Ask the Experts

General Category => Ask the Experts => Topic started by: jbelfior on October 13, 2015, 08:08:44 AM

Title: Lies, Damn Lies, and Statistics
Post by: jbelfior on October 13, 2015, 08:08:44 AM
Interesting stats so far in Keenland route races on dirt. As TGJB said weeks ago, there\'s going to be a lot of ground loss.

Posts 1-4   19 of 25 (76%)  
Posts 5-11  6 of 25  (24%)

Good Luck,
Joe B.
Title: Re: Lies, Damn Lies, and Statistics
Post by: miff on October 13, 2015, 09:07:39 AM
Posts 1-4 26 for 160 (16.2%)
Posts 5-14 14 for 217 (6.5%)


huh?? What posts won the other 77.3% of the routes?? Also impossible that there were 160 runners in  post 1-4 vs 217 runners in posts 5-14 meet is only a 9 days old
Title: Re: Lies, Damn Lies, and Statistics
Post by: P-Dub on October 13, 2015, 09:10:39 AM
Well........Obviously, its a Damn Lie.
Title: Re: Lies, Damn Lies, and Statistics
Post by: magicnight on October 13, 2015, 10:36:06 AM
Mike, \"impossible\" might be a little strong. These aren\'t my stats but from these I\'d say that the 160 horses that started in the 4 inside posts in the 40 2-turn dirt races through the first 8 days of the meet (avg of 5 such races per day - that seems high - maybe this includes the Spring meet) had 26 winners, or, won 65% of these races. The 217 in the posts 5 and out combined to win 35% of the races. 217 & 160 = 377, or, about 9.5 horses per race. Other than the number of 2-turn dirt races so far (and the misleading \"win percentages\") this looks OK.
Title: Re: Lies, Damn Lies, and Statistics
Post by: jbelfior on October 13, 2015, 10:39:43 AM
sorry guys. misread old data from Equibase.
Of 25 main track routes this meet, 19 won by post 4 and under (76%).

15 turf routes so far with no apparent pp advantage.  

 
Good Luck,
Joe B.
Title: Re: Lies, Damn Lies, and Statistics
Post by: miff on October 13, 2015, 10:44:53 AM
Magic,

That\'s my reference, has to be since inception of new surface,\"impossible\" this meet.

Mike
Title: Re: Lies, Damn Lies, and Statistics
Post by: magicnight on October 13, 2015, 10:59:48 AM
Yup. We good.
Title: Re: Lies, Damn Lies, and Statistics
Post by: mjellish on October 13, 2015, 06:50:46 PM
The track had tons of rain early on.  And i mean tons.  Seems to have held up well.  In general the winning move has been to save ground through the turns and then rally out a few paths from the rail down the stretch.  That could mean nothing by the BC, but its worth noting.  Especially since they are now scrutinizing the surface after the recent breakdowns of some notable runners.
Title: Re: Lies, Damn Lies, and Statistics
Post by: RICH on October 14, 2015, 03:36:46 AM
I have 3 for 49, posts 5 and out last 30 days fast dirt mile and over
Title: Re: Lies, Damn Lies, and Statistics
Post by: RICH on October 14, 2015, 03:38:55 AM
... all dirt fast and wet
Title: Re: Lies, Damn Lies, and Statistics
Post by: RICH on October 14, 2015, 03:41:46 AM
and 15-41 posts 1-4
Title: Re: Lies, Damn Lies, and Statistics
Post by: Wrongly on October 14, 2015, 07:20:38 AM
Track appears fast this morning as Runhappy rocketed off  23.4 34 45.3 57.3 out in 110.2
Title: Re: Lies, Damn Lies, and Statistics
Post by: Focus959 on October 14, 2015, 07:50:16 AM
Fall meet, mile and over, main track through October 11:

25 dirt route races:

PP 1-4: 19/25= 76%

PP 5-12: 6/25= 24%

http://www.keeneland.com/sites/default/files/shared/stats/pp101115.pdf
Title: Re: Lies, Damn Lies, and Statistics
Post by: trwhis2 on October 14, 2015, 08:36:02 AM
Those stats aren\'t exactly fair when you factor in field size.

