How to increase stake sizes, without losing all your profit.

What are you betting on Squirrel?

Football

always even money, what is it? odd or evens on goals?

yeah just risk a percentage of your bank and go up as your profits rise

I personally think 5% is a bit too high.....what you should be considering is what it my potential longest losing run at the odds.....

Looks like u have had a fair run so far, but at some point you are going to get stuck with as many losers as winners and you will feel like treading water.....best of luck though

how did you come up with the system? do you have a good reason to believe it has an edge other than its results so far?

To be honest, I am unsure if I have an edge or not hence the trepidation about increasing stakes too much.

All I do is look at a game and see if there will be a late goal or no goal.

Example, stoke v Spurs last year both teams down to 10, Spurs needing a win. I backed Over 2.5 when it hit even money, because I was confident that there would be one more goal in that game. Equally in the Newcastle Arsenal game I backed Under 1.5 as I thought Arsenal would see it through and keep it tight. Reviewing that back it was a winning bet, but I think I got that one wrong and it was a bad bet.

Does this mean I have an edge? I have no idea really, but currently the results show I am performing quite well.

Will it continue, I am doubtful, but would like to make as much as possible from it whilst I am performing well.

When I look at it, I feel that surely the markets should react to this, but after analysing them over the years they almost seem to move in a similar fashion. So maybe I do have an edge?

That is why I want to use a staking plan that grows as quickly as possible, without running the risk of me busting out and losing all my winnings. That is why I thought of the 5% of bank plan.

Do you have an edge? Personally I think no, these goal markets in popular games are almost impossible to beat. and to prove it, I happen to do a season long paper exercise not so long back ( I dont actually bet on goals anyway ), and the actually winners were....Betfair Commission.

167 games is nowhere near enough a sample anyway to decide.....but so far you have found the value without actually looking for it, by your own admission you are sticking a bet on at a single price point. As I said earlier you will eventually start treading water , unless you really do hit a longer term lucky streak. At least you have realised if you do up the stakes too high, then indeed u could be wiped out. Best of luck though however you go forward....

hmmm. personally, I would stick with 2.5-3% in that case. as stated by others, your sample size is far to small to draw any conclusions from, so you definitely want to proceed with caution. you're much more in a testing phase than a maximising returns phase.

essentially, if you were to translate what you're doing into a formalised system, it would seem to be something like:

*look for situations where an outcome is 2.0
*make a judgment as to whether it's more likely to happen or not happen based on specific match situation and your read on what you've seen so far
*bet accordingly
*be quite selective as to when you bet.

i've seen way worse, and you seem to have grasped the idea that a winner can be a bad bet, so you have a decent shout, imo. good luck and let us know how it pans out.

if possible look at paast results for at least 500 events, or take the results you have and find your longest losing runs
see http://www.probabilitytheory.info/content/item/11-winning-and-losing-runs

then whatever the losing run is, double it and divide your bank by that number.

so a losing run of 6 = 12 divide a bank of £100 by 12 = £8.33 = 8.33 % 

Then after every loser increase your stake by the percentage you have worked out. When your bank is back in profit then make your stake the bank divided by the number again.

Maffmatikaly thats wot you does. Let the argument begin.

Joshua tree, do you feel those markets are unbeatable then? If so, is it impossible then to beat the "price" on nearly every market?

I am not sure if I have an edge as I have already stated, but feel the price on these markets is determined by factors that seem more pre determined than match specific.

I can tell by the teams and match situation what time a certain price point will be available, quite accurately. This make me think that maybe these markets follow a set pattern as to what price they will be at a certain time, rather than taking into account each match on its individual merits. But hence the original post, i am not certain of this.

Lex, I like you theory, but adapt this to have more of a safety net.

Squirrel = I ve yet to see anyone come up with a rationale for how many goals in any match. The other thing to bear in mind is the odds in the big games are a result of weight of money. One thing I would highlight is that sometimes you can find better price with the bookies in running - I ve lost count of the number of times I ve seen a better price on  B 3 6 5 website.....

goals and match markets are very tightly modelled, but that doesn't mean those models are accurate in every instance.

especially in play, where you have the advantage of a lot of very specific info that they probably don't factor in.

