Forums

General Betting

There is currently 1 person viewing this thread.

Post your reply

Text Format: Table: Smilies:
Forum does not support HTML
Insert Photo
Cancel
Page 2 of 3  •  Previous | 1 | 2 | 3 | Next
sort by:
Show
per page
Replies: 103
By:
lewisham ranger
When: 13 Jul 16 12:01
I heard that most big football punters for example, they won't bet on betfair as 5 percent commission is just too high.

they'll punt on the asian exchanges where the commission rate is much lower and can take any size of bet.
By:
zooot
When: 13 Jul 16 14:21
Maybe they should scrap the current PC charge, reduce the commission level to 2.5% and have no discounts on commission. Then when people hit a high earning per year threshold they go from 2.5% to 3.5%.

That would make it easier for or less painfull for average punters.  The tilted table effect of low commission towards big syndicates would be gone and the premium charge would be a manageable drain on big winners, not a brick wall.

A major point of difference for BF of better odds is totally undermined by the high commission structure.

In the early days BF was like many new things where people think they can make money. We're gonna be rich was in the air and enough people seemed to be going well to justify the hype.  So commission seemed like a small detail back then.  It is now, in a much tougher and more sober marketplace, a huge issue - either known to punters who analyse their data or understood by most others more in a sense of "something is not right" by people who seem to be betting well but have a steadily declining bank.

BF should be reducing commission over time given that the market is so much more accurate that it can't support 5% anymore.  The moves to 6, 7 and even 8% in some markets just boggles the mind.  8% would strip a bank so fast that any newbie would find the BF experience quite unpleasant.

Given that an exchange attracts those who care about factors being supportive of a fair go and fair chance of making money an exchange must appear to support this concept or else they will not come, or leave quickly and defintely won't tell their friends to try it out.  The last time I suggested to someone to try betfair was about 8 years ago!  Word of mouth not working well out their.

Anyway, as someone once said  "Never in human history have so many lboured for so long for so little"  Just make sure it is still fun and doesn't cost you too much if you stick around.
By:
Thin and Crispy
When: 13 Jul 16 14:57
I gave up trying to make money just backing many years ago.  Now I just scrape a meagre income from trading horses .I've had enough and if I had an alternative I'd be off.
By:
romfordiron
When: 13 Jul 16 14:59
It is a real shame, but there are so many issues with the exchange in terms balancing revenue with long term viability, that I can't see the decline stopping or even decelerating.  I've made a good living off the exchange for 5 or 6 years, but now fully accept that I don't have a future trading full time on here. It's a shame, but such is life.

I'm actually quite looking forward to doing something else now. There's only a certain number of times you can read and input numbers on a screen before it starts to become tedious!
By:
Thin and Crispy
When: 13 Jul 16 15:17
Perversely my problem is I'm not losing money.  Because I can still make a profit it holds out the hope that if I keep at it It'll come good and so I carry on with what feels like death by 1000 cuts.
By:
starship
When: 13 Jul 16 19:05
I heard that most big football punters for example, they won't bet on betfair as 5 percent commission is just too high.

their very little value on the football
whereas like me i bet greyhounds,
and when i dog is say 4/1 and i take 6/1 on here and sometimes a 6/1 is 12/1,
i find 4% on my winnings to be good value.
piont is u make a selection 3/1 and you get 4/1+ on here, the commission will not deteremind whether you win or lose,
being a judge will decide that
By:
Barton Bank
When: 14 Jul 16 00:09
As someone who has paid a lot more in commission than I have in Premium Charge, I can't say I would relish having it "reduced" from 2 percent to 2.5 percent as a matter of course and then increased to 3.5 percent if I win whatever the high earning threshold above is deemed to be. Leave the bottom rate of commission at 2 percent and drop the maximum rate from 5 to 4 or maybe 3.5 percent and you might get an increase in liquidity.
By:
screaming from beneaththewaves
When: 14 Jul 16 00:23
I agree that commission is an issue in attracting punters, but it has been there since the start, and was there during the boom times.

