Another year ends and it is time to sum up the efforts of my bot betting. This year has been a very bumpy ride, and includes both good and bad things. Let’s start with my forecast made in January 2017:
The plan for 2017 is to keep the good models for Home and Away, and do some minor changes to Draw and Over/Under to slightly improve their yield. I aim to get all models +1.0% yield. My total yield goal is to improve from 1.24% to 1.30%. A bigger wallet for 2017 will also mean bigger stakes and more turnover. Projecting the turn for the last few months indicates that 2017 could turn around 10 MSEK in total. If both these conditions hold I guess that the bot will generate around 130 000 SEK. That would be very satisfying!
I hoped to turn 10 MSEK (Approx.: 1.200.000 USD / 900.000 GPB) and yield 1.3%. And I got:
I managed to forecast the turnover almost spot on, but the performance of my models were a disappointment: 0.6 % yield as less than half of what I thought was possible. My total P/L for 2017 was 59.856 SEK (Approx.:7.300 USD /5.400 GPB).
Breaking down the results by model:
When I calibrate my models I aim to hit around a yield of 1.0 – 1.5 %, so the only model that did good enough was the Over/Under model. This will of course generate some new models for 2018!
In previous post during 2017 I have compared the development in turnover/profit/yield between 2016 and 2017 as accumulated weekly figures.
The raise in turnover is mainly a result of using a bigger wallet in 2017, as I use “percentage of wallet” as staking method (currently 2.5% for each bet), the bet size in SEK gets bigger as the wallet increases. There were also an increase in number of markets engaged with a bet (approximately 1300 more markets in 2017), so that also helped to reach 10 MSEK.
The P/L graph is much more interesting! Both in 2016 and 2017 I see the same trend (although more significant in 2017)… Great performance in the first 8-10 weeks, and then a down period until week 23-26 something and then great performance again.
The yield graph just confirms the trend….
So, to sum the good things and the bad things of 2017:
(+) Reached the 10 MSEK in turnover
(+) Green all over my models
(+) Made 60.000 SEK (equivalent to a good summer vacation for me, my wife and our two children!)
(-) Low yield
Unfortunately the “Low yield” is a big problem, especially when the yield turns the P/L from green to red… But there are many actions already implemented during the autumn 2017, which I hope will lead to successful betting in 2018.
How will 2018 turn out? As I need to strengthen the yield I will need to reduce the number of expected bets, I forecast around 10.000 bets. A bigger wallet will compensate for this, but the turnover wont have the same growth as previously years – I guess around 12 MSEK. . The yield? Hmm … I hope to increase it from 0.6% to 1.0%. In total that would make me 120.000 SEK, equivalent to TWO family vacations (or a small car)!
Good luck with your betting in 2018!