what to look for to do my first scalp trade ? advice from established traders please.

[Rocket to the FACE]

I cannot accept that the result of a race and staking are dependent on mathematics alone.

The result of a race is not.
But mathematics has everything to do with it and to argue otherwise is ridiculous, otherwise we wouldn't be betting to odds, because that relates to math. We'd be betting in terms of playdough, or farts or ink blots.

[Rocket to the FACE]

On reflection, I realise that I have always "greened" in over 99% of the trades I have made if I thought that my profit was a reasonable one to take.

How have you judged whether it was a reasonable profit to take or not?
Is it more accurate to say you greened up because it gave you a quick profit and took away much of the thinking?

[rphi6876]

True, would be interesting to see who actually knows my guess is pretty much all of them didnt bother commenting.

Well in the gambling world Alan Woods, Bill Benter and Zeljko Ranogajec, all now billionares from mathematical based gambling have commented -they have all publicly admitted to using Kelly Staking. Perhaps, they have better things to do than post on online forums, or perhaps not. In the financial world you have the likes of Warren Buffet and practically every successful hedge fund manager admitting to using it for asset allocation. But I guess these guys have no idea. Like I said its a widely published notion and accepted method for optimal betting stakes, not withcraft that I cooked up and if its good enough for the biggest and most successful betters in the world it really should be good enough for anybody.

[nevermind]

Rocket: Nothing against value betting. It's a strategy and you can win with it. However, identifying value is not that easy. You really need to know your horses (or whatever you bet on). You should know them better than the crowd, and better than all the bookie experts (for pre-race bets). If you can do that and are a long-term winner, you have my greatest respect! :Thumbs I probably could not do it.

Now if I am not a sport expert, and not a number cruncher, I need a different strategy. I need trading (usually greening up). Here, I can invent thousands of different strategies that may work even if the first bet is not quite value. This is heaven for me, this is exciting!!

[Rocket to the FACE]

Now if I am not a sport expert, and not a number cruncher, I need a different strategy. I need trading (usually greening up). Here, I can invent thousands of different strategies that may work even if the first bet is not quite value. This is heaven for me, this is exciting!!

What you mean is you need to guess which way the market will move.

That is all fine and well but there will always be someone telling you which way the market you are looking at will move.
Anyone can look at a peak and guess it may crash and vice versa. But I doubt there are many account holders doing this who are making it pay big time. Because you are always relying on someone else.

[nevermind]

What you mean is you need to guess which way the market will move.

What you mean what I mean is wrong.

Strategies can be a bit more sophisticated than just guessing which way the market will move... A bit of imagination please!

[rphi6876]

Those who regurgitate what they have read elsewhere, and those who know.

Ah Rocket I am one of the few people who is closest to agreeing with you. And as I mentioned before, my PhD was related to this topic so I am not regurgitating what I read, I am regurgiting what I published and what I and the sophisticated gambling, financial and academic fraternity believe. Furthermore, what I have regugitated is scientifically proven and accepted among the aforementioned communities- what everbody else has said is opinion with no facts/derivations or proofs to back it up.

[Temujin]

Ah Rocket I am one of the few people who is closest to agreeing with you. And as I mentioned before, my PhD was related to this topic so I am not regurgitating what I read, I am regurgiting what I published and what I and the sophisticated gambling, financial and academic fraternity believe. Furthermore, what I have regugitated is scientifically proven and accepted among the aforementioned communities- what everbody else has said is opinion with no facts/derivations or proofs to back it up.

No, you haven't proven anything. All you are suggesting is some pedantic mathematical theory.

The rest of us are talking about the real world.

Kelly formula says that with a 2% edge (assuming you know you have it), you bet 4% of your bank.
50 dollar bank, if that bet loses, you are then out of the game. We are dealing with a world where there are minimum bets. For BF, that is 2 pounds.

And seriously when people are trying to back up that it is possible to bet less than 2 pounds, but are we to take these peoples lack of money to argue that the Kelly formula works so much so they can't afford to bet minimum stakes and they are risking their accounts because they can't afford to bet?