The point about the way commission works on here is that it takes months, if not years, of wins and losses before it becomes evident that nearly everyone, even most winning punters, are paying way more in commission than they make in net profit, thus turning them into losing punters. It took that long for the effects of the commission structure to work through and become the factor which ended the boom times.

Say you're a heavy, regular punter who plays in 5000 markets over a period of a year and gets your commission rate down to 2%. In 2,500 of those markets you make a pre-commission profit of £1,000. In the other 2,500 markets you make a pre-commission loss of £990. Thus overall you show a pre-commission profit of £25,000 over the year. What is your actual net profit after commission of 2%? The answer is nothing. In fact, after commission you will have LOST £25,000 over the year. Even at 2% the Betfair system charges you £50,000 of commission on a £25,000 profit.

But the point is that the winning and losing markets aren't distributed evenly. So it takes a whole year to realize exactly what has been happening. But when you do realize ...Shocked

Yet Premium Charge payers still reckon that THEY are the ones hard done by because they're asked to pay back just a fraction of their POST-commission profit.
By:
romfordiron
When: 14 Jul 16 08:44
Screaming, I fully understand your point. It brings us back to the original post on this thread - the difficulty of beating commission when outright betting. As I previously said, it's down to the bettor to weigh up the attractiveness of betting on here and paying commission, or struggling to get on with bookmakers at usually slightly worse odds. Without a doubt, other exchanges and bookmakers are more attractive a lot of the time.

However, an exchange also needs the PC payers, who are often full time or at least part time. Without people seeding markets and then providing liquidity, there is no exchange. I fully concur that without bettors wanting to punt, the exchange will also cease to exist.

To suggest that PC isn't an issue is to underestimate the importance of liquidity in keeping an exchange healthy.
By:
starship
When: 14 Jul 16 09:14
i know at least 3 people who have have hit the premium charge.
the way i look at it,
i put x amount in, i take a wage every week and i am above my x amount  i put in.
so the amount commission i pay is neither here or there, or am i missing the piont?
By:
zooot
When: 14 Jul 16 10:17
Starship

By being good and form analysis and pricing up individual races you are getting on average way over the true odds.  Backing at 6 when the true odds are 4 is huge value and commission is a non issue.  You get what could be called specific identified value.  You would have a high proportion of bets way over the odds and much fewer poor value bets.

In contrast, my big data set approach has identified what could,be called generalised value with large numbers of bets that are value and lots that are poor value.  Combined they create an edge but a skinny one compared to you.  Also, on the day I have tended to take what odds are there that meet my criteria without any ability to screen out poor prices on a race by race basis.  My criteria say bet, so I do without a judgement on specific value. Big margins make commission a minor issue.

You pick the eyes out of the race card.  I do too but with a much more even mix of good and poor value.  Skinny margins make commission a killer.

Now lots of people either fall into:
- something like my approach of generalised value but through many different paths and so commission is a huge issue
0r
- a version of your approach but without your skill or specialisation so they lose or break even and in either case commission is again a big drain on them. 

I have tried the form and pricing up races a few times for short periods with ratings systems and judgements but the results seemed to show that I was missing something significant.  I remember pricing up dogs and seeing odds movement and results that suggested I was way off.  In know what you do but admit I don't understand how to do it.  There seems to be an element of mystery and alchemy to form analysis.  I've read that if you watch enough races it starts to become something you can see, you just know the hard to measure stuff.

I probably took the wrong path but did so cause I enjoy digging in the data for patterns.  I tried watching lots of greyhound races but didn't really like doing so or seem to get any inspired ability to make better bets from it.

