Trading What I see !?

[michaelhud1]

Scaling:
Forget doubling or x10, let it grow slowely using prior profits. Even nudging it up by 10% every couple of weeks would be reasonable. Most promising strategies will end up being ananomalies that are going to revert and the last thing you want is to have small stakes when it's working and x10 just as it's about to fail.

There's no rush, if it's going to make money it will and you won't make anymore from it long term by being greedy and ramping up stakes too fast.

Thank you

[napshnap]

The DK effect is why I've got box full of random screws and scrap electrical components. I've got a mutilmeter and a soldering iron, how hard can it be to fix an amp/printer/TV. Well it's a piece of cake until about a minute after you take the back off it. And then when you've failed you strip it down for parts as if one day they might be useful. Ditto folders full of spreadsheets.

Cool Shaun, I fix my stuff too, nothing too serious, but my work mouse celebrated its 15y anniversary and not long ago I changed its clickers for new Japanese sexy ones!
The way solder lies on the wires is mesmerizing, it's like alchemy)) and those white burns are super painful! And this feeling after you fixed something is very satisfying, I' don't know... like after sex))
Sorry for offtopic, Goat.

[Kai]

Risk..
Also ive been analysing liquidity, and have some logic in the bot...
However could be in the valley of the stupid !!

Even so, not much wrong with a bit of stupidity Mr Goat, since none of us can really avoid it! As long as it's temporary!

When it comes to trading I actually feel like EQ is far more useful than IQ, just like in real life overall.

Obviously it doesn't hurt to have a bit of both, but arguably too much IQ can be damaging at times and lead to overthinking the simple things etc. As boring as it sounds I feel finding the right balance is key overall, like it always bloody is!

[goat68]

Risk..
Also ive been analysing liquidity, and have some logic in the bot...
However could be in the valley of the stupid !!

Even so, not much wrong with a bit of stupidity Mr Goat, since none of us can really avoid it! As long as it's temporary!

When it comes to trading I actually feel like EQ is far more useful than IQ, just like in real life overall.

Obviously it doesn't hurt to have a bit of both, but arguably too much IQ can be damaging at times and lead to overthinking the simple things etc. As boring as it sounds I feel finding the right balance is key overall, like it always bloody is!

Yeah you're spot on there Kai. I knew some mega boffins at Uni, who were astonishing in their field, but totally lacked the wider aspects of life!

[ShaunWhite]

The DK effect is why I've got box full of random screws and scrap electrical components. I've got a mutilmeter and a soldering iron, how hard can it be to fix an amp/printer/TV. Well it's a piece of cake until about a minute after you take the back off it. And then when you've failed you strip it down for parts as if one day they might be useful. Ditto folders full of spreadsheets.

Cool Shaun, I fix my stuff too, nothing too serious, but my work mouse celebrated its 15y anniversary and not long ago I changed its clickers for new Japanese sexy ones!
The way solder lies on the wires is mesmerizing, it's like alchemy)) and those white burns are super painful! And this feeling after you fixed something is very satisfying, I' don't know... like after sex))
Sorry for offtopic, Goat.

that's a well maintained mouse. If more people did that then the oceans might not be quite so full of crap . Biggest result I've had was fixing my TV. Most of the time it's a power supply problem and lucky PSUs are made of parts you can see without an electron microscope. It's the capacitors that give up (the ltttle cans). Replace any with a bulging top for about 10p each and cross your fingers. My logic is that if it doesn't work I'm not going to make it worse so why not try. Crackly volume knobs are a nice easy fix too.

Brief off topic things are OK I think provided we don't start figting

[goat68]

So just to prove my point slightly, i've had another interesting result today, "live" -£5, but backtest same data shows +£14 Hmm ?!?
Initial check shows "live" partially matched one large win, whereas backtest assumed full match, so very useful info for me...solution hmmm not sure!

[goat68]

Hmm some market dynamics I don't understand why only part of my Lay bet got matched. This is what happened I think:
Price had sharply spiked up from 6.2 to 7ish when bet triggered, Current Best Back 7.2: Lay 7.4:
£32 Lay bet placed at best reverse price 7.2
Market comes down, Current Best Back 7.0: Lay 7.2 => Only £9 gets matched I think
Market comes down further, Current Best Back 6.8: Lay 7.0
Market then rises back up Back 7, 7.2, 7.4
=> Fill&Kill then expires with only the £9 matched, can't understand why rest of £32 wasn't matched given price went lower than Lay entry?

