Quit my job to do this as a living!

[Kai]

I've lost count of the number of ideas I've had that stood up to early testing but not to more comprehensive testing. But the above idea is something most people would see flaws in and not even attempt.

I wish somebody told me there was more than one testing phase when I was starting out

I actually did attempt something similar to OP in my 4th month of trading, although nothing so extreme, but it's without a doubt the lowest point of my trading career and probably a good example of a bad idea I was testing out a random idea of simply laying the current scoreline at halftime at prices of 4-10 on games that looked lively according to stats, so just needed 1 goal in 2nd half to lock in a profit. Don't remember the exact criteria but there were a bunch of games that qualified, I was honestly just looking for an easy strategy for quick profit The initial testing phase went surprisingly well, that period from the screenshot that I've dug up was probably the best run of results, and then you can guess what happened when I tried scaling up afterwards

When it started going terribly wrong I tried to compensate by expanding the price criteria above 10 to cover games with higher goal expectancy like u21s etc. That was my first mistake It quickly went from bad to worse and after running out of selections I was trying to chase losses by including some terrible cheap games to bail me out, and they did not. Wish I had the screenie with the losses but it must have been too painful to record, although I feel that it was a necessary lesson on how not to trade so I could move forward. Even then I felt it was too expensive to lay at prices around 10, can't imagine what goes through people's heads when they're laying at 200 or 500.

There's bound to be a lesson here somewhere, if something looks too good to be true, it probably is?

[SweetLyrics]

Yep, and while it's true you only need a tiny edge to be profitable, if you end up venturing boldly into the market with large stakes and an 'edge' that amounts to a pocketful of nothing, that can be pretty costly.

“There are old traders and there are bold traders, but there are very few old, bold traders.”
— Ed Seykota

There's bound to be a lesson here somewhere, if something looks too good to be true, it probably is?

[JTEDL]

The OP hasn't posted for 4 months so lets hope there hasn't been many 3-3 draws in that time

Well just a few weeks into the season...3-3 once in Championship and 3 times in League one (twice with Coventry!) one in Serie A also.

and before anyone suggests 'Any other draw' instead - there was a 4-4 in La liga too

[SweetLyrics]

I know that overs and unders systems are popular with traders, but I do wonder how easy it is to get an edge.

For example, you might lay 3-3 when one of the teams is a low scoring team like Man U, and think 'easy money', but I imagine that it won't have escaped the market's attention that Man U don't score many goals these days, and that will be reflected in the 3-3 price.

The OP hasn't posted for 4 months so lets hope there hasn't been many 3-3 draws in that time

Well just a few weeks into the season...3-3 once in Championship and 3 times in League one (twice with Coventry!) one in Serie A also.

and before anyone suggests 'Any other draw' instead - there was a 4-4 in La liga too

[Derek27]

The OP hasn't posted for 4 months so lets hope there hasn't been many 3-3 draws in that time

Well just a few weeks into the season...3-3 once in Championship and 3 times in League one (twice with Coventry!) one in Serie A also.

and before anyone suggests 'Any other draw' instead - there was a 4-4 in La liga too

Perhaps he only does the Premiership.

Just goes to show systems can take 5 minutes to think up and 6 months to realise they're futile. But it should only take a bit of thinking to spot the flaws - especially when they involve laying at big prices.

[Kai]

I know that overs and unders systems are popular with traders, but I do wonder how easy it is to get an edge.

I think the key thing for trading football is learning to accept the variance that comes with it, which is something that people often really struggle with. The overs are popular because they are a unique market and in part because once MO becomes untradeable or doesn't offer any good opportunities one can keep trading the game via overs instead.

Also - and I mean this constructively - if you do think that a particular market is too high or too low on average, then I wouldn't recommend advertising it to hundreds of traders on this forum, as the edge may swiftly disappear (or an edge in the opposite direction may be created).

To be perfectly honest nearly all of the football traders that I know of, myself included, are not really interested in any statistical edges because they are often unreliable and simply too minor to be of any great use for manual trading when compared to other types of edges. I know they are the default starting point for most people when they start looking at football but I think they are suited a lot more for automated trading because it can get extremely difficult to blindly trust in stats when they completely conflict with what you're seeing with your own eyes.

All of the edges and trading styles that are based on the ladder itself are considered minor too (scalping etc), because they usually cannot be scaled up too much, they're mostly based on the idea of trying to extract as many ticks as possible for as little exposure as possible, be it in the form of noise scalping or catching swing price corrections etc. For example with these trading styles it's very difficult to create 4 figure results even on the biggest of markets, but it's maybe the only way to somewhat escape the variance of football if you wanted to (although having a lot of losses may not necessarily be a bad thing in terms of commission), and of course because ladder skills are genuine skills that can be practiced to a high level since they're mostly based off muscle memory and your reading of the order flow itself.

