Monday, 6 October 2014

Underlying UK Results for September 2014

Before I review the month of September for all TFA systems, I thought it would make sense to give a quick overview of the Underlying results for September. Even though I track these results closely, I don’t plan on always posting the underlying results before a monthly review as most of the time, the underlying results will fall within an acceptable range and therefore, whether or not there has been a little bias towards Homes or Aways or even Draws shouldn’t impact my ratings and systems too much. 

I felt like I spent a long time last season discussing the Underlying results due to the Home bias we saw early on but as the season progressed, a correction took place and by season end, results were in an acceptable range I think for all outcomes.  It was definitely easier to make money backing Homes last season but it wasn’t impossible backing Aways or Draws as my ratings showed.

It feels like a big talking point each Summer with TFA is the September performance and before this season, in 4 live seasons, two Septembers had been OK and two Septembers had seen massive losses on the systems.  I know a couple of subscribers decided they’d had enough of poor September performances and decided they would start their TFA season in October!  A brave decision based on 4 data points IMO but then again, everyone has their own opinion at this game and this is what makes this game interesting for me. Two others mentioned playing higher stakes on Homes than Aways in September given Homes tend to have a higher strike rate but again, not something I necessarily agreed with myself.

Now we’ve reached the end of the month, how did September play out? Did we see the ‘expected’ Home bias some were anticipating, did the systems struggle like they have done for 2 of the last 3 seasons and should we seriously consider starting betting in October rather than September with my ratings?

Well, we definitely saw a home bias in September and in the main, the systems have struggled again in the opening month of the season but it’s not a disaster. Of course, if you happen to be following the systems that have performed poorly, it may be a disaster for you personally but on the plus side, I’m aware of others who have had an OK opening month of the season and managed a profit. The purpose of this post is to concentrate on the underlying results and we can worry about the ratings and systems in the next post.

Here’s the results so far for this season compared to previous seasons:

Looking at my algorithms then, 3 of the 4 algorithms have made a loss. A small loss on algorithms 1& 2 whereas algorithm 4 has had a shocking month with a loss of 11.6%. As is so often the case, when underlying results are really tough and the other algorithms have a nightmare, algorithm 3 somehow manages to make a profit. Only a profit of 3.1% this month but it’s the best result of all the algorithms.

Trying to understand these results then, we can look at the underlying results so far. A strike rate of 50.7% on Home bets is absolutely amazing and is the highest strike rate in a month since Feb-10 (which was before my systems went live).  It is the highest strike rate in a month for Homes since my ratings went live in Sep-10. Those with a good memory may remember that in Sep-13, I wrote the exact same comment! Sep-13 had a strike rate of 49.5% for Home bets which was the highest strike rate I’d seen for Homes in a month. Well, Sep-14 has just beaten it!

This strike rate has manifested itself into an ROI of 20.5% if backing all Home bets. A truly stunning number when you think about it. The only month that comes close to this was Sep-13 and that had a 11.6% ROI. This month is nearly twice as profitable he next best month if backing Home bets!

On the flip side then, if Homes have performed like this, how have Aways done?  Well, a strike rate of only 24.7% for Aways which is the lowest strike rate in a month for Aways since my ratings went live. Looking further back, it is the lowest strike rate since Feb-10 which won’t be too surprising given what I said above about Homes in Feb-10.

In terms of ROI, a loss of 20.5% if backing all Aways and the worst month for Aways since my ratings went live. Looking further back, the worst ROI in a month since Nov-09 for Aways.

Clearly, a very tough month for Away backers and the size of the loss depends on how many Home bets appeared on your systems or rating algorithms!

Last but not least, the strike rate for Draws was a lowly 24.7%. This equated to a 10.8% loss. September and April last season were worse than this for Draws and therefore, it wasn’t any more difficult for Draws than it was last September. However, as we’ll see when we come to look at the Draws, it hasn’t been a good month at all for the Draw systems.

I always think the overall ROI if backing Homes/Aways/Draws is slightly meaningless as it only makes sense if you had a similar number of bets in each category.  In the case of TFA, Aways dominate and therefore, my results are more swayed towards the -20% than the +20% or -10% but anyway, with Homes and Aways cancelling each other out in terms of ROI, the overall loss was -10.8%. Again, nothing too bad with this and it is much better than the -18.2% we experienced last September. I think the fact Homes did so well to cancel out the Aways loss has really helped us not get into the same difficulties we got into last September.

Putting the results in context of all monthly results since my ratings and systems went live, the underlying results were a little outside of the norm this month for Homes and Aways (best and worst month ever) but in terms of the losses on the algorithms, not too bad. On average, the rating algorithms tend to have 3 losing months a season. Hence, if this is one of 3 losing months, we should be OK this season.

I’m not sure I’ve shared this table before but below is the overall summary of monthly Underlying results since my ratings went live. It summarises the table above. As you will see, September does stand out as being a good month for Homes and the next 3 months (October, November and December) tend to be very good for Aways. Let’s hope the pattern is repeated this season! (NB - the home strike rate is under 40% so far in October after the weekend games)

Next post will be the review of September for all ratings and systems. Will be up at the backend of this week hopefully.


  1. I was sticking to homes last season, which was fine until mid-January, but pretty poor after that. In January I had a 40% ROI for the year and a maximum drawdown of 4 points. But the position changed and I was losing for the rest of the season. There was also a shortage of bets, so the high ROI wasn't converting to a massive profit and it soon withered away. Maximum drawdown increased to 12 points.This year I'm playing the aways, expecting a far lower ROI, a far greater draw down, but a bigger profit from more bets. It hasn't been a disaster. My points profit on home bets so far is greater than my profit for the whole of last season and, thanks to a few big wins last weekend, I'm only down 6 points on aways. It's the draws I'm worried about. I should have left those alone.

  2. Hi Anon.

    Cheers for the comment.

    Interesting strategy last season to only play Homes! Not sure I was aware of anyone doing this but in theory, with the lower average odds and higher strike rate, it should give a smoother P&L journey. Of course, theory is all good and well but as we saw last season, you can get some crazy runs of results and Homes were amazing until January and then useless thereafter.

    This season, Homes have started even better than last season but it’s early days. For a while, I staked more on Homes than Aways to try to smooth my P&L but I moved away from this approach after the losses I suffered backing Homes during a few months spell. I now use level staking and I’m much more comfortable with this.

    I think you make a good point about the Draws. I’m scratching my head a little about their performance so far but it’s early days. I had the exact same reservations early on last season and they made me look pretty silly later on. Going into the season, I was probably as confident about some of these systems as some other systems even though they had a much lower number of proofed results but my confidence is draining by the week!

    Let’s see if the Draw systems can turn it around….