Friday, 29 November 2019

Rebels Push and T-Bone on Full Kelly

The Mississippi Rebels lost their rivalry game yesterday by one point, but unfortunately that was all they were getting, so the game resulted in a Push.

As of right now, we will have five more College selections in Week 14 of the season this weekend, including two today - Miami (Ohio, not Florida) and Kent State who are getting 3 and 4.5 points respectively.

Week 14 is pretty much the final week of the regular College Football season, with just a handful of Week 15 games to come so at 59-37-7 I think I'm safe in recording a nineteenth consecutive winning season for these selections.
On the NFL front, it's looking like a quiet weekend with just a couple of probable Road 'Dog selections, although we do have five Under bets to add a little interest.
 
Triggered by Church of Betting's post on staking referenced in my previous post, I thought I'd show the outcome of using Full Kelly on baseball's T-Bone System this summer. Results have now been updated to include 2019 by the way. 

Note that this isn't every T-Bone selection, but a sub-set comprised of selections where the edge is historically larger.
From 2014 to 2018 (i.e. 5 seasons), the ROI on these selections was 13.1% from 352 selections, and the ROI for 2019 ended up being in line at 13.4%. 
The initial bank size was £1,000 and I used a stop loss at 67%. Fortunately this was never hit although we got close with two losses in the opening four selections, a risk of this extremely aggressive staking system.

Note that when there are more than one selection on a day, the stake on each game is the same and the bank is only updated at the end of a day, not after each bet. Most of us have better things to do with our time, and some games are played simultaneously anyway. 

The final bank was £3,555.31 after peaking at £4,091.26. The stop loss moved to over £1,000 on Star Wars Day, after which we were playing with house money.  

Thursday, 28 November 2019

Football Wash and Ice Hockey Woes

The five NFL Road 'Dogs went 2-3 last weekend, while the College selections went 5-4 so pretty much a wash. 

With the Thanksgiving holiday in the US, there will be three NFL games today but it's not looking like we will have any Road 'Dogs, although there is one likely College selection (Mississippi).

For the season, the NFL selections are 35-25-1 and the College selections 59-37-6.

In the NBA, the while the Road Favourites System is 5-2 following Monday night's win by the Portland Trailblazers in Chicago.

This is a long-winning system that takes advantage of the public underestimating certain teams:
Backing Overs when the Total is 224 or greater is still proving profitable, with a record of 49-35-1 for the season so far.

While it might seem as though all my systems just keep on generating money, that's not always the case, as illustrated by my NHL System which has struggled this season.
Church of Betting has a good post on the topic of staking, with the confirmation that nothing beats Kelly if you have a positive Expected Value (+EV).

Because Kelly has some practical limitations when it comes to sports betting, a topic also covered by Church of Betting, my belief is that most punters are best served by using a 2% of the bank, as discussed in this post here. If you have a larger edge, and if you follow some of my public systems you're in that category, then 3%, 4% or even 5% is reasonable. 

Friday, 22 November 2019

Newtonian Laws, Proof and Irrationality


@Lordlard made a similar point in a comment on my previous post Why Betting Systems Work. He wrote:
Other advantage of systems are on the behavioural side: they completely kill any emotion in betting. Events are ‘numbers’ in a never ending dataset and abstract from what they actually ‘are’...and not even worth watching. You can watch sport for purpose of watching sport only.
As I've said before, if you have a 'proven' system, then...:
To get worked up over individual bets is as unproductive as a casino manager worrying over the outcomes of individual pulls of the slot machines.
'Proven' is in quotes because when it comes to sports betting and markets, there can be no such thing. If a system were 'proven', it would no longer work. 

Even the 'immutable' laws of Physics, are only 'immutable' with respect to defined conditions. If conditions change, for example in Physics, events occurring at atomic and subatomic scales do not follow Newtonian laws, then the outcome may be different.

We can all agree that conditions would definitely change if a betting system was 'proven'. 

The fact is that markets change. Even a system with 19 consecutive winning years will eventually lose. As gamblers, albeit hopefully with an edge, we understand that. 
"The trend is your friend, but you know it will end"
The trick is to identify an edge, and play it until the edge is gone. In the same way that trying to time market tops and bottoms precisely in stocks is a futile task, so is timing when an edge starts and ends being profitable, but that's no reason to miss out on making gains for most of that period.

The one constant would appear to be that humans act irrationally. 

Speaking of winning systems, and the NFL Small Road Dogs moved to 33-22-1 for the season after the Indianapolis Colts covered in Houston last night. 
The ROI for this season is currently the second highest since the league re-organisation in 2002. Hopefully some of you followed me on that one. As a Divisional game, it was also an Unders pick and 14-3 for the season is excellent.
Since revealing that nugget more than two years ago, the ROI on the subsequent 207 bets has an ROI of 9%. I hope someone took note.

By now, everyone must be on the College Football Road 'Dogs who started Week 13 well with Eastern Michigan (+3.5) winning at Northern Illinois by 28 points. The season record now stands at 55-34-6.  With the season almost over for the system, the winning streak almost certainly continues into a 20th season next August. 

Thursday, 21 November 2019

Why Betting Systems Work

Anyone who has ever spent more than a few minutes on a betting forum will no doubt have seen threads with titles along the lines of "Does a football betting system that works exist?".

Remove 'football' or substitute any other sport for 'football' and the question still holds good. 

While the question is somewhat naive the answer, technically at least, is obviously yes - betting systems do work, but the caveat is that they shouldn't work in perpetuity. 

