PDA

View Full Version : Backfitted System ....


wesmip1
26th February 2006, 10:08 PM
All,

Here is a "backfitted" system which might be worth watching. All the back fitted rules are sensible (in my opinion). I developed this today cause I got bored and thought I would give backfitting a go.

It has had 10 bets in the last 10 days and has had 10 winners.

The first bet which is coming up is:

Tuesday @ Grafton Race 1 - Canny Play

Results from the previous 10 races are below :

WIN $ PLACE$
$2.30 $1.40 - 1st
$3.00 $1.80 - 1st
$5.40 $2.20 - 1st
$1.40 $1.00 - 1st
$2.30 $1.30 - 1st
$2.80 $1.30 - 1st
$1.40 $1.10 - 1st
$2.00 $1.30 - 1st
$1.80 $1.20 - 1st
$1.60 $1.10 - 1st

Remember this is backfitted and over a very short time frame so I expect this to fall flat. But if interested you might want to keep an eye on it.

Good Luck.

darkydog2002
27th February 2006, 08:18 AM
Might be worth creating your own ALL - UP win bet with this one.

Cheers.
darky.

dingoboy
27th February 2006, 08:44 AM
WELL DONE !
excellant results

Dingo

KennyVictor
27th February 2006, 11:02 AM
Here's another one.

Rule 1. Select the winner of the fourth race on the card.
Rule 2. Metropolitan meetings only.

You can play around with rule 1 a bit. Ignore rule 2 if you like but a lot of people like to get onto this one so I find it works better on Metro meetings where the prize pool is larger.

KV

wesmip1
27th February 2006, 11:14 AM
KV,

At least I gave a tip with the method.

Tuesday @ Grafton Race 1 - Canny Play
Can you give a tip for your method (before the race) ?

Good Luck.

KennyVictor
27th February 2006, 12:17 PM
Full marks Wesmip you've spotted a flaw in the system. Thank goodness someone did before I started using serious money. :-)

However, in my favour I did post some rules. I don't know about anyone else but I find it a bit frustrating seeing a bunch of results or selections and no idea how they are arrived at. It doesn't teach me anything or give me any grist for the punting system mill.

Best of luck anyway,

KV

wesmip1
27th February 2006, 02:47 PM
KV,

Point taken....

If tommorrows pick wins I will post the gist of the backfitted rules so you can work it out.

I agree seeing results is meaningless without knowing how they are achieved, but its a backfitted system and I doubt it will hold up anyway.

KennyVictor
27th February 2006, 03:00 PM
You're a gentleman and a scholar.

KV

dingoboy
28th February 2006, 12:48 PM
It won!
Protest on ...mmmmm

Nah well done wesmip

Dindo

dingoboy
28th February 2006, 01:01 PM
protest dissmissed

i am drooling here.

wesmip1
28th February 2006, 03:49 PM
Well that is 11 from 11 now.

Ok I promised I would give the gist of the rules for this system if the selection won so I better be a man of my word and pass it on :

The system uses some different ratings which I have developed for different factors on each horse , sort of like the neurals but I found their values weren't always correct or meaningful.

Essentially the system goes the following :

1. Best Average and/or Total Prizemoney.
2. Best Career Rating ( Determined by Win%, Place% and Prizemoney ).
3. Best last 5 start rating ( determined by Last starts and Lengths beaten as well as some minor adjustments for class and track ).
4. Best Jockey Rating OR Best Trainer Rating in the race determined by statistics collected from ********** and AAP.
5. Total Ratings for all four factors ( jockey or trainer only include 1 ) is over 380 (out of 400).

The reasons I use these variables/ratings are :

1. Best average money or total money narrows down to those horses placing and/or winning in better races (paying more prizemoney).
2. Best career rating finds you those that performance consistently.
3. Best last 5 starts rating gives you a current snap shot of how they are performing now.
4. Best Jockey and/or Trainer is the X factor. A good jockey can get something extra out of a horse during the race. A good trainer can select optimum times to race, and the best races to run in. A good trainer can usually get something extra out of the horse.
5. The total rating has to be high to ensure that only the best selections are tipped.

Sorry I couldn't give everyone something nice and easy to follow... but I hope this points people in the general direction.

Good Luck. I'll post anymore selections as they arise but there aren't too many that come up each day to day.

Good Luck.

dingoboy
28th February 2006, 05:47 PM
Wesmip1,

Today i actually had to take your selection as it was one of mine, slightly differant reason though.

