Success psychology

[Emtaxx]

Hi all,

Things are going well for me at the moment with automation, and I think it’s going in the right direction.

Only issue is which is really annoying me now, I can’t stop looking at my phone now, obviously some what addicted.

What do you guys do to just relax, does it pass ?

Thx

[Euler]

You have to train yourself to stop looking. If you keep looking there is a tendency to feel like you can influence things, but in reality you can only do that by looking at results at the end of the day or week and tweaking things.

You have to let it do it's thing and try and detach yourself from everything other than the end result.

It is hard.

[Emtaxx]

Yeah I think looking is important in the early development of things and tweak things to be optimised and thats then Productive work.

But I’m aware I’m just feeding my dopamine levels, which is kind of draining lol, I couldn’t sleep the other night for 2 hours due to the next phase of things, invest into bonds, mortgage, get a Porsche etc etc etc Really
Annoying and stupid lol

Well done on the software tho Peter it’s pretty cool.

But on that note, I’m going to put the phone down today and not look until the day is over and think about how I want to frame/treat this… maybe treat it as it’s just a TV show and it’ll be on later tonight for instance ..

[ShaunWhite]

That passes when you get more confident about it. It's not good for you to be thinking about it 247.
But I think it took me a year or more before I could just go on holiday and not check until I got back, even though I could have checked anytime. But I'd have received an email if it was disastrous so although I'm not watching it, the tech is.

As Peter said the problem with over monitoring is the temptation to change it, whereas it's the week or month that counts. When you finally relax it can be quite intersting/disturbing to see that during the day you might have had a stinker but later got one back. Watching it live would have been nervy but unnecessary stress.

The problem generally is the charts show ups and downs that seem trivial but living it in real time is a different story.

[stueytrader]

Success psychology, or psychology during success, is an oft-neglected area I find.

I wrote a post related to this a while ago on here - about having golden runs (as I had at the time), and how it sometimes affected me, possibly negatively.

I view this in a lifetime type of perspective personally - e.g. how do you deal with building a bank that is bigger than you have ever had before?

It is a far better problem than having no bank, but still a challenge to not slide into bad habits, and if you want to continue to build to even better/higher levels.

[Emtaxx]

Yes I agree with this 100%. Its a really interesting corner of the brain to puzzle out. not a bad problem but still a problem lol

[ShaunWhite]

e.g. how do you deal with building a bank that is bigger than you have ever had before?

Once you get start to get consistent results, figure out how much capital you need to operate, the largest likely drawdown (and double it), leave that amount in the bank and withdraw the rest. It's really more about just doing your best each month and drawing the wages rather than building an ever bigger bank.

In my early years "psychology" became a bit of a catch-all term to describe what I felt was basically just an insecurity about ability, because once you achieve a consistent level of performance the 'psychological' baggage evaporates and it becomes just a job like any other. You just need to get to the point where daily gains or losses genuinely have no effect on you, because you 100% know for certain that over a month or a year you'll be in profit. I don't think anyone who's had long term success would say that have to consciously control their mind or need a yoga session before they sit down to work.

[csewell1987]

e.g. how do you deal with building a bank that is bigger than you have ever had before?

Once you get start to get consistent results, figure out how much capital you need to operate, the largest likely drawdown (and double it), leave that amount in the bank and withdraw the rest. It's really more about just doing your best each month and drawing the wages rather than building an ever bigger bank.

In my early years "psychology" became a bit of a catch-all term to describe what I felt was basically just an insecurity about ability, because once you achieve a consistent level of performance the 'psychological' baggage evaporates and it becomes just a job like any other. You just need to get to the point where daily gains or losses genuinely have no effect on you, because you 100% know for certain that over a month or a year you'll be in profit. I don't think anyone who's had long term success would say that have to consciously control their mind or need a yoga session before they sit down to work.

This may be a daft question but how can you know with 100% certainty that you will profit? I was reading the "Trading what i see" thread from goat68. Some of the more recent comments elude to there being no "secret" and there is evidence all over this forum of edges eroding or completely disappearing and resulting in losses. With this said, surely any strategies you are running at any one time could flip against you? Also, the remarks about no "secret" to this but then 100% certainty on profit seem a little contradictory. Sorry, not calling anyone out, just trying to fathom my way to an edge before attempting some automation with it, and finding it difficult to sift through some of the advice on here sometimes.

