turbobets
Gaining experience
Karma: 8
Posts: 55

Comments appreciated.



Logged





There is not enough information as we need to see where the rest of the market stands, to begin with. Then some other elements might come in too.



Logged




VidaBlue
Has experience
Karma: 50
Posts: 243

I think there are numerous facets to this, so there is no text book answer. I answered 3.00 due to the following personal opinion of mine.
2.90: A small edge in theory, but taking favouritelongshot bias into account, it is not worth it, unless something indicates that sharp odds should be even lower than 2.80. 2.95: Still an edge too small, for such a high odd (variance, I would prefer a bet with probalilities closer to 50%). If there are other market indicators that indicate it has more value, it may be worth it. 3.00: Unless there are market indicators that count against this decision, the bet is worth it. In most cases, I would bet. Maybe, not enough information: Soft odd may in some cases need to be much higher, if there's a long time to game start, if market is very illiquid and so on ... No way: Never reject a possible good opportunity without further investigation.


« Last Edit: March 21, 2018, 08:43:20 AM by VidaBlue »

Logged




DPG
Has experience
Karma: 6
Posts: 155

If it's negative arb, it's not a value bet is it?!? I suggest you look up the definition of a value bet.
You would be simply gambling.



Logged





If it's negative arb, it's not a value bet is it?!? I suggest you look up the definition of a value bet.
You would be simply gambling.
No offense, but I would suggest you continue to read the forum and look up that definition yourself.



Logged




DPG
Has experience
Karma: 6
Posts: 155

"A value betting situation is one where the odds on offer from a bookmaker reflect a probability that is less to the actual probability of that outcome occurring." I value bet and I don't mess around taking small discrepancies. I look for prices at soft books and exchanges which are clearly out of line with the rest of the industry. This is a sure fire way to win consistently. The question in the poll is pointless as there is not enough information given and you can not judge where the correct odds should lie.


« Last Edit: March 21, 2018, 10:20:03 AM by DPG »

Logged




VidaBlue
Has experience
Karma: 50
Posts: 243

Well, a common definition of a value bet is a bet with odds above the true odds. Many bettors calculate the true odds based on a sharp line, often Pinnacle's. Therefore, if Pinnacle has 1.909/1.909 for a certain line, any odd above 2.000 is by definition a value bet (similar to the example 2.90, 2.95 and 3.00).
The comment from DPG awakes a small concern of mine. I am often a bit skeptic by blindly considering the above definition, due to the juice/vig/margin at the sharp bookie. Therefore, I would appreciate opinions on the following:
Let us say that God knows that the true odds for a certain event is 2.00/2.00, so 50% probability for any of the two outcomes. All semigods are the very sharp players shaping the lines with their huge stakes – they know the true odds as well. Sharp odds launch at 1.77/2.17 and the semigods start hammering the 2.17 odd. Ultimately, the line ends at 1.90/2.01, because it would make no sense to trade this line below 2.01, as there is no value any more. All godlike creatures know this, so they don't trade. The 1.90/2.01 line stabilizes and ultimately becomes the accepted reference point for the true odds (which would then be 1.95/2.06). Placing a bet on outcome 1 at odds 1.98 is a value bet according to definition, but if you ask God, he would tell you it is not.
I am considering sometimes, when calculating a value bet based on true odds at Pinnacle, to take into account the juice as well, such that I would not place any bet below true odds PLUS the juice. In this example 2.00/2.11.
Any comments/experiences with this?



