Welcome, Guest. Please login or register.
Did you miss your activation email?
2012-May-23, 06:18 AM

Login with username, password and session length

Racehorse TALK

Pages: 1 ... 5 6 [7] 8 9   Go Down
 
Author Topic: Computer Calculations and Knowledge Rules for Race  (Read 7876 times)
0 Members and 2 Guests are viewing this topic.
TheAnalyser
Listed
user 936
Offline Offline
Posts: 339
Original Post 2010-Apr-25, 05:19 PM

As Requested here is a topic to talk about calcuations/rules/ratings formulas.

As an added bonus I will release my code in the next few days(when I get time to write the install instructions) so that others who can not program will be able to collect their own data and devise their own rules/systems/formula.

If anyone wishes to add comments on ideas they use in their computer programs this would be greatly appreciated.
Logged
 
Wenona
VIP Club
Group 2
user 175
Offline Offline
Posts: 4351
2010-Jul-21, 10:01 AM

+
And the gaps will monotonically decrease with favourite rank.

In turn this means an outright favourite will always underperform Harville for running 2nd.

And the situation will improve as favouritism decreases. At some point 2nd performance will overtake Harville.


So jfc, have you anlaysed where this point is crossed in a large number of races?

If rank is more important than probability you would expect to find a consistency at what rank this point is crossed in races with similar field size regardless of the spread of market prices, wouldn't you?

If probability is more important you would expect to see the rank in which it crosses vary more depending on the differing spread of probabilitites within races with similar numbers of runners.



« Last Edit: 2010-Jul-21, 10:04 AM by Wenona » Logged
jfc
Group 2
user 723
Online Online
Alias: jfc
Posts: 1340
2010-Jul-23, 07:20 AM

+
So jfc, have you anlaysed where this point is crossed in a large number of races?

If rank is more important than probability you would expect to find a consistency at what rank this point is crossed in races with similar field size regardless of the spread of market prices, wouldn't you?

If probability is more important you would expect to see the rank in which it crosses vary more depending on the differing spread of probabilitites within races with similar numbers of runners.






Wenona,

anyone availing themselves of the spreadsheet I just added to the Exotic Calculator thread can fiddle with various market probabilities to note crossover trends or whatever.

Now I note the content of that google books link to that Benter material seems to have vanished like a gambler's lucky streak.

Maybe the work of the aforementioned pair of narks employed to Ustasi libertarian Alan Woods?

Anyway if anyone has alternate permanent free links, why not share them here?

Logged
pegasyber
Group3
user 909
Offline Offline
Alias: comprate
Posts: 580
2010-Aug-24, 07:20 AM

Once again the Warwick Stakes 1400 m. shows itself to be a speed type race with the following information.

Speed Order  . . . . . . . . . . . . [  3  5  4  8
Weight Modified Speed order[  3  5  4  8
Simulated Pace Order . . . . . .[  3  5  8  7

On Statistical Order . . . . . . . [  7  2  4  3  1

Result   [   3   5   2   13  ] with No 3 Metal Bender was top rated,  paying $12.60 win Place $3.40. Quinella 3 and 5 paying $60.70, Trifecta  paid $521,  First Four [ 3  5  2  13  ] paid $10145.40 but No 13 appeared to be unsighted in speed ratings.

There was a Jackpot on the Trifecta pool, so one could expect some unusual results.  
« Last Edit: 2010-Aug-24, 07:26 AM by pegasyber » Logged
Tevez17
Steward
Group 1
user 159
Offline Offline
Posts: 6420
2010-Aug-26, 03:35 PM

What is the general rule regarding first up horses? Especially when Favourite?
Logged
Wyatt Earp
Listed
user 325
Offline Offline
Posts: 338
2010-Aug-26, 03:48 PM

They must have the best last start margin in the race.
Logged
Tevez17
Steward
Group 1
user 159
Offline Offline
Posts: 6420
2010-Aug-26, 03:52 PM

They must have the best last start margin in the race.


