Botty Challenges

[ShaunWhite]

you'll start to trust your statistics more than what you're seeing in the moment.

But should these two breaks of 2.0 (could be a break of any price) (at exactly the same time) be treated the same? Do they have the same probability of holding under 2.0 to post time? Does one have potential to move further? etc etc

Oddly enough I've never actually asked, but are you planning to use all this probability in automation? If it's intended to aid cold trading then imo there's other things apart from the bare stats that are far more important on a case by case basis.

....are you finding this helpful with your daily session of cold trading?

[ShaunWhite]

you'll start to trust your statistics more than what you're seeing in the moment.

But should these two breaks of 2.0 (could be a break of any price) (at exactly the same time) be treated the same? Do they have the same probability of holding under 2.0 to post time? Does one have potential to move further? etc etc

I take my hat off to anyone who could answer those two questions from simply "watching the market", that guy/girl is a gifted trader imo

How does the horse look?
How much more back money is being traded than lay money?
What's the 2nd and 3rd doing?
What time is it?
What type of race is it? Hcap or non-hcap.
Is there any money chasing the move?
Is it Monday at Windsor or Saturday at Ascot?
How did the previous race trade? Late money or a steady flow?

These are what I'm looking at when I 'watch the market'

[ruthlessimon]

....are you finding this helpful with your daily session of cold trading?

Cold. The reason I'm into data, is because I struggle to understand certain discretionary concepts, that people appear to intuitively understand.

For example.

Let's assume I've backed @ 2.82; & the trade goes bad. The price has drifted to 3.0. Do I cut, hold or reverse? Somone could know that instinctively, but personally, I don't (I didn't). Without a spreadsheet, I wouldn't have known which of those 3 options is right call (longterm)

[ruthlessimon]

These are what I'm looking at when I 'watch the market'

I haven't got a clue combining all that as a quant or via discretion! At best I could do a couple, but not the lot!!

I thought you summed up the issue pretty well here:

If you collected the information yourself you will notice you often get days with a bias of strong backing or laying in the market, once one of these becomes obvious to you then you can use it in remaining races

Why does this happen? Without an underlying reason it's hard to believe it's any more than normal variance. To suggest that people who see it should follow it in later races is straying into the irrational realm of conditional probability.

If the propeller heads here think it's a 'thing' then it probably is, but it would be nice to have an idea of what supporting evidence to look for apart from there being six reds or six blacks in a row.

(quite happy for this to be bumped into it's own topic, I suspect it's not an easy answer)

it's often all over the place for no valid reason.

....except on the few days it repeatedly goes where you predict because of the activity earlier in the day.

It's tough being a mere mortal. I think I need a day in Hook at the end of the summer.

There are patterns there. I think you are looking at it too logically, the market isn't particularly logical a lot of the time.

Thanks for taking the time to reply. Too logical, not logical enough....jeez it's a very narrow tightrope

[ShaunWhite]

that people appear to intuitively understand.

Intuition...aka experience.

Let's assume I've backed @ 2.82; & the trade goes bad. The price has drifted to 3.0. Do I cut, hold or reverse?

The simplistic answer is that regardless of your current pl you have to trade what's happening the moment. Ignore the loss as is....would you be opening a new trade now that time has moved on and it's at 3?...and in which direction? If you'd back then stay in, if you'd lay then get out, if you wouldn't trade then get out.

ie just do what you did at 2.82....or do you get attached to your position? If so then you're not immune to the psycology afterall.

Or, When you backed at 2.82, it might have been worth planning to get out at 2.9 depending on how noisy it is (which usually ends up going to 3 anyway) and if you were right afterall then you could get back on when it comes back at 2.82 or 2.9 or even 2.94.

Overtrading is a potential issue but riding losses hoping for a reversal doesn't make sense, especially when you can get off the bus while it goes to the depot, and get back on again when it comes back. There speaks the experience of someone who's used a fair few night busses drunk.

[ShaunWhite]

I thought you summed up the issue pretty well here:

If you collected the information yourself you will notice you often get days with a bias of strong backing or laying in the market,

Why does this happen? Without an underlying reason it's hard to believe it's any more than normal variance.

Jeez Si that was a year ago....I've wised up since.

You don't need to know why or even if it is going to persist.
1.If it is then great.
2.If it's just regular day afterall, then the std ups and downs make it a wash.
3.If it's actually the opposite kind of day that's a problem, but after 1 or 2 you'd lose confidence (score?) on the fact it was a 'backing or laying day''.

I was overthinking it rather than looking at outcomes. Nothing is certain, so in the above just try and make 3 < 1 with appropriate money management (stopping the losses where teh reverse is happening) and you're on the right side.

"you often get days with a bias of strong backing or laying in the market" is the part I accept is fact, or he wouldn't have said it!
Knowing why isn't top of my todo list and can only ever be speculation, but crowd psychology is a funny thing.

