PDA

View Full Version : Systems for Analysis_1



Grey Fox
10-01-05, 08:03 AM
Hi all,
When I get a chance I intend to post a couple of my systems, not to spoonfeed the newer members, but hopefully to stimulate a bit of discussion about the systems, and systems in general. As part of this process I will be including my reason for creating (or pinching!) the systems because I think that it is vital to start off with a well thought out theme or idea for each system, then to sit down and think about which factors would be relevent for that theme.

This first one is named after Neale because it was based on statistics that he published in a newsletter he edited in the 90s -- the stats themselves referred to racing in the 80s, so who says that racing is always changing? ;-)

The stats showed that sticking a pin in the formguide would produce a 10% win strike rate and a POT of about -18% over the long term, so they are the benchmark figures for all our systems. They then showed that Last Start Winners had a higher win rate and lower LOT than average, that Last Start Metro Winners had a higher win rate again, and lower LOT, etc.

So I took these figures, entered them into System Selector along with a few other commonsense (to me, anyway) rules, and voila, a system was born. The rules and results are:



SYSTEM SELECTOR RULES SET FOR: YARDLEY_THEME
Printed by Bet Selector at 16:06:05 on 08/01/2005

Race Field Size: 7 - 24
Day of Week: 7
Venue: SMBW
Track Condition: G
Horse Win %: 25 - 100
Horse Place %: 50 - 100
Career Starts: 6 - 99
Days to Last Start: 1 - 28
Last Start Finish Posn: 1 - 1
Metrop Last Start: Y
Saturday Last Start: Y
Predicted Posn Rank: 1 - 1
Predicted Odds: 0.0 - 3.9




Meetings considered : 996
Win Strike Rate/Seln. : 36.3%
Plc Strike Rate/Seln. : 66.1%
Average Win Dividend : $3.10

WIN PLACE QUINELLA EXACTA TRIFECTA FIRST FOUR
Races Bet : 416 416 416 416 416 230
Races Won : 151 275 179 124 88 36
S.R./Race : 36.3% 66.1% 43.0% 29.8% 21.2% 15.7%
Outlay ($): 416.00 416.00 2080.00 2080.00 8320.00 13800.00
Return : 468.10 413.70 2260.60 2644.10 9242.00 36245.40
$ Profit : 52.10 -2.30 180.60 564.10 922.00 22445.40
% P.O.T. : 12.5% -0.6% 8.7% 27.1% 11.1% 162.6%




Neale, I tried using the pre format, so we will see how they look.

The strike rate is acceptable, a reasonable number of selections each Saturday, and a sensible POT -- even the place betting almost broke even using NSWTAB divvies, so if you wanted you could bet the selections eachway at the track without getting into too much trouble.

BTW, the system works just as well at Public Holiday meetings because the horses are usually up to Saturday standard -- if you tick the box "Ignore Day and Venue" in the System Test window you can get your selections for holiday meetings. The midweek results are fairly poor, which is understandable given the rules, so stick to Saturdays and Public Holidays and you will have a lot of fun with this one.


Bye for now,
Grey Fox

Far Canal
11-01-05, 02:45 AM
G' day Grey Fox,

I've run your criteria through my data back to July 2000 ( gets a bit hap-hazard back there) and come up with about the same figures as you, possibly a better POT....19.8

The last 16 months (from August 03) worry me a little as it is in and out of POT finishing up with these figures....

Races Bet 117
Races Won 38
S/R Race 32.5
Outlay 117
Return 117.30
POT 0.3

5 Selections without Results, since Xmas, we'll get them soon.
By-the-way, Place S/R 62.4 POT -5.5

We'll have a wee play and see what happens.

Cheers mate,

Far Canal.

Far Canal
11-01-05, 05:31 AM
Grey Fox,

Had me wee play and came up with some new figures from 3 or 4 modifications....

Venue. Extend to include "A" some good figures there, "B" not too hot though, I've kept it in but I would not invest there personally.

Horse Win %. Reduce that to 20%. Missing some good winners, I could be persuaded to look at 15% but no lower, but have not had time....later.