How would it work if on has won a week maiden by a lengthy and one has gone down 3ls in a handicap both going into a class one?
Logged
Wyatt Earp
Listed
user 325
Offline Offline
Posts: 338
2010-Aug-26, 04:00 PM

Ah yes, same as a one that's been beaten 3L at a city metro meeting dropping back to the provincials against last start provincial winners.
Logged
Tevez17
Steward
Group 1
user 159
Offline Offline
Posts: 6420
2010-Aug-26, 04:22 PM

Ah yes, same as a one that's been beaten 3L at a city metro meeting dropping back to the provincials against last start provincial winners.



okay, has it been working out well for you?

I'm tempted to not entertain first up horses & maiden races unless trialled impressively,
also I used to be a big fan of class drops but the combo of out of form horses & most being unders is making me think again about how highly i rate a class drop in my thinking.
Logged
Wyatt Earp
Listed
user 325
Offline Offline
Posts: 338
2010-Aug-26, 05:16 PM

I won't back a horse which hasn't got the best last start margin in the race, unless its dropping back in class. I have other rules too. The one's I looked at today were (all had the best last start margin) ...
26-08-2010 GES 1   2 - BROKEN JAW - would've been a bet, but girly jockey
26-08-2010 GES 2   10 - YOUEFFOH - didn't fit my other rules
26-08-2010 GES 3   11 - COSMIC CAUSEWAY - bet
26-08-2010 GES 7   2 - MR O'POSTROPHE - didn't fit my other rules
26-08-2010 GOS 1   3 - EMPEROR BIG NOSE - bet
26-08-2010 GOS 5   13 - WAVE DANCING - didn't fit my other rules
26-08-2010 NOR 1   7 - DOUBLE SWIRL - didn't fit my other rules
Logged
Tevez17
Steward
Group 1
user 159
Offline Offline
Posts: 6420
2010-Aug-26, 05:31 PM

Do they tend to come up short? I guess for the most part you'll come up on a last start winner.
Logged
Wyatt Earp
Listed
user 325
Offline Offline
Posts: 338
2010-Aug-26, 05:35 PM

Yes, a lot of last start winners and maidens in the bush, the city's a different matter. And yes, 99% or them start favourite.
Logged
pegasyber
Group3
user 909
Offline Offline
Alias: comprate
Posts: 580
2010-Sep-26, 11:06 AM

For Jim; Speed Ratings and Calculation of a Rating.

If you look at the four lines that I posted on The Friday Night which included the George Main Stakes race as shown below; it recorded  that even limited to the horses last three runs of form, from the UNITAB site data, can provide some common sense. After all if some of the Racing has nothing to do with SPEED then you might as well pack up and just  buy LOTTO tickets.

Quote
RF210910 rw62509f * RACE NAME ID* [ GEORGE MAIN STAKES 1600 metres.]
RF210910 RW62509F * SPD ORD TABNO [  3  8  7  4  6  1  2  5  ] TOP5 HIHIT *
RF210910 RW62509F * WMSPD ORD TNO [  3  8  4  7  1  6  2  5  ] TOP5 HIHIT GW*
RF210910 RW62509F * L600 FSPD T10 [  4  3  8  7  5  2  1  6  ] TOP6 HI HIT*
RF210910 RW62509F * SIMUL PACE ALL[  8  7  3  4  5  1  6  2  * DBLCHK



Final Placings were  8  More Joyous, 7 Trusting, 4 Emporers Choice, 2 Danleigh, 3 Black Piranha.
To clarify the line headings:

SPD ORD TABNO = HORSE NOS IN SPEED  ORDER High To LOW.
WMSPD ORD TNO= TAB NUMBER IN SPEED ORDER HIGH TO LOW modified by the weight change today from last race.
L600 FSPD T10    = TAB NOs in Order of LAST RACE FINISH SPEED Over the last 600 metres (data available in newspaper)
SIMUL PACE ALL = TAB NOs in Order of A Simulated Pace Rating For the Horses Last Race. ( This is a very crude Approximation from the only easily available data.)

To calculate any Speed type rating, of a previous run, the base formula is:

SPEED RATING =  100 -  ((  RACE TIME secs -   COURSE RECORD secs ) / 5 )  - LENGTHS OFF WINNER  

« Last Edit: 2010-Sep-27, 06:37 AM by pegasyber » Logged
pegasyber
Group3
user 909
Offline Offline
Alias: comprate
Posts: 580
2010-Oct-19, 06:43 AM

The Last Minutes Check on your Selections using 2010 Caulfield Cup as an example:
Quote
Despite the Heavy(10) or (9) Track at Caulfield for the 2010 Caulfield Cup, the Big Boys knew what they were doing. Sure there were about seven starters that had a great chance of winning, but in the end it came down to one simple Ratio.