[ruthlessimon]

ie just do what you did at 2.82....or do you get attached to your position? If so then you're not immune to the psycology afterall.

Not if the entry was a unique event that occurs only once per market (i.e. your pattern occured at a set time/volume range)

Or, When you backed at 2.82, it might have been worth planning to get out at 2.9

Well yeah, that's why I like spreadsheets. Imagine the pain of testing whether getting out at 2.9 differs from 3.0 manually! Unless you're a sadist for manual data capture (which I was btw ).

Overtrading is a potential issue but riding losses hoping for a reversal doesn't make sense

You've just opened a can of worms with that statement Neither does cutting for a guaranteed loss, when there's potential for a retracement. I agree that taking lots of heat on the majority of entries; means the entry is indeed pretty shite. But improving that entry, could, completely eradicate the original edge.

[ShaunWhite]

Neither does cutting for a guaranteed loss, when there's potential for a retracement.

It's just a loss on that interim stage not the whole market.

Take the small loss like a man and make it back when/if the retracment from an even higher level happens.

On at 2.8, out at 2.5 .... i don't need to be on the journey from 2.9 to 3 and back just to stubburnly prove my original idea was correct or not. Besides I can't possibly predict what the world will look like when it's at 3... that's fortune telling.

[ruthlessimon]

Take the small loss like a man and make it back when/if the retracment from an even higher level happens.

This this is why I post

I hadn't actually thought of that!

[ruthlessimon]

"you often get days with a bias of strong backing or laying in the market" is the part I accept is fact, or he wouldn't have said it!
Knowing why isn't top of my todo list and can only ever be speculation, but crowd psychology is a funny thing.

Same, the "why" doesn't really bother me, but knowing how to improve (insert anything tradable) does bother me. I see the "how" as a hedge against a change. I wanna be in this game for a long time

[ShaunWhite]

I wanna be in this game for a long time

The hardest way to make an easy living.

"Three years to make this work, or look a joke and be broke"

It's the only tune I know that mentions spread betting

"In spread betting it's easy to draw a small fortune, start with a big fortune and lose a small fortune"

[mcgoo]

Bloody hell-just watched a pace runner go from 2.80 to 12 pre-off My bot was on the wrong end of a late move but only losing 44c (predominantly backing)on that express train is a win I reckon.
Loving these stored values ..add the signal(s) combination and you can microwave yer brain . Horse botty looking even more beautiful -let's hope that translates to

[jimibt]

Bloody hell-just watched a pace runner go from 2.80 to 12 pre-off My bot was on the wrong end of a late move but only losing 44c (predominantly backing)on that express train is a win I reckon.
Loving these stored values ..add the signal(s) combination and you can microwave yer brain . Horse botty looking even more beautiful -let's hope that translates to

i guess what you're saying is that by watching the races from a few hours out, you can look at a range (say 2.00 to 3.50 odds) and monitor if they escalate out to more than double their initial values. from there, you can look to place a medium sized back bet to take it into the pre race 5-10 mins in the expectation that it will steam back down, even if only by a few odds...

[mcgoo]

Bloody hell-just watched a pace runner go from 2.80 to 12 pre-off My bot was on the wrong end of a late move but only losing 44c (predominantly backing)on that express train is a win I reckon.
Loving these stored values ..add the signal(s) combination and you can microwave yer brain . Horse botty looking even more beautiful -let's hope that translates to

i guess what you're saying is that by watching the races from a few hours out, you can look at a range (say 2.00 to 3.50 odds) and monitor if they escalate out to more than double their initial values. from there, you can look to place a medium sized back bet to take it into the pre race 5-10 mins in the expectation that it will steam back down, even if only by a few odds...

Normally would agree-but should have added timescale to my description..this was within roughly 2 mins.Horse must have shot itself in the hoof

[jimibt]

Bloody hell-just watched a pace runner go from 2.80 to 12 pre-off My bot was on the wrong end of a late move but only losing 44c (predominantly backing)on that express train is a win I reckon.
Loving these stored values ..add the signal(s) combination and you can microwave yer brain . Horse botty looking even more beautiful -let's hope that translates to

i guess what you're saying is that by watching the races from a few hours out, you can look at a range (say 2.00 to 3.50 odds) and monitor if they escalate out to more than double their initial values. from there, you can look to place a medium sized back bet to take it into the pre race 5-10 mins in the expectation that it will steam back down, even if only by a few odds...

Normally would agree-but should have added timescale to my description..this was within roughly 2 mins.Horse must have shot itself in the hoof

sounds like it threw the jockey or got itself into a sweat... either way, good opportunity perhaps, even if it were one taken with a view to taking it IP, given that the odds jump was irrational and the calculation being that it may even settle down to intial values... singal and SV's to test i guess