Quote:
Originally Posted by PaceAdvantage
I don't think people are betting after the bell. For the record.
But when 40% of the pool doesn't show up until after the gates open, even at a huge venue like Keeneland, never mind the tiny tracks, we have a serious problem on our hands.
|
PA,
I know you are steamed. Welcome to the club.
Seriously, the answer is not that complicated but it does take a change in thinking.
What I have done... and many of my users have done... is to simply stop using the tote altogether. Almost 4 years since I stopped.
Amazingly, my handicapping life is actually BETTER.
What we do is create what I used to call "phony odds." Now we call them "object odds" or simply "aML," which stands for "artificial Morning Line."
It is based upon the idea that we can predict
How the Public USUALLY bets the money on a horse like this one.
The driving force is what we call a handicapping object. That is nothing but 5 or 6 factors weighted together, and then normalized to 100%. The individual horse percentages become the "phony pool."
All the factors should be "high-level" factors. Just think in terms of factors that have high hit rates because they will
always correlate with the tote board.
The primary factor is almost always Morning Line. Others include factors like BRIS Prime Power or HDW's Projected Speed Rating. Think of which factors you might add beyond that.
BTW, I actually use the ranks in these factors as opposed to raw ratings.
Not only is there great freedom in this, but there is a bit of logic as well, in the sense that horses SHOULD be bet based upon HOW THE LOOK (in terms of important factors).
Now, races with FTS and foreign horses present a problem. Even HSH does not handle that well. Much easier to overcome in a spreadsheet.
BTW, do not think for a minute that this approach is going to do a near-perfect job. There is no such thing. Not even the whales, who have the benefit of not only a much more complex approach than a 5-factor object
and also tote board patterns.
But trust me when I say that
it really doesn't matter. If you implement this you will find that you are no longer hitting a moving target.
Will you bet a horse expecting to get 6/1 and wind up at 5/2? Absolutely. You will also see 12/1 pay 4/5. Why? Because the horse
really shouldn't have been 4/5 based upon how good he looked.
In fact, I actually built a seminar called
The Renegade Handicapper around this concept:
1. Create the Phony Odds.
2. Pick Contenders based upon your handicapping factors.
3. Create probabilities from the REAL ODDS.
4. Penalize the Non-Contenders' probabilities hugely and make new probabilities.
5. Bet into the Phony Odds with your Probabilities.
If there is sufficient interest, I will arrange a free seminar to teach people how to do this. One MUST expect that it will take some work.
Dave
PS: I've not talked about it in a long time but I am really within 6 weeks or so of releasing the beta version of my new SMALL DATA software. That software will have a BUNCH of these aML objects built in.