This is the real world, where there are minimum bets, and maximum bets. Not sume Excel formula that allows you to stake bets to 100 decimal places. That is not how currency works.

This whole argument came about when the idea of never greening up was suggested as more profitable and it carries no risk, and can be done with a minimal bankroll.

Either prove it, or walk.

If we all wanted to be wankers, we could suggest something like:
If you only ever bet 99.99999999999% of your bankroll on any selection, you can never end with a 0 bankroll.

Now you can do that with an expected return on turnover at level stakes of .00000000000000001%.
The worst possible strike rate, and here is a system that allows you to bet, at any starting level, and NEVER hit a point where you have lost the lot.

Nobel prize, coming my way.

[Temujin]

To put all this in persective.
When I was betting, I had a total of just under 1% advantage, over 18 months of betting, call it 1%. Kelly would have me betting 1% of bank on all bets.
For the purposes of this, assuming 50% strike rate, gives me that 1%.

Now for a period of 4 months, my "system" was running at 75%. 200 bets per week, call that 800 a month, 3200 bets in that time.

Kelly formula in the spreadsheet, says with a 10k bank, after 4 months I would be left with 2.80 in my account, and betting 2.8 cents, hoping to get that back up to the 10,000, let alone the .01% my bank should have increased on average after each bet.

It isn't practical, and there is no amount of "but you didn't go broke" rubbish arguments to show that as profitable, with 0 chance of going bust. Your bank can't possibly grow to the level it started at unless you are prepared to spend the next 50 years going at it.

It is all good having an advantage, but variance will bite you in the ass, if you don't have the bankroll to make it work.

Now to put that further in perspective, I had about 15k bets, and a run of 20% of that running at less than 25%, and had a straight stake profit of around 1%. If I had have used Kelly, and that happened any time in the first 8 months, I may not be here today..

The reality is, start off on the wrong foot, and your holy grail could have easily set my bank back 99.9972%.

But then, I would have still had 2.80 left, so you are right. Wouldn't have gone broke...

[rphi6876]

No, you haven't proven anything. All you are suggesting is some pedantic mathematical theory.

The rest of us are talking about the real world.

.

You are asked for a mathematical proof I gave it to you, Seems you didnt like it as it didnt agree with what you think- so you have dismissed it as pedantic. Probably like the rest of the people here without even going through the derivations. If you read the thread closely you will see I also provide a proof using real world values and with minimum stake sizes that probability of ruin is essentially zero.

My first contribution to this thread was purely highlighting that if you are taking punts on value, then if you know what your edge is you can start with a small bank and build it without any chance of going bust. I then provided you all with the theoretical proof. The argument of minimum bet size was rasied, and yes when that is accounted for the probability is not zero. But I then showed that with a minimum bet size of 5cents and a bank of $100 your probability of ruin is 0.0001 so while not zero it is effectively zero.

So as I previoulsy stated if you blow an account using value bets its because one of two things 1) you have poor risk management or 2) You dont really have an edge. But if some mug wants to blame variance then by all means keep coming back to the market with a freshly topped up account, as that simply means more free money for those with an edge.

The thread then turned to if you are trading to not green. As I suggested, I wouldnt recommend that as you know what the "trading value" is but dont know what the true "fundamental" value is, the horse may be dramatically over or underpriced, you just dont know. The problem if you were to leave an open stake under this scenario using Kelly Stakes is that you will grossly misestimate your expected return, and are succesptable to overbetting and hence can go bust.

And the nobel prize is awarded for prooving that using the Kelly Criterion you will go bust, not coming up with an alternaitve risk management tool that also ensures you want go bust. I have provided a real life proof using a realistic minimum bet size and small balance which shows the probability of ruin is essentially zero. I have stepped up to the plate and provided both theoretical and real world proofs that it is possible to prevent going bust, yet I AM STILL YET TO SEE A PROOF SHOWING OTHERWISE.

So perhaps its is you who should walk away. You appear hypocritical telling me to prove it or walk. I have you havent.

