negapo
-
Posts
22 -
Joined
-
Last visited
Posts posted by negapo
-
-
Re: Some good betting system websites? Good description of the process of transforming ideas into models Areme.
-
Re: How to calculate ROI increase or decrease with odds going up or down I cannot edit the previous post (that has a lot of line breaks) unless i make this post and get to the 20 posts required to edit posts :D So here t goes
-
Re: How to calculate ROI increase or decrease with odds going up or down I think this is correct: The punter beats the odds on average by 10%. If he tips you on a odd of 3 and the odd has the average 10% value then the true odd would be: 2.73 = 3.00/(1 + 10%) So anything above 2.73 would be good. I think zbrochu answer was the same stated in another way.
-
Re: High Odds...Medium Strike...Big ROI Good luck christos83
-
Re: 3 categories of punters zbrochu Let me try to clear some of my earlier thoughts: If there is no new relevant information coming to the market (like an injury, or something similar) and the odds are volatile to the point that the Highest historical odd on every selection made a "sure bet" (the highest odds summed more than 100%) then there was a moment where a selection presented value (this is by definition). This happens many times. Another thought is that a Bookie prefers to make a perfect book knowing that it will win money no matter the outcome than to leave a selection open even if he thinks it has value. With this two thoughts (and the other arguments on the previous post) i think someone has to conclude that the Bookie will aim to publish the odds in a way that he expects to be inline with the demand and not with the fair odds. He can, if he thinks the balanced odds are to far from the fair odds, not publish the odds, increase the margins or wait to see some action in the market before he publishes. And i think you are saying the same:
If punters will hit one of the selection then that's not dividing the action. What i think you are saying is that punters are now more sophisticated and there is a bigger alignment between what the bookies/experts think are the fair odds and the equilibrium odds.Unfortunately because it would be easier for punters to beat bookmakers if bookmakers' goal was to divide action instead of making accurate odds. Then sophisticated players would just blindy bet against "public" teams and bets to make profits. It was possible in the past, sharp players mostly bet unders and dogs, because bookmakers shaded few points knowing people will always take favs and overs anyway.
That idea of changing odds is part of what i think supports my argument. The discussion went off topic (Three Styles of Punters).Yes, this is obvious bookmakers change odds as a result of money wagered. If they get more action on one team, they lower odds to balance the action to secure margin-profit regardless of the result. This is some obvious stuff. -
Re: 3 categories of punters AlanJenkins2, I was answering zbrochu in my post (i cant edit posts :( )
-
Re: 3 categories of punters If that was true then bookmakers would not change lines frequently when there is no new relevant information coming in, this is, they would not move their lines with the pressure of punters betting unevenly in one selection. What you see today is bookmakers subscribing to odds API that continuously check their prices against the competition, they are in an effort not to model the game properly but to move rapidly with the market. This is a global world where information flows continuously, and it's more comfortable to make a book and go with the flow than it is to win money by betting against punters (on a head to head of who knows more). I give you this, there are in fact a few bookies that when odds move a lot from the initial published prices that suspend the market (they stop offering odds). This could signal that they are not willing to make a book on those odds but what is more probable is that they don't want to participate in a market where volatility is high. The bookies final goal is to make money by making a market, not by betting against clients. If they have a superior way of modeling the game they don't need to be a bookie, spend a lot of money on marketing, have a crew of employees on a payroll to model the odds, give costumer assistant, share revenues with money wallets, have the entire liability of running a business. (They need some skill in modelling the odds of course if they want to publish their prices early). You can find parallel in financial markets where when you sell something (make an IPO, you are trading stocks) you usually don't ask yourself how much is this worth but how much are people willing to give to me for this.