I think you under-estimate how rare your skill is among the general BF population.  Or perhaps we did not focus enough?
By:
starship
When: 14 Jul 16 10:54
I WAS at the stow picking up videos when i noticed a bookie picking up the videos.
suprised me as i did not think he got the vids.
as he walked away i said, i did not know you got the vids, he replied i do, as he walked further away,
i said, do u look at them.
because we got the same vids, look at the same race and yet he barred me from betting with him.
so we can all look at the same thing, it is how we inturpted it.
phil ivy [ great poker player ] came out with a classic line,
we all play the same game, well almost.
By:
zooot
When: 14 Jul 16 11:18
That is why form experts can maintain their edges over time. 

In contrast, some things I have found that are in the data and measureable just take one big punter or syndicate to stumble on / find and they will soon be gone.  However, the attraction was the aim to find the viens of gold and then mine them with automated bet placement while I lived life away from the computer.

Perhaps I should drive the family crazy watching endless racing videos?  Or perhaps not as over the last few years a few relatively small stock market picks have made me way more than BF over the same time period with 1/1000 the time commitment.
By:
zoom_top
When: 14 Jul 16 15:04
"However, the attraction was the aim to find the viens of gold and then mine them with automated bet placement while I lived life away from the computer."

That's the dream, Zooot, that's the dream. When you find out how this is done please private message me, I would love to get my hands on that one.
But joking aside, the one critical element I have found on here (betting BF) is connected to what Starship has said. Not only do you need to be very selective in your choice of wager, but find betting propositions that are priced strongly in your favour. That is a human skill not something that a computer can necessarily determine. Bots may be able to "skim" but they still need to combat the takeout. All the you tube videos I have seen talk about very small profits on a race using a number of techniques. Well that's fine if you want to work for a very small hourly rate. I also agree strongly with Starship's point of specialisation. Your betfair a/c needs to be 3 steps forward and one step back. 2 steps forward and one step back will see too much erosion due to takeout. (I'm probably repeating what has already been said !) I've also learnt that patience is the most important quality a punter can possess. And secondly the ability to have absolutely no emotional attachment to the endeavour.
By:
fixed
When: 14 Jul 16 16:38
also next time one is transformed from loser into winner by reading "study, look for edge, be patient" tips in a forum would be the first time
By:
Ericali
When: 15 Jul 16 21:33
I'm currently making money on a horse racing system, but have only been doing it for 47 days so far.....
By:
askari1
When: 28 Jul 16 22:54
I've never been able to formalise my knowledge in a commission-beating system.

I know when a horse is likely to come down the hill well at Goodwood well enough to make money (there was one today). But this is dependent on the analysis of specific horses; the draw; the trainer's historical record in races at the track, meeting and race; even the likely pace bias and whether a horse will get cover.

The stage I'm at is that I'd like to have 100-150 bets a year and not punt otherwise (I have a part-time job and have mostly laid according to my own tissues for 15 years on here on the big horse days). The slog of market-making without automation isn't worth it, and isn't where my ability lies; I have a slight form edge.
By:
stu
When: 30 Jul 16 12:18
There are load of edges, but not necessarily ones that could be written as a set of strict rules, as Askari suggests above.
By:
stu
When: 31 Jul 16 08:44
I sometimes think the idea of 'automation' is to blame for the idea that markets are too efficient - they probably are too efficient if all you want to do is write a program and sit back without using any effort on your own part. Pretty mad to imagine it would be that easy really.
By:
Mr Magoo
When: 02 Aug 16 17:27
zooot: Going back to your original post, how are you calculating your matched odds? Are you taking the starting prices, or looking at the wide range of odds matched on here?

The wonderful feature of horse racing is that the odds move so much, and while some of it will be due to new information (going changes, or perhaps astute judges watching a horse as it parades), most of it (IMO) is surely speculation. If a horse's price has changed by say 20%, there must have been value available at one end of the prices or the other.

