Let’s start by looking at my banks at the start of the season and how I bet the model.
Betting Banks
I followed both the Head to Head and the Line models. Both had a $125,000 bank dedicated to them but the Kelly fractional was different for each one. The Head to Head betting will result in a lot more variance as we will be betting on longer odds. This season my Kelly for the H2H bets was 1/10th. Line betting has much less variance as the average odds are around $1.95. My Kelly for Line betting was 1/6th.
This is the only model I use a moving bank with. At the end of each week I tally up the profit and loss and increase or decrease my bank accordingly. Although I do not go below the $125,000 initial bank. So if I lose in the first week, the bank remains at $125,000.
The average bet per week was about $34,000 and as the bets were split into 3 separate days, I never needed more than $30,000 each week to place the bets.
Placing Bets
As I have mentioned numerous times, this model is the world’s best and there are a lot of big (much bigger than me) bettors following this model. What that means is as soon as the tips come out, millions of dollars bombard the bookies. This causes very quick price movements. I have a system in place to make sure I can get the best possible odds I can at time of release and it works fairly well. I was able to match or get better lines in the majority of cases this year.
The service simply records the odds at Pinnacle at the time of release. Pinnacle this season increased the limits to $21,000 for line bets. But by planning early, many aussie books have better lines.
There are some aussie books who will still take a decent bet on the AFL (with bowler accounts) on game day. But the vast majority of my bets were placed at Pinnacle. The reasons are pretty simple, I know for a fact my bet will be taken (if I am quick enough) and don’t have to sit around and wait for a trader to accept or limit my bet. If I have to wait, all other books will have moved their lines by the time I get an answer, so I simply won’t bet at any place that does this.
With Pinnacles increased limits, this season was probably one of the easiest it has been to get bets down. I bet line bets first and then move onto the H2H bets, so my H2H results should be a little worse.
If you can’t be at your computer at 11am Friday through Sunday you won’t be able to match the lines and odds. But a few forum members said they did very well with rebounding lines this season.
It’s a shame all the members are competing with each other every weekend, as if we worked together we would all average much better results. Take for example the Pinnacle limits. They are $21,000. My average line bet is about $4,000. If I get in before everyone else and place that $4,000 at say $1.952 on Pinnacle, the price will drop to $1.92, each subsequent bet also drops the price. The price eventually moves to one that makes the other books look much better. Whereas we as a whole could have bet $40-60,000 at very nice prices, instead we as a whole only maybe put $10-20,000 on. There is a lot of wastage there. I believe a Syndicate model is still the future of this service.
Results
The combined results were a total bet amount of $802,203.00 for a profit of $67,818.51. That’s a return on investment of 8.45%. I dream of a long term ROI of 2-3% so as per usual, the AFL model far exceeds my dreams.
AFL H2H:
Bet amount: $184,837
Profit: $16,134.64
ROI: 8.73%
Bets won: 47
Bets Placed: 100
Win %: 47%
AFL Line:
Bet amount: $617,366
Profit: $51,683.87
ROI: 8.37%
Bets won: 89
Bets Placed: 165
Win %: 53.9%
As we can see, both the H2H and Line models both returned over 8% this season. I bet 3.3 times less on the H2H bets, something I will need to correct for next season as I would like to bring that closer to 2 times less. The Kelly for H2H bets will come down to 1/8th next season.
By looking at the graphs we can also see that July seems to be the month where profits stop. This is a tiny data set, but I believe I have seen similar patterns in the past, and we know other SP models struggle later in the season. It might be an option to stop following in the final 2 months of the season. I would like to see a full breakdown of each months performance over the last 10 years before I make this decision.
How did I compare to official results?
The official results are recorded at the Pinnacle price at time of release. The Pinnacle price are not usually the best odds available. This is shown when we compare my odds to the official results. It is always pleasing when your results beat the official results, as it means the odds recording is more than fair. Something very rarely found in the Tipster world.
Also note that the official results include all bets and I choose not to bet in the final round of the season, which saved me a fortune. Jonno did tell all subscribers not to bet, but does record the bets. It also includes extra bets that I placed when I found better odds and lines. As you can see from the line betting results, I found almost 30 more bets over the season.
The official results for H2H betting was a 2.2% ROI over 108 bets. My return was 8.73% ROI over 100 bets.
The official results for Line betting was a 6.3% ROI over 138 bets. My return was 8.37% ROI over 165 bets
Conclusion and next years plan
I know it sounds outrageous, but this season was satisfactory when comparing it to the last few seasons. The H2H bets have averaged a 10.7% ROI over 16 seasons and the Line bets have averaged 12.1%. You won’t here me complaining about an 8.45% return though.
It’s a shame there are not more games over a season. If this models performance was replicated on an american sport that has games everyday, I would have made my millions years ago. I can increase my stakes, but I believe at best I can double my bets before it becomes a real issue to get on at the current lines. Off course with such a huge edge, I can bet at lower odds for higher stakes and make a larger return overall.
Next season I will roll over my profits and the starting banks for both system will be $200,000.