Losing my marbles......

As the title suggests, I’m clearly in need of a break very soon!

I made an error last weekend where I somehow (don’t ask me how!) managed to only post the bets for 3 systems instead of all 20 systems.  Considering it’s a copy and paste job, I must have just selected the systems and thought I was at the top of the page and copied and pasted and somehow missed the fact there were very few bets. I’ve got Steve to thank for pointing that out although I suspect a few others may have done so on Friday evening but considering I was out, they would have been left in limbo until Saturday morning.

Yesterday, I sent an email out to subscribers confirming there were no bets this midweek but also that I was going to have to change the time of the email this week to 8pm on Friday.  No issue there but I then came on the blog and said I’d post up the bets on here at 6pm on Friday!  

I’ve Les to thank this time as he kindly pointed out that one of the timings is clearly wrong. :) 

The 8pm time is correct as I won’t even have the bets done by 6pm on Friday.  Therefore, I’ll need to post the bets on the blog on Saturday morning.

I’m hoping that if I give the susbcribers a full 14 hours of a headstart, everyone will be on the bets by then (even my Aussie followers!). Therefore, the bets will appear on Saturday morning at 10am on the blog.

In case you don’t read the comments to the posts, there was an interesting comment on my recent results post from Fizzer

“Good analysis as always Graeme.

I think algorithm 2 has a history of giving good results. My view is that the selections it has that don't go into the 6-21 combo are generally better than the selections algorithm 1 makes that don't go into the combo, so it’s the one I follow.

Good luck with final weeks.”

Being honest, until this season, I hadn’t really noticed a big difference between algorithm one and two.  Last season, System 6 had an ROI of 4.3% across 573 bets and system 21 had an ROI of 4.7% across 489 bets. 

However, now I look back at my notes and the changes made in the Summer last year, dropping all short priced aways and restricting Premiership Aways was always going to impact the second algorithm more as this algorithm played with shorter away bets than the first algorithm anyway. In addition, the home bets on the second algorithm have always been better than the home bets on the first algorithm, so now that I’ve improved the aways a bit on the second algorithm, it makes sense that it has performed even better than algorithm one this season.

As for Fizzer’s comment about the bets on 21 that don’t appear on 6-21 being better than the system 6 bets that don’t appear on 6-21, he’s spot on.  However, I’d not recommend anyone using this strategy if I’m honest as the results for the teams on either system that don’t appear on 6-21 aren’t great which isn’t surprising.

This season, there have been 212 bets that appeared on system 6 that didn’t appear on system 21.  Can you guess the profit from backing these?  No profit I’m afraid but a mighty big loss. A loss of 25pts from these 212 bets.

There have been 81 bets that have appeared on 21 that haven’t appeared on system 6. It’s a profit of 4pts from these 81 points (they were loss making before last weekend!). 

Therefore, although Fizzer is right that the system 21 is better than 6, backing teams that don’t appear on 6-21 isn’t a strategy that anyone should be using looking at the returns this season.  Looking back historically, 21 and 6 were both loss making using the same idea last season, so it’s definitely not a strategy anyone should be using!

I remember in the first year of my systems last season, Andrew (The Value Bettor) decided to take a small interest in the systems.  His idea was to back 1pt on a team that appeared on system 6 or 21.  As you can probably tell from reading the above, this isn’t a good idea!  Hence, he was staking 1pt on a team that only appeared on system 6 say and yet, he was staking the same 1pt on a team that appeared on every system.  We all know that this sort of level staking can never work as the bad bets will pull down any profits you make from the good bets.  Needless to say, system 6 was in profit, system 21 was in profit and yet, Andrew’s strategy was losing!

I’ve discussed this before on the blog but it is VERY easy to turn a winning system into a losing system if you try to do something that goes against what the systems are trying to do. Andrew’s strategy in effect was picking out the worst bets every weekend (along with some good bets) but clearly, this strategy is never going to work as there are more weak bets than strong bets in a season!

If any people reading this do subscribe to the service for next season and want to look at things like this during the summer, I’m happy enough to help out with the analysis in order for them to find a strategy that’s optimal.  I’d much rather they used my analytical skills to get something that worked rather than making something up themselves and end up losing whilst others are winning.