Following on from the Premier League Fund, I have adapted the algorithm to the Championship.  As well as adding profits, the addition of a second Fund should also reduce the overall volatility of total invested funds.  This is Markowitz portfolio theory working its statistical magic.  Volatility is reduced since the performance of the Premier League and Championship models are completely uncorrelated.

Continued

Still a few weeks to go until the start of the Premier League for 08/09 but I thought I would lay out the expectations for the performance of the Funds for next season. 

After improving the algorithm, the greatest decision i’ve had to make is what level of risk to take.  On the one side, increasing stakes lead to higher returns but at the same time adds to the volatility of the Fund.  The biggest change in the current model over the model used for last season is a vastly reduced stake size.  The question is, how much volatility am I now comfortable with?

Continued

Following the success of last season, my thoughts obviously turned to whether this can be repeated. Or more accurately, what can be expected from the algorithm betting model in future seasons both in terms of return and the variability of those returns.

“Monte Carlo” is the name given to simulations which make use of computer generated random numbers to identify the range of possible outputs a model may generate in the ‘real world’.  Each random number generates an input from a user defined probability distribution which is run through the user’s model to produce a simulated outcome on each run.  Run this simulation thousands of times and you generate a probability distribution for the output of your model.

Continued

Well, that’s it; the end of the season. A very profitable one apart from this final week. Whilst there was no financial gain for the Balanced Fund in this last week there was a lot to learn from the divergence in the market’s and model’s odds on the relegation and title games.

Quite simply the Balanced Fund model took no account of the impact of the final week relegation and title battle. Hence it was drawn into betting against Birmingham, Fulham and Reading who fought against relegation. The same went for Everton and Villa who were fighting for the last European place. It also backed Wigan and Bolton against Man U and Chelsea as they fought for the title. In all, 7 matches were influenced by the relegation/title race which the market took account of in it’s odds.  The Balanced Fund lost on 5 of these 7 bets. Continued

The Newcastle fight-back against Spurs cost me this weekend. Looks like Kev’ means business after all.

The Aggressive Fund stood by it’s name as it lost almost 12% this week, though it did gain 24% for the month of March. This fund is an incarnation of the algorithm which started in September 07. It hasn’t changed much since and I don’t amend it now, I keep it going for the sake of history. I don’t think it will ever get back on target as, in hindsight, it was too structured toward historic results; i.e. ignored ‘what if’ scenarios.
Continued

How useful is yield as a performance measure?

There are many variables to measure the success of a betting system.  These range from the simple ’strike rate’ showing the percentage of bets won to ‘yield’ which, by incorporating the potential loss on a bet, includes some measure of risk.  It goes without saying that the longer a system has been in place and the more bets it has covered the more reliable it is in terms of predicting future results.   However given the same number of bets, a system with a high yield is not necessarily better than a system with a lower yield.  Betting against Derby is a low yielding bet since the risk of loss is low.  However if such bets offer positive value (i.e. net profit over time) and we exclude them simply because they are low yield we are giving up potential profits.

Continued

I’ve been tinkering with the algorithm model over the past week or so.  The result is lower risk and higher yield - the perfect combination.  This is of course based on the results output from the new algorithm.  In some way this goes without saying but I think it’s always good to remember this point.  If the model’s wrong then performance will be worse than predicted.  How much worse?  The trouble is you cannot measure what you don’t know!
Continued

There are no changes in the top 9 ranking teams this week with Liverpool remaining top and Derby bottom. The most significant move is Newcastle falling to 19th place after their 6-0 loss to Man U. This may seem overdone but while they are without a manager they are there for the opposition to knock over. So, whilst the algorithm doesn’t incorporate the ‘no-manager’ status the rating does reflect the form.
Continued

With the season half over I have taken a look at how the model’s ranking of teams compares with their Premier league position.

As can be seen in the table below most teams’ league position is similar to their algorithm ranking.  A ranking above the league position suggests all flare and no substance or plain bad luck.  For a team to be unlucky over 20 mathces however is unlikely.  It is more likely to suggest therefore a good team with lots of chances but lacking the finishing ability or unable to close out a match.  Conversely a team with a league position above their ranking shows an ability to get results without creating a lot of chances or dominating a game.
Continued

The Alpha Fund used level stakes and was a pre-cursor to the percentage stakes Beta Fund.  Results over the past month have shown the Beta Fund to be superior and continuing to monitor the Alpha Fund serves little purpose.  I am therefore going to drop coverage of the Alpha Fund and the Beta will just be referred to as ‘the Fund’.

This week once again saw moderate gains.  The 6 winning match bets were soured mainly by away wins for Everton and Man City.  These two teams do appear underrated by the model as their relative rankings are low versus the rest of the league consisdering their recent results.
Continued