soule wrote:

Some of the stats seem lightweight.Like 9/7 @BM shows Charismatic with only 8 starts.Thoroughbredtimes.com shows him with 71 starts.This large discrepency is why I thought the sires starts may be for the local track.

Upper Nile wrote:

How often are the trainer and sire stats updated? Every day?
Once a week? How much of a lag time is there between what's on
the sheets and what the real stats are. Let me give you two
examples of what I'm asking, one for the trainer and one for
the sire. For the last 90 days trainer stats for example how
many days (I'm assuming it's only days) behind the previous days races is it falling? If I'm looking at a set of sheets for last Saturday at Belmont does the last 90 days include races run that week? I hope I'm explaining this well enough to
get an answer.

Now for the sire stats, there positivly is either a greater lag time or some serious errors. Let me give you one example I've seen. The sire stats for Distorted Humor. On Derby day he's listed as having 10 stakes winners. On Belmont Day he's still listed as having 10 stakes winners. We all know what happened in between the derby and Belmont. Distorted Humor got at least 3 more stake winners I know of. 2 with Funny Cide and 1 with Go Rockin Robin who won if I remember the Peter Pan or was it the Dwyer. So how often are the sire and trainer stats
updated? How often is the sire TGI updated.

I've been meaning to ask this for a while but I've been too busy. This is not intended to be critical but more to know especially for the trainer stats just how far behind recent racing are they. Then we can factor in for ourselves any recent days racing results.

Thanks in advance.
Phil


Soule and Upper Nile raise some general and specific questions concerning our trainer and sire statistics.

Let me detail the procedures we use to update our data and then respond to the specific questions asked.

First, Upper Nile you are correct there is greater lag time vis-a-vis updating sire than trainer statistics. Trainer information is updated on a daily basis, usually in the morning after we've received all the previous day's race information via electronic feed from our data source. However we purposely lag trainer data one day to allow for corrections which may come in the day after the races. So for example
today Tuesday, we updated our trainer statistics to include all information received on Sunday. Now we call that a one day lag but later on in the day, now, we're processing entries for Thursday. (More on this in a moment.) So if you download or buy hard copy for Thursday, you'll see trainer data updated through Sunday. Some tracks draw 72 hours in advance and we may process them later after taking care of the 48 hour draw tracks but before updating the trainer the next morning to include Monday data. Derby day is good example. Churchill Downs draws Saturday's card on Wednesday and we process it as soon as possible. In this case you're looking at a Saturday card with trainer data updated through the preceding Monday.

Now occasionally things go wrong. Electronic files get corrupted; the update is aborted and corrections aren't made until later on in the day or the next day. However, these instances are few and far between, rare. The trainer update procedure can be characterized as rote, one of our daily tasks.

Sire data are updated on a monthly basis. Usually the second Saturday of every month, I update the sire statistics to reflect inclusion of the preceding month's racing and Thoro-Graph data. Sire data take longer to update than trainer data. But the primary reason why we don't update sire data daily is that we need the Thoro-Graph figures in addition to the race results, and they're not available a day or two or three after the races. Given this and the fact that limited sire data can be misleading and statistically insignificant we decided a month was an appropriate lag period for sire data. We'd have all the correct race information, the Thoro-Graph figures, and enough of a sample particularly for new stallions to be useful as a handicapping tool.


Now specifically soule mentions Charismatic sire data as shown in the sheets for Bay Meadows 9/7/03. soule correctly notes that Thoro-Graph sire data for Charismatic shows 8 starts. More precisely that's 8 starts through July 2003. August data will be added to and reflected by the database this coming Saturday, 9/13/03. soule also cites thoroughbredtimes.com as showing Charismatic progeny making 71 starts. I believe this is incorrect. I think the number soule cites is actually the number of 2yr-olds. The link is given below. (soule correct me if I'm wrong.)

http://www.thoroughbredtimes.com/sirelists/sirelistdetail.asp?Section=Freshman

Note that the table is entitled "Freshman Sire List--Worldwide Statistics" with subtext. "For sires who stand in North America (or stood in North America if pensioned or deceased), or foreign-based sires who have a minimum of 10 starters in North America. Freshman sires are those sires whose first crop of foals are two-year-olds in the current year. North American statistics complete through 09/08/2003"


According to this table, Charismatic stands in Japan, has 71 2yr-olds, 20 starters and 4 winners. Thoro-Graph sire data for Charismatic through 9/7/3, unpublished, shows 29 starts, 17 starters, and 2 winners, in addition to 81 registered 2yr-olds--they have Jockey Club numerical codes although they don't all have names. We know the Thoroughbred Times data is updated through 9/8/3 for North American (NA) starts but we don't know how updated the foreign data are and it seems reasonable to assume Charismatic progeny have raced outside NA since he currently stands in Japan.

Upper Nile, I answered your question concerning update frequency above.

Specifically since Belmont is almost always a 48 hour draw track, Saturday sheets will most often will include updated, all categories, trainer data through the preceding Tuesday.


On Derby day sheets, sire Distorted Humor showed progeny having 10 stakes wins, not stakes winners. That reflected sire data updated through March, and Belmont day sheets reflected sire data updated though April, hence the same number of stakes wins.


With hope this response answers all the questions raised. But feel free to ask more should more questions arise.



TGAB