[rphi6876]

To put all this in persective.
When I was betting, I had a total of just under 1% advantage, over 18 months of betting, call it 1%. Kelly would have me betting 1% of bank on all bets.

Well there is your first error by rounding up to 1% from under 1%, you have violated the main assumption of Kelly and guaranteed the probabbilty of going bust by overstaking so all you have done is proved my point. Understand the theory before trying to apply it. This hack attept to disproove has to be taken with a grain of salt as you didnt even get the fundamental assumption right.

[rphi6876]

T call it 1%. Kelly would have me betting 1% of bank on all bets.:

Umm no it wouldnt, the fraction of your bank will primarily depend on the odds you are betting at. For example, if you are offered odds of 1.01 and the probabilty of winning is 100% Kelly would have you stake 100% of your balance as its a sure thing. No Risk. Do you even understand what Kelly Staking is? Go Back to school

[custard]

How have you judged whether it was a reasonable profit to take or not?
Is it more accurate to say you greened up because it gave you a quick profit and took away much of the thinking?

No, it isn't "more accurate". You seem to rely on mathematics and probability to determine your own actions. Horses, football teams, etc. do not live according to a world which revolves around statistics.

If I have read you correctly, once you have determined your own "value" you pursue it regardless of all other factors. I see this as a recipe for disaster.

There are two sides to "greening". Yes, I may take too little value out of a market when I have a winning bet, but I have the willpower to reduce my liabilities by distributing my losses and taking "reds" when necessary.

The most important thing in trading is not "winning". Not "losing" is infinitely more important to most people who have financial responsibilities towards others.

Most people would see my thinking as being sound, I believe.

[WhyAlwaysMe]

To put all this in persective.
When I was betting, I had a total of just under 1% advantage, over 18 months of betting, call it 1%. Kelly would have me betting 1% of bank on all bets.
For the purposes of this, assuming 50% strike rate, gives me that 1%.

Now for a period of 4 months, my "system" was running at 75%. 200 bets per week, call that 800 a month, 3200 bets in that time.

Kelly formula in the spreadsheet, says with a 10k bank, after 4 months I would be left with 2.80 in my account, and betting 2.8 cents, hoping to get that back up to the 10,000, let alone the .01% my bank should have increased on average after each bet.

It isn't practical, and there is no amount of "but you didn't go broke" rubbish arguments to show that as profitable, with 0 chance of going bust. Your bank can't possibly grow to the level it started at unless you are prepared to spend the next 50 years going at it.

It is all good having an advantage, but variance will bite you in the ass, if you don't have the bankroll to make it work.

Now to put that further in perspective, I had about 15k bets, and a run of 20% of that running at less than 25%, and had a straight stake profit of around 1%. If I had have used Kelly, and that happened any time in the first 8 months, I may not be here today..

The reality is, start off on the wrong foot, and your holy grail could have easily set my bank back 99.9972%.

But then, I would have still had 2.80 left, so you are right. Wouldn't have gone broke...

Temujin I have to take issue with your figures. You mention a 50% strike rate which implies you did 3200 51/49 bets. If this is so it would be impossible to run at anything like 75% over a sample size like this.

I have looked up my probability of loss tables for Blackjack (which are adjusted for zero house edge) and over 3000 hands the chances of losing just 200 hands are 0.05%. And Blackjack has a much higher standard deviation than tossing a coin. Your figures are uterly impossible on even money bets. Even more so when we use a 51/49 win rate.

Your 75% win rate tells me you were betting far higher than evens and your average odds would have to be in the 3 to 10 range. It sounds very much to me that you were overbetting relative to your edge and it is this which has caused the problem.

[WhyAlwaysMe]

If I have read you correctly, once you have determined your own "value" you pursue it regardless of all other factors. I see this as a recipe for disaster.

He will have backtested though. He also stated he was prudently discounting his edge by using a fractional version of Kelly. I'm sure he simulated the effects of the staking plan in excel.

If when you are running it and you are running towards a low point on the simulations you have a difficult decision to make. So you can plan to avoid disaster.