Indeed .....amd in fact I m at the stage now where I m actually finding more success with a " what do I fancy to win " approach - which is actually what I used to do pre betfair !!!

SquirrelJ, I think you are onto a winner with looking at the formulated movement of the market. If you have an edge or not will depend on your ability to read the game.

Does anyone know where i could get hold of a modelling chart, book etc that shows in play prices?

Jul 21, 2013 -- 9:49AM, SquirrelJ wrote:

To be honest, I am unsure if I have an edge or not hence the trepidation about increasing stakes too much.All I do is look at a game and see if there will be a late goal or no goal. Example, stoke v Spurs last year both teams down to 10, Spurs needing a win. I backed Over 2.5 when it hit even money, because I was confident that there would be one more goal in that game. Equally in the Newcastle Arsenal game I backed Under 1.5 as I thought Arsenal would see it through and keep it tight. Reviewing that back it was a winning bet, but I think I got that one wrong and it was a bad bet. Does this mean I have an edge? I have no idea really, but currently the results show I am performing quite well.  Will it continue, I am doubtful, but would like to make as much as possible from it whilst I am performing well.When I look at it, I feel that surely the markets should react to this, but after analysing them over the years they almost seem to move in a similar fashion. So maybe I do have an edge?That is why I want to use a staking plan that grows as quickly as possible, without running the risk of me busting out and losing all my winnings. That is why I thought of the 5% of bank plan.

To nit pick.  What you have isn't really a "system".  Sure you do the same thing, but your bets are derived by an opinion you have at the time.

As stated, these markets are quite accurate.  But that is a statement based on averages.  It is no more accurate than the horse markets which all average to 100%. 
Opinion is what makes you money, if you are right more often than not when betting at those odds.

The point of this:
You very well could have an edge with the bets you have placed.

But, can you guarantee that given the same scenarios, would you have placed exactly the same bets each time.  I mean in those 167 bets, what guarantee do you have that the 167 bets you placed, would happen every time.  Given that it is a thought process based on current game scenarios, I feel you cannot say those bets are a lock. 
Hence, it not being a system.

As such, it is impossible to say if you have a long term winning "system" as your state of mind is never going to be exactly the same when making decisions.

Other things to note are runs of outs as suggested.  I have personally seen 21 losses in a row of even money bets.  I think 23-24 is what we could expect to see in a lifetime, depending of course on frequency of betting.
But then your point is that if you have value, you are not betting on even money shots, but betting closer to 1.80-1.90 shots, just getting paid 2.00 for them.

167 bets is way too low to actually know anything.  I have personally run a winning system for 18 months betting around 230 bets a week.  It certainly didn't have the sort of edge you are talking about, but was profitable after commission all the same.  I had a period in there of over 1500 bets, where it was returning less than 75%.  1500 bets running over 30% under what is needed to profit, yet the system was still profitable.  167 could simply mean you are on a good run. 
Which of course you have pointed out and questioned.

The other thing to take into account, is that a losing sequence really only paints some of the picture.  The chance of losing 23 bets in a row is one thing, but that is easier to overcome than losing 700 bets out of 1000.

Another thing to consider, is that if you simply bet a % of your bank, you are ALWAYS betting more money on your losing bets.  This costs you money.
Say for example you are betting 10% of your bank.  You place a bet of 1,000 at 2.00 that wins.  Your bank goes up to 11,000 and your next bet is 1100 which then loses.
You have made 2 50/50 bets, paying 2.00, and you are 100 bucks down.  Before commission is even taken into account.

What I would suggest, is to have bank targets, and bet a % of your bank and commit to it between these limits.
Say your bank is 10,000, and you choose maybe 2% of your bank.  Your targets maybe every 500 dollars profit or something.
So you bet 200 until your bank hits 10,500.  If you start with a losing run, you are still betting 200 a time.  Your stake should only change when you hit your next target.
When on 10,500 (or whatever your bank and target levels are), your stake would then change to 210, until your bank hit 11,000.

I personally wouldn't be betting much over 2% of your bank.  3% is most certainly the limit I would think.

