Draw laying system
- Thread starter paulb164
- Start date
paulb164
Colt
It wasn't a particularly good pre-christmas set of results, but the 5 from 5 results on Boxing day put the P/L lines back up again. The addition of the N Irish league looks so far to be a good one with 100% win records for the 2 games played today.
Results to date since last update;
Results to date since last update;
paulb164
Colt
Friday and Saturday proved to be a very frustrating period for the strategy. Only 6 games played and of which half lost, bringing the two chart lines below £300 again. To compound the issue, the 3 games from the Bolivian league were not included in the Betfair games covered, and therefore I will have to remove them from the results (which was doubly annoying as they all won, and would have kept the lines above £300).
Also it seems to me that whenever I have bets on games that Anderlecht feature, always seem to lose ! They have featured twice in this system, and have resulted in losses for both games. It isn't just here, but in some of my other systems as well - so much so that when I see them crop up in qualifying games I always feel the need check further !
Fingers crossed for today's 6 games.
Also it seems to me that whenever I have bets on games that Anderlecht feature, always seem to lose ! They have featured twice in this system, and have resulted in losses for both games. It isn't just here, but in some of my other systems as well - so much so that when I see them crop up in qualifying games I always feel the need check further !
Fingers crossed for today's 6 games.
paulb164
Colt
Also worth mentioning is that lots of leagues in this system have winter breaks, so for the next week or 2 we won't see as many games qualify. Also, I tend to be wary of leagues returning from their breaks as in most cases they return full strength and with injury-free players that have benefited from having the time off (a bit like injured players not picked for internationals returning for their domestic teams following international breaks).
paulb164
Colt
Liverpool's game on the 2nd Jan almost qualified, and on another day I would expect them to win comfortably and pick them for this system, but they are so due a loss by now, and considering the congested fixture list of the past month (especially since they also played in the Champions cup), I think they are about ready for a fall. Sheffield Utd pushed them hard in their fixture earlier in the season and only just lost by a goal to nil.
I fancy the idea of laying a Liverpool win at 1.31 - incentive for Liverpool though is that if they don't lose, they will have remained unbeaten in the PL for a whole calendar year !
I fancy the idea of laying a Liverpool win at 1.31 - incentive for Liverpool though is that if they don't lose, they will have remained unbeaten in the PL for a whole calendar year !
paulb164
Colt
Well, I got that angle wrong ! Liverpool had a massive 75% possession, and just surprising that they only managed to score 2 goals ! I still think Liverpool are due a loss at some point, and it looked like a good opportunity for Sheffield Utd in 8th place to inflict that (or at least steal a draw). It says a lot for Klopp's motivational management that he's kept the team's desire as high as it has been so far during his stewardship.
paulb164
Colt
Thanks........ Always a nice, neat presentation and a credit to your skills. I had noticed that the whole six came in the other day, and that was great going Paul.
Like all systems, it has good and bad days, but the chart lines are generally going in the right direction. Currently on a nice 11 game win streak. It's run for a little over a month now, and with £10 stakes (not forgetting that the liability is much higher with laying) it would have returned over £300 profit. I'm also approaching 200 results, and another chance for me to analyse the results to see where it can be improved.
paulb164
Colt
Unbelievable ! Barcelona top of the table versus the bottom team Espanyol... winning by a goal and then get themselves a red card. Espanyol gets an equaliser in the 88th minute ! I wish I was watching the stats online and saw the red card when it happened... I'd have greened up there and then !
paulb164
Colt
I mentioned this in a previous post on this thread - if you are keeping an eye on any of these games, and the favourite earns themselves a red card in a winning position, then green up as soon as possible for a little profit. I guess you will also need to use your judgement say for instance the red card comes in the 88th minute on whether you think the favourite can hold out for the win, but if there's at least 10 minutes still to play, my advice is to get out. I would also say the same if the game is still a draw when the red card happens... that might be a small loss, but that doesn't compare to the big loss of the full liability of the bet if the favourite with reduced men don't get a goal because of it.
paulb164
Colt
200 games now played, so another chance to analyse the success rate against the results from around 100 games ago. I have put the most recent comparison chart to the right of the same chart from 100-odd games ago.
100-odd games ago, I identified that my own picks that just fell out of the Q/QQ criteria were losing money. Since I adjusted my own choice criteria, the non-Q rated games have improved greatly and have gone from negative into positive, so I will continue with that new picking process.
Also, I couldn't understand then how the QQ-rated games were performing worse than the Q-rated games, so I adjusted the filters which has now clearly highlighted the QQ-rated games better. I also think that some of the former Q-rated games now qualify as QQ-rated, so that partly explains the proportional reduction in P/L of the Q-rated.
These charts endorse what I mentioned above - better win rate with my own picks, and the system being better at identifying QQ-rated games. Unfortunately Q-rated games dropped in accuracy (from 87% to 83% which because it makes up a high percentage of the system games, is having a significant impact on P/L). I now need to look at the basic filters to improve the Q-rated qualifiers.
The only thing I identified from this chart 100 games ago was that leagues with draw averages of 18% and under performed really well, and this updated chart continues with that assumption. 100% from all 17 games then is now 96% from 21 games. Because of such a good hit rate from just those 21 games, they are responsible for £159 of the current P/L ! I would even say now that it would be advantageous to bet double points on these games in future. Had we done that from the start, the P/L would be just under £500 !
Just out of interest, the 1 game that was a draw was Dundee Utd v Dundee - hard fought local derbies often don't follow the rules and on reflection I maybe shouldn't have included it in this system (though it was a Q qualifier) ?