[goat68]

So from examining my log, what I think happened, which sounds very odd and I didn't know was possible, is maybe something like this:
1. A big Lay order sent price out from 6.0 to 7.4, thus taking all Back orders out in that range
2. My bot triggers placing a Lay bet at 7.2, as a result of the Lay spike there is not many Backers available to take it, only £9 matched.
3. Price falls back to 6.8/7.2(gap), then 6.8/7.0 and Back/Lay volume is taken from their, but because my Lay order is at 7.2 it is not taken, even though new Layers are placing Lay orders at 7.0... I didn't know that was possible, doesn't seem fair!?
4. Price then "gaps" back up from 6.8/7.0 to 7.2/7.4, so a Lay order takes all the backers at 7.0 and 7.2, again leaving none for my Lay order at 7.2
5. Price continues up without my remaining Lay matching...

So basically you can be unlucky and not get your bet matched if the market goes back and then reverse sharply back over your bet without taking it... oh well!

[LinusP]

So from examining my log, what I think happened, which sounds very odd and I didn't know was possible, is maybe something like this:
1. A big Lay order sent price out from 6.0 to 7.4, thus taking all Back orders out in that range
2. My bot triggers placing a Lay bet at 7.2, as a result of the Lay spike there is not many Backers available to take it, only £9 matched.
3. Price falls back to 6.8/7.2(gap), then 6.8/7.0 and Back/Lay volume is taken from their, but because my Lay order is at 7.2 it is not taken, even though new Layers are placing Lay orders at 7.0... I didn't know that was possible, doesn't seem fair!?
4. Price then "gaps" back up from 6.8/7.0 to 7.2/7.4, so a Lay order takes all the backers at 7.0 and 7.2, again leaving none for my Lay order at 7.2
5. Price continues up without my remaining Lay matching...

So basically you can be unlucky and not get your bet matched if the market goes back and then reverse sharply back over your bet without taking it... oh well!

Not possible, would only assume the fill kill activated just before the price drop to 6.8/7.0.

[ShaunWhite]

It's all in the timing, unless you're logging your bets and market data to the millisecond you won't see what's really happened. What's guaranteed though is it's never 'not fair' the matching engine never lies and if your bet looks like it was skipped, it's because it wasn't there. Using 1s data is like watching 1second of match every 50seconds, the ball has moved and you didnt' see why.

The important thing to realise about your strategy is that you're reacting to a situation rather than pre-empting it, so speed will be a big factor. I mentioned the importance of bet latency when testing before but I don't think you were ready to understand the implications, or believe it. 50ms and you're rich as Croesus and at 400ms you're losing.

Everyday's a school day.

[Anbell]

We're all rooting for you GOAT, but I've had this tab open for a month waiting for an appropriate time to use it, and I really want to close the tab - so here you go (the file is too large to embed apparently)

https://pbs.twimg.com/media/EzPKnhwVgAU ... me=900x900

[wearthefoxhat]

Hmm some market dynamics I don't understand why only part of my Lay bet got matched. This is what happened I think:
Price had sharply spiked up from 6.2 to 7ish when bet triggered, Current Best Back 7.2: Lay 7.4:
£32 Lay bet placed at best reverse price 7.2
Market comes down, Current Best Back 7.0: Lay 7.2 => Only £9 gets matched I think
Market comes down further, Current Best Back 6.8: Lay 7.0
Market then rises back up Back 7, 7.2, 7.4
=> Fill&Kill then expires with only the £9 matched, can't understand why rest of £32 wasn't matched given price went lower than Lay entry?

I think SEGA had it covered back in the day.....

sega.png

[goat68]

We're all rooting for you GOAT, but I've had this tab open for a month waiting for an appropriate time to use it, and I really want to close the tab - so here you go (the file is too large to embed apparently)

https://pbs.twimg.com/media/EzPKnhwVgAU ... me=900x900

Ah brilliant, really like that, cheers!

