Author Topic: profitable strategy  (Read 13099 times)

Tags:
  • All members
  • Posts: 145
  • Gender: Male
Re: profitable strategy
« Reply #15 on: 13 May 2019, 22:07 »
There are more than one extreme promissing strategies, not that useless that most available. I may tell you and ask you to test some. But would you spend hours daily, staring at your PC? Nothing is free. I may introduce you into my projects but if I find out you don't want to work hard and you just want everything on golden plate, we are done. You won't get better offer anyway.

  • All members
  • Posts: 126
Re: profitable strategy
« Reply #16 on: 13 May 2019, 22:24 »
There are more than one extreme promissing strategies, not that useless that most available. I may tell you and ask you to test some. But would you spend hours daily, staring at your PC? Nothing is free. I may introduce you into my projects but if I find out you don't want to work hard and you just want everything on golden plate, we are done. You won't get better offer anyway.
analyzing data from bigdata is my job, I'm an analyst I might be surprised by the results I can extract.  I would be in favor of being part of a project by sharing all my experience as an analyst and developer

  • All members
  • Posts: 44
Re: profitable strategy
« Reply #17 on: 20 May 2019, 02:25 »
Hi SartuX,

[I've been trying marketfeeder for a few months and haven't found a profitable strategy yet. do you have any advice?]

I guess my advice would be similar for someone who asked: "How do I make money doing dentistry?"

Intensive study in the field and attain the necessary theoretical and practical knowledge.

You are on the way already as a data analyst (the theory) now next stage is how to apply the theory in practice to your betting interest.

First propose a model with some common sense features/variables and the response/outcome variable. Set up training, testing and validation datasets as you are already used to doing this in other areas.

Try to find input variables that the market has not already factored in. Don't make the mistake of throwing a huge number of of features into a model and hoping for the best. Start with a very small number of variables that make sense in your area of interest. IGNORE their relationship to the response at this early stage of the model building exercise. For example I know very little about football/soccer but if I was asked to develop a model I might start with: Avg cost of players, Avg home crowd size, League position last couple of years, distance travelled to match.

In my case I use the R statistical language and mostly gradient boosting algorithms like xgboost.

If you follow the competitions such as those run by Kaggle mostly these algorithms feature amongst the winners. Winning competitors often provide detailed methods on how they approached the competition task. You may be able to get clues as to how to apply their methods to your area of betting interest.

  • All members
  • Posts: 126
Re: profitable strategy
« Reply #18 on: 30 May 2019, 23:22 »
the studio pays off :) the big data is my daily bread

  • All members
  • Posts: 126
Re: profitable strategy
« Reply #19 on: 30 May 2019, 23:27 »
carry an excerpt of the play logs (I don't send them all because they would be many and I wouldn't bore you)

  • All members
  • Posts: 44
Re: profitable strategy
« Reply #20 on: 31 May 2019, 02:14 »
Hi SartuX,

The graphics look interesting but my tiny brain struggles to understand them.

Are they profit/loss results from arbitrage? In any case they look very good!

Are you using a statistical or machine learning model to make betting decisions?

I am watching the French Open but have never bet on any tennis match.

The women's draw looks unclear but in the men's how does anyone stop Nadal?
Tres difficile (excuse my high school French, disastrous)

  • All members
  • Posts: 126
Re: profitable strategy
« Reply #21 on: 31 May 2019, 06:51 »
Hi SartuX,

The graphics look interesting but my tiny brain struggles to understand them.

Are they profit/loss results from arbitrage? In any case they look very good!

Are you using a statistical or machine learning model to make betting decisions?

I am watching the French Open but have never bet on any tennis match.

The women's draw looks unclear but in the men's how does anyone stop Nadal?
Tres difficile (excuse my high school French, disastrous)
Hi,
as you can see I have selected to display only the results (yellow) otherwise we would not have seen the global trend.
I do arbitration, I have studied in depth over 3000 games with the relative share trends and I have created the necessary triggers.
to verify the strategy I first launched the test and then with the montecarlo method verified that it was not variance ... it will not be the holy grail but it goes close
the errors that you unfortunately see derive from a bug that I didn't continue to follow the game but it stopped only in step 1, in practice they would be 0 errors in 2 days on over 150 markets played, but I left them just to show that it is a sustainable strategy.
However I am adopting the strategies learned in years of financial trading on the movements of tennis, which seems unreachable but instead is the sport that lends itself to trading.
The next step, when the roland garros will end, will be to no longer enter fixed but progressively, to optimize profits

  • All members
  • Posts: 126
Re: profitable strategy
« Reply #22 on: 31 May 2019, 12:05 »
Report Day#3

  • All members
  • Posts: 126
Re: profitable strategy
« Reply #23 on: 31 May 2019, 13:24 »
I improved the input and output signals, I think that on 140 markets (over 400 inputs) it is going well :)

  • All members
  • Posts: 23
Re: profitable strategy
« Reply #24 on: 02 Jun 2019, 11:21 »
Hi SartuX,

Are you placing you bets in-play?

  • All members
  • Posts: 23
Re: profitable strategy
« Reply #25 on: 02 Jun 2019, 12:47 »
If a should speculate there would be more opportunities to find value in the inplay market relative the premarket. Although I personally try to find value in the more static premarket.

  • All members
  • Posts: 126
Re: profitable strategy
« Reply #26 on: 02 Jun 2019, 13:11 »
Trading only tennis inplay

  • All members
  • Posts: 23
Re: profitable strategy
« Reply #27 on: 02 Jun 2019, 13:37 »
I wounder how efficient is the soccer match market?  Has anyone checked it with time machine for exemple by placing manny bets for home team for markets having back_book value relativly close to 100 and check the profit, for a fully effective market the expekted vaue should be around zero.

  • All members
  • Posts: 12
  • Gender: Male
Re: profitable strategy
« Reply #28 on: 02 Jun 2019, 13:48 »
I wounder how efficient is the soccer match market?  Has anyone checked it with time machine for exemple by placing manny bets for home team for markets having back_book value relativly close to 100 and check the profit, for a fully effective market the expekted vaue should be around zero.

I personally don't think so, as the match odds market is so popular and once the over-round starts to exceed 99.00%+, the market is going to self correct itself by either traders or Betfair themselves.
If this was like 2009, then you probably would likely find blips that exceed 100% now and again.

  • All members
  • Posts: 23
Re: profitable strategy
« Reply #29 on: 02 Jun 2019, 13:59 »
I wounder how efficient is the soccer match market?  Has anyone checked it with time machine for exemple by placing manny bets for home team for markets having back_book value relativly close to 100 and check the profit, for a fully effective market the expekted vaue should be around zero.

I personally don't think so, as the match odds market is so popular and once the over-round starts to exceed 99.00%+, the market is going to self correct itself by either traders or Betfair themselves.
If this was like 2009, then you probably would likely find blips that exceed 100% now and again.
I am not talking about arbitrage, I am intresting to see how good the market odds is in predicting the "true odds", by simply placing lots of back bets for example the home team in the soccer market and check the net profit. In theory(fully efficient market) the the net profit should be around zero, but is it actually that?

 

Please note, BetFair is seems to be currently OFFLINE