Betting short odds

[mrmuzeman]

Bit of a strange question perhaps but hear me out I might manage to get across what im talking about with any luck.

My usual betting is singles on 6/5 type shots. Thats my typical kind of bet. So for instance a normal bet for me would be to place 4pts on a 5/4 shot to make a 5pt profit. I like to bet for a fixed profit.

If I was to back short odds which I occasionally do but really not often I struggle to know how much to put on for the bet. Using my usual 5pts profit method if I were to bet a 1/2 shot id have to bet 10pts which is a pretty big bet to be honest and it just feels a bit weird to me.

If you guys back different prices such as what I describe do you have a seperate betting bank for different types of prices? For instance one bank for odds against and one for odds on.

Or maybe its best to bet for level stakes whenever picking an odds on selection?

[zonker84]

I have a standard sized bet (I'm not too keen on fixed profit betting, although I can see the logic). I stake 0-20pts, and my standard bet is 5pts. If I'm taking a 10/1 outsider, I'd be more likely to place say 1pt on this rather than 5. People can argue that if you think you have a big edge (Kelly theory), then you should stake more, so really the price shouldn't matter particularly.

From a fixed profit point of view, if you were to take a 1/10 shot, you'd be staking 50pts, which I assume you'd not be too keen to do! This is why I limit myself to 20pts on any bet. Even with this 20pt limit, I think I've only placed 3 or 4 bets this size in the past couple of years.

I did stake 200pts on Australia winning the Ashes in 2002/03 though Was never in danger though, and when you are totally certain about something as I was then, you might as well cash in

[mrmuzeman]

So youd stay to your agreed fixed limit for a bet and not exceed it basically.

Why dont you like fixed profit betting by the way? Is it because you take a range of priced bets and it just makes it a bit weird with a possible out of proportion huge bet as I described?

[Beanie]

Just to pre-empt Doc here ...

http://www.punters-paradise.com/news...ng-systems.php

Well worth a read...

But as you and zonker mention, such a fixed profit system can 'force' you to stake an amount you aren't comfortable with if you are to stick to the plan rigidly.

My staking plan is generally done by the edge I feel I have over the bookie - so its closest relative is probably the Kelly Criteria.

My biggest bet thus far on a 10pt scale was 8pts on Hearts away at Kilmarnock on day 1 of the season at 2.68, closely followed by Motherwell at Livi last week at 2.50 for 6pts.

Likewise, I often stick maybe just 2 or 3 points or 1.40 shots if I feel I've not got too much of an edge.

[zonker84]

Originally Posted by mrmuzeman

Why dont you like fixed profit betting by the way? Is it because you take a range of priced bets and it just makes it a bit weird with a possible out of proportion huge bet as I described?

A few reasons. Firstly (this may sound silly) is that I've always done it this way. That's not to say I'm not open to new ideas and staking plans, but it works for me so no need to change it too much.

I did test whether I'd make a bigger profit using fixed profit betting over the course of a whole season's results, and found that actually using my own variable staking plan works better.

I don't use Kelly Criteria, but if I saw what I regarded as a fantastic bet @ 10/1, I would still put a big stake on it. I didn't make this clear before when I said I'd put 1pt on a 10/1 shot. Normally for higher odds I would stake less, but if I think I have a big edge I'll do more.

My staking plan isn't very scientific, I just stake whatever I feel is right after giving it a bit of thought. Hence the 200pt as I thought it was as close to a certainly as you could get. I certainly wouldn't make a habit of placing such huge bets

[Chaz]

Zonk and Beanie when you look at a football match do you actually put a figure on your selections. By that I mean, take Hearts for an example, you may rate them an even money (2.00) chance and 6/4 is available (2.50) which would be a good value bet, or is it more of a feeling you have for certain matches. On the flip side of that do you find yourself sometimes with what you think is a very solid chance, lets say even money again, and then find the price on offer is 8,9 or 10 to 1, do you ever think to your self, have I missed something here? At the end of the day Betfair and the like, are money markets formed on opinions of thousands of people. Could they all have missed something that you are seeing?
Fot the record I use a fixed stake method between 1 and 8 points purely based on my feelings mainly because I dont know how to frame prices for football or most other sports. Any thoughts guys.

[Jezza]

I would have to say (purely from a mathematical gambling aspect and ignoring the fact that it is sports betting) that good old prof kelly is king here.

Thus the bigger your advantage, the more you stake. In sports betting of course this is subjective, but those that can truly judge the value or lack of are the only ones that long term profit, and are the only ones that could make use of this theory anyway (a long term loser obv should bet nothing at all or as little as they can).

I have to say fixed stake betting seems incredibly silly and misguided to me given that you have such a varying range of bet advantages.