Logged





Well, a common definition of a value bet is a bet with odds above the true odds. Many bettors calculate the true odds based on a sharp line, often Pinnacle's. Therefore, if Pinnacle has 1.909/1.909 for a certain line, any odd above 2.000 is by definition a value bet (similar to the example 2.90, 2.95 and 3.00).
The comment from DPG awakes a small concern of mine. I am often a bit skeptic by blindly considering the above definition, due to the juice/vig/margin at the sharp bookie. Therefore, I would appreciate opinions on the following:
Let us say that God knows that the true odds for a certain event is 2.00/2.00, so 50% probability for any of the two outcomes. All semigods are the very sharp players shaping the lines with their huge stakes – they know the true odds as well. Sharp odds launch at 1.77/2.17 and the semigods start hammering the 2.17 odd. Ultimately, the line ends at 1.90/2.01, because it would make no sense to trade this line below 2.01, as there is no value any more. All godlike creatures know this, so they don't trade. The 1.90/2.01 line stabilizes and ultimately becomes the accepted reference point for the true odds (which would then be 1.95/2.06). Placing a bet on outcome 1 at odds 1.98 is a value bet according to definition, but if you ask God, he would tell you it is not.
I am considering sometimes, when calculating a value bet based on true odds at Pinnacle, to take into account the juice as well, such that I would not place any bet below true odds PLUS the juice. In this example 2.00/2.11.
Any comments/experiences with this?
It seems obvious you would add the juice to reflect the true odds...but add MORE juice on top of the true odds makes no sense to find a value bet. It would simply make your edge bigger, give you a higher ROI and reduce variance slightly as every bet you place would have a higher edge. In larger markets, considering you never know the "true God" odds...Pinnacle/Betfair/sharps are the next best thing to "God odds" as it's proven their closing odds reflect the true odds if your sample size is big enough. Obviously, they can be off for any given individual game but that doesn't matter as we're talking about a large sample and long term. Regarding DPG's comment about a negative arb never being a value bet: that's simply not true. If Pinnacle has odds of 1.95/1.95, a soft odd of 2.04 clearly has value but it would provide a negative arb.



Logged




VidaBlue
Has experience
Karma: 50
Posts: 243

It seems obvious you would add the juice to reflect the true odds...but add MORE juice on top of the true odds makes no sense to find a value bet. It would simply make your edge bigger, give you a higher ROI and reduce variance slightly as every bet you place would have a higher edge.
In larger markets, considering you never know the "true God" odds...Pinnacle/Betfair/sharps are the next best thing to "God odds" as it's proven their closing odds reflect the true odds if your sample size is big enough. Obviously, they can be off for any given individual game but that doesn't matter as we're talking about a large sample and long term.
Thank you for your answer Alfa. The aim of the hypothesis was exactly to challenge this "textbook approach" as the one and only truth. After all, sharps disagree by a juice's margin all the time. It would be reasonable to believe that there could be certain groups of markets where line movements on Pinnacle are often stuck and do not move the last tiny bit aligning with true odds, as it in principle would make no sense for any sharp money to move the line via stakes on odds below or equal to true odds. So had it not been for the juice, the line could be pushed further. Like when a high liquid line is stable on Betfair such as buy 2.04/1.95, sell 2.06/1.97 while Pinnacle is at 2.04/1.87. Comparing the two, they are sort of skewed to each other by a "juice's" margin  odds at outcome 1 are equal, but outcome 2 differ by 2 x juice. Could it be that it makes no sense for bots or traders to push outcome 1 at Pinnacle? Are the true odds really 2.05/1.96 (Betfair) or 2.09/1.92 (Pinnacle)? It is inevitable not to wonder about this when studying the odds dynamics between exchanges/asians/sharps. The discussion is maybe a bit too academic and irrelevant for shop betting on twodigit edges, but I suppose it could be valid for high frequency trading on sharps/asians with very low edges, like 12 %.



Logged




turbobets
Gaining experience
Karma: 8
Posts: 55

Thanks for the replies.
I am using pinnacles no juice odds or at least as best as I can determine: probA/(probA+probB).
Currently betting to win 0.5% of bank on edges 2% and greater: ((soft/sharp)1)*100.
I was betting every edge no matter how small for while, what a wild ride. The year started decent with a profitable January but I gave it all back and more in February which has left me a little timid and still trying to find my comfort zone. So far I am up a little this month but haven't put as much time into it as previous months.



