Why not a system with few bets?
 

I've just been fiddling around with my systems. My MAIN system (not the one that drives me spare with Vic & NSW class), one I've followed for decades, with Herald Sun form guides, a pencil and paper, once again came up trumps. It's uncanny in it's consistency given that it has few bets ..... I've used Chrome Princes database for several years now so it's easier.

It got me thinking that there must be some shrewd punters out there who are extremely selective when they bet and only have a few bets each month. Why not with systems?

My system focuses on Open Hcp, Listed, Group 11 and Group 111 races only, so the bets are mainly in springtime in Melb and Sydney, and over the year there will be approx. 30 bets, that's all! The return is nearly 60% on a win strike rate of 25% and the maximum number of outs is 11. It shows a place profit also, but I only bet for the win.

It's not been backfitted because when I started way back when I had hair I was drawn to one main filter (there are no weights filters, days between runs, track condition, margins) and added 5 more which includes Metro as one of them. Will it keep working?, I don't know. How many would follow it?, only me probs :) It's certainly NOT the holy grail

I recall Shane Dye saying if he was a punter he would only have one bet each year on the favourite in the Golden Slipper. Not sure how that would have gone, but the comment was interesting. If you're a Vic you could back Lloyd Kennewell's horses when they come over ..... who knows? Food for thought at least. There'll be a couple of well credentialed NZ horses come over and clean up this spring as they normally do.

Depending on the rule there is no reason why it can't continue, the main thing with a system like that is for most people it would not provide enough action, what people do that follow systems like this with say one or two rules or even 20 rules that target a specific type of horse is they might have 10 of these such type systems.

many many years back a friend of mine gave me a book it was labeled the Pegasus System this book was similar to the approach i have just outlined and in fact was probably written back when you came up with yours, he wanted me to see if i could turn it from paper to computer selections, back then i didn't have the knowledge or the form, maybe in this day i do.

I agree with Shaun. People want more action then a few bets.

I run lots of systems. Each one only has a few bets a day. Some have only a few bets a week or month.

Systems work if they are properly put together and tested.

Hi Tipsy

The issue with this type of strategy is that the nature of probability and statistical variance means that it will take a long time to know whether or not your system is actually successful, you simply can’t avoid it. I'm not saying it won't be, i'm simply pointing out some facts associated with the math.

So lets make some assumptions:

Betting Bank : $10,000
Strike rate : 25%
POT 10% (most professionals struggle in today's environment to achieve this
in reality before factoring in any rebates)
2 lines of deviation (95% of the time your returns will be in this range)

If you bet 30 selections in a year like you suggest, then the range of returns will could fall between a loss of $1,430 to a profit of $1,930. In fact it doesn't matter whether it takes you a year or a week to make those 30 bets the figures remain the same.

Back on point, you would need to make 1,452 bets in total using the above assumptions before the range of your returns would be somewhere between break even and a profit of $23,232. Importantly at some stage during this cycle of 1,452 bets you will experience draw down on your betting capital. The above scenario factors in a maximum of 50% draw-down which means that at some stage you will have a draw-down figure of approximately $5,000 assuming you have in the first place a betting edge of 10%. It could happen when you start or at any other stage during the cycle. In fact it could reach the implied maximum more than once during the cycle.

In summary, we believe that the most successful betting strategies are ones that provide a balance between an adequate number of bets and a reasonable strike rate. Getting the combination right largely depends on what type of punter you are. I wish you all the best and hope this helps.

In the main i am a Multi - bettor as in the main there are many chances in most races.

However if I was relying on single selections they would need to be based around Professional ratings as the 3 I use are.
The drawback is most are short priced as one would expect with the occasional surprise.

Cheers
darky

PaulD01

"The issue with this type of strategy is that the nature of probability and statistical variance means that it will take a long time to know whether or not your system is actually successful, you simply can’t avoid it."

There's more than 25 years of data, a Race Census data base for the past 15 and before that at least 10 years of hand written results. Given the small amount of bets it wasn't difficult to keep for those 10+ years. I've only been punting it for the past few years and you can go a fair while without a winner.

Shaun you're spot on with your assessment, and maybe a gem with the Pegasus idea.

Thanks for all the responses. :)

Hi Tipsy,

As I said i'm not saying it won't be successful and I take your point regarding the data that you have showing that it's currently successful. The relevant key point is that based on the number of bets required to ensure that it is and will continue to be successful overall means that at 30 bets p.a and 1,452 bets required to prove its success, then that equates to more than 48 years. The bottom line is that such a method is not feasible as its simply does not provide enough action. In any case I hope it continues to work for you.

PaulD01 ..... "The bottom line is that such a method is not feasible as its simply does not provide enough action."

That's your opinion but you appear to have stated it as factual ..... :)

If I've interpreted your posts correctly then regarding your criteria to have a certain number of bets to prove that a system is workable, then where do you stand on backfitting a system to meet your criteria?

