|
By:
Forgot to mention that I had a huge profit after around 50 bets
|
|
By:
A lucky run no doubt.
|
|
By:
Maybe.
But I'm still a bit confused about these clean ****, sorry clean sheet, bets. The way I handicap them I experience the following: When to equally strong or almost equally strong teams meet I never find any value. My lines agree with the market. When there's a clear favourite, where the odds on the favourite is around 2,00 or lower, I often find value betting on clean sheet for the favourite. When there's a huge favourite I often find huge value betting on clean sheet for the favourite. However my lines on clean sheet for the underdog always agree with the market. I use poisson props. And that seems to work well as long as my lines agree with the market, which is always the case when betting on clean sheet for the underdog. The market seems to do the same, except when the expected number of goals is low, which is the case when betting on clean sheet for the favourite (when there is a clear favourite). Then the expected number of goals for the underdog is often below 1. My only explanation is that according to the market, poisson underestimate the number of goals scored by the underdog, and this is the reason why I find "value" here. But I've done some research here. Gone through a huge number of games (several thousand), and poisson works pretty well on low numbers also. Poisson underestimate the expected number of goals a bit when the numbers are low, but far as much as the market means. |
|
By:
Correction: but NOT far as much as the market means
|
|
By:
I'm puzzled as to how you manage to check so many past results, without
knowing the 'clean sheet' betting on these matches, but whatever way you do it, 10/10 for persistence. I simply work out my own way the goal expectancy for each side, then refer it to a chart I have developed. From that point, I only bet if there is at least 20% up or down, indicating a back or a lay, needless to say I don't get many which I am not unhappy about, as I've said before that if I am close to the market, I think I am right. So far this season, from tv matches, I have had just 16 bets, 10 successful, for a very decent profit.GL. |
|
By:
Lets take an example. Tomorows game between Chelsea and Marseille.
Match odds. Around 1,35 on Chelsea and 13 on Marseille. Assuming the odds are fairly correct one can conclude that Chelsea will score 81 % of the goals in the match and Marseille 19 % of the goals. It does not matter how I reach these numbers. The point is that my numbers always agree with the market when there's no clear favourite. Also my numbers almost always agree with the market when betting on clean sheet for the underdog, which also is the case in this bet. Let's assume I'm correct so far. Maybe you also agree with me so far? Accoring to the over/under market, total goals index market and correct score market the game will produce a total of around 3,05 goals. This again means that Chelsea will score 2,47 goals (81 % av 3,05) and Marseille 0,58 goals (19 % of 3,05) Assuming Chelsea is expected to score 2,47 goals, there is according, to poisson, a 92 % probability that Chelsea will score at least one goal. Wich again means there is a 8 % probability of clean sheet for Marseille. And so far the market totally agree. The odds on clean sheet for Marseille is around 12. Assuming Marseille is expected to score 0,58 goals, there is according, to poisson, a 44 % probability that Marseille will score at least one goal. Wich again means there is a 56 % probability of clean sheet for Chelsea. But here the market totally disagree. Correct odds should be around 1,79 when the probability is 56 %. But he odds on clean sheet for Chelsea is around 2,15. The only explanation is that the market don't trust poisson when numbers are low, in this case 0,58. According to the market there actually is a 54 % probability that Marseille will score at least one goal, meaning a 46 % probability of clean sheet for Chelsea. Any comments? I don't expect you to reveal your methods. But are you doing this very differently? I check past results at www.nifs.no |
|
By:
Based on your goal superiority (1.6) and expectancy,(3.05)
I would have Chelsea up for 2.325 and Marseille on 0.725, indicating (to me) clean sheets 2.06 and 10.25 respectively for Chelsea and their opponents. |
|
By:
At www.nifs.no you can for example check the average away score for e team over a whole season. This number is often low for bad teams in low scoring leagues. It could for example be 0,60 goals on average in away matches. I then find what poisson says about the probability of at least one goal being scored when the expected number of goals cored by a team is 0,60, and compare that with the facts, meaning percetage of games in which the team scored 0 goals when average score is (in this case) 0,60. I've done this for more than 6000 matches. And the results are clear. Poisson slightly underestimate the probability of at least one goal being scored when numbers are low. (When the average numers of goals scored by a team is between 0,40 and 0,90 goals). But not by far as much as the market assume.
|
|
By:
What does poisson say for 2.325 and 0.725 respectively ?
|
|
By:
I don't know how you reach the numbers 2.325 for Chelsea 0.725 for Marseille.