So if you have a strategy that seems to have no edge when using BSP, it could well be profitable if you managed to catch earlier bets that were at better odds. Of course, it works both ways - if you're not betting right at the off then you can end up with much worse odds than the SP. There's nothing worse than watching a horse's odds plummet from 10/1 down to evens when you've laid it at double figures, and then it goes on to win easily. Even if it loses, you still end up kicking yourself for not laying the lower odds :(

The other 'gotcha' when looking at the wild odds movements is that you need a decent estimation of how much you'll actually get matched at early prices. Naturally this also can work against you, and you may finding yourself struggling to fill high-odds bets on fancied horses, while your bets on the unloved runners get taken in full.

But still, the fact that the odds vary so much surely indicates that there was value for the taking, and the pots of gold are still out there to be won... at least, I hope so!
By:
stu
When: 02 Aug 16 20:17
Adapt, update, analyse and respond - all constantly required IMO in order to play with an 'edge', something related to the above post too.
By:
Gallivanter
When: 02 Aug 16 23:44
Stu, you say "There are load of edges, but not necessarily ones that could be written as a set of strict rules."

Have you considered looking at these edges in relation to each other? I don't mean totting up points for each runner and betting the best. I'm thinking more of examining the differences between all the variables for every runner and calculating a percentage chance of winning for each one. That gives you a fairly good tissue to work with.
By:
stu
When: 03 Aug 16 12:31
Interesting comment gallivanter - not sure I entirely understand your idea of looking at differences without assigning these as 'points', but I guess you mean generally make it more numerical in nature.

My only problem with that is that I see that different factors affect each market (my personal view), so the possible different permutations of factors would be rather huge to do this in such a direct way, or have a simpler structured system. As my points above were meant to imply, I think finding edges is a constantly changing process rather than stationary.
By:
kincsem
When: 03 Aug 16 13:25
I don't often look at the General Betting forum but this thread looks very interesting.  I'll read it in detail later to pick up fresh ideas (fresh to me).

Everyone seems to be intrigued by the one example the OP threw out - horse names.  I'm sure he has tested dozens of other angles with his large data set.
When he said some ideas only produced 15 bets a week I saw a different world.  I have less than 15 proper bets a year (and a few entertainment bets).

My interest is thoroughbred pedigrees.  My database has 389,370 horses this morning.  I also have other databases that I compare to the pedigree database.
I see thousands of horse names every day.  I even analysed all names in my database and extracted a count of each word used in horse names.
The most common were Miss; Lady; La; The; Royal; My; Blue; Golden; Red; Sweet; Princess; Little; Silver; Queen; Star; High; Bold ...
I believe well named horses are better horses, probably because the person naming the horse is better educated, wealthy, well read, and can afford better stock.
My preference is for a one word name about seven or eight letters, or a name that rhymes, or makes me smile: Prominer; Hittite Glory; Darn That Alarm.
However, I do not use names to select bets.

My idea is to analyse pedigree in volume (using programs I write) to identify good pedigrees and poor pedigrees.
I have some results that I will expand when my data collection/tidying is complete (very nearly).
Over the winter I will automate the ideas I have been collecting.
My preference is high odds in Group races (flat), 10f and over.
If I can find one negative in a favourite I become interested in the outsiders.

I am with the OP in thinking you must work in areas other people do not, and oppose beliefs others hold.
By:
fixed
When: 03 Aug 16 17:16
and by laying it all out here your subconsciousness indicates that all this shielded-from-reality static-database-from-past juggling may be good entertainment for some but will not lead to anything meaningful in a dynamic live now/future environment.....


your subconsciousness is right about that
By:
zooot
When: 04 Aug 16 23:16
Magoo,

I've tried many approaches to odds selection over the years when doing both stats based approaches and odds movement based approaches - SP, early prices, some based on movements of odds.  The simple answer to what I found is that the market has the overwhelming tendency to reflect value overall.

