First I want to report what was the result of Chi-Square testing for my over/under strategy. For 272 matches which qualified for the strategy, test has shown us that there is slightly over 5% chance that my strategy is the result of luck (normal distribution) rather than edge.
That doesn't sound like a lot but in statistics it is actually on the edge of statistical relevance.
But we have to take in the account that this was a simple Chi-Square test which doesn't take in the account what is the edge for each individual bet!
So for example it doesn't take in the account when fair odds are 1.4 and market odds are 1.9 and Kelly bet is 39% of the bank or when indicated fair odds are 1.4 market odds are 1.6 and we should bet only 20% of the bank. It looks only at the average edge and treats both tips only either as a win or loss.
So taking this in the account and adjusting results for edge differentiation we can be confident that actual chance of no statistical edge in our strategy is under 1%.
On 15th of March we had first bet on the match which qualified for our strategy. Indicated bet was on over 2.5 with odds of 1.83 between Man U vs Tottenham and already after 34 minutes of play when Man U increased their lead to 3-0 the bet was settled as a win.
I write these posts at least once per two weeks, primarily for myself as a part of my accountability.
But I was asked to share more about my bank management. So from now on I will not only post tips for this strategy but also exactly how many units you should bet.
I use Kelly criterion formula for this betting strategy.
So for example last tip was on over 2.5 at odds of 1.83 and according to Kelly criterion bet had such an edge that it was worth 0.36 (36.1%) of our starting bank.You can do partial Kelly bet to decrease risk to 1/4 or more as I do.
How do I calculate my bet?
I decided that starting bank for this will be 3000€ and I will always bet 1/4 of full bet according to Kelly criterion.
So 3000/100*36.1 = 1083€ which would be the full bet according to Kelly criterion.
I divide that by 4 which in my case gives me bet of 270.75€ I should risk.
You can see from my P&L statement from a certain bookie that my bet for this match was 250€ and the corresponding profit was 207,98€. I did bet another 21€ at another bookie with a profit of 17.64 for a total bet of 271 euros.
Starting bank: 3000€
Total profit on 15/3/2015: 226€
ROI: 83.3%
Final bank on 15/3/2015: 3226€
Remember, ROI will drop a lot with a time BUT our bank will also grow.
One more thing: the odds before the match actually moved out from 1.83 to 1.95 on Betfair. I don't know the reason of the movement but remember this: one thing is odds movement and another thing, completely different, is actual chance of winning!
So don't expect that the odds will always come down on this strategy, they might go in our favor and make the edge even bigger.
I will always calculate the edge and amount to bet at least one day in advance so there is still a lot of time for bet placement.
Also I'm always interested in valuable edges and strategies!
All the best!
Saturday, 28 March 2015
Saturday, 14 March 2015
Looking for the edges
In last two weeks I made quite a good progress. I did some further statistical studies and testing and I have found 2 solid tipsters which I started follow.
Tomorrow will be there the first match (Man U - Tottenham) which will qualify for my own betting strategy I was working on recently. According to my model fair odds for over 2,5 should be in 1.5-1.6 range, they are 1,83 instead so I will take an advantage.
What kind of betting management will I use?
I will go for partial Kelly stake which means I will bet about 1/3 of what full Kelly stake would be.
My plan for week ahead is to find some more edges, especially external which are much easier to find than going through backtesting. When I exhaust external sources I will dig through the stats.
There is so much opportunity in this industry, you just have to know what you want to do, do your work, be disciplined and the rest will take care of itself.
Good luck
Tomorrow will be there the first match (Man U - Tottenham) which will qualify for my own betting strategy I was working on recently. According to my model fair odds for over 2,5 should be in 1.5-1.6 range, they are 1,83 instead so I will take an advantage.
What kind of betting management will I use?
I will go for partial Kelly stake which means I will bet about 1/3 of what full Kelly stake would be.
My plan for week ahead is to find some more edges, especially external which are much easier to find than going through backtesting. When I exhaust external sources I will dig through the stats.
There is so much opportunity in this industry, you just have to know what you want to do, do your work, be disciplined and the rest will take care of itself.
Good luck
Saturday, 28 February 2015
Where I stand?
So as I have mentioned I was sports arber for 8 years. While that was working very well for me I didn't really looked into other betting strategies. Since 2011 my profits decreased, I was spending more than I earned and I started looking into other areas.
I was hoping to build a betting bot but I didn't reach anything consistently profitable. To be honest I was looking for quick and easy fix and was too much spoiled by my previous successes.
Since last summer I decided to follow value betting approach, and started looking more into difference between prices offered by the markets and fairer prices deducted from statistics.
I have developed statistical system for English Premier league and back-tested it for last 6 seasons. Using Kelly formula as a staking method it came with profit of 900%, if starting every year with the same bank.
What is my plan now?
I want to find out what is the chance, if at all, that profit from this strategy (272 matches) could be achieved just by a chance. Then I want to put this strategy to practice.
Another edge I have is external, I'm following one prediction service which gives yearly about 30% on bank.
My bank diminished over the years to about 4000€. Now I want to find another edges which will give me the option to grow it.
For now I'm focused only on statistical analysis and pre-match odds. I will do a blogpost about my progress once per 2 weeks.
Good luck to all!
I was hoping to build a betting bot but I didn't reach anything consistently profitable. To be honest I was looking for quick and easy fix and was too much spoiled by my previous successes.
Since last summer I decided to follow value betting approach, and started looking more into difference between prices offered by the markets and fairer prices deducted from statistics.
I have developed statistical system for English Premier league and back-tested it for last 6 seasons. Using Kelly formula as a staking method it came with profit of 900%, if starting every year with the same bank.
What is my plan now?
I want to find out what is the chance, if at all, that profit from this strategy (272 matches) could be achieved just by a chance. Then I want to put this strategy to practice.
Another edge I have is external, I'm following one prediction service which gives yearly about 30% on bank.
My bank diminished over the years to about 4000€. Now I want to find another edges which will give me the option to grow it.
For now I'm focused only on statistical analysis and pre-match odds. I will do a blogpost about my progress once per 2 weeks.
Good luck to all!
Friday, 16 January 2015
Welcome!
Hi, I'm Alchemist and this is my betting blog. For last 8 years I have been sports arber and now I'm dedicaded to find another ways how to profit from betting. This blog will document my progress. I will report here on my P/L and also post about strategies.
Subscribe to:
Posts (Atom)