Anyway your numbers indicate value betting on clean sheet for Marseille. You should get 12 or maybe 12,5. |
|
By:
The funny thing is that you should also get value betting on clean sheetfor Chelsea. At least 2,15. Maybe 2,20. If your numbers indicate that you can find value betting on clean sheet for both teams, then something is probably wrong.
|
|
By:
How do you arrive at 81% and 19% respectively ?
|
|
By:
Sorry, but I feel I've told enough. Some "secrets" I'd like to keep.
These precentages are not the "clue" here anyway. As I've told, my numbers always agree with the market when there's no clear favourite. Also my numbers almost always agree with the market when betting on clean sheet for the underdog, which also is the case in this bet. Lets say that these were not correct. It would only mean that I either would find even more value betting on clean sheet for Chelsea, OR that value could disappear and I've would insted find value betting on clean sheet for Marseille. For example if I reached the probabilities 75 % and 25 % insted, then the odds on clean sheet Chelsea would agree with the market, but I would insted find value betting on clean sheet on Marseille, as the numbers would indicate correct odds to be 9,90. |
|
By:
The reason I ask about the 81 and 19 percentages is because that is
where we differ, as by your 1.35 Chelsea and 3.05 goals suggest to me 73 and 27, not that I actually want to know your method. For what it's worth I slightly disagree with you on the goal expectancy also but not by much. Re Chelsea clean sheet, I would be a backer at around 2.25, Marseille a backer at 12.5. GL. |
|
By:
If you differ from the market so much, have you not
considered that the market may be more correct than you are ? Just a thought. |
|
By:
I guess you reach the percentages 73 and 27 by taking 1 and divide by the odds on Chelsea, 1/1,36 = 73. Meaning Chelsea has a 73 % winning chance, and Marseille 27 %. Is that correct?
|
|
By:
If you differ from the market so much, have you not
considered that the market may be more correct than you are ? Just a thought. -------------- Of course. If I was sure I was right and the market wrong I would put everything else aside and BET. The ting is that I'm really trying ti find out what I might do wrong, so far with no results. Everything I do seem to confirm that I'm right. But I could'nt possibly be that right, and the market that wrong. |
|
By:
Suddenly all the clean sheet bets on Chelsea vs Marseille are gone..
|
|
By:
Yeah, most of the markets have been cleared for that game atm
|
|
By:
Lets take another example. Tomorrows match Ajax vs Milan.
Using the same procedure, my lines totally agree with the market. How about yours? |
|
By:
I have not looked at what is on offer but I would
have Ajax 3.60 with Milan at 3.75. Am I anywhere near ? |
|
By:
Re your post at 11.06 the answer is no.
|
|
By:
One of the problems of Poisson is, that it assumes that the number of goals of the favorite is independent from the number of goals of the underdog. That isn't true. So if there is a clear favorite, there is a big probability they will score some goals. And when the favorite scores some goals, it is more likely that the underdog also scores, compared with the Poisson distribution.
Or, when the favorite has a clear advantage (2-0, 3-0, ...), their defensive capacities go down, because of that advantage. |
|
By:
I have not looked at what is on offer but I would
have Ajax 3.60 with Milan at 3.75. Am I anywhere near ? ------------------------ According to my lines correct odds for Ajax is 3,53 and for Milan 3,82. What precentages do you use in this match? I use 52 and 48. |
|
By:
[b]JosTheelen[/b], I agreewith you entirely. A good example was last
week Chelsea v Blackpool. I backed the Chelsea for a clean sheet and collected, but watching the match I felt incredibly lucky in the second half. When a goal being scored is meaningless, it's more likely to happen for the reasons you mention. |
|
By:
One of the problems of Poisson is, that it assumes that the number of goals of the favorite is independent from the number of goals of the underdog. That isn't true. So if there is a clear favorite, there is a big probability they will score some goals. And when the favorite scores some goals, it is more likely that the underdog also scores, compared with the Poisson distribution.
