Bris Race Summary: Avg Race Rating question
 

In the recent Travers Stakes I was puzzled when I saw 117 as the Average Race Rating for the #7 horse KING ZACHARY. Bris's Ultimate Race Summary Explanation states that Avg. Race Rating is "an average of most recent Bris Race Ratings at today's distance/surface category." Since King Zachary never ran a race with a RR higher than 116 (see attached) how could his Average Race Rating be 117?

I asked Brisnet.net support this question and to their credit they responded very quickly stating: "You would think it would be an average, but it is not. From what I have seen from handicapping, they match maybe 60-70% of the time. This is because the average race rating is for just this distance and surface. The numbers listed for the last 3 races are the true last 3 races, which may or may not be the the same surface/distance."

Again, since the horse has never been in any race at any surface/distance where the Race Rating was higher than 116 so why 117?

Color me puzzled.

Category is probably not a specific distance. It is probably just dirt routes.

Even if they are only considering dirt routes the average is still 116. If they are excluding the race in the slop, considering it a different surface, the average is still 116.
I thought that maybe they were using average class ratings by mistake then the average would be 118. If they are excluding the slop race again the average would be 118.67. There is no way I can see how they got an average race rating of 117. I can only see this as a typo.

All figures should be taken with a grain of salt and that's how I use them. The anomaly may be an adjustment because of the uncommon race distance.

Also there is potential for confusion with the ARR and the Avg Dist/Surf figures.

King Zachary has a 117 ARR but only an 89 Avg Dist/Surf. So those two figures although defined by Bris as similar, are different. Also note that King Zachary never ran 1 1/4 distance so there again Bris leaves you to wonder how they came up with an 89 for him at the Avg Dist/Surf when he never ran the distance.

I use the race summary section as my main form of handicapping data, but only in races with a common distance, surface and age where the data would be sufficient to minimize errors. No way can you rely on any data for 3yo's going 1 1/4 miles and the results proved that.

At the risk of drawing out a relatively trivial issue, it seems odd that people would confuse ARR, which is a class rating, with AVG. Dist/Surf ratings which are based on speed/pace figures. They measure different things and are on different scales. The only thing they have in common is the word average.

Maybe the confusion is between the Average Class Rating of last 3 races(118) and the Average Race Rating (116), so they are splitting the difference and averaging the 2 to 117.

Or maybe it was simply a typo, :)

PP....ARR..........ACR
1..... 115.00..... 118.83
2..... 115.67..... 119.33
3..... 119.00..... 124.00
4..... 117.33..... 120.33
5..... 117.50..... 119.67
6..... 113.00..... 115.00
7..... 113.50..... 116.67
8..... 116.00..... 117.00
9..... 117.50..... 121.67
10... 116.17..... 118.50
11... 116.67..... 119.83


From my spread sheat on the Travelers.

Do you average previous RRs and CRs? If so, how deep do you go? Going 3 deep on the RR for this horse would return a 115 average which in my opinion seems reasonable. Going 4 back to the Wood Memorial would return a 115.25 average but further than that we get into maiden contests where the RR is understandably lower and would give us a RR closer to that which is reflected in your work.

I have no other explanation other than the 117 being a typo. I find that disheartening. Brisnet says the Avg RR category describes "an average of most recent Bris Race Ratings" yet I think we can all agree there is no way the average of anything can be greater than any of the elements in the sample being scrutinized. I hope such errors, if indeed it is a typo, are few and far between as I rely on Bris numbers exclusively.

My pension amount was derived, in part, by reviewing my annual salary history over 30 years and averaging the highest 3 years. Unfortunately for me, the bean counters at the State pension system didn't use Brismath.

I don't think it can be a typo. These are clearly generated by a computer program. So either there is an error in the program or something is not visible that is being included. One example, maybe maturing 3yos are given extra credit? A race in March with a 114 might count as 118 in August or something like that.

I really don't know, just throwing it out there. If you are on Twitter, ask Ed DeRosa, he might be able to help.

Analysis paralysis is deadly to handicappers.

While I've certainly suffered from AP in the past, the syndrome only remotely applies here.

My handicapping process is fairly straightforward. It always begins with Bris PPs and an excel spreadsheet that I've programmed to execute simple commands. The end result is an odds line. I've been back-testing and tweaking the spreadsheet for years.

Warren Buffet talks about investing only in what you know. I rely on PPs to tell me much of what I need to know about every investment option in the starting gate. I know too that "garbage in, garbage out." This is an effort on my part to allay my concerns about garbage going in. I've always held Brisnet PPs in high regard and I want to do the best I can to have a full understanding of the numbers that I'm using, or not using, to arrive at "investment decisions."

I'm going to reach out to Brisnet again for further clarification.

Please let us know when Brisnet gets back to you with something that makes sense. Or something that doesn't make sense.

You definitely need to get back to Brisnet and speak to someone who knows what they're talking about. This nonsense that their Average figure is not an average seems to be from the Rudy Giuliani "The truth is not the truth" school of logic.

Question Resolved
 

To their credit, Brisnet got back to me fairly quickly to clarify their first response with something that was reasonable, at least to me. Apparently when there isn't data available to calculate an Average Race Rating (in this case, the horse never ran the distance, yet showed an ARR of 117) the default is the Bris Pedigree Rating. Sure enough, the horse's Pedigree Rating is ... 117.

Interesting to note that the Bris Pedigree rating must consider the dam's sire (or both dam and dam's sire, or just dam?). While CURLIN has sired KING ZACH along with 3 other horses entered in the Travers this year, only ZACH sports a Bris Pedigree Rating of 117 (the highest of the four entries sired by CURLIN).

Thanks to Tammy S from Brisnet Player Services for clearing this up and restoring my faith in Bris Race Rating figures.