Horse Place %. Wipe that completely as it restricts your selections. 50% is neither one way or another.

Career Starts. Drop that to 4 to 99. The horse won last start out so why restrict it further.

Barrier. 1 to 9. As there are not restrictings as to Distance we are reducing the wide gates for the shorter Distances.

Weight. 53.0 to 75....reduce the claiming hoops.

I've run the new criteria back through the exactly same data as previous post and the July 2000 results have a better POT, 21.4....Place, 1.9.

August 2003 figures are....

Races Bet 190
Races Won 70
S/R Race 36.8....Place 67.4
Outlay 190.00
Return 232.90
Profit 42.90
POT 22.6....Place 1.9

6 Selections without Results....they have just turned up but I'm not going to redo this. There is no doubt that this system can be improved upon but time restricts.

Have a good day mate

Slow N Easy
11-01-05, 03:36 PM
Appreciate you putting up Sys Analysis 1 and 2 for discussion.
Liked the themes, we tend to forget the lessons as we get older.

I think the guys respect your experience so much that they are a little reluctant to get in and try to improve one of your creations.

Come on fellas have a go. The best way to learn is to give it a shot.

Heres a real challenge if you can improve one of Grey Fox's systems you'll have something to tell your grandchildren about.

I haven't had a chance to get at them yet but I will.

Cheers Mel

Slow N Easy
12-01-05, 07:06 PM
Hi Grey Fox

Ití me again.

Note that I have included Venue A as I didnít have very many W in my Base records. So these findings are not a true reflection of the System presented by Grey Fox but they should be somewhere in the ball park.
Grey Fox kindly put forward this system to create some discussion that we may be able to benefit from, so any comments (sensible or not) should not be taken as anything but constructive.
I ran the rules on my small database to establish my start point.

ZZYARDTHEME SMBW7G Bet Selector at 15:22:34 on 11/01/2005

Race Field Size: 7 - 24
Day of Week: 7
Venue: SMBAW
Track Condition: G
Horse Win %: 25 - 100
Horse Place %: 50 - 100
Career Starts: 6 - 999
Days to Last Start: 1 - 28
Last Start Finish Posn: 1 - 1
Metrop Last Start: Y
Saturday Last Start: Y
Predicted Posn Rank: 1 - 1
Predicted Odds: 0.0 - 3.9

Meetings considered : 377
Win Strike Rate/Seln. : 33.1%
Plc Strike Rate/Seln. : 62.1%
Average Win Dividend : $3.13

WIN PLACE QUINELLA EXACTA TRIFECTA FIRST FOUR
Races Bet : 145 145 145 145 145 123
Races Won : 48 90 61 39 27 18
S.R./Race : 33.1% 62.1% 42.1% 26.9% 18.6% 14.6%
Outlay ($): 145.00 145.00 725.00 725.00 2900.00 7380.00
Return : 150.40 137.54 705.60 711.50 2152.20 6883.10
$ Profit : 5.40 -7.46 -19.40 -13.50 -747.80 -496.90
% P.O.T. : 3.7% -5.1% -2.7% -1.9% -25.8% -6.7%


Summary of Starting Price of Winners (Applicable to above)
Collated from running SP Price on Bet Selector a number of times.
Note 4 Losers lost in fractions or in 12+
SP price range 0-0.9 1-1.9 2-2.9 3-3.9 4-4.9 5-7.9 8-12 12+
AvWin Div : $1.54 2.40 3.31 4.60 6.25 10.05

P SR./Sel : 94.7% 73.8% 57.1% 62.5% 35.0% 37.5%
Races Bet : Nil 19 42 28 24 20 8
Races Won : 15 15 7 5 4 2
W SR./Sel : 78.9% 35.7% 25.0% 20.8% 20.0% 25.0%
Outlay ($): 19.00 42.00 28.00 24.00 20.00 8.00
Return : 23.10 36.00 23.20 23.00 25.00 20.10
$ Profit : 4.10 -6.00 -4.80 -1.00 5.00 12.10
% P.O.T. : 21.6% -14.3% -17.1% -4.2% 25.0% 151.3%