[ShaunWhite]

100% is probably an overestimate but it's the state of mind that gets you through varience. And yes a strategy can fade or die, but a) once you learn how to find strategies then finding another is something you can be more confident of, and b) you'd usually have a portfolio of strategies, some in dev, some working well and some fading. That's why automation is rarely a case of setting it up and then living the easy life,you're always working on the next thing and when it's your income then you can't have all your eggs in one basket.

But time is the answer really, if you've made steady money for a few years then you gain confidence that it's going to continue.

Practically, there's several metrics that put some numbers on your confidence. Firstly only measure your EV not your actual cash (they always converge eventually). And also look at the t-statistic and p-value of your last couple of months bets. Strategies that suddenly die often aren't profitable anyway and were maybe just randomness, those stats help to establish that.

The strategy type also has a bearing, something based on sentiment (ie the behaviour of favs or traps etc) are less likely to be sustainable than those which are based on the fundamental mechanics of markets, such as EMH which can work equally well across several market types. Markets have differences but also similarities, they're all driven by supply and demand so ideas based on that are more resilient too.

That just scratches the surface as its a massive subject but the bottom line is experience. People think automation is just about waiting until the computer says yes, but it's much more nuanced and aside from the metrics you have to get to the point where you learn what feels right, or just smells wrong. It's that which give you the 99.1% confidence.

[csewell1987]

100% is probably an overestimate but it's the state of mind that gets you through varience. And yes a strategy can fade or die, but a) once you learn how to find strategies then finding another is something you can be more confident of, and b) you'd usually have a portfolio of strategies, some in dev, some working well and some fading. That's why automation is rarely a case of setting it up and then living the easy life,you're always working on the next thing and when it's your income then you can't have all your eggs in one basket.

But time is the answer really, if you've made steady money for a few years then you gain confidence that it's going to continue.

Practically, there's several metrics that put some numbers on your confidence. Firstly only measure your EV not your actual cash (they always converge eventually). And also look at the t-statistic and p-value of your last couple of months bets. Strategies that suddenly die often aren't profitable anyway and were maybe just randomness, those stats help to establish that.

The strategy type also has a bearing, something based on sentiment (ie the behaviour of favs or traps etc) are less likely to be sustainable than those which are based on the fundamental mechanics of markets, such as EMH which can work equally well across several market types. Markets have differences but also similarities, they're all driven by supply and demand so ideas based on that are more resilient too.

That just scratches the surface as its a massive subject but the bottom line is experience. People think automation is just about waiting until the computer says yes, but it's much more nuanced and aside from the metrics you have to get to the point where you learn what feels right, or just smells wrong. It's that which give you the 99.1% confidence.

Cheers Shaun. I completely agree that automation isn't just waiting till computer says yes, based on what I have read on here. In fact, I have been put off writing any code as I think I would just lose money. I am trying to fathom out an edge before heading down this route. At least manually looking at markets I will lose money at a less rapid rate than a bot taking positions here, there and everywhere.

One thing I am struggling with is the concept of actually predicting this in the first place (guess the majority do). The "simple" aim is to beat BSP (plus comm). However, I could price up a horse at 10/1 but the market has it down as a 5/1 shot. I may actually be closer in terms of the horses ability but that doesn't make a single difference if the market doesn't think so. The market is made up of all the varying opinions on each selection so in effect you are trying to predict the masses. I have no idea if I am looking at this the complete wrong way but surely that's impossible? Unless you are a pricewise or someone who has a following in the market.

I appreciate, having said the above though, that people do predict this and are taking money out of the markets. Not sure if I have worded this great but what I am trying to say is when you place a back bet at 6, how could you ever have even 1% confidence that the market will agree that this is over priced and will contract?

[jamesedwards]

One thing I am struggling with is the concept of actually predicting this in the first place (guess the majority do). The "simple" aim is to beat BSP (plus comm). However, I could price up a horse at 10/1 but the market has it down as a 5/1 shot. I may actually be closer in terms of the horses ability but that doesn't make a single difference if the market doesn't think so. The market is made up of all the varying opinions on each selection so in effect you are trying to predict the masses. I have no idea if I am looking at this the complete wrong way but surely that's impossible? Unless you are a pricewise or someone who has a following in the market.

