Quit my job to do this as a living!

[ruthlessimon]

I didn't find your explanation of why the system was profitable over the course of the selection to be at all compelling.

I'm literally lost for words. No-one is ever gonna be more concise!! I was just reading back what I wrote; I've literally explained it lock, stock & barrel - what I wrote there, is what I've got written on a A5 card which I use everyday!! When I need to get involved, the variables I need to watch, & to top it off a fundamental explanation alongside the technical one.

Yeah it could fail, but why worry yet

[SweetLyrics]

I'm not going to get into the clarity of your explanations here.

However, I would like to make a point about the equity curve (even though I've said I was leaving the conversation, as this is a point that I think beginners would do well to be aware of).

Have a look at the equity curves below.

You might think that one of these charts shows a really good system, and the other shows a losing system.

However, they both represent backing at odds of 1.76, where the true odds are 1.76 (I have a spreadsheet that simulates randomness).

Beware being fooled by randomness.

2.JPG

1.JPG

[ruthlessimon]

I'm not going to get into the clarity of your explanations here.

Appreciated; we'll save it for the after party - assuming you don't tell me to get lost!

ever, they both represent backing at odds of 1.76, where the true odds are 1.76 (I have a spreadsheet that simulates randomness).

Beware being fooled by randomness.

My instant reaction to that is that it's a really unfair comparison. if you're gonna test a strategy @ a specific price, makes sense the prices around it also conform - if they don't then I'd agree, random

[SweetLyrics]

I don't understand what this means? Conform to what?

Here's what my spreadsheet did.

1.76 equates to a win chance of 0.5682.

My spreadsheet generated a series of random numbers between 1 and 10,000.

If the number was 5682 or less, the horse wins, and you make 76p profit.

If the number is > 5682, the horse loses, and you lose a pound.

I'm assuming zero commission to keep things simple.

My instant reaction to that is that it's a really unfair comparison. if you're gonna test a strategy @ a specific price, makes sense the prices around it also conform - if they don't then I'd agree, random

[SweetLyrics]

PS My final (final!) thought on the matter.

Get some historical data from https://www.betfairpromo.com/betfairsp/prices/ and compare the BSPs of the selections against your prices (there is a spreadsheet posted somewhere in the forum that allows you to download them en masse).

If you would have made a profit had you closed at BSP, then you might be onto something.

[ruthlessimon]

I don't understand what this means? Conform to what?

I see what you've done, & i guess yeah, it proves the dangers of a low sample size. But I just have a completely different outlook on the market. If something passes an out of sample, the equity curve looks 'nice' - that's enough for me. A discretionary trader, to me, is basically saying - "there's so many variables I can't define it" - but if they could define it, it'd be hundreds of low frequency strategies, imo. An example being the lay system

[SweetLyrics]

Found it! viewtopic.php?t=14395

There is an Excel formula you can use that will speed this up, but I'll let you work it out yourself, as I'm not a big fan of spoon feeding! Oh, and you also need to work out how to combine the spreadsheets into one large spreadsheet - again, I don't wish to deprive you of the satisfaction of working that out!

PS My final (final!) thought on the matter.

Get some historical data from https://www.betfairpromo.com/betfairsp/prices/ and compare the BSPs of the selections against your prices (there is a spreadsheet posted somewhere in the forum that allows you to download them en masse).

If you would have made a profit had you closed at BSP, then you might be onto something.

[Kai]

the equity curve looks 'nice' - that's enough for me.

Simon you strike me as the type of guy that gets more excited about equity curves than other types of curves

[ruthlessimon]

the equity curve looks 'nice' - that's enough for me.

Simon you strike me as the type of guy that gets more excited about equity curves than other types of curves

Unfortunately... yah that's kinda right. Actually no it's completely right

[eightbo]

...
(I have a spreadsheet that simulates randomness)
...

not sure how it affects what you've done but useful to know:

When simulating a coin flip you need to simulate randomness, which Excel's not great at:
    - Excel 2007/2003 is worse than newer versions for this
    - Excel 2010 saw improvements to the RAND() function by switching out the '07 algorithm but it's still not truly random
    - VBA's rnd() function uses a different algorithm and is known to use the same seed number each time meaning you'll produce the same numbers & cycle through them

[eightbo]

the equity curve looks 'nice' - that's enough for me.

