JCapper Message Board

          JCapper 101
                      -- Track Profile report not working?

"Easy enough...

Let's start with a sql expression in its simplest form:

SELECT * FROM STARTERHISTORY WHERE (X)

Note that in the above template, even though the expression is as simple as it gets, I am using parenthesis characters.

It's also implortant to note that in sql expressions, everything inside a set of parenthesis characters, even if it is hundreds of lines long, gets evaluated as if it were a single parameter.

Therefore, placement of parenthesis characters in sql expressions is critical.

In the above sql expression 'X' can represent either a single sql command or even hundreds of lines of sql commands.

Because 'X' is wrapped in parenthesis characters 'X' is evaluated as if it were a single line of sql.

---

Let's expand the above sql expression a bit...

What happens if we replace 'X' with individual sql commands? (Substitution.)

Instead of SELECT * FROM STARTERHISTORY WHERE (X) if we replace 'X' with '(A) OR (B) OR (C) OR (D)' our sql expression now looks like this:

SELECT * FROM STARTERHISTORY WHERE ((A)

OR (B)

OR (C)

OR (D))

---

The above sql expression can be used as a basic template to combine 4 multiple sql UDMs into a single sql UDM.

In the above sql expression (A) is a single sql UDM.

In the above sql expression (B) is a second single sql UDM.

In the above sql expression (C) is a third single sql UDM.

In the above sql expression (D) is a fourth single sql UDM. "

"Track bias can shape race results in the near term - Imo, to a greater degree than most players realize.

In the past I've posted about the influence of track surface (track maintenance, materials composition including water, depth of harrowing, degree of banking, and placement of banking, etc.) on track biases.

Here's a link to one such thread:
http://www.jcapper.com/messageboard/TopicReader.asp?topic=1480&forum=Private

However, I don't think I've ever posted any of my thoughts about weather --

Weather systems that bring cycles of wind, rain, and sunshine have the ability to create cycles of track bias.

The winds of low pressure weather systems in the Northern Hemisphere rotate in a counterclockwise direction.

The couterclockwise rotation of a large low pressure system traveling west to east across the US would (likely) create winds out of the southwest as its first edges reach (say) Lexington, KY where Keeneland is located.

Winds out of the southwest at Keeneland can translate to presser types getting a boost from a tailwind on Keeneland's far turn. When that happens, early types can be at a disadvantage compared to normal circumstances.

But once the "eye" of a low pressure system traveling west to east across the country passes over Keeneland, that same counterclockwise rotation would now (likely) create winds coming from the opposite direction: the northwest.

Winds out of the northwest at Keeneland can translate to presser types facing a headwind on Keeneland's far turn. When that happens, early types can be at a slight advantage compared to normal circumstances.

On top of that, a large low pressure system traveling west to east across the country will (likely) bring rain to the area.

Depending on how much rain, the grounds crew at Keeneland may (or may not) decide to seal the dirt surface.

Of course, each surface is different. The amount of moisture in the surface, as well as how tightly compressed the surface is, has an effect on track bias.

As the low pressure system moves out of the area the sun comes out. The weather warms, and surfaces go through a drying out period. Of course this drying out period can bring its own unique short term track biases.

Eventually, after enough drying out, things get back to normal. The dirt course becomes fast and the turf becomes firm.

But a few days later, maybe a week, maybe several weeks --

The cycle repeats itself (not just at Keeneland, but at every track you can name) as the next storm system moves into the area.

My point?

The development samples that we use to create UDMs, UserFactors, UPR, or MLR models are influenced not just by track maintenance but by weather patterns too.

It is not only possible (but highly likely) that the validation samples we use to forward test and tune UDMs, UserFactors, UPR, or MLR models are influenced not just by variations in track maintenance but by variations in weather patterns too."