[goat68]

So from examining my log, what I think happened, which sounds very odd and I didn't know was possible, is maybe something like this:
1. A big Lay order sent price out from 6.0 to 7.4, thus taking all Back orders out in that range
2. My bot triggers placing a Lay bet at 7.2, as a result of the Lay spike there is not many Backers available to take it, only £9 matched.
3. Price falls back to 6.8/7.2(gap), then 6.8/7.0 and Back/Lay volume is taken from their, but because my Lay order is at 7.2 it is not taken, even though new Layers are placing Lay orders at 7.0... I didn't know that was possible, doesn't seem fair!?
4. Price then "gaps" back up from 6.8/7.0 to 7.2/7.4, so a Lay order takes all the backers at 7.0 and 7.2, again leaving none for my Lay order at 7.2
5. Price continues up without my remaining Lay matching...

So basically you can be unlucky and not get your bet matched if the market goes back and then reverse sharply back over your bet without taking it... oh well!

Not possible, would only assume the fill kill activated just before the price drop to 6.8/7.0.

So no, that's what I thought, however I have a number of "facts" that would question it, as follows... Shaun, I realise my 1second data capture won't capture the whole story, but the 1second data are "facts" to this particular "unique" situation. So here goes the following are some "facts", lets see if we can find a "reason" please.
When looking at this keep in "mind" the following things particular to this issue:
- "Large Lay order at 13:11:06"
- following low "liquidity" after large Lay order
- "gap" in market due to low liquidity at 13:11:08
- Another Lay order jumping market from Current Lay 7.0 to 7.4 at 13:11:22

Guardian Automation log:
28/04/2021 13:11:06: [G_Auto 1] : £ 32.26 Lay bet placed on Bollin Neil at 7.2. Entirely unmatched when it initially reached the market. Ref: 231588143062 ( Fill or kill bet with 55 seconds delay. )
28/04/2021 13:12:02: [G_Auto 1] : £ 9.85 of unmatched bet (Ref:231588143062) was filled before remainder was killed.

Betfair Account Betting activity:
13:15 Wolverhampton
Bollin Neil - Win Betfair Bet ID 1:231588143062 |
Placed: 28-Apr-21 13:11:06 Lay 7.20 9.85

Some entries from my Guardian 1second data:
28/04/2021 13:11:05,Wolverhampton 28th Apr - 13:15 2m Hcap,G_AUTO1:
f3_back_price {S},6
f3_lay_price {S},6.2
28/04/2021 13:11:06,Wolverhampton 28th Apr - 13:15 2m Hcap,G_AUTO1:
f3_back_price {S},7.2
f3_lay_price {S},7.4
28/04/2021 13:11:07,Wolverhampton 28th Apr - 13:15 2m Hcap,G_AUTO1:
f3_back_price {S},7
f3_lay_price {S},7.2
28/04/2021 13:11:08,Wolverhampton 28th Apr - 13:15 2m Hcap,G_AUTO1:
f3_back_price {S},6.8
f3_lay_price {S},7.2
...same 6.8/7.0 until...
28/04/2021 13:11:22,Wolverhampton 28th Apr - 13:15 2m Hcap,G_AUTO1:
f3_back_price {S},6.8
f3_lay_price {S},7
28/04/2021 13:11:23,Wolverhampton 28th Apr - 13:15 2m Hcap,G_AUTO1:
f3_back_price {S},7.2
f3_lay_price {S},7.4

I can also get various "queued" and "traded" volumes for these levels.

You can also see the chart price action on my chart here, the "Green" line, Lay spike is obvious, i've annotated Lay levels 7.0/7.2/7.4:

market_Wolverhampton 28th Apr - 13_15 2m Hcap.jpg

Appreciate your thoughts please?
thanks

[goat68]

My thought is even when the price was down at 7.0 up until 13:11:22, no "Backers" wanted to back at 7.0 or 7.2, they ran scared from the large Lay order, so the price just pushed out at 13:11:23 to 7.4 due to "evaporation" of Backers, hence my remaining £23 Lay order at 7.2 was not taken, until which point at 13:12:02 it expired and was Cancelled.