But the football edges that I consider to be the most valuable are those that can be scaled up to the max, a true sports edge where it doesn't matter whether you use a stake of 20 or 20k because the stake size doesn't really impact the market, those edges come from deeper understanding of both football and the prices around it. Contrary to popular belief it is not about predicting results and anticipating goals whilst completely ignoring the prices on offer, it's about reading the game situations themselves whilst working out your own prices for any given situation in comparison to the ones on offer, it's simply about working out the risk/reward of those situations, and if that ratio is very favorable then why not get involved. So the real edge for me is about identifying those favorable situations and capitalizing on them, some may be unique and some may repeat regularly. If the market says there is 25% of something happening while you think it's actually closer to 50% then that is probably worth a trade. You cannot really expect to win on this one but you should win if you repeat such trades 10-20 times.

For me that is what trading really is in a nutshell no matter the sport or market, about actively looking for the upside and then working out how to position yourself to catch it whilst at the same time trying to limit your downside. If a football trader can identify 30 opportunities during the week that have very similar risk/reward ratios and he is convinced that all of them offer some type of +EV in the long term, then even if he was wrong about half of them he should still end the week in profit. That is literally what a lot of people do all day long for years, actively scanning the games for opportunities or simply trying to work out whether a goal is going to be scored or not and whether the price for that is cheap/expensive. If one practices this skill for thousands of hours then one should eventually be able to scale up significantly, although I see a lot of people that already posses such skills but are still trading with smaller amounts for years. People often run into all sorts of mental issues when they start scaling up so they naturally retreat to their comfortable staking zone, but that is a different topic altogether.

[SweetLyrics]

Hi Kai

Interesting, thanks for sharing.

Re: 'If the market says there is 25% of something happening while you think it's actually closer to 50% then that is probably worth a trade.' -

Definitely, but - and I appreciate that you only used those figures to illustrate the principle - the markets are rarely that wide of the mark.

Let's say, for the sake of argument, that you are exploiting maybe a 5% difference between where the market is and where it 'should' be. For most people, it won't be at all obvious that the market is wrong. So without getting a spreadsheet to work out the true odds based on Poisson and all that, how do you know that the market is wrong?

I know you can't give your specific technique, but in general terms is it based on reasoning like, 'The market seems to think United have won this, but my gut tells me they look shaky at the back, so I reckon they are worth a lay'?

Or is it based on something along the lines of, 'Statistically, when Event X happens, the market over-reacts, so this is a good opportunity to capitalise on that'?

[ruthlessimon]

Definitely, but -

Vry nice post; those are exactly my issues too

[Kai]

Let's say, for the sake of argument, that you are exploiting maybe a 5% difference between where the market is and where it 'should' be. For most people, it won't be at all obvious that the market is wrong. So without getting a spreadsheet to work out the true odds based on Poisson and all that, how do you know that the market is wrong?

That's the thing, there is no mathematical model that can calculate the real price at any given moment, not to my knowledge anyway. That's the beauty of it if you ask me, because it's completely subjective and spotting price misalignments (aka value) can only be done through experience like any other skill, which means that almost anyone can do it if they practiced it enough, but barely anyone will even make a genuine attempt to try and learn the skill, for one reason or another. If anything you try to use the fact that the markets use pricing models based off stats etc against them, so the only way that you can know whether the market was wrong is through your results. The really good opportunities are usually pretty obvious and clear cut so even the market knows about them and will try to adjust if possible but is very constricted because of time decay, I've seen the unders being held up for 10+ minutes at a certain price point without any decay on some games because the market was anticipating a goal at any second. Value can be temporary too, it may be there for a 10 minute period and it may be gone after that, so you need to constantly evaluate.

Say there is a 1.80 favorite that started the game very badly and it's painfully obvious to the market that he does not justify the 1.80 price (which is a good price for a drifter because of this weak volatile area from 1.80 to 2.0). But the market cannot instantly correct his price from say 1.80 to 2.20, it has to do it tick by tick and therein possibly lies the first opportunity to lay them to take advantage of their drift over the next X amount of minutes, so even if you don't get rewarded with a goal during this period where they're getting absolutely battered without threatening you may still get 30ish ticks, so try and work out if there's any +EV in this situation, or literally any situation. If you analyze all of the situations that come your way during games you will probably find some that you really like, it's just trial and error and a lot of practice like anything else.

