Daily Bias, a simple solution?

[jay68802]

Ran across this article, found it interesting. Pretty simple to use, but I do question the use of position instead of lengths back. Any comments?

http://www.troyrecord.com/news/calcu...fc1d67882.html

[cj]

Originally Posted by jay68802

Ran across this article, found it interesting. Pretty simple to use, but I do question the use of position instead of lengths back. Any comments?

http://www.troyrecord.com/news/calcu...fc1d67882.html

It is ok, good starting point. The rating we put in TimeformUS PPs for bias had their origins here though they have evolved quite a bit.

[jay68802]

I am doing this for one track now. In one race a horse is 6th by 4.75 lengths but is in the top half of the field. According to the method, does contribute to the bias by being in the top half of the field. But a horse on the same day is 4th by 1 length and does not contribute because he is not in the top half of the field. Does not make sense.

[jay68802]

My thinking is, that may be the running style shown in the race might replace the position..........

[classhandicapper]

IMO, most of the time position is going to be adequate, but obviously there's a difference between being 3rd in a 3 horse duel and 3rd sitting 6 lengths off a blazing duel between 2 other horses.

I find that my own bias numbers are a place to start.

If they flash something far enough away from average to get my attention, I know to look at the details like the odds of the horses, the flow of the races, inside/outside a little more carefully. IMO there is no magic formula that's going to weigh all that stuff perfectly.

The bigger problem is sample size. What are you supposed to do if there are 5 dirt races (3 sprints and 2 two turn routes) and 4 turf races (2 sprints and 2 routes or races on different turf courses). You don't have a large enough sample of anything to be sure what was going on that day. Sometimes if a small sample is flashing a bias, I'll look at the previous day and next day to see if there was any trend.

In any event, I guess what I'm saying is that imo numbers like that are good for flashing possible biases, but you should still probably look at the details.

[PhantomOnTour]

When making my Humble Numbers (HumNums as I call them) I note the running position of the top three finishers with the following notations:

e = on a clear lead
E = within a length of the lead/dueling
p = between 1-3 lengths behind
m = between 3-5 lengths behind
c = more than 5 lengths back

When i am done i can scan the races to see if a certain position was dominant on any particular day...

R1: cmp
R2: ccc
R3: mce
R4: ccm
R5: cpc

**This day was not good for speed and possibly a negative rail/inside also. When I see this type of array I read each chart comment to find out what paths the top finishers were taking. The current Aqu meet is a good example of how this can be helpful.

Hope this helps!

[Tom]

Originally Posted by jay68802

I am doing this for one track now. In one race a horse is 6th by 4.75 lengths but is in the top half of the field. According to the method, does contribute to the bias by being in the top half of the field. But a horse on the same day is 4th by 1 length and does not contribute because he is not in the top half of the field. Does not make sense.

Try expressing position by dividing running pos by field size.

4th in an d 8 horse field is .50, 4th in a 12 horse field is .33.

[jay68802]

Thanks Tom, I am going to do that, and a few other thoughts. Your post made me realize something. I have not been entering field size in my data base.....

It is only 670 races.......

I can be such a DUH! once in a while.....

[JustRalph]

no wonder I could never hit a race at Keeneland

[cj]

Originally Posted by jay68802

My thinking is, that may be the running style shown in the race might replace the position..........

Here is an idea for you, I use it in some other situations.

Instead of using one or the other, use the one that is lesser of these...

position
lengths back + 1

So a horse that is 4th by 1 length is actually considered 2nd (1 + 1)

A horse that is 4th by 6.5 lengths is still considered 4th.

A horse that is 2nd 5 lengths back is still 2nd.

[thaskalos]

I add the horse's position to its lengths behind, and then divide by 2. A horse that's running 2nd by 4 rates a 3...and a horse that's 4th by 8 rates a 6. A bit radical perhaps...but it works for me.

[Robert Fischer]

i may be doing something wrong, I get a 'paywall'...


sounds like cool ideas.


I look at race flow, and how certain seemingly consistent/known horses perform relative to their expectations given certain trips.

If it's not extreme, I am not good enough with bias to include it in my handicapping.

A numerical system that can be used in a database sounds really cool.

[Tom]

Originally Posted by Robert Fischer

i may be doing something wrong, I get a 'paywall'...

Me too. I just copied and pasted it into Word.

[cj]

Originally Posted by Robert Fischer

i may be doing something wrong, I get a 'paywall'...


sounds like cool ideas.


I look at race flow, and how certain seemingly consistent/known horses perform relative to their expectations given certain trips.

If it's not extreme, I am not good enough with bias to include it in my handicapping.

A numerical system that can be used in a database sounds really cool.

Gotta turn whatever form adblocker off that you may have running.

[Robert Fischer]

Originally Posted by cj

Gotta turn whatever form adblocker off that you may have running.

good call

ahhh......
disabled it, and refreshed, - then I was informed that I've reaced the free limit of views

that's OK.. I follow the 'gist' in this thread.

Personally??
my own insight,
some of the Track Handicappers of certain circuits,
and the TimeformUS designations of track bias.
All these things I just take as a 'consideration' and see if the odds, and other big things are moving in that same direction to encourage a play, or moving against it as a tradeoff where I may pass.