In an efficient market with rational participants, systems will have a limited shelf-life, although as in the case of College Football's Road 'Dogs, this 'limited' shelf-life is 19 years and counting. As I've mentioned before, the good news is that market participants are not rational, but are influenced by biases, often long-held.

SmarterSig had an interesting perspective on the subject of "do betting systems work" concluding a very readable article with:
Systems do work, they work in the sense that there are rule based methodical betting methods that produce consistent profits and they work because they give people a valuable apprenticeship in understanding data.
Probably not the answer most readers looking for easy money want to read, but the writer makes a good point.

The same point applies to making a budget of any kind of plan. However well you try to accurately predict the future, invariably you're wrong, but:
A plan is important because it's the foundation to help you helping you project objectives and achieve your ultimate goals. Having a plan helps you define the full scope of a project but it also helps you stay focused, set goals and objectives, meet deadlines, measure success and debrief the entire project.
Or as Dwight D Eisenhower put it:
"I have always found that plans are useless, but planning is indispensable"
 SmarterSig's article also makes the point that:
There are benefits to system development beyond the uncovering of a successful system, one of which is the journey itself. Analysing data from a system perspective is a great introductory way to get a feel for the terrain you face as a punter. The amount of loss you need to overcome in order to become profitable and more importantly what factors move you closer even if in isolation or combination they do not quite achieve long term profit. The study will give you a greater feel for the biggest hurdle all punters have to overcome, namely variance. In other words the kind of winning and losing periods once can expect to experience even with a decent looking system.
What makes a 'decent looking system' is subjective, but because sports markets have inefficiencies, opportunities are out there for those willing to do the work. Just don't expect them to last for ever, which is why, if you should uncover an edge, you need to act on it. By the time you have statistical proof that a 'system' works, it will likely have stopped working - if you see what I mean.

In a liquid market, as I've written before, simply "picking your selections at random means you should break even over the long term, excluding commission of course", so even if your edge turns out to be an illusion, you shouldn't lose your shirt. Keep an eye on the sleeves though. If they are gone, it might be time to cut your losses.

This "@SmarterSig" chap (aka Mark Littlewood) might be worth following if you are into horse racing.

A quick look at some of his other posts show some quality content, something that is something of a rarity these days. He appears to be of a similar vintage to myself, although we differ in our opinions of Farage. In one recent post on horse racing, he (Mark, not Nigel) writes:
I had begun my betting career with a strong bias against handicaps which I now realise was a mistake. My belief that there was no point arguing with the handicapper who’s full time job was to make all the horse’s theoretically dead heat was incorrect but it took me a few years to realise this.
I recall coming to the exact same conclusion on handicaps myself back in the 1970s, although the direction I ultimately took was away from horses altogether. Mark writes in one post:
My belief that there was no point arguing with the handicapper who’s full time job was to make all the horse’s theoretically dead heat was incorrect but it took me a few years to realise this.
Ignoring the incorrect apostrophes, I remember avoiding handicaps for this exact reason. Larger fields were another reason, and for a few years in the late seventies, backing the favourite in non-handicap Novice Hurdles and Novice Chases with 6 to 10 runners was a system I followed, and I see variations are still in vogue 40 years later! Somewhere, quite possibly in the attic at my childhood home, are some notebooks with the results neatly documented in my 'just left school' handwriting. If my parents ever move, and 64 years after moving in they are still there so don't hold your collective breath, I shall put on my hazmat suit and attempt an extraction.       

Sunday, 17 November 2019

Bullseye

In the good old pre-Premium Charge days, trading sports in-play was where the easy money was, but it could be quite stressful. The relatively infrequent four figure losses are not pleasant, even if the edge meant that in the long run the bank kept growing. In the five years from 2006 to 2010 I had 59 such bad days, with one loss taking 275 days to recover (financially) from. Emotionally, it only took me 250 days... The point is that when something is stressful like this, taking a break can be beneficial.

With the less stressful 'bet and forget' approach that I have now adopted, taking a break is just annoying as effectively, every day that you are not active is a missed opportunity.

So it's always with mixed feelings when I come back from a holiday, run the numbers and see what I could have won, as "Bullseye" would put it.


I missed three weeks of the College Football season, and of course the Small Road 'Dogs had winning weeks in all three, including a 7-0 sweep in Week 11. Overall a 21-8-1 record, and when I'm back for Week 12, they can only manage a 3-3-2 week. For the season, the system is now at 54-35-7 and looking very good for a 19th consecutive winning season. 

Earlier this year, I wrote about being unconvinced by an 18-0 record. Barring a disaster, the unconvinced will soon need to explain away a 19-0 record, but yes, of course the market here is efficient. It's just a pattern... 

The same three weeks were missed in the NFL also, plus the Thursday night game for Week 11, but here I missed out on a losing 4-7-1 record. This weekend the system has four selections all currently in play - Jacksonville Jaguars, Atlanta Falcons, Houston Texans and New York Jets looking to improve on a 31-20-1 overall record for the season. 

I've also missed the start of the NBA season, which in recent seasons has been interesting to follow from the totals perspective. Bucking the trend of recent season, it appears the points total isn't increasing, at least not yet:
The strategy of backing Overs on high totals does continue to be profitable, with the 224 total (58.3%) looking optimal so far.

In other news, this year's Fields Medal could be heading to this guy who has somehow discovered that coin tosses are NOT independent events...
Now you know. Good to know that such people are out there competing against us, and with such confidence. 

One final note is that while I was away, this blog passed the 2,000,000 hits mark.
It only took about 11 and a half years to get there, but as with accumulating wealth or losing weight, it's a slow and steady process (barring a lottery win or amputation of course).