Ansell home win system

[KThom]

I did try and start a new system thread a while back but have been waylaid with work, personal life and lack of internet connection for a few months. It has given me some more time to concentrate on the dynamics of my system anyhow. The system is based around picking home teams to win only, generally at odds of evens or better. I have some score prediction software which has enabled me to select games which fit a certain criteria generated by recent home form, h2h meetings and league statistics. From these selections (across a multitude of leagues) I have been able to back test the results using odds from football-data. Originally the plan was to start the system using flat stakes (which shows profits as below)

Flat Stakes Using Bet 365
Tot Stake	£53,800.00
Avg Stake	£100.00
Profit	£3,148.00
Yield	5.85%
S/R	55.39%
Games	538
Wins	298
Avg Odds	1.97
Lose Streak	8
Win Streak	10
Low	-£80.00
High	£3,148.00
Season Low	-£1,023.00
Low Year	2007
Season High	£1,398.00
High Year	2009

Flat Stakes using best odds
Tot Stake	£53,800.00
Avg Stake	£100.00
Profit	£5,747.00
Yield	10.68%
S/R	55.39%
Games	538
Wins	298
Avg Odds	2.06
Lose Streak	8
Win Streak	10
Low	-£75.00
High	£5,747.00
Season Low	-£774.00
Low Year	2007
Season High	£1,719.00
High Year	2009

But then I started considering other staking plans. It is clear from the above that the strike rate is a pretty constant 55.5% - and this is consistent from season to season by league to league. The above stats are pretty good. Using the best odds available (according to football-data) a yield of 10.68% shows after 538 selections giving a 57 point profit (since 2004-05 season). There were / are however long “dry” periods where the bank remains flat, or doesn’t increase at a quick enough rate for my liking. I then started measuring the same selections using the Kelly and Ansell staking criteria. In both instances I used 55.5% as the “true chance” figure – which basically means any selection with odds of 1.8 or lower was excluded. It also meant that the higher the odds – the higher the stake. The results are more promising. Even using the bet365 odds alone a significant profit and yield is noted, which is steady season on season.

Ansell Stakes using bet 365
Tot Stake	£55,627.55
Avg Stake	£175.48
Profit	£14,674.21
Yield	26.38%
S/R	51.42%
Games	317
Wins	163
Avg Odds	2.26
Lose Streak	6
Win Streak	7
Low	£182.33
High	£14,674.21
Season Low	-£2,228.54
Low Year	2007
Season High	£4,457.30
High Year	2005

I know there are a lot of pro’s and con’s where the Kelly v Ansell criteria is discussed, but I personally favour the Ansell criteria – unless anyone can sway me otherwise. Now I understand that what has happened in the past has no reflection on what will happen in the future – and I can possibly be accused of back fitting. The last time I half started this thread with flat stakes the few selections resulted in a poor return – but had I had the time and resource to continue a profit would have been made from that point until now. The results to me look smooth, consistent and more importantly profitable. There may be a more significant way of measuring my selections and data. For the mean time I will post my selections, showing the flat stake (always £100) and the Ansell stake (if best odds exceed 1.80). At the moment the system is in the middle of a fantastic run, so even more reason for my selections to go belly up once posted on here :-P Anyway, one selections for tonight. Blackpool (H) vs Liverpool. Odds 3.80 £100 level stake, £762.44 Ansell Stake I will run the bank – and results as of this fixture because running it from 2004 is not fair as the results (although accurate) were never published.

[Pepen]

Re: Ansell home win system Hi KThom! First of all good luck with this (semi)new system of yours, I remember following your correct score thread and the previous Ansell home wins some time ago..was sad to see 'em disappear If you don't mind me asking, is the software you mention of your own making or something you've gotten elsewhere? I seem to recall you being full of statistical and programming knowledge. Another question for your selection criteria is the H2H meetings. Personally, I don't see why the H2H should be significant in a system, based on the heavy fluctuations in some clubs with regards to the managers (thereby the changing tactics) and players etc? For example, wouldn't a lot of your selections with Liverpool being one of the sides be biased towards Liverpool, if the H2H's have anything to say (due to them being much better just a few years back)? I know there are other variables influencing your calculation of the 'true odds' but I'd imagine there was some negative effect from the H2H's.. Sorry for the load of questions, I just thought I'd block your thread before the horde of worshipping followers arrive when the system gets profitable :)