Stebbo,
Have to put my little bit in, and I certainly ain't any expert on staking,ups & downs etc. I personally flat stake, sometimes I win sometimes I lose BUT when I get a $10 winner it more than makes up for the $3 losers.
I see subscriber services that give you smaller bets on some selections than others.
My argument is why have a smaller bet on some, if they dont think they have a good chance of winning why pick them in the first place.
As far as ups and downs I dont think anyone can predict this.
You can run 10000 races through a program and come up with X put your hard earned on and you will come up with Y , every race is different.
I know I sound a bit cynical but to me these are facts.
If it was all so easy we wouldnt have 100+ ratings, system services. dial a bet etc etc etc etc.
I have had good results from this site, and some rotten ones too.I am well in front at the moment but who knows how long it will go (how long is a piece of string).
What I am trying to get across is you have all these professional sites around who have access to things that we probably dont and they sell their selections.
Why? just bet them themselves and rake in the profits, It cant be done racing is racing and every race is different.
Hope I have'nt spoiled the thread.

Cheers

On 2003-10-20 18:57, kenchar wrote:
Stebbo,
Have to put my little bit in, and I certainly ain't any expert on staking,ups & downs etc. I personally flat stake, sometimes I win sometimes I lose BUT when I get a $10 winner it more than makes up for the $3 losers.
I see subscriber services that give you smaller bets on some selections than others.
My argument is why have a smaller bet on some...snip...

Hi Kenchar,

With all due respect I think you missed the point of this thread... The aim of this thread is to discuss risk management techniques, not staking techniques. The two are interrelated of course, but can be treated separately.

Of course, the best risk management technique I know of is to not bet losers, but I haven't quite mastered that one yet

:lol: :lol: :lol:

Cheers,
Chris.

Becareful,

Wouldn't stepped bets often produce the bigger bets hitting loosers and smaller bets hitting winners producing no advantage other than allowing more bets for your $ [perhaps] but not necessarily a greater chance of winning due to ever smaller collects ?

Cheers.


Crash,

In my case I am currently getting ever bigger collects as the bank is going up not down. Obviously with this system you first need a selection system that is showing a profit not a loss! (But in my opinion that is a basic requirement to make a profit anyway - I don't believe chasing losses with a staking plan will work in the long run). For example in the last 3 months of betting I have gone up 3 "steps" but have yet to have to step backwards.

To illustrate lets say I start with a bank of $1000 with a bet size of $50. In this case I would step "up" to a $60 bet if the bank gets to $1200 or "down" to a $40 bet if the bank falls to $800. Once I get above $1200 (so bet size = $60) then I will step "up" to $70 if I get to $1400 or back "down" to $50 only if the bank falls back below $1000. In this way I am not constantly changing bet sizes each time I get a loss so I have a very good chance of hitting a winner with my higher bet size before I have to drop back down.

Now the system I am using with this 5% approach has a strike rate of around 60% so I am fairly confident of getting a winner within 4 bets each time - with a lower strike rate system you need to use a lower % of bank otherwise you would constantly be changing bet sizes (not to mention the risk of going broke).

Just for the interest I compared my POT for the last 3 months using this stepped system v's a pure level stakes method (with the same starting bet size). At level stakes I would have had a POT of 30% and a profit of 33.8 "units". With my stepped bets the POT was 28% but the profit was 35.5 "units".

[ This Message was edited by: becareful on 2003-10-20 21:55 ]

This discussion is already flawed because it didn't start optimally, then quickly descended into darkness with DarkyDog's effort.

It should have started with the Kelly Criterion, but I'm prepared to bet that none of you understand it. I didn't, until finding this over the weekend.

http://www.jimgeary.com/poker/letters/KELLY.HTM

All of you should read through it a number of times to milk as much wisdom from it as your mathematical capabilities can handle.

Using its definitions the optimal Bet to Bank ratio = Edge/Odds

So if you have a 35% Edge for a 7/4 shot. You should stake 35*4/7 = 20% of your Bank.

Despite the simplicity of the formula, to my amazement it gives the precise optimal figure.

So anyone who talks about "half Kelly" (whatever that is) is clearly deranged. Why would you replace the optimal stake with a sub-optimal one?

Before exploring some case studies found here about abuses of the Criterion, I need to introduce this formula for converting probabilty to odds (in this context)

Odds = (1+Edge)/Prob - 1

Also, I should add that as by definition Kelly needs edge, odds and a proper bank to work, and typically I have none of the 3, so I never use it.


[ This Message was edited by: jfc on 2003-10-21 14:17 ]

With all due respect to every one it sounds like a lot of academic ******** to me.
Cheers.
Darky.