Hi Tipsy,

I'm not a systems punter but an ardent form analyst, but I accept that systems can be a viable way of grafting a profit. If you are building a system then I do not advocate that you should back fit data to suit the purpose of seeing what appears to be a profitable method.

Hi PaulD01 ..... "I'm not a systems punter but an ardent form analyst"

I can see with that statement why you're sceptical, and fair enough too. :)

Thanks for your responses, good info!!

I think there are two interesting points here.

1. Yes there is a defined set of bets per system which tells you if its workable or not. Its all based on both statistics and common sense.

2. Backfitting is an interesting problem for any system punter doing research. If you understand how to do step 1 properly then it doesn't matter. The problem is that 99.99% of punters don't know how to determine the viability of a system going forward.

They should have a fair idea of the liklihood if using sensible form factors
and stake correctly (i.e Kelly )
cheers

Here is the thing though.
In this day and age almost everyone has some sort of Data Base including the Racing Journalists so why not take advantage of their expertise.
I.e In the Tipster Poll if they have only 2 horses on the top line then the indication is that these 2 are considered the best 2 for that race.
One then only has to get a minimum $3 about both to make a good profit.
Of course not all will be at that minimum price so why not just take the better priced.
Over a period it should even itself out.

My belief is that we can outsmart ourselves looking for complications.

Cheers
darky

One small difference between the system you've just posted and my system is that mine shows a profit over the past 25+ years. The system you just put up would lose. No offence meant, but those types of systems are based on the office junior doing the tips.

They have to tip in every race, we don't, as far as ratings go everyone has there own perspective of what form factors are important, that's what makes the difference with good ratings and average one's.

Hi UB,

Any chance you could share the number?

Tipsy,
They used to be before .the advent of Computer ratings but not now.

My belief is that overhandicapping leads to the poor house based on my 40 years as a successful punter.

We dont have to bet in every race - only the standout selections (the key to my success as a punter.)

Cheers

Why is it that computer ratings in the newspapers have such an abysmal record? There are many days where there's not one winner. How do journo's overcome this hurdle when it's their own newspapers that publish these (useless) ratings ..... do they go against their own newspaper?

By the way.
Isn,t it great that we are finally coming back to sharing Handicapping ideas like we used to do.
The last couple of years I,ve felt we have been taken over by fantasy posts.

Cheers
darky

Now what will win the 6th at Wyong - the 3 or the 5.

I wouldn't mind an answer to my premise that newspaper ratings, the top rated horse, have an abysmal record. And I believe that the
Herald Sun uses one of the better ratings services, the name which escapes me at the moment. I do have a bias against ratings but I would welcome those who believe in ratings putting up a decent argument for them, maybe including some data.

Some of the top rated horses are double figure odds ..... How can that be?

@ $1.95 to take out $500 I will have $256 on it and the 5 @$6.50 I will have $77

Not that mine are the best, but lets see.

No Selection R/Price
3 SECOND OBSESSION 2.32
2 JUPITER 4.20

I don't rate the 5 in the top 3 my next best is No1

I recall Dye saying exactly that, but it was in reference to the Cox Plate.

Quite a nice profit there of $167 / ROI 33.4 %

Hi Chrome,
For me I would put $20 a week away until the Melbourne Cup then spread it between 5 Horses.

In general I have found that if a horse has not at least done 2800 then I can put a line straight through them.

darkydog, you've sort of ************ed up the potential of this thread .....

BTW darkydog, your (fantasy) bet shows a return of 50% not 33% as you calculated.

You hijacked this thread and refused to answer my question about ratings.

Funny on this very same thread you posted ..... "By the way. Isn,t it great that we are finally coming back to sharing Handicapping ideas like we used to do. The last couple of years I,ve felt we have been taken over by fantasy posts."

What's the point of anyone posting anything here when you come on and tell us all how good you are !!, twice in the same post "I'm a successful punter"..... I'll go, you stay!

Ok with me if you want to go.
You appear to just want to create trouble anyway.
I,m well over that thanks very much.

I,m only hear to share what works for me.

Not that mine are the best, but lets see.

No Selection R/Price
3 SECOND OBSESSION 2.32 (W-2.0)
2 JUPITER 4.20 (2ND S/P $9.0)

I don't rate the 5 in the top 3 my next best is No1

The five ran unplaced.

Guess that answers the question(s) above.

Cheers.

Hi Tipsy,

The Herald Sun ratings are the same as used on Unitab.

I have found the Zipform ratings to work quite well on Saturday Melbourne racing.

All horses >=Zip rating 56 has made 6.6% POT since 5th October 2013.

Obviously a good place to start if you want to fine tune your selections!

Let's see if we can get this thread back on track!

If only it was that easy. A starting point is chi results. A second is by examining the distributions of results. The third would be based on an analysis of time framed results. There are other ways but this covers some basics.

Glad to be of help Tipsy and pleased to be able to answer your question with a practical answer.

Cheers

Ps.Dont forget Darciwood and Lyric in the 7th @RW