I was a tad concerned with the protest though.

It is a brave man that sticks his head out and to you i take my hat off, well done.

And thanks for the info, now its just a matter of following your leads i guess.

Hope your process keeps up and you win an absolute motza !

Thanks dingo

Truckie
2nd March 2006, 03:42 PM
I suppose there are many like me who are computer illiterate, and although we know much more after Crash's experience, much of what we read is beyond our comprehension. Have to admit to being over 80, so don't consider it worthwhile to improve my knowledge. Also, punting with $1 stakes, and trying to find the perfect system stops the mind from rusting.

Wesmin was good to reveal his system, which I have no possibility of following, but on trying to make something of the winner he predicted days in advance, I noticed the form of his selection contained 3 starts placed 4 to 6 followed by either a 2 or a 3. On looking over one month's papers, this picked quite a few short priced winners, without too many selections. Is this the seed for a method, selecting horses priced at $11 or less? Is anyone interested enough to check? Hope you continue with the good work, and maybe follow up Darky Dog's suggestion of all up.

dingoboy
2nd March 2006, 06:45 PM
If one looks at unitab (sorry dont have data base yet but working towards it)
have a look at the last three starts.
I have a place thing going on trial at the moment where i dont worry at the odds if they are under 20s.
As long as its like this,...kind of.

742, 641,6X2, 32X, 322 etc.
It often comes up trumps, have a look.

I target price bands for win, 3.5-12, over these i dont go win, place only but its worth looking at "improvers", even if they are up in class and weight.

Cheers Dingo.

Also have a look at Bhagwans Neural plan, he excludes all factors bar CF, the reverses it.

wesmip1
2nd March 2006, 07:11 PM
Truckie,

Its probably more of a coincidence that the tips I have are improving in form. But looking at the inputs horses that are improving will get a better score then those who are racing poorly.

Thanks for the post dingo boy. I'll check it out.

Good Luck

wesmip1
2nd March 2006, 07:29 PM
All,

Here are a couple more tips which are running on Friday:

Ipswich Race 2 - Mr Clang
Tatura Race 5 - Swit

My quick analysis shows that Swit is really the only horse in form in the race. Could be upset by one of the returning horses. While Mr Clang has ran second a lot and hopefully can break through and get that win.

Good Luck.

wesmip1
3rd March 2006, 03:56 PM
Hey,

Well today was the first loss of this backfitted system. Swit lost by a nose ( came 2nd). I never saw the race but the commentator indicated there was a bit of interference from the winner as they entered the straight. I might try to catch the race tonight on the replays.

Ipswich Race 2 - Mr Clang - SCRATCHED
Tatura Race 5 - Swit 2nd

Results
Races : 12
Winners : 11
Seconds : 1

Still to only be beaten by a nose (or head) is a good effort.

I'll post anymore tips when I find them.

Good Luck.

Duritz
3rd March 2006, 11:12 PM
I love systems, they're the promised land. And this one is so bare-facedly, shamelessly backfitted, with the smallest sample size in the history of statistics, that it just might work dammit!! (If we're in a movie, it'll work).

Looking forward to seeing tomorrow's.

wesmip1
4th March 2006, 07:25 AM
Duritz,

I totally agree but I'll keep posting it while it wins (its always a bit of fun). There is only a selection every few days and none today.

I don't use backfitted systems myself and this was my first attempt at it.

michaelg
4th March 2006, 07:50 AM
Even with eleven selections it looks very promising.

Have you thought of slightly varying it to form an additional system that would provide more selections?

Michael.

KennyVictor
4th March 2006, 10:17 AM
Here is a "backfitted" system which might be worth watching. All the back fitted rules are sensible (in my opinion). I developed this today cause I got bored and thought I would give backfitting a go.

It has had 10 bets in the last 10 days and has had 10 winners.
Hi Wesmip,
When I read this at first I assumed the 10 bets for 10 winners was the result of backfitting but reading some replies I get the impression some others think you backfitted it and then had the 10 winners using the rules.

Which is it? How many bets have you had since you devised the system?

KV

wesmip1
4th March 2006, 11:38 AM
KV,

ok to avoid confusion I did the following :

Took some data I had stored from Dec - Jan and essentially backfitted the results so that it only had horses which had won or placed. Any horse which didn't at least place I found a reason to exclude ( hence backfitting ).

I then checked it over the previous 10 days and found it had 10 selections for 10 Winners. But I didn't test these real time ( but they were not part of the data which was used for backfitting).