Here is a table (not very detailed, but given enough info to be handy) showing the chances of losing sequences.
http://www.free-bet.co.uk/betting-article-display.asp?id=22


If your current strike rate is golden and continues forever then the table shows:

58    6    8    9    13

Where 58 is your strike rate, 58 winners per 100.
Where 6 is a losing sequence more than 75% likely to happen.
Where 8 is a losing sequence more than 50% likely to happen.
Where 9 is a losing sequence around 25% likely to happen.
Where 13 is a losing sequence around 1% likely to happen.


The more bets you make, the more you will find this to be shining through in your results.

Given 13 outs is showing around 1% of the time which is pretty high (for a system you wish to run presumably forever), you straight away know that 7.5% of your bank as your staking, would wipe you out on average 1 time ever 100 series of bets, where a series of bets is from a bet to a winning bet.  The goal is obviously not to wipe your bank out, so I would more than half the 7.5% to try and last as long as possible.

All based on the assumption that the results are as they should be based on your opinion, and that your opinion will continue the same.  Of course you could be on a bad run now, and your strike rate should be much higher, or a good run, and the opposite, or you could get better, or you could get worse.

Really can't hope for any better answer than that, as the reality is, the only way you can possibly question bank size and such, is by your opinion of what your edge is.

Squirrel.....BJT has just hit the nail smack bang firm in the middle of the head, thats probably the best post on this thread.

There was a run of something like 19 overs which lost last year for one team, I think it was either Swansea or Swindon, so the proof really does exist in reality.

Good luck with your future efforts though, keep us posted how u get on.....

Imo if you have a true objective formulated system then it will earn at a consistent rate over a certain number of wagers.
To establish what this rate and number of wagers are for any system, you can analyse past results.
Once you know this expected targeted rate of return, then you can set your upward bet size adjustments accordingly.
Say you establish that generally over a minimum of 200 bets, as an example, your system always shows reasonably consistent returns, then you can reasonably safely upwardly adjust your bets every 200 games/races.
Simplistic perhaps, but it works for me.

What an excellent post from BJT. Worth anyone's time to read it carefully.

And if anyone's looking for a staking system that's reasonable while at the same time offering the potential for big profits in the long term, it's there in the middle:

What I would suggest, is to have bank targets, and bet a % of your bank and commit to it between these limits.
Say your bank is 10,000, and you choose maybe 2% of your bank.  Your targets maybe every 500 dollars profit or something.
So you bet 200 until your bank hits 10,500.  If you start with a losing run, you are still betting 200 a time.  Your stake should only change when you hit your next target.
When on 10,500 (or whatever your bank and target levels are), your stake would then change to 210, until your bank hit 11,000.

The only comment I'd make is that personally, I'd increase the gaps between raising the stakes. In other words I'd wait until I'd doubled my bank before I doubled my stakes and trebled my bank before trebling my stakes. Otherwise I'd be running the risk of frittering the way the profits from a good but short winning run via a small but significant increase in stakes on a subsequent losing run. It can all start to unravel mentally in that sort of scenario.

If you have the cushion of profits big and long-term enough to have doubled your bank, it's a lot easier to retain confidence that you're still doing the right thing when sod's law dictates that results start to go against you the moment you double your stakes.

agree with the posters above, brilliant post BJT , one of the best ive ever seen on this forum

Yes BJT's post is brilliant in as much as it highlights the "non system" the OP probably has.

Jul 28, 2013 -- 10:29AM, screaming from beneaththewaves wrote:

What an excellent post from BJT. Worth anyone's time to read it carefully.And if anyone's looking for a staking system that's reasonable while at the same time offering the potential for big profits in the long term, it's there in the middle:What I would suggest, is to have bank targets, and bet a % of your bank and commit to it between these limits.Say your bank is 10,000, and you choose maybe 2% of your bank.  Your targets maybe every 500 dollars profit or something.So you bet 200 until your bank hits 10,500.  If you start with a losing run, you are still betting 200 a time.  Your stake should only change when you hit your next target.When on 10,500 (or whatever your bank and target levels are), your stake would then change to 210, until your bank hit 11,000.The only comment I'd make is that personally, I'd increase the gaps between raising the stakes. In other words I'd wait until I'd doubled my bank before I doubled my stakes and trebled my bank before trebling my stakes. Otherwise I'd be running the risk of frittering the way the profits from a good but short winning run via a small but significant increase in stakes on a subsequent losing run. It can all start to unravel mentally in that sort of scenario. If you have the cushion of profits big and long-term enough to have doubled your bank, it's a lot easier to retain confidence that you're still doing the right thing when sod's law dictates that results start to go against you the moment you double your stakes.