-
Re: 3 categories of punters To add a little something Bettors do not race against bookmakers, they don't need to beat them. Bookmakers don't care (or they shouldn't) if in the prices they offer there is a huge miss valuation on one selection. They just want to find the equilibrium between the different bettors, that sweet spot. Take an example. Bookie A feels the true probabilities of a soccer match are: Team A to Win: 36% Team B to Win: 34% Draw: 30% So they make the odds (decimal) like: Team A to Win: 2.77 Team B to Win: 2.98 Draw: 3.30 They run it to expert A who tells the math guy: People are crazy about Team A, they have scored 5 goals on their last game and people will inflate their probability of winning (he is basing he's analysis on some bias that bettors have). We don't want to get hit on that selection so squeeze that odd. I think punters will settle for a market like this, says the expert: Team A to Win: 2.38 Team B to Win: 3.23 Draw: 3.70 So the bookie takes it and then applies their 3.5% margin across selection and final odds look like this (they can also take their margins unevenly): Team A to Win: 2.32 Team B to Win: 3.11 Draw: 3.55 So the bookie thinks is offering the odds for Team B with value, 3.11 against the fair odds estimate of 2.98, but he doesn't care as long as people find it in equilibrium and bet accordingly on every selection. When you bet on team B that has a lot of value that value is coming from the suckers that bet on Team A (and good punters that bet on draw that has value also). This is the magic of betting and the reason why there can be bookmakers like Pinnacle. Not every bookie is like this, there are bookies who just don't like winners, and there are bookies who take risks and like to be open on some selections that they find to be valuable. So, following the line of thought, there is a subtle variation of the stats/math guy: You don't need to model the game, you can model the bettor. Imagine that people undervalue favorites (like the favourite longshot bias) in a regular basis. So bet the favourite, ignore the game, the rules, etc. In financial markets they call it behavioural finances (thet study of how people act and not how should they be acting). It's a little bit like the difference between how the world should function (the true odds of a game, the true value of a stock, etc.) vs. how the world functions (people overreact to big wins, people overvalue things they love like Apple and it's stock, etc.) If you know where people make their mistakes and they make it on a regular basis then you can make a living out of it (the marketing guy make that).
-
Re: Tennis Forecasting system I tested your site, i went straight to the ferrer by chance and saw the 1.31 at pinny. It's seems that you have put a lot of work on the site and on the models. Have you ever wondered of automating the bets (retrieving prices from your model, check them against BetFair and placing bets)?
-
Re: Nice selections, but I'm losing money ... I rather have 10 selections (bets, trades, business opportunities) with an expected value of 5% than 6 with an expected value of 10%
-
Re: Nice selections, but I'm losing money ... I wrote this on another thread that i think is relevant in you case:
In essence you will achieve your excepted value (you will find out if you make money or lose) in the long run after a lot of bets so grab every opportunity that you find value even if it inst a lot. Bet small on every selection that you find value (reduce the roll of luck). PS: It's a little bit like diversification in financial assets.When you trade (both in sports betting or financial markets) or even when you open a business your expected profit is dependent on the number of opportunities you encounter and the value that they provide. When you account for luck you get to the conclusion that you rather have a lot of opportunities with small value than few opportunities with big value. An example: You get to back Real Madrid to win the game at odds of 2.0 against the team that is in last. You compute that the fairodds will be something like 1.30 so this is a chance in a lifetime. What do you do, how much do you bet? If you follow your guts or something like the Kelly Criterion you will find out that you should not stake too much because Real Madrid can still end not winning (in 23% of the times), so you can only grab in safety a little amount of this huge opportunity. Id rather have a way of identifying a lot of small imbalances and take them all with small stakes. And to find a lot of opportunities a human is not a good thing, it will get tired, it will get fed up and risk a lot more than is should (i believe that poker players can relate with this). -
Re: Betting System Suggestions No staking plan (position sizing) will make an unprofitable system (a negative EV) into a profitable one (a positive EV). It will only at tops hide the luck factor. O good staking plan will reduce the chance of bankruptcy (always present even with a profitable system) and at the same time extract the most possible value of your hedge (increasing the size when your hedge is higher). The Kelly criterion is an example of such a plan (not saying it's good or bad). For example, if you think the fair odds for an event is 2.00 and a bookie is offering you 2.20 then according to this method you should bet: % of Bankroll to risk = ((2.00-1)*(1/2.20)-(1-(1/2.20)))/(2.00-1) = 8.3% (All odds must be net of Fees, commissions, etc.)