Additionally, favorites have won 11/25 (44%) of these races at the current meet...Higher than the average when you look at Fall 2014 - Present.

9 of the 11 favorites to win have been in post position 1-4...So are they the favorite because they\'re in post 1-4 or are they just the fastest horse in the race and happen to be in post 1-4?
Title: Re: Lies, Damn Lies, and Statistics
Post by: miff on October 14, 2015, 08:39:49 AM
Public don\'t make horses favorite solely because of post position but will knock a horse down a bit for drawing well inside, two turn races mainly.
Title: Re: Lies, Damn Lies, and Statistics
Post by: jbelfior on October 14, 2015, 12:49:09 PM
Blue Grass post position winners (yes, I know it includes several years of polytrack but this is from 1937).

1. 13
2. 12
3. 10
4. 14
5. 10
6.  4
7.  3
8.  5
9.  2
10. 2
11. 2
12. 0
13. 1
14. 0

So of the 78 BG winners,76% have won from post 5 and in. Pretty close to the 76% winners from post 1-4 so far this meet.

Good Luck,
Joe B.
Title: Re: Lies, Damn Lies, and Statistics
Post by: kmart4503 on October 14, 2015, 02:12:32 PM
A comment for most of the post position numbers being thrown out:

Not sure that purely wins by post position is a great indicator.  Perhaps, wins as a percentage of starters from each post might be more meaningful.

Just my 2 cents.
Title: Re: Lies, Damn Lies, and Statistics
Post by: TGJB on October 14, 2015, 02:38:54 PM
Absolutely. Better yet, ROI of each post. ITM also worth looking at.

But basically, post matters most on races with a short run to the turn. Don\'t have to be Sherlock Holmes.

That part of the seminar writes itself. Other parts, not so much.
Title: Re: Lies, Damn Lies, and Statistics
Post by: miff on October 14, 2015, 02:39:21 PM
K,

True,but believe you will find the result close to the same if every post had the same number of starters.

Long string for a basic point,outside posts are nfg in two turn races generally, more so with a short run to the first turn.

BC Pre-entries in 4 days.

Mike

P.S. JB said that a minute ago
Title: Re: Lies, Damn Lies, and Statistics
Post by: jbelfior on October 15, 2015, 10:15:47 AM
Mike: What\'s the definition of an \"outside post?\"  6 and out? 7? 8?  The point from the long thread is that there appears to be a huge dropoff at KEE as you go past 4.

Good Luck,
Joe B.
Title: Re: Lies, Damn Lies, and Statistics
Post by: miff on October 15, 2015, 10:38:01 AM
Joe,

Depends on field size and the ability of an outside posted horse to make early position avoiding getting caught wide.Cant put a post position number on it, tend to look over the field and how a race appears on paper before tossing outside horses.

Mike
Title: Re: Lies, Damn Lies, and Statistics
Post by: SoCalMan2 on October 16, 2015, 09:22:52 AM
miff Wrote:
-------------------------------------------------------
> Joe,
>
> Depends on field size and the ability of an
> outside posted horse to make early position
> avoiding getting caught wide.Cant put a post
> position number on it, tend to look over the field
> and how a race appears on paper before tossing
> outside horses.
>
> Mike


This just occurred to me, why not do the stats from the outside in rather than the rail out?  For example, what is the percentage of the outside horse?  One from the outside, two from the outside, etc?  Every race has that, so there will always be an outside horse.  That may be a more elucidating way to run these stats, no?
Title: Re: Lies, Damn Lies, and Statistics
Post by: horsegoer on October 18, 2015, 03:16:42 AM
Why can\'t it be that just about 76% of the time the better horse was in post 1-4?
Title: Re: Lies, Damn Lies, and Statistics
Post by: Molesap on October 18, 2015, 01:40:05 PM
Because if 95% of the horses were from posts 1-4, that would not be very good would it?  If there were 10 horses in every race, it would make more sense in general. Or if 80% of the favorites were from those posts that would not look so hot either. In addition, I think you need you have to consider the gate number versus the post position - they are not the same thing. If there is a full field of 14 in a race and there are 14 stalls, every horse loads up in each one, but if there are six horses, they might skip the first couple of gates and load the first horse in gate #4, but that horse will be listed as post position #1. Might make a difference.