In another example a team can score an early goal and their price will barely move a tick for most of the first half before time decay starts kicking in, so if the other team is dominant during this period because they're looking for the equalizer that very well may turn out a semi-free trade where you don't really lose anything if you get it wrong and there's no equalizer, and if it goes to 2-0 you would lose less ticks than you would gain in case of catching a goal for 1-1 etc, although a lot of traders really like 2-0 scorelines because they can offer very good value on the cheap.

[ruthlessimon]

That's the thing, there is no mathematical model that can calculate the real price at any given moment, not to my knowledge anyway. That's the beauty of it if you ask me, because it's completely subjective and spotting price misalignments (aka value) can only be done through experience like any other skill, which means that almost anyone can do it if they practiced it enough, but barely anyone will even make a genuine attempt to try and learn the skill, for one reason or another. If anything you try to use the fact that the markets use pricing models based off stats etc against them, so the only way that you can know whether the market was wrong is through your results.

But again, this all comes back to auditing/journaling. How would we work out what was working & what wasn't, if we're trying hundreds of 'ideas' during a/several matches?

With data/spreadsheets, I could work it out - but I'm baffled about how to do it without spreadsheets. Trust me, I don't wanna be a data nerd, but the market forces me - cos usually my "gut instincts" are plain wrong

[ruthlessimon]

it's completely subjective and spotting price misalignments (aka value) can only be done through experience

Actually my immediate reaction was this

[SweetLyrics]

I know that some people swear by quantitative methods, but if someone's subjective approach has yielded a positive overall return over thousands of trades, they are probably on safe ground even if they can't specify exactly where the profit comes from or explain what they did.

I suspect the key is to choose a basic, no frills strategy like range breakout or swing trading, and experiment using £2 stakes. You might find it takes off, and you're soon able to up your stakes. Or you might find that you keep losing money, in which case you ask yourself 'What might be doing wrong here? Let's see what happens if I (for example) only enter when the WOM is more firmly in my favour'.

I get the impression that that's along the lines of what Peter did.

You may find that, after a few months of grind, you are good enough to make a small profit, at which point you up your stakes. And you probably won't lose much money in the process - if you are manually trading with £2 stakes, and never going in play, your losses will probably be the cost of a pint of beer or two, if that.

If you want to be a quant, that's fine, but anyone who doesn't have a computing background will probably have to invest quite a lot of hours getting to a point where they can properly backtest and fire in bets using a bot. I've been spending quite a few hours learning Java ahead of a masters course that I'm about to start, and I'm still nowhere near that point.

Subjective trading might feel riskier, but if you want a guarantee, I suggest you buy a toaster! There are no guarantees in this game. I've known people whose methods have worked for years, but there is nothing to say that their edge won't disappear next week.

But again, this all comes back to auditing/journaling. How would we work out what was working & what wasn't, if we're trying hundreds of 'ideas' during a/several matches?

With data/spreadsheets, I could work it out - but I'm baffled about how to do it without spreadsheets. Trust me, I don't wanna be a data nerd, but the market forces me - cos usually my "gut instincts" are plain wrong

[SweetLyrics]

BTW, this is a good thread for beginners interested in manual trading, written by a guy who knows his stuff - viewtopic.php?f=42&t=2021

The advice is nine years old, but still totally relevant in my opinion.

I suspect the key is to choose a basic, no frills strategy like range breakout or swing trading, and experiment using £2 stakes.

[ruthlessimon]

I suspect the key is to choose a basic, no frills strategy like range breakout or swing trading, and experiment using £2 stakes.

I get the impression that that's along the lines of what Peter did.

& how would you audit the 50 variations of said breakout strategy? - which Peter also advocates

I am fascinated by how that analysis is done subjectively

[SweetLyrics]

With respect, I think you are making heavy weather of this.

At the start of the day, you say, for example, 'Today I am going to swing trade. If I see a market that meets x, y and z criteria, I'll go long. I'll close if a, b or c applies'.

It will be boring. Trading isn't supposed to be exciting. If you make a profit, it will be in the pennies. But after a day of doing this, you'll be able to look back and go either say:

- 'Nice one! I made a profit there. I'll keep this up for a few more days and see if I'm still ahead at the end of the week. If I am, I'll consider upping my stakes'.

or

- 'I made a slight loss. Not to worry, maybe if I add criteria d, I'll do better tomorrow'.

Rinse and repeat.

This isn't 3d chess. You aren't trying to find some hidden formula known only to the gods. You're just prodding and probing till you find an edge, and when you do you can think about upping your stakes.

Keep it simple. Cut through the noise.

I suspect the key is to choose a basic, no frills strategy like range breakout or swing trading, and experiment using £2 stakes.

I get the impression that that's along the lines of what Peter did.

& how would you audit the 50 variations of said breakout strategy? - which Peter also advocates

I am fascinated by how that analysis is done subjectively; unless that person is a masochist