-
Re: Betting on under 6.5 goals, in play at odds of 1.01 Thanks allanmac83 Yes, i think you should know your hedge, but that's not very consensual and there are good arguments against it (or favoring that you don't necessarily need to know why you are winning money). I think you can take two roads in general, the systematic one, where you build rules (with or without the help of past data) and try to automate the execution and the discretionary one, where you try to find value in situations where you brain can assess a multitude of variables and identify an opportunity. There are advantages and disadvantages in both ways of trading and there are also traders that mix both dimensions. There is one particular thing that the totally discretionary road is in disadvantage (i believe): When you trade (both in sports betting or financial markets) or even when you open a business your expected profit is dependent on the number of opportunities you encounter and the value that they provide. When you account for luck you get to the conclusion that you rather have a lot of opportunities with small value than few opportunities with big value. An example: You get to back Real Madrid to win the game at odds of 2.0 against the team that is in last. You compute that the fair odds will be something like 1.30 so this is a chance in a lifetime. What do you do, how much do you bet? If you follow your guts or something like the Kelly Criterion you will find out that you should not stake too much because Real Madrid can still end not winning (in 23% of the times), so you can only grab in safety a little amount of this huge opportunity. Id rather have a way of identifying a lot of small imbalances and take them all with small stakes. And to find a lot of opportunities a human is not a good thing, it will get tired, it will get fed up and risk a lot more than is should (i believe that poker players can relate with this). That are also a lot of limitations in going the totally systematic way. In essence, don't risk too much, try to put your thoughts and ideas into rules if you can (at least parts of it) and try to get rid of luck (bad and good). That's what i try to follow (Sorry cooliyog, I don't want to hijack the thread)
-
Re: Betting on under 6.5 goals, in play at odds of 1.01 If you are betting against an event that has a 1% chance (you think that percentage is lower and that's why you Back it at 1.01) then you have to get a huge sample in order to understand if the market is effectively miss priced. What i mean is that your balance is irrelevant unless you have a huge track record. It's a little bit like believing that you can tell when a car accident will happen by every day saying that it will not happen today. It's a improbable event so you will be right most of the times. Another thing you have to worry is that even if you really have an opportunity and you are successful at discovering events trading at 1% that have a lower probability of happening you have to protect yourself from bankruptcy. If the true probability of the event is something like 0.50% there is still a chance that the event happens 3 out of 30 times (what we call bad luck). In essence, account for bankruptcy (stakes must be relatively small), wait for a lot of bets before concluding that you have a winning system and try to understand why the odds would be miss priced (people lay the 1.01 to net their positions and maintain the 1.01 for longer than it should? people like big payouts and believe the impossible? etc.). You can have a winning system and not knowing why it is paying off (you discover by accident, you do some data mining on large datasets or you use neural networks to build a system) but it's always better to understand it. Good luck
-
Re: Estimations based on Odds Data of local book I wish you very good luck christos83. This thread had a very interesting discussion at the beginning and i believe that you are on the right track (the way you think). Statistics are useful (not sufficient) and backtesting can be a very useful tool (believing in yourself, in your perception, can be the ultimately mistake, and a well done backtest with a good sensibility analysis and so on can clear a lot of questions). Sure you cant forget that the future doesn't have to imitate the past and that when you build a system you have a memory that you cannot detached from (ex: don't backtest backing nadal after he loses a game or buying apple stock after it falls 5%. They are winners so you know they recouped.). Ill disagree that you need to understand complicated math to do something profitable (i work at building financials trading models and can assure you that not knowing complicated math isn't a problem when you have the right mindset, it can be a virtue sometimes). Good Luck
Laying Fav Systems don't work! Have tried it and failed everytime!
in At The Races - Racing Forum
Posted
Re: Laying Fav Systems don't work! Have tried it and failed everytime! Check your tracking error between your backtest live (paper trading) and your actual trades with just a few pounds. What can happen a lot of times is that you only get matched on the non value trades and the value trades run away from you or only get partially matched.