I haven't bothered going back further.

The backfitted results were :

Races : 45 ( 2 months of data)
Wins : 30
Places : 45
WIN ROT : 183%
PLACE ROT : 153%

I don't believe the above figures are worth anything except to show the system might have some potential. Hence I only put up the last 10 days worth as these were not part of the backfitted data.

The first "live" selection where I actually picked the winner before the race was the first one I posted here. Regardless of that the system had selected 11 winners in a row.

Therefore new stats are (with those 11 winners and a second) :

Races : 57
Wins : 41
Places : 57

Hope that clears it up.

wesmip1
4th March 2006, 11:44 AM
michaelg,

Probably could vary it, I just had a quick look over the data and found it was only one rule that really mattered. I might be able to just use this one rule, but it is the main filter and was the one that really cut out the majority of losers.

Anyway I will have a play aorund and anything I come up with I will post.

BTW there were no selecitons today.

Good Luck.

KennyVictor
4th March 2006, 12:59 PM
Thanks Wesmip, that makes it much more valid than my first impression. I look forward to more results.
KV

Duritz
4th March 2006, 05:47 PM
Yay Wesmip, our new leader!! Let us follow you, as children into the Piper's caves!! Lead on!!

wesmip1
4th March 2006, 07:53 PM
No Selections for Sunday.

Duritz
4th March 2006, 10:12 PM
Oh Wesmip, this is torturous!! We want to unload!!

We want to walk up to the bookie and say, "Give me EVERYTHING on the Wesmip moral!!!!!"

KennyVictor
5th March 2006, 04:59 PM
And the bookie shall say unto us "Nay, nay heathen rabble, get thee behind me unto the piper's caves from whence ye came. I shall bet against no such certainties."
And the horses shall say "Neigh, neigh.........

No, I've changed my mind. I refuse to be part of this Tomfoolerey.

dingoboy
5th March 2006, 06:09 PM
Are there any women here ?
No ...No....SHe is, she is,...no,...he is ,....he is.

Its all good fun.

Duritz
5th March 2006, 06:43 PM
And Wesmip shall calleth from the mountain to us, delivering unto us his ten Commandments for punting glory, and we shall be glad.

wesmip1
5th March 2006, 06:56 PM
My Ten Commandments are :

1. Pick the winner
2. Don't pick the horse which runneth second
3. Don't pick the horse which runneth third
4. Don't pick the horse which runneth fourth
5. Don't pick the horse which runneth fifth
6. Don't pick the horse which runneth sixth
7. Don't pick the horse which runneth seventh
8. Don't pick the horse which runneth eigth
9. Don't pick the horse which runneth ninth
10. Don't pick the horse which runneth tenth or worse

Its easiest and more spirtually enlightening to follow Commandment 1.

wesmip1
5th March 2006, 06:58 PM
For those who care ..

Selection tommorrow at Geraldton Race 2 - Lillan Min

Duritz
5th March 2006, 07:00 PM
I've just had $100,000 on it with Read.

dingoboy
5th March 2006, 07:26 PM
Good luck to you wesmip,

I shall be having a look, do not judge a man until one judges ones self,...i say.

All the best, good on you for sticking the neck out !

Dingo

wesmip1
5th March 2006, 07:36 PM
dingoboy and anyone else interested,

I am trialling a completely different system in the horse racing forum as well. It has had a lot more backtesting completed and after 2 weeks of live testing it is still showing promise. Still not enough data to confirm the selection method yet but I think its on the right track. And it throws a lot more selections than this method.

Duritz,

How about a % of those winnings when it wins tommorrow .. say 5% of the profit..

Good Luck.

Duritz
5th March 2006, 09:33 PM
Wesmip if I had $100,000 on it, I'd give you 5% for sure.

Duritz
6th March 2006, 01:20 PM
LOL early STAB price, with $757 in the pool, $1.60 the Wesmip moral, someone's already had a bet!

And it's OVERS at that quote!!

wesmip1
6th March 2006, 05:12 PM
Geraldton Race 2 - Lillan Min - 3rd NTD

I never saw the race so can't comment on the horse performance.

wesmip1
6th March 2006, 05:27 PM
There are 2 runners tommorrow :

Port Lincoln Race 1 - Perficio
Wyong Race 8 - Nisp

Good Luck.

Duritz
7th March 2006, 01:11 PM
Perficio was a moral.

wesmip1
7th March 2006, 01:24 PM
Duritz,

I wish I had $100,000 on him. I was happy to get the $1.60 I got for him.

