Trading What I see !?

[CallumPerry]

You can invert your signals but the points the other traders have mentioned still occur. As you're already aware by now you have commission eating away at profits and the spread to deal with and noise in the markets. Nonetheless, I still believe it's not a bad idea to try because what you will learn is which markets to deploy your strategy in normally and which work with the inverted signals and from there you are starting to learn about matching your strategy to the correct market.

Differentiate between tight races (odds are all very similar), ones with a clear fav, one with two favs, one with three favs etc. Break horse racing into handicaps and non-handicaps, jumps etc. Problem you then face is your sample gets split rather this and you can fall victim to over-fitting the data. It's this constant grey area that you have to battle when automating. I find it sadly thrilling and it's important to know there is NO right answer. Just find your own path and tests, fiddle with the system as much as you need to. Be aware of doing too much though. Keep good records and approach problems from many angles.

[ruthlessimon]

Here's a two trader example @ 0% commission:

Trader 1 is a losing swing trader, & trader 2 thinks he'll be profitable by trading against him (but has to cross the spread)):

Market 1:
Trader 1: Backs £175 @ 2.86, Exits @ 2.70 = +£10.36
Trader 2: Lays £175 @ 2.88, Exits @ 2.68 = -£13.05

Market 2:
Trader 1: Backs £170 @ 2.94, Exits @ 2.68 = +£16.50
Trader 2: Lays £170 @ 2.96, Exits @ 2.66 = -£19.18

Market 3:
Trader 1: Backs £189 @ 2.64, Exits @ 3.10 = -£28.10
Trader 2: Lays £189 @ 2.66, Exits @ 3.05 = +£24.22

Trader 1 result: -£1.04
Trader 2 result: -£8.01

[Tetras]

Just watched Kemp 18:50 and there were some decent moves, but my bot picked none of them!! Interesting watch Elmjor, and it came in quite a bit, but I could see no "indicator" it would do that, so not surprising bot didn't either.
I am thinking maybe entering at random is better, as if you create a bot based on indicators they are lagging and as far as I can see no indicator. I'm guessing with so many 1000s of traders/bots, anything obvious is going to be -ev.

I don't automate, but I don't believe that's true, I think there's still plenty of room in the market for 'obvious' strategies. It just needs to be something that happens 51% of the time and you can either control the downside, or anticipate enough upside to compensate. There are many more of those scenarios than I'm able to take advantage of, but slowly learning.

[goat68]

Here's a two trader example @ 0% commission:

Trader 1 is a losing swing trader, & trader 2 thinks he'll be profitable by trading against him (but has to cross the spread)):

Market 1:
Trader 1: Backs £175 @ 2.86, Exits @ 2.70 = +£10.36
Trader 2: Lays £175 @ 2.88, Exits @ 2.68 = -£13.05

Market 2:
Trader 1: Backs £170 @ 2.94, Exits @ 2.68 = +£16.50
Trader 2: Lays £170 @ 2.96, Exits @ 2.66 = -£19.18

Market 3:
Trader 1: Backs £189 @ 2.64, Exits @ 3.10 = -£28.10
Trader 2: Lays £189 @ 2.66, Exits @ 3.05 = +£24.22

Trader 1 result: -£1.04
Trader 2 result: -£8.01

As i said above when I say invert i dont mean exiting at the same time, that would be stupid. So inverting the entry, but same exit strategy.
So in your example market3, trader 2 Lays @2.66 and market runs all the way to 3.5 for a nice win.

[Trader724]

It seemed to me that you actually want to reverse the whole strategy in the hope that you will be successful, you did not mention that you only want to reverse the entry criteria. If you took into account what I said earlier, you would understand that you only want to reverse if you are strongly negative, more than twice the commission paid. It is obvious that you are looking for a methodology to reach a profitable strategy, but if you do not take into account the valuable advice that others give you, you will not succeed too soon. Today you received some exceptional advice and if you do not consider it then others definitely will, and they are the ones to whom you will give your money soon.

[goat68]

You can invert your signals but the points the other traders have mentioned still occur. As you're already aware by now you have commission eating away at profits and the spread to deal with and noise in the markets. Nonetheless, I still believe it's not a bad idea to try because what you will learn is which markets to deploy your strategy in normally and which work with the inverted signals and from there you are starting to learn about matching your strategy to the correct market.

Differentiate between tight races (odds are all very similar), ones with a clear fav, one with two favs, one with three favs etc. Break horse racing into handicaps and non-handicaps, jumps etc. Problem you then face is your sample gets split rather this and you can fall victim to over-fitting the data. It's this constant grey area that you have to battle when automating. I find it sadly thrilling and it's important to know there is NO right answer. Just find your own path and tests, fiddle with the system as much as you need to. Be aware of doing too much though. Keep good records and approach problems from many angles.