Bet very early, for example, and if backing you will tend on average to back at lower prices as the layers start low and slowly move their prices up as the market develops.  So of course you should lay early?  But then on average the value bets will tend not to get matched or the only way to get consistently matched is to take prices at the top of gaps which overall equal poorer value. If you take odds available early as they are then to ensure an instant match you will tend to consistently get matched at poor value.  If you place a bet at a more value price and wait for it to get matched, you will tend to get matched by steamers (when laying) or drifters(when backing) and in both cases you tend to get matched at poor value. 

What you want (on average) is to get matched early on steamers when backing before they steam much or early on layers when before they drift much.  If you do that you will get rich.  But unless you have a factor that tells you ahead of time what is going to happen you are left to work wit the average tendencies just described which do not favour you.

That is where true form analysis can create powerful edges as you have an extra dimension to work with and spot value in the context of that race, that time of placement and that horse.

Also, my approach was dependent to a large extent on data based around the SP price.  That was the reference point.  Value often exists at SP - just with the challenges discussed earlier.  So if I tried a placement approach at odds away from SP I had lost my point of reference and any judgement on vale at that point was lost.

I've seen incredible stuff in analysing my historical bets where if only I backed the ones that moved in say 4% from the price I took I would have made a killing.  But there was no way without a time machine to identify those that would do this.  It can get super precise.  Made up example here - but it may be that 5 minutes before the off prices below 10 and above 5 that then steam in 3-4% offer good profits.  But if they steam in over 8% they don't.  But at the steamed in price the value is gone.  At the point when the value existed I had no way to tell if they would be in the 3-4% category (I tried to look for signals but none where reliable).

So in short, you must assume as a guiding principle that overall the market will tend to match bets, move, shift and morph in ways that tend to minimise your opportunities to spot and take value. They rarely (overall) do so to your advantage. Value will be apparent all over the place in hindsight though.

A generalised value approach such as I pursued for reasons described is at the mercy of this market tendency to a fair extent dampening the value level to below commission. 

Inside knowledge or being very good at form analysis can give people that extra dimension to pick the value before the moves occur.

Gallivanter,

I always overlay market edges (filters) on top of one another.  This improves results a lot but I found the number of bets dropped dramatically.  I have some things left I could pursue which are highly filtered and in theory would make money.  But doing all the stuff to place say 6 bets a week seems questionable ROI.  If I knew those edges would sustain for the next 2 years and could handle scaling up to higher stakes it would be okay but with 6 bets a week and normal losing streaks thrown in, there would be months and months of losses at times even in winning systems.  Uncertainty and doubt settles in strongly when you plug away for months going backwards. The "fun" completely goes out of it and it becomes a negative in your life.  What is a thousand pounds won after 8 months of dispiriting grind?  Is that worth it?

High volume offers more potential but also shortens in time the losing streaks.

That is a question for the exchange - high commission may make more money in the short term but in terms of an "amusement ride" the site ends up with players who don't enjoy the ride and don't tell anyone else that it is fun.  That is a big limiting factor for BF.
By:
stu
When: 05 Aug 16 10:23
zoot - I think there's another important factor here, which is size of stake.

If you're a big player vs a smaller stake player, you may have less opportunities. The guy looking for smaller, but good, returns will do better these days (IMO).
By:
stu
When: 05 Aug 16 10:23
By good, I mean regular and consistent profits (just not very high for each market on average)
By:
zooot
When: 05 Aug 16 12:03
Out of interest, given how form analysis and specialisation seems the answer, I set up a rating system for greyhounds including form data and observed form from watching races to run through the process of compiling data and my own unique observed factors.

After a week I could see how it could work and I got a sense of where it could create a long term edge but, core blimey, the time it took to watch all the races, update data and then on the night try to roughly price up 10 races on two nights of the week was massive.

Hours and hours to update data and also to price up and then hours to place bets at the best prices on the night. Like taking on a second job after hours with no light at the end of the tunnel in terms of the workload declining much over time.

So one day when I retire I might look at that again if I am lost for things to do but till then - no thanks.