What I am doing at the moment, is waiting for my results to sort of level out.  So not really simply waiting to hit a profit margin then raising stakes, but waiting for the profit margin to seem settled then raising stakes.
Personally don't see too much difference once you hit a certain number of bets though.
For example, my current project has just hit 113.65% after around 7,000 bets.  I didn't raise the stakes until that level seemed consistent. If your percentage bet is low enough however, then you have the freedom to raise the bet once it hits.  After all, money management is key.
If your percentage is borderline high in regards to risk, then raising too early risks that.  If it is low, then as long as you commit, then it should still see you profit as you should.



The main point in anything, and certainly in regards to this thread, is money management is EVERYTHING.  Well, of course you need a winning strategy, but if you have that, then money management decides if you profit.
One thing I guarantee, is that there has been many many people come through Betfair with a winning strategy, and lost the lot.  If you are too aggressive, based on your strike rate and value percentage, then you are guaranteed to lose at some point.

The other thing I will guarantee, is that I have nowhere near the best strategy in regards to winning.  In fact, my return on money bet would be the laughing point of many many winners on here if they knew.  But my strength is money management.  And that ability combined with a slight edge, keeps me in the game.  Of course, this is before my new project showing 113.5%, but facts still remain.  And given ~150 bets a day, betting to win ~3% of my bank per bet is absolute maximum I am prepared to risk as I know this will keep me in the game.
A promise I make right now to myself, is that this % will lower to 2%.  This is purely because I know this is around optimum, and the bank I have assigned to it is loseable to risk til I get to that point.



BTW.  Thanks for the kudos guys.  You may read a lot of siht about certain things on here that I post.  Some because I am bored, but most because my profitability relies on understanding the psychology of betting that everybody else employs.  But be sure when I make points such as this is because I understand it and am passionate about it.

Keep the posts coming, love a decent thread with legitimate questions/concerns/etc.....

I like this BJT a lot more than the BJT on the cricket forum

BJT
I believe the extremely important point you are making, is that most people are far too aggressive and impatient in their desire to win long term.
Unless you have discovered some miracle ability to pick winners at will, then it is really just a hard grind at pretty small margins, with money mgt. becoming the key issue.
My profit on turnover is far lower than even the figures you quote you are achieving, but it matters not a wit.
I know pretty exactly what I will earn over a certain volume of bets and so I just grind consistently away and wait patiently for the returns to accumulate and build.
The excitement foe me is never in the immediate moment of winning and losing certain bets, but in the knowing that I am building volume by placing bets and that will return me profits in due course.
Of course, as you say, you have got to find a winning strategy in the first place, which I believe has to be primarily based on objective numerical factors and have zero or very low opinion based parameters.

Any updates on how you've progressed?

Be wary of any staking strategy that is too sensitive to the sequences of winners and losers.

You might be able to say with some confidence that over 1000 best you'll have a strike rate of x% , but what about the spread of winners and losers?

In addition , I've experienced a losing run , had a winner then another losing run and then hit a purple patch. Could your staking strategy cope with two losing runs very close together?

I play NHL markets which are midnight UK time 80% of the time.  Usually the 1x2 market, minimum 1.80, usually 2.20 +-

I didn't want it to impact on the rest of my trading, betting, etc.

I used the reverse labouchere used that based on a 'fixed figure' to start, regardless of what my horses, other betting, trading did.

Found it was fine.

Sorry to move the debate sideways slightly - but does anyone have any views about this staking level issue when it comes to spread betting? i.e. where the potential win/loss is a variable amount rather than a fixed amount. I've been trying to work out how much I should be staking, but find it confusing in the spread markets I must admit.

Of course, there's still a potential 'max loss' for each, but that is less common and more likely to be less in each case.