Simon you strike me as the type of guy that gets more excited about equity curves than other types of curves

[SweetLyrics]

If something passes an out of sample, the equity curve looks 'nice' - that's enough for me.

So are you really ramping up your stakes then? What stakes are you trading with, out of interest?

If you do take up my suggestion of comparing the prices you enter with against BSP, please let us know how you get on. Given that BSP is ultra-efficient, that will give a very good indication of whether you are entering at value prices.

A discretionary trader, to me, is basically saying - "there's so many variables I can't define it" - but if they could define it, it'd be hundreds of low frequency strategies, imo. An example being the lay system

Maybe it's just me, but I'm not sure this makes any sense!

[ruthlessimon]

So are you really ramping up your stakes then? What stakes are you trading with, out of interest?

Mine vary from 10p a tick to £5 a tick. But to be professional pre-off swinger, min stakes need to be £5 a tick (at least) if we’re trading a very rigid strategy.

Here’s the irony, I’m taking your advice to the letter (god knows why tbh, because most decent traders advocate max stakes the moment something is discovered). But that by the by, the majority of my stuff is very low stakes (cos it’s very new, & the techniques used to spot them are very new). I just disagree with how those trade ideas are generated (data > videos). & I really don’t understand why you’re disagreeing, considering your taking an MSc in Computing. Why are you doing that if you think data’s useless - & videos are the answer?

If you do take up my suggestion of comparing the prices you enter with against BSP, please let us know how you get on. Given that BSP is ultra-efficient, that will give a very good indication of whether you are entering at value prices.

*sigh*

Let’s assume we discover an edge that works exiting @ bsp – how are you gonna execute it? Only 1 person in the world I know who can do that, & it’s bloody messy; bloody complicated!

Read that email again, I’ve detailed the exit – & it’s very close to Bsp, but easily achievable (entry & exit, @ market ).

If I could exit @ Bsp, my profitability is boosted by like 5%/10% - although I’d need to reanalyze that – but I’m pretty sure it’ll give me a boost.

A discretionary trader, to me, is basically saying - "there's so many variables I can't define it" - but if they could define it, it'd be hundreds of low frequency strategies, imo. An example being the lay system

Maybe it's just me, but I'm not sure this makes any sense!

Because I bet you Peter/Dallas etc trade a highly correlated strategy to the one I sent you – even though they label themselves as discretionary. I’ll put £1000 on it – they are aware of the bias, & they work around that bias - will you do the same & commit £1000 - payment for the system & I'll be the 1st to sell a profitable system

[SweetLyrics]

Who you remind me of with the stroppy attitude - https://www.youtube.com/watch?v=dLuEY6jN6gY

Look mate, you do whatever you want to, and I wish you well. Genuinely.

Maybe you'll get somewhere with the conspiracy stuff. Who knows? However, I reckon you'll be no further forward in a year's time. That's not me being negative - that's actually meant as encouragement, and I'd dearly love to be wrong!

I think that you are too set in your ways and closed minded to move forward (having been that guy, I recognise the signs!). But please, please prove me wrong.

But this conversation is going nowhere, so in the nicest possible way, I'm out, as they say on Dragons' Den.

Bye for now and all the best.

[ruthlessimon]

But this conversation is going nowhere, so in the nicest possible way, I'm out, as they say on Dragons' Den.

Bye for now and all the best.

Put your money where your mouth is.



Massive value backing Yes imho, should be like 1.20 . Hell it's such good value I'll let you off for nothing

Go on, free shot @ £1000 - but the caveat is, Peter has got to explain the fault. If he can explain the fault, & that the idea is random, I will pay you £1000 towards your tuition