Logged





Thanks for the replies.
I am using pinnacles no juice odds or at least as best as I can determine: probA/(probA+probB).
Currently betting to win 0.5% of bank on edges 2% and greater: ((soft/sharp)1)*100.
I was betting every edge no matter how small for while, what a wild ride. The year started decent with a profitable January but I gave it all back and more in February which has left me a little timid and still trying to find my comfort zone. So far I am up a little this month but haven't put as much time into it as previous months.
Interesting staking plan. Value betting has ups and downs. While you can expect to be a winner in the long run, persistent losing streaks could make many lose their faith in mathematics. We have seen a very negative November 2017, that must me the worse month of at least the last 2 years. December was positive and January was an even month followed by a negative February. February switched to greens after 15th and March looks great so far. So great that makes me think that the next losing streak is around the corner.



Logged




pythonic
Gaining experience
Karma: 6
Posts: 61

Thank you for your answer Alfa. The aim of the hypothesis was exactly to challenge this "textbook approach" as the one and only truth. After all, sharps disagree by a juice's margin all the time. It would be reasonable to believe that there could be certain groups of markets where line movements on Pinnacle are often stuck and do not move the last tiny bit aligning with true odds, as it in principle would make no sense for any sharp money to move the line via stakes on odds below or equal to true odds. So had it not been for the juice, the line could be pushed further. Like when a high liquid line is stable on Betfair such as buy 2.04/1.95, sell 2.06/1.97 while Pinnacle is at 2.04/1.87. Comparing the two, they are sort of skewed to each other by a "juice's" margin  odds at outcome 1 are equal, but outcome 2 differ by 2 x juice. Could it be that it makes no sense for bots or traders to push outcome 1 at Pinnacle? Are the true odds really 2.05/1.96 (Betfair) or 2.09/1.92 (Pinnacle)?
It is inevitable not to wonder about this when studying the odds dynamics between exchanges/asians/sharps.
The discussion is maybe a bit too academic and irrelevant for shop betting on twodigit edges, but I suppose it could be valid for high frequency trading on sharps/asians with very low edges, like 12 %.
Your model is maybe a bit too simple because in reality you typically won't have a set of players knowing the true odds, but rather a population of (more or less) sharp players with maybe somewhat different opinions about the true odds. But you are right that you should be careful to assume that juice is always incorporated proportionally on both sides into the pinnacle odds. That is simply the best guess you can make without further information. In your example the betfair odds might be a more accurate reflection of the true odds.



Logged




Ando
Gaining experience
Karma: 3
Posts: 2

At 1.5/2.8 (assuming these are the actual sharp book closing line) you are looking at 2.38% juice. At 2.90 you're at just over 1% novig odds, almost 3% at 2.95 and above 4.5% at 3.00. So it all depends on the minimum you are willing to take. I went ahead and downloaded the data (over 43 000 matches) from footballdata from season 201213 to the current season (the latest data they have) and all fixtures that didn't have Pinnacle's closing odds were discarded. That includes top 5 divisions from England, top 3 for Scotland, top two for Spain, Italy, Germany, France and top division for Holland, Belgium, Portugal, Turkey and Greece. For Scotland, Holland, Belgium, Portugal, Turkey and Greece, they didn't have Pinnacle,s closing odds for earlier seasons. Using, Pinnacle's novig odds as reference, I compared BetBrain's average odds against that reference and if BetBrain's average odds were above the reference odds by at least X%, a flat stake bet would be made. If in there was more than 1 value bet in a single match, the bet would always be the one with the highest %. I know that no all of the leagues included are high markets and therefore it would be assumed that Pinnacle's closing odds would be less accurate as a reflection of the "true" odds but I went ahead anyway. At >=1% above Pinny novig odds we are looking at a yield of 6.75% for 12962 bets 2% and above: 7.60% yield for 10426 bets 3% and above: 8.43%yield for 8330 bets


« Last Edit: March 24, 2018, 02:26:37 AM by Ando »

Logged




turbobets
Gaining experience
Karma: 8
Posts: 55

Great info Ando. So in theory I should be safe betting 2% edges on the leagues you posted that are major events. I assume this is true for most sport leagues where a high volume is wagered, time will tell.



Logged




Ando
Gaining experience
Karma: 3
Posts: 2

For the top 2 divisions in England, Spain, Italy, Germany and France, it would've been a yield of almost 7% yield but a couple of leagues are in the red in that bunch.



Logged