Or, when the favorite has a clear advantage (2-0, 3-0, ...), their defensive capacities go down, because of that advantage. ---------------------------- Good point. I've given this a thought. But the thing is that the numbers I use is derived from the over/under market. Let's take the Chelsea-Marseille match, discussed above. My numbers nidicate that Chelsea will score 2,47 goals (81 % av 3,05) and Marseille 0,58 goals (19 % of 3,05). According to the over/under market the game as a total will produce 3,05 goals. The Clean sheet market seem to agree with me that Chelsea will score 2,47 goals. But if Marseille actually should score more than 0,58 goals, due to what you wrote, then why is the game still to produce 3,05 goals, according to the market? Why not more goals? |
|
By:
My numbers nidicate that Chelsea will score 2,47 goals (81 % av 3,05) and Marseille 0,58 goals (19 % of 3,05). According to the over/under market the game as a total will produce 3,05 goals. The Clean sheet market seem to agree with me that Chelsea will score 2,47 goals. But if Marseille actually should score more than 0,58 goals, due to what you wrote, then why is the game still to produce 3,05 goals, according to the market? Why not more goals?
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Because it works both ways. When Chelsea has scored, it is more likely that Marseille will score. But when Chelsea does not score, it is also less likely that Marseille will score. They will try to keep it 0-0, and attack even less than normal and defense more rigorous. So on the one hand, the average number of goals will increase, because of the first effect. But the average number of goals will decrease, because of the second effect. The two effects work in opposite directions. |
|
By:
My numbers nidicate that Chelsea will score 2,47 goals (81 % av 3,05) and Marseille 0,58 goals (19 % of 3,05).
---------------------------------------------- I have total goals 3.01 with Chelsea 2.318 and Marseille 0.696. Those figures are simply derived from match odds and over odds. |
|
By:
Thanks. This was very helpfull.
According to Kevin Pullein (The definitive guide to fotball betting) the probability of a team scoring goes up with 3 % when being under with one goal, compared to when the result is 0-0. If that is correct this effect is not big enough to explain why the market odds is that far away from the numbers I reach. Have you explored the cean sheet market? Do you feel it's mostly correct? |
|
By:
One of the problems of Poisson is, that it assumes that the number of goals of the favorite is independent from the number of goals of the underdog. That isn't true. So if there is a clear favorite, there is a big probability they will score some goals. And when the favorite scores some goals, it is more likely that the underdog also scores, compared with the Poisson distribution.
Or, when the favorite has a clear advantage (2-0, 3-0, ...), their defensive capacities go down, because of that advantage. ------------------------------------ Even if there's not a clear favourite there's always a big probability that a goal will be scored. But I guess that once there is a pretty high probability that a team will lead with two or more goals they will start to relax. Then they can accept a goal being scored against them and still win. |
|
By:
According to Kevin Pullein (The definitive guide to fotball betting) the probability of a team scoring goes up with 3 % when being under with one goal, compared to when the result is 0-0. If that is correct this effect is not big enough to explain why the market odds is that far away from the numbers I reach.
Have you explored the clean sheet market? Do you feel it's mostly correct? +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ There is some dependence between the number of goals of both teams. When there is a big difference between both number of goals (like 3-0, 1-4, ...) the probability that the difference is reduced is bigger than with 2 independent Poisson distributions indicate. How much the differences will change, I don't know. I don't know much about the clean sheet market. I just look at what I think are the 4 main markets (Match Odds, Under Over 2.5, Correct Score and Half Time/Full Time). Mainly because those markets are so liquid. When loads of people bet, the odds give a good indication what the market thinks. I assume however, that a less liquid market like the clean sheet market has flaws in it. And people also seem to make errors when they bet on games with big favorites and underdogs. |
|
By:
I have total goals 3.01 with Chelsea 2.318 and Marseille 0.696. Those figures are simply derived from match odds and over odds.
--------------------- Right now my numbers are Chelsea 2,39 and Marseille 0,62 goals. Derived from match odds and over/under odds. Total goals around 3. The market seems to agree on clean sheet Marseille. On Chelsea however the market totally disagree. My odds says 1,85 on yes for clean sheet Chelsea and 2,18 on no. The market odds are in fact the excact opposite. I realize that I'm not doing this correctly. At the same time this seems to be very hard to do correctly. There seems to be no excact way to do this correct. It depends on so many factors. I wonder how correct the market is.. |
|
By:
I realize that I'm not doing this correctly. At the same time this seems to be very hard to do correctly. There seems to be no excact way to do this correct. It depends on so many factors. I wonder how correct the market is..
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Maybe you found something profitable. Just like you wrote, it is very hard to correctly estimate it. So a good way is to gamble 100 times with your method and see how much you will earn (or loose). |