I appreciate, having said the above though, that people do predict this and are taking money out of the markets. Not sure if I have worded this great but what I am trying to say is when you place a back bet at 6, how could you ever have even 1% confidence that the market will agree that this is over priced and will contract?

I may well be wrong, but I doubt many traders find long term success by using form to price up their own markets. It's unlikely that a single trader is able to consistently outprice the data, knowledge, insider info, and skill of the overall market "hive mind" (love that term).

I expect most successful traders use some sort of edge to predict market movements and eke out value that way.

[ShaunWhite]

The "simple" aim is to beat BSP (plus comm). However, I could price up a horse at 10/1 but the market has it down as a 5/1 shot.

BSP is only a benchmark over very large samples, in fact the accuracy of an individual BSP is unknowable to any meaningful accuracy because they're animals not dice. Look towards beating BSP over maybe 1000 bets and don't pay too much attention to individual markets or days.

It's this needs for meaningful sample sizes that makes a strategy that bets more often easier to prove. Automation is pretty hopeless at trying to emulate a manual trading strategy but it excels where humans can't. So consider something that might generate 4 or 5 bets on every selection in every market and it soon racks up the information. Whereas if you have a strategy that has a bet on the fav if X Y and Z are all true will take months before it tells you much.

But the way to get a foothold is historic data, and use half of it look for a strategy and the other half to prove you haven't just backfitted it. That can be quite basic at first, but specialists store the Betfair data stream and can run trading simulations so there's lots of scope depending on how serious you are, or become.

[ShaunWhite]

I may well be wrong, but I doubt many traders find long term success by using form to price up their own markets. It's unlikely that a single trader is able to consistently outprice the data, knowledge, insider info, and skill of the overall market "hive mind" (love that term).

100%. And that’s exactly where the edge can lie, efficiency can be an advantage, not a barrier. Taking prices is just paying the people smart enough to offer them. But you can piggyback off their work and offer at the same prices they do. It won’t show much over a handful of bets, but over volume, small margins can start to emerge.

I expect most successful traders use some sort of edge to predict market movements and eke out value that way.

Again, 100%. 'Movement' is how you access structural EV even if you don’t know the EV of any single bet. It’s just supply and demand playing out, a tug-of-war between everyone convinced they’ve got the price right.

yeah so do as little as possible and leave all the hard (boring?) work to the others.

[ruthlessimon]

I may well be wrong, but I doubt many traders find long term success by using form to price up their own markets. It's unlikely that a single trader is able to consistently outprice the data, knowledge, insider info, and skill of the overall market "hive mind" (love that term).

At the risk of a derailing I do wanna offer a counter here.

Every day at 6pm, 10–20 horses will arb. Then again at 9–10am, 30–40 horses. Assuming the “person” gets their 50 bets matched on the exchange, they will smash the SP - despite offering a “bad price” (i.e. offering 7.8 when it’s 8.0 at Paddy). To me, that suggests there’s some fundamental factor the market can’t price in.

What really baffles me is the timing. Why 6pm and 10am? That feels arbitrary and human. What exactly changes between 7am and 10am — does the going suddenly shift at 10? I’d love to reverse-engineer it one day, just for the challenge. But it feels like a PhD-level project. For the mo, backing 8 for £1 at Paddy is a shite ton easier

[jamesedwards]

I may well be wrong, but I doubt many traders find long term success by using form to price up their own markets. It's unlikely that a single trader is able to consistently outprice the data, knowledge, insider info, and skill of the overall market "hive mind" (love that term).

At the risk of a derailing I do wanna offer a counter here.

Every day at 6pm, 10–20 horses will arb. Then again at 9–10am, 30–40 horses. Assuming the “person” gets their 50 bets matched on the exchange, they will smash the SP - despite offering a “bad price” (i.e. offering 7.8 when it’s 8.0 at Paddy). To me, that suggests there’s some fundamental factor the market can’t price in.

What really baffles me is the timing. Why 6pm and 10am? That feels arbitrary and human. What exactly changes between 7am and 10am — does the going suddenly shift at 10? I’d love to reverse-engineer it one day, just for the challenge. But it feels like a PhD-level project. For the mo, backing 8 for £1 at Paddy is a shite ton easier

How is the arb created? From Exchange price shortening?