Post From PaceAdvantage:
Link to Post (actual response):
Adjusting the Betting Line

There are lots of intelligent analysis here, but I have a problem with the strategy behind the bets. When I'm going through the horses the first time in a race I'm looking at characteristics in a horses pps that will make his odds long. I avoid the morning line until I have fully handicapped the race. I don't base my opinions of a runner on the "market". I'm looking at the shape of the race. Is there a stand out? Are there lots of evenly matched runners? What's the comfortable pace of all the runners? What's each runner's comfort zone on the track in relation to make up of the field. How does the race figure to play out? At this point I have a general opinion of the odds line. Does the race have a chance of playing out that gives an overlooked horse a chance to run a strong race? Overlooked horses have a chance to win if they get favorable race conditions. Now it's time to look at the morning line. This is when I decide whether to play the race or not. The morning line will reflect the pecking order of the market's opinions. The market is each handicapping method's opinion on win probabilities. How does my opinion of an overlooked horse fair in the pecking order? If he's not high in the pecking order of the morning line, it's time to find some likely runners to pair with him in the exotics. The make up of the other likely top 3 or 4 finishers will decide how I bet the race. I guess my problem with the strategy behind the formulas in this thread is the market decides whether I bet the race, and how I bet the race, but it doesn't decide who I'm betting on... — bisket

____________________
(My Response - would have I responded there rather than here.)
LOVE THIS!
Indirectly, you nailed the very thing that could locate the Achille's Heel of the whales.

____________________
The whales use a form of iterative regression.

The data could best be described as:
Each Row = One Entry (i.e. one start from one horse's history)
Each Column = One Factor from that Entry's past performances, tote (both real and projected), and finish/prices in that start.

In other words, the ENTITY that determines who/how they bet is based upon producing an accurate output for individual horses.

BASICALLY, they have one giant model.
Obviously, those "factor columns" contain Local Optima (i.e. Turfway post positions have different values that Saratoga).


____________________
This is what MOST PLAYERS do, in a far less sophisticated manner.
The impact of this is why it is so easy for the Whales to calibrate how the public plays as a whole: the public basically plays one global system, too!

We, the players (and this includes me), all believe that our handicapping is so unique because we have assembled the factors together in a unique way.

What we've ignored is that we're still using the same factors that the whales have calibrated values for to such a degree that the top 150 to 200 factors are pretty much nullified.

IOW, those factors have been made EFFICIENT.


____________________
WHAT YOU HAVE DESCRIBED...
... could be called SITUATION-BASED handicapping.
Basically, you have asked GOOD QUESTIONS, and decided to alter your STRATEGY based upon the ANSWERS.

Example: "I handicap a a race with a lot of early speed differently than a race with less early speed."

Or: "I handicap a a race that contain few or no horses with a recent good race differently."

In other words, you're producing SITUATION OUTPUT.

This takes you out of the MAIN STREAM, ONE-SIZE FITS ALL handicappers who make up 90% of the handle.

You have given yourself an honest chance to win!


____________________
The remaining Stumbling Block is...
Verification that your strategies are correct.

This is an almost impossible task for a player that does not build his own data tables, with his own decisions as "columns" in that table.

For an artful player - one who does different things in the similar/same situations - it is literally impossible.

Such a player must truly depend upon HIS ART to carry him thru.

He can then validate his strategies using his results. The best question in such a situation is, "Am I winning?"

Sadly, most players refuse to ask nor answer this question, instead opting to pretend that all is well.

When it comes to artists, not everyone can be a Dutch Master.