewpoints:

1. The race to the 1st call becomes a battle of its own, and those 2 or 3 horses will run around the track together to some degree.

2. The rest of the horses are in their own race, and the profile of what wins in such as race is far different.


The key change in the thinking process is to treat every race as a #1 race - SPEED WILL WIN - then ask the question who will win if it becomes a #2 race?

Conceptually, this sounds much like the principles behind the Match-Up, a variation within the Sartin Methodology. In the words of Jim "The Hat" Bradshaw, the race doesn't start at the 2nd call, the race starts when the gate pops.

The Match-Up is based on projecting an early pace and determining whether that pace and the make up of the field will result in a race that will run early, with an "E" running style horse winning, or OTE (other than early) with a presser or sustained style winning.

If one of the "E"s in #1 above can dominate the others, it will likely win. If not, the "E"s will fight each other for the lead and burn out, leaving the others in #2 to take over in the stretch.