Thanks Callum, your points on differentiation are great, at the moment I only pick volatile type races, so Mdn, Nov, or volatile Hcaps, so the other variations are a great idea. Thanks

[goat68]

It seemed to me that you actually want to reverse the whole strategy in the hope that you will be successful, you did not mention that you only want to reverse the entry criteria. If you took into account what I said earlier, you would understand that you only want to reverse if you are strongly negative, more than twice the commission paid. It is obvious that you are looking for a methodology to reach a profitable strategy, but if you do not take into account the valuable advice that others give you, you will not succeed too soon. Today you received some exceptional advice and if you do not consider it then others definitely will, and they are the ones to whom you will give your money soon.

Sorry I should have explained what I meant by invert. It's actually something Peter mentions in his great oak video, he tweeted again recently, where he mentions reversing for discovery.
I have taken note of Alex's and Callum's great ideas, it's why I ask the questions.
Sorry if I come across as grumpy, just a bit fed up of not seeing positive results...
Today's bot results 100% losses again! However, just watched it play out Sthl 16:05, and it layed NickVedder at 3.15, only to exit for a loss, but the price actually went on up to 3.5. I think I can see an improvement in my exit strategy, as it's getting taken out a lot by noise, hence I am getting a lot of small losses.

[Trader724]

Knowing all the content of this thread from the beginning I saw how much advice was given and you had to be at least on break-even so far so I drew your attention a few times that you do not consider what you are told even though it comes from people who are better in this field because they have worked more or smarter and have more experience, and sometimes we have something to learn even from those we consider inferior. I think one of the biggest enemies of a trader's success is ignorance, because you don't have to ignore even the most seemingly stupid idea you have, but the ones you get here are invaluable because they are the result of a lot of work and wisdom. When you apply what you have learned here it will definitely be seen in your results.

[Kai]

sometimes we have something to learn even from those we consider inferior

Absolutely, I too learned a lot from all sorts of unsuccessful traders, by learning from their mistakes etc.

But some mistakes you do have to make yourself, in order to learn.

[goat68]

sometimes we have something to learn even from those we consider inferior

Absolutely, I too learned a lot from all sorts of unsuccessful traders, by learning from their mistakes etc.

But some mistakes you do have to make yourself, in order to learn.

As a software engineer, i know learning from my own mistakes frequently. Your mind thinks this is definitely the way this thing works, then someone points out a gaping hole and realization suddenly dawns on you that the circle is really a square!

[vhdgkl]

The hardest thing of all is to be focused.

If you are not focused you will lose..everything is built to make you lose..there are few moments when the pressure of money on more horses can be exploited..most of the time it is noise..if you put money on noise, they will turn the price against you immediately.
You can only put money when a horse's market is not manipulable ..find those points there.

Sitting and waiting, even for hours...very difficult.

[goat68]

The hardest thing of all is to be focused.

If you are not focused you will lose..everything is built to make you lose..there are few moments when the pressure of money on more horses can be exploited..most of the time it is noise..if you put money on noise, they will turn the price against you immediately.
You can only put money when a horse's market is not manipulable ..find those points there.

Sitting and waiting, even for hours...very difficult.

Thanks, you sound as though you know the markets very well. I must admit however, you make it sound impossible to me! It's sounds a difficult environment to trade manually as you describe it, and what you would suggest is near impossible to automate...?

[vhdgkl]

The hardest thing of all is to be focused.

If you are not focused you will lose..everything is built to make you lose..there are few moments when the pressure of money on more horses can be exploited..most of the time it is noise..if you put money on noise, they will turn the price against you immediately.
You can only put money when a horse's market is not manipulable ..find those points there.

Sitting and waiting, even for hours...very difficult.

Thanks, you sound as though you know the markets very well. I must admit however, you make it sound impossible to me! It's sounds a difficult environment to trade manually as you describe it, and what you would suggest is near impossible to automate...?

I don't know how to use computers... this is my biggest limitation.
I don't know if bots work...at the current level of development they might not be that great...but I'm not so sure.

I'm pretty sure that the goldmine is excel..excel allows you to do research, which is what good traders do.

[goat68]

The hardest thing of all is to be focused.

If you are not focused you will lose..everything is built to make you lose..there are few moments when the pressure of money on more horses can be exploited..most of the time it is noise..if you put money on noise, they will turn the price against you immediately.
You can only put money when a horse's market is not manipulable ..find those points there.

Sitting and waiting, even for hours...very difficult.

Thanks, you sound as though you know the markets very well. I must admit however, you make it sound impossible to me! It's sounds a difficult environment to trade manually as you describe it, and what you would suggest is near impossible to automate...?

I don't know how to use computers... this is my biggest limitation.
I don't know if bots work...at the current level of development they might not be that great...but I'm not so sure.

I'm pretty sure that the goldmine is excel..excel allows you to do research, which is what good traders do.

Thanks I appreciate your honest comments. I've been looking through my bot's logs for this weekend as I output quite a lot of log information, and i've made some alterations, which I feel confident about. So we shall see how it does this week.
Cheers
Goat

[goat68]

This weeks bot results:
Markets: 52
Strike Rate: 30% (oh!)
Net: -£4.06