For example, a card counter in a casino may have an advantage on the next hand of 1% due to his count and place a reasonably bigger bet than usual on the next hand. However the same card counter in another situation knows for sure he is going to get an ace first card on his box next hand. This is a 50% advantage situation approx and as such would call for as big a bet as he could afford or the casino would allow. To have a fixed big bet in these situations would be ignoring the chance to take advantage of the second dont you think?

Jez

[Beanie]

Originally Posted by Chaz

Zonk and Beanie when you look at a football match do you actually put a figure on your selections. By that I mean, take Hearts for an example, you may rate them an even money (2.00) chance and 6/4 is available (2.50) which would be a good value bet, or is it more of a feeling you have for certain matches. On the flip side of that do you find yourself sometimes with what you think is a very solid chance, lets say even money again, and then find the price on offer is 8,9 or 10 to 1, do you ever think to your self, have I missed something here? At the end of the day Betfair and the like, are money markets formed on opinions of thousands of people. Could they all have missed something that you are seeing?
Fot the record I use a fixed stake method between 1 and 8 points purely based on my feelings mainly because I dont know how to frame prices for football or most other sports. Any thoughts guys.

Its a bit of both really Chaz - I price up games based on form, H2Hs, team news etc, but this only really gives a ball park figure of the %age chances of each outcome, which could realistically still be a long way out, and the key is then in the manual alteration based on the more subtle things that might affect the game.

Nevertheless though, particularly in Scotland a lot of my bets are down to 'feel' rather than based on scientific pricing, but even here I'll be using some sort of figure I believe to be a fair price in my head.

However, you probably already frame prices for matches and outcomes a lot more than you realise, but if you want a very basic way to start:

Take Dunfermline v Hearts this weekend, and take the home record for Dunfermline and the away record for Hearts.

That is 1-1-8 and 4-4-2.

Add the no. of Dunfermline home wins to the number of Hearts losses - 1 + 2 = 3

Add the no of draws - 1 + 4 = 5

Add the number of Dunfermline losses to Hearts wins - 8 + 4 = 12

That gives

3 - 5 - 12

You need these to equal 100, so divide 100 by the number of matches you have used, which is 20 here, and multiply all values by this number.

That means.

15 - 25 - 60

These then need to be altered based upon all the factors you believe will influence the game - and this is the important bit for which there is no formula. If a key player is out, or H2H record is particularly strong you must judge how much of an edge you believe that that gives a side, and adjust accordingly.

Once you have done this, translate your probabilities into prices, which for the above would be:

6.66 - 4.00 - 1.667

So you then have a framework to start judging whether prices represent value or not.

I should stress again that this is very basic, and the key lies in the manual adjustment, so if you want to develop this make sure you do a lot of trial runs on paper until you are comfortable with it.

There are also a lot more complex ways to generate your own prices, but I think this is a good entry point.

[Chaz]

Thanks Beanie, it may only be basic to you but it is more than I knew two hours ago. Look forward to tinkering with it this week end. Your probably right in saying I frame prices in my head with out realising it. I think I have a rough idea of what represents value but never been able to put any sort of figure on it which is important if you are going to bet to prices.
I grew up betting horses and betting to a given rating/price to return a set amount eg:
Makybe Diva is rated 6/4 (2.50) to win and provided you can get 6/4 or better on track you would bet 40 units for a return of 100 units (100 divided by 2.50 =40), the bigger the overlay the more return. (hope that made sense)

Im pretty much off the horses now as I really like what sports betting has to offer me so thanks again mate

[crowie]

I have mucked about with staking here and there but I am most comfortable with fixed stakes.

Out of say a scale of 1-20, my normal stakes would be 2pts but for certain bets I would stake 5pts. I have gone bigger on occasion but not very often. I am happy with that method whether my returns are going to be 0.5pts, 1pt, 2pts or whatever. I would also drop down to staking 1pt where I see a price that is attractive but rate the chances lower than what I would rate a 2pt bet ie Bolton at 4.10 at home to Liverpool.

Interesting thoughts about the Kelly method and staking more if you feel your getting a bigger edge. My only negative about that is I feel you would have to be pretty damn certain your own prices were near enough spot on to start adjusting your stakes to take advantage of any incorrect prices. Myself, like a few people on here, doesn't price game's systematically but in many cases forms a view on a game of roughly what price you would, or wouldn't back a fancy. By using that sort of method I would be reluctant to start increasing stakes based on what advantadge I felt I had.

There are however, certain stick out cases where its worthwhile doing. The last example I can think of is Globets 1.67 on over 2.5 in last seasons Gretna v E Stirling game. Based on the available stats for the fixture, I thought it was approx 1.10 correct price and had my biggest bet in approx 3 years on it. Course it finished 1-0 but given that same situation again, I would have the exact same bet again.