The above charts further show that firstly just how much games with a perceived 3:0 result are more a part of the qualifying games than before, and also just how good the P/L is when it thinks it's going to be a 4:1 result with a 100% strike rate. Just following this score prediction alone would have given a P/L of £142.50. Another set of games where double points should be bet with ?
The second chart also shows the away predictions which wasn't being looked for in the first 100 selections, which also is showing a profit.
100-odd games ago, I identified that my own picks that just fell out of the Q/QQ criteria were losing money. Since I adjusted my own choice criteria, the non-Q rated games have improved greatly and have gone from negative into positive, so I will continue with that new picking process.
Also, I couldn't understand then how the QQ-rated games were performing worse than the Q-rated games, so I adjusted the filters which has now clearly highlighted the QQ-rated games better. I also think that some of the former Q-rated games now qualify as QQ-rated, so that partly explains the proportional reduction in P/L of the Q-rated.
These charts endorse what I mentioned above - better win rate with my own picks, and the system being better at identifying QQ-rated games. Unfortunately Q-rated games dropped in accuracy (from 87% to 83% which because it makes up a high percentage of the system games, is having a significant impact on P/L). I now need to look at the basic filters to improve the Q-rated qualifiers.
The only thing I identified from this chart 100 games ago was that leagues with draw averages of 18% and under performed really well, and this updated chart continues with that assumption. 100% from all 17 games then is now 96% from 21 games. Because of such a good hit rate from just those 21 games, they are responsible for £159 of the current P/L ! I would even say now that it would be advantageous to bet double points on these games in future. Had we done that from the start, the P/L would be just under £500 !
Just out of interest, the 1 game that was a draw was Dundee Utd v Dundee - hard fought local derbies often don't follow the rules and on reflection I maybe shouldn't have included it in this system (though it was a Q qualifier) ?
The above charts further show that firstly just how much games with a perceived 3:0 result are more a part of the qualifying games than before, and also just how good the P/L is when it thinks it's going to be a 4:1 result with a 100% strike rate. Just following this score prediction alone would have given a P/L of £142.50. Another set of games where double points should be bet with ?
The second chart also shows the away predictions which wasn't being looked for in the first 100 selections, which also is showing a profit.
Attachments
Last edited:
paulb164
Colt
LOL - I did an analysis of what the P/L would have been had we did 2X points on all games from leagues with 18% or less draw average, and also with 4:1 predictions from the start (remembering that 1 point = £10)...
Interesting reading !
All games with the 4:1 prediction were Q/QQ-rated, as is the vast majority of the under-19% draw average games... which is the main reason that the green line is much higher than the red line. As this is the results from 200 games through the system (i.e. a decent amount of results to ascertain it's effectiveness rather than still trialling it), I think from now on, I will post lists which will show Q and non-Q games, and also the 2X point matches.
Interesting reading !
All games with the 4:1 prediction were Q/QQ-rated, as is the vast majority of the under-19% draw average games... which is the main reason that the green line is much higher than the red line. As this is the results from 200 games through the system (i.e. a decent amount of results to ascertain it's effectiveness rather than still trialling it), I think from now on, I will post lists which will show Q and non-Q games, and also the 2X point matches.
Last edited:
paulb164
Colt
Further thoughts as a consequence of reviewing the results over this weekend...
It's very clear that following the cold hard data that the system returns with the Q-rated games is the way to go for the most profitable returns. I have also further refined my non Q-rated picks and that should hopefully add more value to the system. The figures in the last chart show all the games so far included, and if I just include the games that conform to the checking procedures I want to use from now on, both of those lines would be higher still.
The next thing I want to look at is capturing the best odds for the system. With laying, you are open to much higher liabilities (certainly for this system anyway), and so it is extremely advantageous to get the lowest possible lay odds. The below graph taken from Betfair 45 minutes before the game kicks off shows that while it is currently 11.0 odds to lay the draw, it was possible for a long time to get the draw at odds of 9.0 or less. If you have a decent enough pot to allow for tying up the funds ahead of a match, it may be worth placing the bets in the morning of the game. The odds can of course shorten in that time as well, but my instincts tell me where there are overwhelming favourites involved, that occurrence would be very rare.
The game in the below graphic is exactly the reason why this is important to look into - it finally finished 3:3 and so lost as far as the system is concerned. It would obviously be better to lose with lay odds of up to 9.0 rather than the 11.0 it was at the start of the match !
It's very clear that following the cold hard data that the system returns with the Q-rated games is the way to go for the most profitable returns. I have also further refined my non Q-rated picks and that should hopefully add more value to the system. The figures in the last chart show all the games so far included, and if I just include the games that conform to the checking procedures I want to use from now on, both of those lines would be higher still.
The next thing I want to look at is capturing the best odds for the system. With laying, you are open to much higher liabilities (certainly for this system anyway), and so it is extremely advantageous to get the lowest possible lay odds. The below graph taken from Betfair 45 minutes before the game kicks off shows that while it is currently 11.0 odds to lay the draw, it was possible for a long time to get the draw at odds of 9.0 or less. If you have a decent enough pot to allow for tying up the funds ahead of a match, it may be worth placing the bets in the morning of the game. The odds can of course shorten in that time as well, but my instincts tell me where there are overwhelming favourites involved, that occurrence would be very rare.
The game in the below graphic is exactly the reason why this is important to look into - it finally finished 3:3 and so lost as far as the system is concerned. It would obviously be better to lose with lay odds of up to 9.0 rather than the 11.0 it was at the start of the match !