[Dorianmode]

Originally Posted by mrhug

I have really enjoyed this post, and as a beginner (14 months) wish we would have more like them. I really enjoy reading about the successes of other members and how they are accomplishing their success. However, it can become confusing very quickly.
DM
I have a question about how you are using NP. I see where each horse has a range posted within RDSS NP. However, are you establishing a speed rating for the current race and then comparing this rating to each horses range in NP? If so would you please explain in depth how you are doing it and if not then how exactly do you eliminate non contenders with NP.
Just trying to understand.
Thanks
Melvin

I do not project a speed rating for a horse in today's race. The range you see "posted" in NP, is a calculated range using SOME of that horse's past SR's. Under some circumstances, that range can be predictive, to a certain extent. Have you watched/listened to the introductory NP videos in which Mr. Schwartz lays out his "confidence interval", for the SR "range" in NP ? The "confidence interval" that NP calculates, in general, "contains"/"projects", (statistically), to about a 90 %, ("confidence") level what an animal will do today. We have been talking about the problems of NP, and about that "other" 10% of the probability (total) spectrum, (90+10=100%). If NP drops the last 2 races, (as Schwartz says in the 5th video), then "things might have been happening", in the last two races, (which were dropped, ....by his definition), in the calculation of the NP range, that you see on the screen. That means, I have to look at each horse, and see if it has changed it's behavior in the last two races. Assuming it hasn't, the NP range can be predictive, at around the 90 % level.

BTW, I now have my own version of a "range", which calculates a slightly higher "confidence" level. It's almost the same, but instead of 90 %, (which Schwartz uses in his "confidence level), by definition, (statistically), mine is 95%, because 2 Standard Deviations, (under the "Bell Curve"..or statistically..what is called a "normal distribution"), contains within it 95% of the horse's behavior, (and mine uses all the SRs I can get, and weights them towards the recent ones).

I don't have a "magic" system by which I use NP for eliminations. I use it quickly, visually, to see the horses which could not possibly run up to par in today's race, and, ((after making sure they have not "improved" in the last two, (or whatever) races which were, by definition, dropped in the calculation of the NP range)), either toss them, or leave them in for further analysis, both by RDSS, and my own final "match up" handicapping.

I'll try to keep my eye out for an example race from today, which demonstrates how I used NP for eliminations, practically, today, and post it later today.