[zonker84]

I go based on feel more than anything. Just by looking at a game in the UK on a Friday for example, I can usually guess very closely what the best price will be on each outcome. If this price is way out, then that's obviously a good bet.

For me, I just know when I think that I have a good edge. Sometimes I find myself really excited looking at the odds

I don't think I need to do it all scientifically using Kelly.

[GETT1NLOTS]

Originally Posted by crowie

Interesting thoughts about the Kelly method and staking more if you feel your getting a bigger edge. My only negative about that is I feel you would have to be pretty damn certain your own prices were near enough spot on to start adjusting your stakes to take advantage of any incorrect prices.

Spot on mate. This is the problem with the Kelly method when betting on sports. Jezza is translating a sound betting method from poker (where the pot odds can be calcualted exactly) and trying to apply it to another type of punting where supposition, opinion and gut feeling play a huge part.

To make it simple.

If I am waiting for one card to be shown at texas holdem and I know the two cards in my hand and the four cards already out in the middle there are 46 cards that I still don't know. If I need an Ace to win the pot, and an (ace isn't in the six cards I know) I have a 46/4 chance of hitting it = 11.5% chance of the ace showing. If there is £1,000 in the pot and the raise from my only opponent is £50 I should call because I have a 11.5 to 1 chance of getting the card for a 20 to 1 pay off. Given this situation it would be correct to bet as there is no disputing mathmatically that in the long term I will be a big winner making this bet.

However if I think Arsenal are about 6/4 to beat Man Utd at home I wouldn't have a big bet just because a bookie priced them at 7/4. The reason is that I might be influenced into thinking this is value based on

a) My mood.
b) How much I want a bet.
c) How much I dislike Man Utd.

etc, etc, etc.

In other words searching for 'value' is all well and good but to increase your bet like it was a science depending on the size of the value you perceive there is to be had, is in 99.9% of cases, giving oneself a bit too much credit.

I stick to fixed stakes mainly because I know that searching for value is one thing, but taking it to the next level where I increase my stakes because I think I have the deepest level of understanding of 'mathmatical probability'? Not for me.

In poker yes as it IS an exact science. In sports betting no.

PS - I do exclude bets if I don't think the odds to be had are value. But I don't tend to increase bets if I think there is value.

Hope all that made sense.

[mrmuzeman]

The last example I can think of is Globets 1.67 on over 2.5 in last seasons Gretna v E Stirling game. Based on the available stats for the fixture, I thought it was approx 1.10 correct price and had my biggest bet in approx 3 years on it. Course it finished 1-0 but given that same situation again, I would have the exact same bet again.

Yeah and youd be right too. I remember that and it fcukin sucked I wasnt on but it really could have been 8-0 couldnt it.

Good thoughts here in this thread. Personally I use the 'Kelly' type of theory in my usual singles betting (which typically ranges from 4/5 up to about 6/4 normally). I price up games based on what I think the price should be then see what edge im perceivably getting. The bigger the edge the more fixed profit I bet for. I didnt know Kelly existed though but I seem to be following his kind of pattern I guess.

For instance my normal bet is to get a 5pt profit if it wins. Say this was a 5/4 shot - I would bet 4pts if I felt I was getting a decent edge. If I thought I was getting a very good edge, bigger than I would bet normally for, id bet for a 10pt profit so would bet 8pts and so on.

[Beanie]

I think you've contradicted yourself a little there Gett1n IMO.

Imagine you see a particular bet, and given your expectations you believe your expected return to be 0.9 x your stake.

Imagine a different bet where you believe your expected return to be 1.1 x your stake - if correct a 'value' bet.

And a third where you believe your expected return to be 1.3 x your stake.

Consider the first two - you don't bet on the first, but do on the second - to me that is increasing your stake based on perceived value.

In other words, you are saying that between the expected returns of 0.9 and 1.1 you are prepared to trust your judgement, that your calcultion of a positive return on one compared to the other is correct, but not between 1.1 and 1.3 by increasing your stake.

Detecting the difference between negative and positive expected return bets, and taking only positive, IMO is no different to detecting bets with a lower or higher and positive expected return, and increasing your stake accordingly.

I didn't word that too well, but I think my point is in there somewhere.

[mrmuzeman]

I think Gett1n just means sticking to fixed stakes on a day to day betting basis where maybe you have little more of an edge on 1 bet than another but not enough that he could trust it was any better a normal bet.

Gett1n still places bigger bets when he thinks theres more value Im pretty sure, so in a way he does use the Kelly type of thinking.