Good Luck.

wesmip1
7th March 2006, 02:39 PM
All,

I have decided to only display the stats for the selections which have been posted:

Races : 4
Winners : 2
Seconds : 1
Thirds : 1
Unplaced : 0

I think that is correct.

Good Luck.

KennyVictor
7th March 2006, 02:50 PM
Races : 4
Winners : 2
Seconds : 1
Thirds : 1
Unplaced : 0
I know it sounds like I am but really I'm not trying to be picky here, however -
I think something like Win bets 2 from 4, Place bets 3 from 4 and / or the return on win bets and the return on place bets would be more meaningful.
I say that because although none of the horses so far has been unplaced on a first to third basis, one didn't score a payout. In a system which is probably targetting very short priced horses most of the time the difference is probably important.

Cheers,
KV

wesmip1
7th March 2006, 03:33 PM
KV,

I'm only betting wins .. So not really interested in place divis. Just want to know if the horse was at least placing which usually means it had some sort of chance of winning.

I really only care about Races, Wins and WIN ROT.

I have deliberatly kept the ROT not recorded as people can get the results themselves and everyone gets different prices. I am more interested in this study as a backfitting test and I am conentrating on the strike rate performance. If it holds the strike rate it had during the backfitting then I will investigate the POT and ways of improving it.

Good Luck.

Duritz
7th March 2006, 09:07 PM
That's fair enough Wesmip but why not put the ROT up nonetheless, just using your local tote dividend, just as an indication?

wesmip1
8th March 2006, 07:27 PM
ok guys I'll post the POT next time.

Today had 2 runners but I never got a chance to put them up. Both were very short priced:

Ipswich Race 1 : Capeline
Ascot Race 7: Raw Metal

Sorry I couldn't paste them earlier but I was just too busy last night and today.

Results for the races were :

Ipswich Race 1 : Capeline WIN $2.20
Ascot Race 7: Raw Metal WIN $1.70


I'll post POT next time....but the results so far are :

Races : 6
Winners : 4
Places: 6

wesmip1
8th March 2006, 07:36 PM
There are a few tips coming up tommorrow. I haven't checked the fields or anything but I suspect the will all be short priced.

Tips for tommorrow :

Kembla Grange Race 2 - The Free Stater
Port Lincoln Race 7 - Patarchie
Rockhampton Race 7 - King Hunter
Werribee Race 4 - Martian Dust

Good Luck.

wesmip1
9th March 2006, 08:43 PM
Results from today :

Kembla Grange Race 2 - The Free Stater WIN $1.60 / $1.40
Port Lincoln Race 7 - Patarchie LOST
Rockhampton Race 7 - King Hunter 2nd $1.40
Werribee Race 4 - Martian Dust 2nd $1.10

Results so far :
Races : 10
Winners : 5
Places: 9

I will post pots in my next post.

wesmip1
9th March 2006, 09:01 PM
OK I seem to not be able to add up as I have only posted 9 tips which have raced (1 was scratched) and only 4 have won.

New results are :

Races : 9
Wins : 4
Seconds : 3
Thirds : 1 (NTD still counted as Place)
Places : 8 (NTD still counted as Place)

WIN ROT : $8.00
PLACE ROT : $8.60

Looking through these results the top jockey criteria seem to be the ones failing. From only top trainer we have had :

Races : 3
Wins : 3
Places : 3
WIN ROT : $5.80
PLACE ROT : $3.80

Good Luck.

marcus25
10th March 2006, 07:04 AM
Looking through these results the top jockey criteria seem to be the ones failing. From only top trainer we have had :


Hi wesmip1!
I am not surprised about your finding that top rated jocks, not doing as well as they should.
If you think about it, it is quite logical, by the time the jockey rating reaches us, they are at least a week old, if not more.
If you want to follow them, do it on a daily basis, they can be in-form one day and, not feeling quite right the next, but still rate well at the end of the week or month.
Trainers are far more reliable, mainly because they have long term input.
Good luck.

wesmip1
10th March 2006, 04:42 PM
OK here are some tips for Saturday with the Trainer or Jockey criteria defined.

Ascot Race 2 - Shy Wild (Trainer)
Ascot Race 6 - Scenic Shot (Jockey)
Eagle Farm Race 2 - Pure Energy (Trainer)
Pakenham Race 6 - Spirit Of Capri (Jockey)
Warwick Farm Race 2 - Wasp (Jockey)


Good Luck.