This little exercise was for my appreciation of what lay under the surface of the win results which produced 48 winners at a strike rate of 33.1% and a profit of only $5.40
at 3.7%. Itís interesting to occasionally do this sort of thing for understanding.
As it presents an enlightening illustration of poor value, so I thought it would be good to show it is important to be aware of the need to seek fair value.
In the first 2 columns we have 61 bets 30 wins with SR 49.2% NET LOSS of $ 1.90.
Oh I almost forgot the extraordinary place SRate of 94.7% (col 1) had 6 @ $1.04 and 9 @ $1.10 and from memory there were about 16 or 17 consequetive?? Placers.

Here is what I came up came up with on a project that I had difficulty in dealing with because of the myriad of such short prices. I was frustrated because I wanted to keep to the theme presented by Grey Fox and try not to tinker outside of the fields that he had introduced.

SYSTEM SELECTOR RULES SET FOR: ZZYARDTHEME SMBW7G
Printed by Bet Selector at 00:44:55 on 12/01/2005

Race Field Size: 7 - 24
Day of Week: 7
Venue: SMBAW 1.
Track Condition: G
Horse Win %: 15 - 45 2.
Horse Place %: 40 - 70 3.
Career Starts: 6 - 30 4.
Days to Last Start: 1 - 28
Last Start Finish Posn: 1 - 1
Metrop Last Start: Y
Saturday Last Start: Y
Predicted Posn Rank: 1 - 1
SP Price: 3.0 - 999.9 5.

Meetings considered : 377
Win Strike Rate/Seln. : 24.8%
Plc Strike Rate/Seln. : 47.9%
Average Win Dividend : $5.64

WIN PLACE QUINELLA EXACTA TRIFECTA FIRST FOUR
Races Bet : 121 121 121 121 121 109
Races Won : 30 58 33 18 11 7
S.R./Race : 24.8% 47.9% 27.3% 14.9% 9.1% 6.4%
Outlay ($): 121.00 121.00 605.00 605.00 2420.00 6540.00
Return : 169.20 124.30 610.90 612.70 1577.90 5005.30
$ Profit : 48.20 3.30 5.90 7.70 -842.10 -1534.70
% P.O.T. : 39.8% 2.7% 1.0% 1.3% -34.8% -23.5%


A small improvement but nothing to write home about.

My thought process on why I changed the things I did.

1. Added A to boost data Qty.

2. I like to include a min of 10% to 20% to add value because they outside the thought processes of others. I tend to vary this depending on sys type. The maximum I like to keep around 40%-60%. Higher and every man and his dog is getting on the band wagon. Keep your life jacket on, donít go overboard.

3. 40-70 is nice number. Under Ė most times you wouldnít want to be on them. Over - can be perennial placegetters or they a getting into the short price range because they have won more than their fair share of races.

4. I donít want to be on horses that have had 99 starts. I can vaguely remember one once. Iím looking for young or lightly raced that are on the way up.

5. Thanks Reggie you are a star. I like this one but it has the problem of how you can estimate whether it will be above your limit. Systems are not rocket science you know. Have a guess, how many have tipped it, does it get a heavy wrap on radio, check paper price or even ask your young toddler he may know. But make some estimate and go with it. I have introduced a bottom limit here because I have bothered to look behind the scenes. 61 bets 30 win @ SR 49.2% Loss $ 1.90. Now thatís on the way to the moon you rocketeers.

So you can have valid thoughts about what you choose to do. Deliberate, think about how you would tackle the problem if you were considering an action on item if you were on course. What is your normal thinking in this area? If you make a decision on this item in this system you have the convenience of making it now and forgetting it until your next system.

Donít get carried away and cut and slash just for the sake of it.

Take a little break and find a valid reason for your action.
I have many things that I am opposed to on the racecourse. So when I sit down to compile a system I bring these things to the table and try to apply them to the system as I would in the field.
It is for this reason that I like to tinker with some areas of systems.

Egís Odds On Ė If I can find a way they are off the menu.

Anyone for snitzel itís at the roundabout.

Cheers for now Mel