The seven possible winners were ( Smallest Ratio Best ) seen by the professionals ( or the market as a whole ), to be :

6.  Harris Tweed  0.59  Second. but Was the winner just before the post and just after the post.
12.  Alcopop       0.60
15.  Descarado     0.64  Winner.
20.  Red Ruler     0.64
10.  Triple Honour 0.74
17.  Faint Perfume 0.75
1.  Shocking . .  0.76  Fourth.
9.  Monaco Consul 1.10  Third.
The two Improving 4 year olds that had very good last race runs, although well disguised, were 15 and 9.

By the way the calculations showed that 4 Tokai Trick and 18 Dariana had  little chance of even placing, at least as far as the market was concerned, and the market is usually close to the mark.

Although these were the actual ratios on the closed pool, the figures are not that different for some time before the pool is closed, giving enough time for final adjustments to the exotics. However once  the Win Pool is above $10000 the calculation can be quite revealing and therefore useful in modifying selections for the tipping comps, that need a decision before 12.00 on raceday.  

« Last Edit: 2010-Dec-13, 07:53 AM by pegasyber » Logged
pegasyber
Group3
user 909
Offline Offline
Alias: comprate
Posts: 580
2010-Oct-22, 03:16 PM

Speeed and The Geelong Cup 2010:

RF201010 GL82010F * RACE NAME ID* [ GEELONG CUP 2406 metres.] PLACINGS [  1   5   9   10  ]
RF201010 GL82010F * SPL TJ*COMBO* [  10 ] MR CHARLIE Peter G Moody Jock: L NOLEN ] *STAR COMBO* F4  !!
Speed:
RF211010 GL82010F * SPD ORD TABNO [  1  11  5  4  7  10  15  3  8  9  14  13  ] TOP5 HIHIT *
RF211010 GL82010F * WMSPD ORD TNO [  1  11  15  3  4  5  7  10  8  9  14  13  ] TOP5 HIHIT GW*
RF211010 GL82010F * L600 FSPD T10 [  6  9  5  12  4  14  10  1  2  8  7  11  ] TOP6 HI HIT*
RF211010 GL82010F * SIMUL PACE ALL[  11  14  3  8  15  5  4  13  10  7  6  1  12  * DBLCHK
Statistical:
VBFZ0710 GL82010W * TOP6 FACTORS  [  2  5  14  9  10  1 ] [    12   CONST$. ]  TOP 6 VERY HI PLACE HIT
Logged
pegasyber
Group3
user 909
Offline Offline
Alias: comprate
Posts: 580
2010-Oct-22, 03:46 PM

Mooney Valley COX PLATE 2010:

Speed Calculations:
Quote
RF211010 MV82310F * HIGHEST EARNER[  1  ZIPPING WITH $ 3630195 CANT DROP F4
RF211010 MV82310F * SPD ORD TABNO [  3  4  7  8  10  9  5  1  6  2  ] TOP5 HIHIT *
RF211010 MV82310F * WMSPD ORD TNO [  4  3  8  7  10  9  5  1  6  2  ] TOP5 HIHIT GW*
RF211010 MV82310F * L600 FSPD T10 [  5  9  2  3  10  8  7  6  1  4  ] TOP6 HI HIT*
RF211010 MV82310F * SIMUL PACE ALL[  4  7  10  8  3  5  9  1  6  2  * DBLCHK


STATIST FACTORS  IN ORDER [  5  10  4  6  7  3  2  8  ].
« Last Edit: 2010-Oct-22, 04:07 PM by pegasyber » Logged
Pages: 1 ... 5 6 [7] 8 9   Go Up
 
 
Jump to:  

 - Links - Luxbet Horse Racing Betting - Racenet - Horse Racing - Noms and Acc - QTAB - TAB Horse Racing Betting - Racing and Sports - Horse Racing Only
Powered by SMF 1.1.15 | SMF © 2011, Simple Machines
TinyPortal v0.9.8 © Bloc | Adagio design by Bloc