Thoughts on September

I thought it was worth doing a quick follow up post to my usual monthly results post discussing September. I initially set out with the intention of looking at the performance of September and comparing it to the historical months but being honest, September is so bad, I'd be wasting my time. When you have systems that are running at an ROI of -50% and worse, trying to do any sort of statistical inference is a waste of time.

Everyone knows I'm playing on a market where the bookmakers have an edge of around 6%-7% and therefore, over an infinite number of bets, a trained monkey will lose 6%-7%. In the short-term though, a trained monkey and a system with a big edge can lose or win much more than 6% or 7% and I guess that's what we've witnessed in September. Either that or I've created a great lay system after 2 years of work and analysis!

I know there are a lot of systems and ways to play the bets I post on here but in the Summer, I posted up a table in the SBC forum which went down quite well with the members across there. Ultimately, the SBC felt I stepped over the mark a little by comparing my systems last season to all the other subscription footie tipsters (fair comment!) but I think this table is something that punters can relate to. Last season, I ended up with some of the highest performing systems (based on ROI) in the whole tipster market and therefore, it allowed a fair comparison to be made between my free systems and other subscription systems in the market.

Admittedly, September was a shocking month for my systems with all systems losing but September counts for just over 10% of the total season. Therefore, it's too early to draw any meaningful conclusions compared to previous seasons and much too early to say that whatever worked last season, isn't working this season.

The table below is an update of that table I showed in the SBC forum in the Summer. Of course, I've removed all reference to other football tipsters but if anyone is following other guys, you can slot in their September results for comparison. I know from guys I correspond with that the main football ratings services struggled in September, so I expect their results to look as bad as mines!

There are lots of ways to look at this table. I've shown the results for each proofed system from last season and simply greyed out the systems that weren't alive last season.

I've added a Total at the bottom which doesn't mean much (there are more systems this season than last season) but the Average of the systems is a good indication of what happened in September I think and allows comparisons to be made to last year.

I've also ranked the results by the ROI from September, so the better performing systems are at the top and the poorer performing systems are at the bottom.

So, what does it show?

Well, firstly, the systems with the least number of bets performed the worst in September. Although you don't see it immediately from looking at the table, the systems which performed worst last year (the less selective systems) performed best in September. Again, this is linked to the number of bets and over a smaller number of bets, anything can happen.

Secondly, due to the fact the ROCE is negative for each system, then ROCE is therefore also the proportion of each bank that has been lost by the system in September.

System 8 DNB lost 38% of its betting bank in September. On average, the systems lost 17% of their betting bank in September.

Lastly, the strike rate in September was around half of the strike rate last season. Simply, not enough teams won in September but importantly, they didn't draw either!

Overall, I'll leave others to decipher how the systems performed in September. Whatever way you look at it, it isn't pretty but losing 20%-25% of your betting bank isn't a disaster.