Hi all,
To me risk management is crucial to success at the punt.
Over many years I have tried a variety of approaches with varying success.
I have now lifted the "retirement staking plan" from the Grandstand Publishing website and have been using it with great success for some time.
I aim to achieve 30%+ winners at an average of 3/1 or better and with a seperate bank for each selection process I use all are well in front.
The other problem ,I find is the old case of removing your head and putting on a pumpkin when you are in front , to give an example sat , Delzao biggest and first bet of the day recouped total outlay for the day plus a small profit. Off goes the head on goes the pumpkin an additional 5 bets that all lost luckily 3 of my original bets won and I made a profit.Nonetheless I had thrown away $500 on stupid bets.
I am usually in control but one slip and down $500 !!!!
What is needed is good risk management plus discipline....
Crasher

JFC,

Welcome to the discussion (flawed as it may be)!

In THEORY Kelly is fantastic and, as you point out, gives the optimal staking to increase you bank at the fastest rate. In PRACTICE however there are several major problems with it that limit its usefulness for horse racing and sports gambling.

Firstly the Kelly formula requires you to know the “edge” you have on each bet. As far as horse racing goes this means you must be able to say with certainty that your selection has say a 57% chance of winning the race so that you can work out your edge based on the odds available. Now maybe you are such a good racing analyst that you can come up with specific numbers like that but I am ************ed if I can. The best I can do is come up with an approach that says if I back certain horses within certain odds ranges then in the long run I will end up with say a 25% advantage IN THE LONG RUN. Now whether I have a 25% advantage on a specified runner, or a 50% advantage or only a 5% advantage I really don’t know with any certainty so what value do I plug into Kelly? At one time I was trying to come up with a specific price for a runner and then only bet when it was over that price but do you know what I found – when runners started well over that price (so in theory I had a big advantage) then my strike rate was poor but when they started below that price (so no bet as no advantage at all) the strike rate was very good. Had I been using Kelly I would have gone broke real fast!

The second problem (in a way linked with the first) is that the Kelly formula can give huge bets when your selections are at short odds (so place betting, sports betting, etc). For example lets take a sports bet which you assess as a sure thing but the bookies are offering even money on. Lets say you assess your edge as 50% then your bet is .5/1 = 50% of your bank Do you really want to be risking 50% of your bank on a single bet? Are you that sure of your advantage? Using full Kelly on bets where the odds are frequently in the odds-on range results in your bank jumping around like a hyperactive kid with a sugar rush – you can easily double you bank one day only to see it down to half of the starting bank a couple of days later. (This where the half Kelly and quarter Kelly come in).

So I would be interested to hear how you accurately assess the edge for every single bet and if you have done any past analysis on how good these assessments have been in practice?

BC you are correct in your ways even thought your bet percentage of 5% is quite high for me it is still a safe option.
I have used many different types of stakeing but come back to the old percentage of bank bet...it is the safest bet of all with the ability to get big returns as your bank grows.....no matter how good your selections are and what your percentage of winners are...no one can determine when there wins will come...and this is why other types of stakeing lose with the exception of flat stake bets....even if you have a 60% win strike and have won the last 5 years...if you get a bad loseing run...and it will happen....you are in trouble even if you don't lose the bank you will have a hell of a time getting it back to where you were makeing good money.

In principle Kelly or even Half Kelly is a way of taking advantage of overs, but for mine using overs to determine bet size in percentage of bank is dangerous and increases the chances of going bust.

Whether this thread should have started with the Kelly Criterion or not, it is one of the most dangerous and obviously lucrative but depends entirely on the roll of the dice.

Becareful's percentage of bank (increasing /reducing) is the safest way of minimising risk and taking advantage of good fortune.

If you know the fair price you should be receiving about your bet and can plainly see overlays (your assessment must be correct), then using MODIFIED Kelly is the best option.

I would bet according to the overlay, however, using this to determine the percentage of bank to bet is not, in my opinion, the correct way to do it.

If anyone has a spreadsheet of bets where it shows a level stakes profit over at least 100 bets and the average dividend is more than $2.00, I have a plan I'm willing to share and show how I can increase the POT usually by at least 5%.
I'm not interested in your system, you can leave out the horse names and race numbers if you wish, I just want to demonstrate something for my own benefit as well.
They must be realtime bets though.

send to: racestats at hotmail dot com and I'll return my findings within a day with explanation.

Shaun,

The 5% is quite high because that particular system has a very high strike rate and I am willing to put up with some volatility in the bank. Obviously you need to work out the appropriate percentage for each “system” depending on the strike rate (higher strike rates can have much higher percentages) and how much variance you are willing to have in your betting bank. So a low percentage will give you quite a slow but steady growth (assuming you have a profitable system!) whilst a high percentage will accelerate the growth but you will have more “ups and downs” as you go. In a way it is all linked back to the Max Drawdown figure that Chris mentioned initially and working out how many “steps” backwards you are willing to take in this situation – you can then work out what % will give you that result (in my case the 5% works out at about 2-3 backwards steps in the worst case drawdown scenario I have estimated).

The scary thing is if I used Kelly I would be betting around 20% of my bank on every bet (maybe that would give me a better profit in the end but the other week I would have lost 50% of my accumulated bank when I had 3 losers in a row – a bit too aggressive for me!).

__________________
"Computers can do that????" - Homer Simpson