Makes it all pretty easy to walk away from.
By:
henryluca
When: 12 Aug 16 22:06
So I leave after quite a few years, in significant profit still but with an average hourly earning rate over the years that would make a cleaning lady cry. I also have much improved Xcel skills that have helped me a lot at work.  Although I genuinely enjoyed the analysis I did, and trying all sorts of stuff, so am not bitter, ultimately it had to pay off in terms of profits as I am not a gambler and want a decent ROI for my time.

The point of this post?  Well mainly to spell out just how far you have to go to make money on BF long term and to highlight how what often seems like a system with simple analysis and short data runs  is probably not or will not last. 

So remember you have to go way beyond clever analysis of readily available data into detailed form analysis or highly sophisticated automated trading to make good money.  Or you have to get good at trading and spend a lot of tedious hours doing so.


More accurately this should read
'spell out just how far I have to go  etc...'




data on 22000 horses, spreadsheet with 130,000 rows - lots of data.

There will be many profitable patterns....you just 'surprisingly' couldn't find one

22000 horses ....
ConfusedConfused
By:
boom322
When: 21 Mar 19 19:47
Excellent Article...
By:
Dr Crippen
When: 23 Mar 19 18:27
Very useful information in the opening post for anyone considering that approach.

I've never taken the view that there's a magic formula or a system that will work long term.
If there were, these days with computers and some very clever people around (such as here on this thread in fact )with the ability to exploit the data, then almost every winner would be at shorter odds in races where form is exposed.
By:
roadrunner46
When: 09 Apr 19 11:40
so the computer geeks have got zero chance with their edge based data systemsLaugh
By:
ZEALOT
When: 11 May 19 14:41
There are edges in horseracing that will always exist . There is no substituting VALUE . I take it you are talking of BSP where edges are eroded .

Take for example are horse that has been matched at a low of 3.2 and a high of 4.8 and goes off at a BSP of 3.9 . It doesnt matter if the horse wins or loses or falls or pulls up . The aim if to beat the price . If you layed at 3.2 or backed at 4.6 you have obtained VALUE and do this over a vast number of bets and you win . There are folk on here who play the horses who would find it impossible to lose in the long run .
By:
lewisham ranger
When: 15 May 19 00:13
why do you need systems? if you are intelligent, and just apply common sense, you can approach every race on it's individual merits...
By:
DFCIRONMAN
When: 15 May 19 11:37
Systems can enable CONSISTENCY in arriving in a selection.....whereas, the brain can operate at different levels from day to day. You do NEED to be reasonably "intelligent" to devise such a system that takes account of the multitude of relevant FACTORS that apply to each race chosen....Common sense can vary again from day today.
By:
DFCIRONMAN
When: 15 May 19 11:41
Taking on every race is a good learning experience.....but certain race types provide better VALUE , where your "edges" are stronger. Filtering the race types is important.
By:
lewisham ranger
When: 15 May 19 15:48
so what you are essentially saying is systems are the only way to maintain discipline.

I really don't believe in systems. Sure, I believe in accumulating knowledge and applying that knowledge to races. but systems? I don't believe they exist.
By:
lewisham ranger
When: 15 May 19 15:52
Filtering the race types is important.

this is the flaw in systems I'm getting at. you have to back every runner that meets your criteria, even if logic dictates you shouldn't back it.
By:
DFCIRONMAN
When: 15 May 19 22:02
Not true LR.

There is "subjective " input as well, and this , together with all the other FACTORS can provide a strong system of ratings.

If ratings are made on a CONSISTENT basis, the holy grail lies before yoooooooooo......if you can identify how to use the system!
Page 2 of 3  •  Previous | 1 | 2 | 3 | Next
sort by:
Show
per page

Post your reply

Text Format: Table: Smilies:
Forum does not support HTML
Insert Photo
Cancel
‹ back to topics
www.betfair.com